JP2016045667A - Control system and control method, exposure device and exposure method, as well as device fabrication method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To accurately and stably perform a drive control of a drive object relative to an action of disturbance.SOLUTION: In a control system (basic configuration) 50, the amount to be input to a control object 52 is obtained on the basis of a nominal mode (model reproducing an actual responce of the control object 52) Pn expressing the control object 52 from a controlled amount y to be measured by a sensor 52, the amount to be input to the control object 52 is obtained on the basis of a model (model expressing a desired responce of the control object 52 relative to disturbance d) Grendering a desired impedance characteristic, the amount of correction for correcting the amount of operation is obtained from a difference between the two amounts, and the amount of operation U is corrected by use of a result of obtaining the amount of correction. The amount of operation U is corrected by use of the amount of correction, and thereby the responce of the control object 52 is modified to the desired responce to be expressed by the model G, which in turn enables the control object 52 to accurately and stably drive even when the disturbance d acts.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a control system and a control method, an exposure apparatus and an exposure method, and a device manufacturing method, and more particularly, a control system and a control method for driving and controlling a control target by giving a control input (operation amount). The present invention relates to an exposure apparatus provided, an exposure method using the control method, and a device manufacturing method using the exposure method.

  In a lithography process for manufacturing electronic devices (microdevices) such as liquid crystal display elements and semiconductor elements, a step-and-repeat type projection exposure apparatus (so-called stepper) and a step-and-scan type projection exposure apparatus (so-called so-called stepper) are mainly used. Scanning steppers (also called scanners)) are used. With respect to exposure apparatuses (liquid crystal exposure apparatuses) for liquid crystal display elements, scanning projection exposure apparatuses such as scanners have become mainstream as substrates become larger. Electronic devices (microdevices) are manufactured by forming a plurality of layers of patterns on a substrate (glass plate, wafer, etc.). For this reason, the exposure apparatus is required to accurately and quickly superimpose and transfer the mask pattern onto the pattern previously formed on the substrate, that is, high overlay accuracy and high throughput.

  In a coarse / fine movement type substrate stage comprising a coarse movement stage driven on a base with a relatively large driving force, and a fine movement stage supported on the coarse movement stage and driven precisely on the coarse movement stage, In order to drive the fine movement stage to be held at a high speed, a catapult type drive device has been developed in which the coarse movement stage is brought into contact with the fine movement stage and the driving force of the coarse movement stage is applied to the fine movement stage (see, for example, Patent Document 1). . However, chattering (mechanical vibration) occurs when the coarse movement stage is brought into contact with the fine movement stage, which impedes the precise driving of the substrate stage (fine movement stage) and lowers the overlay accuracy, and also damages the substrate stage. There is also a risk.

  In contrast to the open control system in which the drive target (fine movement stage) crosses the boundary and contacts the outside (coarse movement stage) like the substrate stage having the above-described configuration, when the drive target contacts the outside, Chattering can be suppressed by controlling the viscosity and elasticity (hereinafter referred to as viscoelastic characteristics). The representative approaches are roughly classified into a control approach and a mechanical approach.

  As an example of a control approach, impedance control is known that adjusts viscoelastic characteristics (and may include inertia) between a driven object and the outside. In the impedance control, first, an input (disturbance) acting on the drive target from the outside is obtained using a force sensor or the like. Next, the position command is obtained from the external force so that the viscoelastic characteristics obtained by feeding back the command are adjusted to the desired characteristics. Finally, the position command is fed back to control the drive target (control the position of the drive target). Since the accuracy of the drive control depends on the accuracy of the response of the object to be controlled with respect to the position command, the impedance control is not necessarily suitable for an external force having a quick response such as a force caused by a collision.

  On the other hand, as an example of a mechanical approach, a method using an electrorheological fluid (ER fluid) whose viscosity changes by applying a voltage, that is, an ER fluid is interposed between a driven object and the outside, and the viscosity is There is a way to control. However, in an exposure apparatus that requires high drive accuracy (positioning accuracy) of the drive target, the presence of ER fluid may be a non-linear factor in feedback control. Therefore, it is usual to eliminate the viscosity acting on the driven object by adopting an air guide or the like. Furthermore, a change in viscosity due to aging of the ER fluid is also a problem.

  Therefore, the feedback control that virtually adjusts the viscoelastic characteristics of the drive target with respect to the external force with a quick response by using the high responsiveness of the drive device mounted on the exposure apparatus or the like is promising.

JP 2003-45785 A

  According to the first aspect of the present invention, there is provided a control system that generates an operation amount of the controlled object from a controlled amount of the controlled object, the input unit to which the controlled amount is input, and the controlled amount or From a generation amount generated by calculation using the controlled amount and a first amount obtained from the first model corresponding to the controlled object, and a second model different from the generation amount and the first model A control unit that generates the manipulated variable using a difference from the calculated second quantity, and the second model represents the disturbance applied to the controlled object and the controlled quantity as a predetermined function. A control system is provided.

  According to this, it becomes possible to drive a controlled object precisely and stably.

  According to a second aspect of the present invention, there is provided an exposure apparatus that exposes an object with an energy beam to form a pattern on the object, the first moving body holding and moving the object, and the first A moving mechanism having a second moving body that is movable relative to the first moving body while supporting the moving body, and a control system according to the first aspect in which the first moving body is the control target; An exposure apparatus is provided.

  According to this, it is possible to accurately and stably drive the moving body that holds the object, and consequently, it is possible to expose the object with high accuracy.

  According to a third aspect of the present invention, there is provided a control method for controlling a controlled object that inputs an operation amount and outputs a controlled variable, wherein information relating to the position of the controlled object is obtained as the controlled variable. Obtaining a first amount to be input to the controlled object from the controlled amount or a generated amount generated by a calculation using the controlled amount and a first model corresponding to the controlled object; Determining the second amount inputted to the control object from the generated amount and a second model different from the first model, and using the first amount and the second amount, the operation Generating a quantity, and the second model is provided with a control method in which the disturbance applied to the controlled object and the controlled quantity are represented by a predetermined function.

  According to this, it becomes possible to drive a controlled object precisely and stably.

  According to a fourth aspect of the present invention, there is provided an exposure method for exposing an object with an energy beam to form a pattern on the object, the first moving body holding and moving the object, and the first A driving method according to a third aspect in which the first moving body is the control target among the moving mechanisms having a second moving body that is movable relative to the first moving body while supporting the moving body. An exposure method to be utilized is provided.

  According to this, it is possible to accurately and stably drive the moving body that holds the object, and consequently, it is possible to expose the object with high accuracy.

  According to a fifth aspect of the present invention, using the exposure method according to the fourth aspect, forming a pattern on the object, and developing the object on which the pattern is formed A device manufacturing method is provided.

It is a block diagram which shows the basic composition of the feedback control system which concerns on 1st Embodiment. It is a block diagram which shows the deformation | transformation structure of the feedback control system which concerns on 1st Embodiment. FIGS. 3A to 3C are diagrams (Nos. 1 to 3) for explaining derivation of the modified configuration from the basic configuration of the feedback control system. 4A and 4B are diagrams (Nos. 4 and 5) for explaining the derivation of the modified configuration from the basic configuration of the feedback control system. It is a Bode diagram which shows the frequency response characteristic of a plant. It is the table | surface which put together the physical parameter employ | adopted in the disturbance observer which comprises a feedback control system. It is a figure which shows the simulation result of the time response characteristic of a plant. It is a block diagram which shows the structure of the feedback control system which concerns on 2nd Embodiment. It is the table | surface which put together the physical parameter employ | adopted in simulation. It is a Bode diagram which shows the frequency response characteristic of a amendment part. It is a Bode diagram which shows the frequency response characteristic of a plant. It is a figure which shows the sensitivity characteristic of a control system. It is a block diagram which shows another structure of the feedback control system which concerns on 2nd Embodiment. It is a figure which shows schematically the structure of the exposure apparatus which concerns on 3rd Embodiment. It is a block diagram which shows the structure relevant to the stage control of exposure apparatus. It is a figure which shows the structure and arrangement | positioning of a mask stage, a mask stage drive system, and a mask interferometer system. It is a figure which shows the structure of a gas flow apparatus. FIGS. 18A and 18B are block diagrams for explaining control switching in the feedback control system of the fine movement stage constituting the catapult type stage device, respectively. FIG. 19A and FIG. 19B are diagrams showing reaction forces and position errors received by the fine movement stage when the coarse movement stage collides with the fine movement stage while accelerating.

<< First Embodiment >>
The first embodiment will be described below with reference to FIGS.

FIG. 1 shows a basic configuration of a control system 50 according to the first embodiment. The control system 50 is a feedback control system for controlling driving of the control target 52 by providing a control input U, the sensor 52 ₂ for measuring the position y of the control object 52 as the controlled amount, the output of the sensor 52 ₂ ( and a controller 54 that the measurement result) of the position y performs an operation gain-multiplied (-1 times) to obtain the correction amount from the output of the controller 54 (i.e., the output of the sensor 52 _2), controlled as a result the control input U a correction unit 53 for transmitting the drive unit 52 ₁ for driving the target 52, and a. Note that the value of the gain multiple in the controller 54 is not limited to that of the present embodiment, and can be set to an arbitrary value.

Drive device 52 ₁ in accordance with the received control input U (having a dimension of the force in the present embodiment), emits equal driving force to the drive quantity. Thereby, the drive target is driven and controlled. In FIG. 1, the driving device 52 ₁ and the sensor 52 ₂ is shown as being included in the control object 52. Further, d shown in the figure is a disturbance acting on the control object 52 from the outside and is unknown. The control input U and the position y are respectively an input and an output for the control object 52, and both are known. The actual input to the controlled object 52 is a control input U which is output from the correction unit 53 (equivalent to the driving force driving apparatus 52 ₁ emitted) to the sum of the disturbance d.

  Here, controlled amounts, control inputs, and the like are defined as functions of time. However, in the description using FIG. 1 and the like, their Laplace transforms are used according to customs.

As the sensor 52 _2, interferometer, adopting a position sensor such as an encoder. Sensor 52 ₂ is always measured position y of the control object 52, and outputs the result to the correction unit 53. Sensor 52 ₂ is not intended to be limited to the interferometer and the encoder, it is also possible to use another measuring device.

The controller 54 is actually realized by a microcomputer and software, but may be configured by hardware. The control unit 54 has been inserted between the sensor 52 ₂ and the correcting unit 53, instead of this, may be inserted between, for example, the correction unit 53 and the controlled object 52. In this case, the controller 54 multiplies the output of the correction unit 53 by a gain (-1 times) and outputs it to the control object 52 (drive device 52 ₁ ).

The correction unit 53 includes two controllers 53a and 53b and a subtractor 53c. Note that each of these components is actually realized by a microcomputer and software in the same manner as the controller 54, but may of course be constituted by hardware. The controller 53a is a transfer given by the product of the filter Q representing the phase delay and the inverse of the transfer function Pn corresponding to the nominal model representing the controlled object 52 (a model reproducing the actual response of the controlled object 52). Has characteristics. The controller 53a receives the controlled amount y (strictly, the amount −y via the controller 54) and inputs the amount actually input to the controlled object 52 (referred to as a first amount). Ask. The result is output toward the subtractor 53c. The controller 53b has a transfer characteristic given by the product of the filter Q and a reciprocal of the transfer function G _imp corresponding to a model (a model expressing a desired response) exhibiting a desired behavior (impedance characteristic described later). The controller 53b obtains an amount (referred to as a second amount) to be input to the controlled object 52 by inputting the controlled amount y (strictly, the amount −y via the controller 54). The result is output toward the subtractor 53c. The subtractor 53c calculates the difference (second amount-first amount) between the outputs (first and second amounts) of the controllers 53a and 53b, and the result is the control object 52 (drive device 52 ₁ ). Output to.

In the control system 50 configured as described above, the sensitivity function S = 1 / (1 + PC ₅₃ ) is given, and the transfer function of the correction unit 53 is C ₅₃ = Q (G _imp ⁻¹ −Pn ⁻¹ ). Assuming that the ideal condition, that is, the response characteristic P of the controlled object 52 is equal to the nominal model (transfer function corresponding thereto) Pn and an ideal filter Q = 1, a sensitivity function S = P ⁻¹ G _imp is derived. From the response characteristic of the control target (characteristic from the external force d to the output y) Pd = y / d = SP, Pdn = G _imp is derived under ideal conditions. This means that the response characteristic of the controlled object to the disturbance d (relationship between the disturbance d and the controlled amount y) can be changed to a desired characteristic (desired relationship) expressed by the transfer function G _imp .

  In a modified configuration equivalent to the basic configuration of the control system 50 described above, the performance of the control system 50 according to the first embodiment is evaluated. In addition, while using the same code | symbol for the component same as the basic composition of the control system 50, detailed description is also abbreviate | omitted.

FIG. 2 shows a modified configuration (control system 50 ′) of the control system 50 according to the first embodiment. The control system 50 ′ is a feedback control system that drives and controls the controlled object 52 by giving a control input U, as in the basic configuration described above, and the driving target of the controlled object 52 (having a driving force dimension). generates R, the result calculates the control input U using (and correction amount dU described later), the result, the stage controller 51 to be transmitted to the drive unit 52 ₁ for driving the controlled object 52, a controlled a sensor 52 ₂ for measuring the position y of the control object 52 is an amount comprised of, a correction unit 53 'for determining the correction amount dU with respect to the control input U from the output of the sensor 52 ₂ (measurement results of the position y).

Stage controller 51 includes a target generator 51 ₀ and a subtractor 51 _2. Note that each of these components is actually realized by a microcomputer and software constituting the stage control device 51, but may be constituted by hardware. Target generator 51 _0, drive target of the controlled object 52 (in this case the target has dimensions of driving force) to generate R, supplied to the subtracter 51 _2. Subtractor 51 _2, the output from the driving target R and the correction section 53 (correction amount) to calculate the difference R-dU and dU seek control input U, toward the control object 52 (driving device 52 ₁₎ Output To do.

Correcting unit 53 'is configured as a disturbance observer in two stages, including a two disturbance observer _53, 53 ₃ and the adder 53 _2. Note that each of these components is actually realized by a microcomputer and software constituting the stage control device 51, but may be constituted by hardware. First stage of the disturbance observer 53 _1, the output of the sensor 52 and _second output (the controlled variable) y and stage controller 51 (control input) U and are inputted, estimates a disturbance d applied from outside to the control target 52 To do. Result of Estimation <d> is output to the adder 53 _2. The adder 53 ₂ calculates a sum of the output of the disturbance observer 53 ₁ (estimation of the disturbance d) and Between <d> and the output (the control input) U of the stage controller 51, to output to the disturbance observer 53 _3. Second stage of the disturbance observer 53 _3, the output of the sensor 52 ₂ (the controlled variable) y and the adder 53 ₂ outputs U + <d> is input, calculates a correction amount dU. The result is output toward the stage control device 51 (subtractor 51 ₂ ).

  Here, it will be described that the control system (modified configuration) 50 ′ (included in the correction unit 53 ′) is equivalent to the control system (basic configuration) 50 (included in the correction unit 53).

FIG. 3A shows a control system (basic configuration) 50. However, the relative positions of the correction unit 53 and the controller 54 are interchanged. The correction unit 53 divides the filter Q included in the two controllers 53a and 53b into two filters Q ₁ and Q ₂ (given by the product Q ₁ Q ₂ of the two filters). In response to this, as shown in FIG. 3B, the controller 53a is divided into two controllers 53 ₁₁ and 53 ₃₂ in series. Here, the controller 53 ₁₁ has a transfer characteristic given by the product of the filter Q ₁ and the inverse of the transfer function Pn corresponding to the nominal model (a model that reproduces the actual response of the controlled object 52). The controller 53 ₁₁ obtains the first amount actually input to the controlled object 52 by inputting the controlled amount y (estimates the sum of the operation input U and the disturbance d). Controller 53 ₃₂ is provided by filter Q ₂ . The controller 53b is divided into two controllers 53 ₅ and 53 ₃₁ in series. Here, the controller 53 ₅ is provided by a filter _{Q 1.} The controller 53 ₃₁ has a transfer characteristic given by the product of the filter Q ₂ and the inverse of the transfer function G _imp corresponding to the model (a model expressing a desired response). The controller 53 ₃₁ obtains a second amount to be input to the controlled object 52 when the controlled amount y is input.

As shown in FIG. 3 (C), the correction unit 53, two filters ₅₃ 12 are in a relationship where the output of each other to cancel, 53 _4, subtracter _{53 13} adds adder 53 _2. The filters 53 ₁₂ and 53 ₄ are given by the transfer function Q ₁ . Here, the transfer function Q ₁ represents a phase delay that occurs when the controller 53 ₁₁ estimates the disturbance d. Through the filter ₅₃ 12, 53 _4, the control input U is inputted to the respective subtractors _{53 13} and adder 53 _2. The subtracter _{53 13} calculates the difference between the outputs of the filter _{53 12} of the controller _{53 11,} i.e., estimates the disturbance d, the result is outputted to the adder 53 ₂ <d>. The adder 53 ₂ obtains a sum of the output of the subtracter _{53 13} output of the filter 53 _4, and outputs toward the result to the filter _{53 32.}

In FIG. 3C, the controller 53 ₁₁ , the filter 53 ₁₂ , and the subtractor 53 ₁₃ receive the controlled variable y and the control input U, estimate the disturbance d, and output the result <d>. constitute a disturbance observer 53 _1. Further, the controller 53 ₃₁ , the filter 53 ₃₂ , and the subtractor 53 c obtain the correction amount by inputting the controlled variable y and the sum of the control input U and the disturbance estimation result <d>, and output the result dU. constitute a disturbance observer 53 _3. Therefore, the block diagram shown in FIG. 3C can be equivalently converted into a simple block diagram shown in FIG.

Finally, in FIG. 4 (A), the controlled variable y and the control input U, respectively via the filter ₅₃ 4, 53 ₅ having a filter function _{Q 1} equal to each other, and are input to the disturbance observer 53 _3. Therefore, the filters 53 ₄ and 53 ₅ are deleted. Thereby, the block diagram shown in FIG. 4B is derived. This control system is equivalent to the control system 50 ′ shown in FIG. 2 with the target R = 0. In other words, in the control system 50, by further adding the target generator 51 _0, it can be derived a control system 50 'shown in FIG. Therefore, the control system 50 ′ (correction unit 53 ′) is equivalent to the control system 50 (correction unit 53) shown in FIG.

The control system 50 'in the above FIG. 4, the disturbance observer 53 _1, for example, a disturbance as a step-like disturbance can be designed based on the minimum dimension observer design according to the method of Gopinath. Here, the motion of the nominal model used for the design (a model that reproduces the actual response of the controlled object 52) is given by the function P _n1 (= Pn) = 1 / Ms ² representing the free translational motion of the rigid body having the mass M. . (Employing equal characteristic properties (actual plant characteristic) P = 1 / Ms ² controlled object 52 employed in the simulation to be described later.) To estimate the disturbance d by using the disturbance observer 53 _1, to the controlled object 52 The actual input U + d can be estimated. However, the set to be higher than the bandwidth of the disturbance observer 53 ₃ band of the disturbance observer 53 _1.

Disturbance observer 53 _3, like the disturbance observer 53 ₁ can be designed based on the minimum dimension observer design according to the method of Gopinath. Here, the motion of a nominal model (model expressing a desired response) used for the design is represented by a transfer function P _n2 (= G _imp ) representing the translational motion of a rigid body having mass M _n , viscosity C _n , and elasticity K _n. = 1 / (M _n s ² + C _n s + K _n ). Deviation of the response of the control object 52 from the response represented by the nominal model P _n2 by using the disturbance observer 53 _3, that is, the correction amount dU obtained for the control input U.

By correcting the control input U using the obtained deviation dU, it is possible to suppress only the plant fluctuation without suppressing the disturbance d. Incidentally, by designing the disturbance observer 53 ₃ using any model, the response characteristic of the controlled object 52 from the disturbance d to the output y is within the band, the desired properties are expressed by the arbitrary model Present.

Note that the design methods of the disturbance observers 53 ₁ and 53 ₃ are not limited to the Gopinus method described above, and various methods that are actually proposed in many documents may be adopted. In the above description, the two disturbance observers 53 ₁ and 53 ₃ are designed in a continuous time system, but can also be designed in a discrete time system.

FIG. 5 shows a simulation result of the frequency response characteristics of the controlled object 52 from the disturbance d to the output y. Here, the response characteristic of the controlled object 52 (actual plant characteristics) were given property P = 1 / Ms ^2. The table given in FIG. 6 shows the values of physical parameters adopted in the simulation. The poles of the disturbance observers 53 ₁ and 53 ₃ were 150 Hz and 50 Hz respectively. In order to simulate the response of the control object 52 to the disturbance d, the drive target R = 0.

In FIG. 5, the solid line (black) is the response characteristic P of the actual controlled object 52, the solid line (gray) is the response characteristic P _n2 of the desired controlled object expressed by the model, and the broken line is the controlled object 52 from the disturbance d to the output y. The response characteristic (d → y with prop) is shown. Since the response characteristic P of the control object 52 is given by a model (P = 1 / Ms ² ) representing a free translational motion of a rigid body having a mass M as a simulation condition, the amplitude monotonously decreases with increasing frequency, and the phase Will remain constant. On the other hand, the desired response characteristic P _n2 is a function (P _n2 = 1 / (M _n s ² + C _n s + K _n ) representing the translational motion of a rigid body having mass M _n , viscosity C _n , and elasticity K _n. viscosity _{C n} in order given, in the low frequency band (band below approximately 0.01 Hz) by the effect of the elastic _{K n} to maintain a substantially constant amplitude and phase, middle frequency band (band of approximately 0.01～10Hz) by As a result, the amplitude is gradually decreased and the phase is delayed, while the response characteristic of the controlled object 52 shows a slight deviation in the high frequency band (approximately 10 Hz or more) in both the amplitude and phase. shows a behavior substantially equal to the desired response characteristic (P _n2). this result, by the control system 50 of the present embodiment, the response of the controlled object 52 against an external force d, outer Suggesting that can be modified to respond with any viscoelastic properties represented by the model used in the observer 53 ₃ design (transfer function P _n2).

FIG. 7 shows the simulation result of the time response characteristic. FIG. 7A shows the external force d acting on the control object 52. As the external force d, an impulse-like force having an amplitude of 10 (N) and a time width of 0.01 (s) is applied to the control object 52 at time t = 1 (s). FIG. 7B shows an output (position y) from the control object 52 (sensor 52 ₂ ) when the control system 50 of the present embodiment is not employed. In this case, since the control object 52 does not have viscoelasticity characteristics, it continues to move at a constant speed to infinity due to the action of the disturbance d. In FIG. 7 (C), showing the desired response characteristics to expect the control target 52, i.e. the response characteristic represented by the model (transfer function P _n2) used in the design of the disturbance observer 53 _3. The controlled object 52 moves to a certain distance (approximately 1 mm) by the action of the disturbance d until time 0.5 (s) elapses due to the addition of the viscoelastic characteristics (C _n , K _n ). Gradually return to the original position. FIG. 7D shows the response characteristics of the controlled object 52 when the control system 50 of the present embodiment is employed. It can be seen that the desired response characteristics shown in FIG. FIG. 7E shows a control input (driving force) U applied to the control object 52 when the control system 50 of the present embodiment is employed. The control input (driving force) U shows an impulse-like force (reaction force) and subsequent overshoot (hunting) in a direction to cancel the external force d at time t = 1 (s). By inputting the control input U shown in FIG. 7 (E) to the control object 52 with respect to the disturbance d shown in FIG. 7 (A), the control object 52 virtually has viscoelastic characteristics (C _n , K _n ). It turns out that it responds as a rigid body which has.

As described above in detail, according to the control system (basic configuration) 50 of the first embodiment, the nominal model (controlled object 52 that represent the controlled amount y, the controlled object 52 to be measured by the sensor 52 ₂ A model that reproduces the actual response of the control object 52 based on Pn and obtains a desired impedance characteristic (a desired response of the control object 52 to the disturbance d, that is, the disturbance d and the disturbance). A model that expresses a desired relationship with the control amount y) _Obtains an amount to be input to the control object 52 based on G _imp , obtains a correction amount for correcting the control input from the difference between these two amounts, and obtains the result In this way, the control input U is corrected. By correcting the control input U, the response of the control object 52 is corrected to a desired response expressed by the model G _imp . Thereby, even if the disturbance d acts, it becomes possible to drive the control object 52 precisely and stably.

Further, according to the control system of the first embodiment (modified configuration) 50 'obtains the disturbance d applied to the controlled object 52 by the disturbance observer 53 _1, using the control input U to be input to the controlled object 52 as a result and obtains the deviation of the response of the control object 52 from the model (transfer function P _n2) representing the desired response by the disturbance observer 53 ₃ Te, the control input U is corrected using the result. The response of the control object 52 is corrected to the desired response expressed by the nominal model P _n2 by correcting the control input U using the deviation from the response expressed by the nominal model P _n2 of the control object 52. Is done. Thereby, even if the disturbance d acts, it becomes possible to drive the control object 52 precisely and stably.

Further, according to the control systems 50 and 50 ′ of the first embodiment, the model G _imp and the nominal model P _n2 are given as models representing the translational motion of a rigid body having mass, viscosity, and elasticity. It is possible to drive and control the control object 52 as a rigid body having virtual viscoelastic characteristics (C _n , K _n ) with respect to the disturbance d acting on.

Further, according to the control system 50 ′ of the first embodiment, the correction unit 53 ′ for obtaining the correction amount dU for the control input U can be simply configured using the two disturbance observers 53 ₁ and 53 _3. . Here, first one of the disturbance observer 53 _1, and an output (control input) U of the sensor 52 and _second output (the controlled variable) y and stage controller 51, and estimates the external force d of the controlled object 52 from the outside is subjected . The second disturbance observer 53 _3, the output of the disturbance observer 53 ₁ (disturbance estimation) <d> and the output of the stage controller 51 (control input) and a U, virtually controlled object 52 is desired viscoelastic properties A correction amount dU exhibiting a response having

Further, the control system 50 'of the first embodiment, it was decided to estimate the disturbance d acting on the controlled object 52 using the disturbance observer 53 _1. Thus, by utilizing the high response characteristic of the driving device 52 _1, the viscoelastic properties of the driven object against an external force d can be adjusted virtually fast response. Further, as the nominal model used in the disturbance observer 53 ₁ design, the actual response of the controlled object 52 by adopting a model that accurately reproduced, it is possible to estimate the disturbance d more accurately.

Incidentally, in the control system 50 'of the first embodiment is not limited to estimating the disturbance d by using the disturbance observer 53 _1, it is also possible to measure the disturbance d for example, using a force sensor. A case, the measurement result is input to the disturbance observer 53 _3. Furthermore, the disturbance observer 53 ₃ contains a predicted or measured disturbance d results (or their corresponding amount) correction amount dU with respect to the measurement result (or equivalent amount to) the control input U from the of the controlled variable y If desired, it is not always necessary to design as a disturbance observer.

<< Second Embodiment >>
Hereinafter, the second embodiment will be described with reference to FIGS. Here, the same reference numerals are used for the same components as those in the first embodiment described above, and detailed description thereof is also omitted.

FIG. 8 shows a configuration of a control system 50 according to the second embodiment. The control by the control system 50 having this configuration is called IPFB (Impedance Parameterization Feedback) control. The control system 50 is a feedback control system that drives and controls the controlled object 52 by giving a control input U, as in the basic configuration and the modified configuration described above, and a sensor that measures the position y of the controlled object 52 that is a controlled amount. 52 and _2, a controller 54 for performing an operation to gain multiple (measurement results of the position y) output from the sensor 52 ₂ (-1 times), (and that has dimensions of position) the target position of the controlled object 52 R It generates a stage controller 51 and outputs the sum of the output of the target position R and the control unit 54 (i.e. the deviation between the measurement result of the position y of the control object 52 from the sensor 52 ₂₎ e, the stage controller 51 output obtains a correction amount from a correction unit 53 for transmitting the drive unit 52 ₁ for driving the controlled object 52 the result as the control input U, composed.

The control system 50 having the above-described configuration has a configuration in which the stage control device 51 in the control system (modified configuration) 50 ′ is added to the control system (basic configuration) 50 according to the first embodiment. However, the subtractor 51 ₂ is replaced by a combination of the adder 51 ₃ and the controller 54. Therefore, assuming that the above ideal condition, that is, the response characteristic P of the controlled object 52 is equal to the nominal model (transfer function) Pn and the ideal filter Q = 1 is assumed, the response characteristic of the controlled object (from the external force d) Characteristic up to output y) Pd (= y / d = SP) = G _imp is derived. That is, the control system 52 according to the first embodiment is generated by generating the target position R by the stage control device 51 and driving the control target 52 to follow the target position R and correcting the control input U by the correction unit 53. Similarly to 50, the response characteristic of the control object 52 can be modified to a desired response expressed by the model (transfer characteristic) G _imp .

The performance of the control system 50 (IPFB control) according to the second embodiment is evaluated. As a nominal model that expresses the controlled object 52 (a model that reproduces the actual response of the controlled object 52), a rigid model that expresses the free translational motion of a rigid body having mass is adopted. Its transfer function is given by a function Pn = 1 / Ms ^2. (A characteristic equal to the actual plant characteristic described later is adopted.) As a model exhibiting a desired behavior (a model expressing a desired response), a model expressing a translational movement of a rigid body having mass, viscosity, and elasticity is adopted. . The transfer function is given by the function G _imp = 1 / (M _n s ² + C _n s + K _n ). As the filter Q, a secondary low-pass filter Q = ω ² / (s ² + 2ζωs + ω) is employed. The table given in FIG. 9 shows the values of physical parameters included in these models.

FIG. 10 shows the frequency response characteristic C ₅₃ of the correction unit 53. Response characteristic C ₅₃ of the correcting unit 53, the amplitude increases with increasing frequency in the frequency band below the cut-off frequency 50Hz, has a maximum at the cutoff frequency 50Hz, a frequency at a cut-off frequency 50Hz or more frequency bands Decreases with increasing. Also, its phase increases with increasing frequency, has a maximum at about 2 Hz, decreases with increasing frequency, has zero at a cut-off frequency of 50 Hz, and further decreases.

In FIG. 11, the simulation result of the frequency response characteristic of the control object 52 from the disturbance d to the output y is shown. Here, the response characteristic of the controlled object 52 (actual plant characteristics) were given property P = 1 / Ms ^2.

In FIG. 11, the solid line (gray) indicates the response characteristic P of the actual control target 52, the solid line (black) indicates the response characteristic G _imp of the desired control target expressed by the model, and the broken line (substantially overlaps the solid line) indicates the disturbance d. The response characteristic Pd of the control object 52 from output to output y is shown. Since the response characteristic P of the control object 52 is given by a model (P = 1 / Ms ² ) representing a free translational motion of a rigid body having a mass M as a simulation condition, the amplitude monotonously decreases with increasing frequency, and the phase Will remain constant. In contrast, the desired response characteristic G _imp is a function (G _imp = 1 / (M _n s ² + C _n s + K _n ) representing the translational motion of a rigid body having mass M _n , viscosity C _n , and elasticity K _n. viscosity _{C n} in order given, in the low frequency band (band below approximately 0.1 Hz) by the effect of the elastic _{K n} to maintain a substantially constant amplitude and phase, middle frequency band (band of approximately 0.1～100Hz) by As a result, the amplitude is gradually decreased and the phase is delayed, whereas the result of the response characteristic Pd of the controlled object 52 shows a behavior that is substantially equal to the desired response characteristic G _imp in both amplitude and phase. this result, by the control system 50 of the present embodiment, the response of the controlled object 52 against an external force d, Ren represented by the model (transfer function G _imp) which exhibit the desired behavior Suggesting that it can be modified to respond with viscoelastic properties of.

FIG. 12 shows the sensitivity characteristics of the control system 50. The sensitivity function S increases in amplitude as the frequency increases, and maintains the amplitude constant in a band of about 10 Hz or higher. The behavior of the sensitivity function S substantially matches the function P ⁻¹ G _imp . That is, the response characteristic P of the control object 52 is equal to the nominal model (transfer function corresponding to) Pn, and the relationship S = P ⁻¹ G _imp derived under the ideal condition assuming the ideal filter Q = 1 is a good approximation. It is made up of. This suggests that the position control of the control object 52 is also possible depending on the plant characteristic P and the impedance characteristic G _imp .

FIG. 13 shows another configuration of the control system 50 according to the second embodiment. The control by the control system 50 having this configuration is referred to as PSFB (Plant Shaping Parameterization Feedback) control. The control system 50 is a feedback control system that drives and controls the controlled object 52 by giving a control input U, as in the basic configuration and the modified configuration described above, and a sensor that measures the position y of the controlled object 52 that is a controlled amount. 52 and _2, a controller 54 for performing an operation to gain multiple (measurement results of the position y) output from the sensor 52 ₂ (-1 times), (and that has dimensions of position) the target position of the controlled object 52 R It generates a stage controller 51 and outputs the sum of the output of the target position R and the control unit 54 (i.e. the deviation between the measurement result of the position y of the control object 52 from the sensor 52 ₂₎ e, the stage controller 51 and of generating a control input U from the output, so that the controller 51 ₄ to send a control object 52 to the drive unit 52 ₁ for driving the control from the output of the controller 54 (i.e., the output of the sensor 52 ₂₎ A correction unit 53 for obtaining a correction amount dU to forces U, composed.

The control system 50 having the above-described configuration includes a correction unit 53 in the inner loop with respect to the control system (basic configuration) 50 according to the first embodiment, and includes a stage control device 51 in the control system (modified configuration) 50 ′ and the controller 51 ₄ is configured by adding at outer loop. Here, the controller 51 ₄ generates a (measurement results of the deviation of the position y of the control object 52 according i.e. target position R and the sensor 52 ₂₎ controlled from e input U output of the stage controller 51, the result to output to the adder 51 _5. Correcting unit 53 calculates the correction amount dU with respect to the control input U from the output of the controller 54 (i.e. measurement result of the position y of the control object 52 from the sensor 52 _2), is output to the result to adder 51 ₅ . The adder 51 ₅ adds the correction amount dU outputted from the correction unit 53 to the control input U, which is outputted from the controller 51 ₄ is transmitted to the drive unit 52 ₁ for driving the controlled object 52 results.

The control system 50 configured as described above assumes that the above-described ideal condition, that is, the response characteristic P of the control target 52 is equal to the nominal model (corresponding transfer function) Pn and the ideal filter Q = 1 is assumed. Response characteristic (characteristic from control input U and external force d to output y) Pd (= y / U = y / d = SP) = G _imp is derived. That is, the control system 52 according to the first embodiment is generated by generating the target position R by the stage control device 51 and driving the control target 52 to follow the target position R and correcting the control input U by the correction unit 53. Similarly to 50, the response characteristic of the control object 52 can be modified to a desired response expressed by the model (transfer characteristic) G _imp . Controller 51 ₄ is designed to precisely and stably driving the model (transfer _{function) G imp.} As a result, the control object 52 can be driven accurately and stably.

As described above in detail, according to the two control systems 50 according to the second embodiment, the target position R is generated by the stage control device 51 to drive the control object 52 following the target position R, and By correcting the control input U by the correcting unit 53, the response characteristic of the controlled object 52 can be corrected to a desired response expressed by the model (transfer characteristic) G _imp . Controller 51 ₄ is designed to drive precisely and stably to the correction model (transfer characteristic) G _imp. As a result, the control object 52 can be driven accurately and stably.

<< Third Embodiment >>
Hereinafter, a third embodiment will be described with reference to FIGS.

  In the third embodiment, the control system (deformation configuration) 50 ′ in the first embodiment is applied to an exposure apparatus 110 having a catapult type stage apparatus with some modifications.

  FIG. 14 shows a schematic configuration of an exposure apparatus 110 used for manufacturing a flat panel display according to the present embodiment, for example, a liquid crystal display device (liquid crystal panel). The exposure apparatus 110 drives the mask M and a glass plate (hereinafter referred to as “plate”) P at the same speed in the same direction with respect to the projection optical system PL, thereby causing the pattern formed on the mask M to move to the plate P This is a scanning stepper (scanner) to be transferred onto. In the following, the direction (scanning direction) in which the mask M and the plate P are driven during scanning exposure is the X-axis direction, and the direction in the horizontal plane perpendicular to the Y-axis direction is orthogonal to the X-axis and Y-axis. The rotation direction around the Z-axis direction, the X-axis, the Y-axis, and the Z-axis is the θx, θy, and θz directions, respectively.

  The exposure apparatus 110 includes an illumination system IOP, a mask stage MST that holds a mask M, a projection optical system PL, a body 70 that supports them, a plate stage PST that holds a plate P, and a control system thereof. The control system is composed of a main control device (not shown) that controls each component of the exposure apparatus 110 and a subordinate stage control device 51 (see FIG. 15 and the like).

  The body 70 includes a base (anti-vibration table) 71, columns 72A and 72B, an optical surface plate 73, a support body 74, and a slide guide 75. The base (anti-vibration table) 71 is disposed on the floor surface F and supports the columns 72A, 72B and the like by removing vibration from the floor surface F. Each of the columns 72A and 72B has a frame shape, and the column 72A is disposed inside the column 72B. The optical surface plate 73 has a flat plate shape and is fixed to the ceiling portion of the column 72A. The support 74 is supported on the ceiling of the column 72B via a slide guide 75. The slide guide 75 includes an air ball lifter and a positioning mechanism, and positions the support 74 (that is, a mask stage MST described later) at an appropriate position in the X-axis direction with respect to the optical surface plate 73.

  The illumination system IOP is disposed above the body 70. The illumination system IOP is configured in the same manner as the illumination system disclosed in, for example, US Pat. No. 5,729,331, for example, light (illumination light) IL emitted from a light source (not shown) such as a mercury lamp. Is applied to the mask M via a reflecting mirror, a dichroic mirror, a shutter, a wavelength selection filter, various lenses (all not shown), and the like. As the illumination light IL, for example, light such as i-line (wavelength 365 nm), g-line (wavelength 436 nm), h-line (wavelength 405 nm) or the like (or combined light of the i-line, g-line, and h-line) is used. Further, the wavelength of the illumination light IL can be appropriately switched by a wavelength selection filter according to, for example, the required resolution.

  Mask stage MST is supported by support body 74 constituting body 70. A mask M having a pattern surface (lower surface in FIG. 14) on which a circuit pattern is formed is fixed to the mask stage MST by, for example, vacuum suction (or electrostatic suction). The mask stage MST is driven with a predetermined stroke in the scanning direction (X-axis direction), for example, by a mask stage drive system MSD (see FIG. 16) including a linear motor, and in the non-scanning direction (Y-axis direction and θz direction). Is driven slightly.

  Position information (including rotation information in the θz direction) of the mask stage MST in the XY plane is measured by the mask interferometer system 16 (see FIG. 16). The measurement result is supplied to the stage controller 51 (see FIG. 15). The stage control device 51 drives the mask stage MST via the mask stage drive system MSD according to the measurement result of the mask interferometer system 16.

  Projection optical system PL is supported by optical surface plate 73 constituting body 70 below mask stage MST (on the −Z side). The projection optical system PL is configured in the same manner as the projection optical system disclosed in, for example, US Pat. No. 5,729,331, and a plurality of (for example, staggered projection areas of the pattern image of the mask M) (for example, A rectangular image field including the projection optical system 7) (multi-lens projection optical system) and having the longitudinal direction in the Y-axis direction is formed. Here, four projection optical systems are arranged at a predetermined interval in the Y-axis direction, and the remaining three projection optical systems are arranged at a predetermined interval in the Y-axis direction, spaced apart from the four projection optical systems to the + X side. ing. As each of the plurality of projection optical systems, for example, a bilateral telecentric equal magnification system that forms an erect image is used. A plurality of projection areas of the projection optical system PL arranged in a staggered pattern are collectively referred to as an exposure area.

  When the illumination area on the mask M is illuminated by the illumination light IL from the illumination system IOP, the circuit pattern of the mask M in the illumination area is illuminated by the illumination light IL transmitted through the mask M via the projection optical system PL. A projection image (partial upright image) is formed in an irradiation area (exposure area (conjugated to the illumination area)) on the plate P arranged on the image plane side of the projection optical system PL. Here, a resist (sensitive agent) is applied to the surface of the plate P. The mask stage MST and the plate stage PST are driven synchronously, that is, the mask M is driven in the scanning direction (X-axis direction) with respect to the illumination area (illumination light IL), and the plate P is moved to the exposure area (illumination light IL). On the other hand, by driving in the same scanning direction, the plate P is exposed and the pattern of the mask M is transferred onto the plate P.

  The plate stage PST is disposed on a base (anti-vibration table) 71 below (-Z side) the projection optical system PL. The plate stage PST is driven on the base 71 with a predetermined stroke in the X-axis and Y-axis directions by a stage drive system PSD including a linear motor and the like, and is finely driven in the Z-axis direction, θx direction, θy direction, and θz direction. Is done. On the plate stage PST, the plate P is held via a plate holder (not shown).

  Position information in the XY plane of the plate stage PST (including rotation information (yaw amount (rotation amount θz in θz direction), pitching amount (rotation amount θx in θx direction), rolling amount (rotation amount θy in θy direction))) Is measured by a plate interferometer system 18 (also referred to simply as an interferometer (FIG. 15)). The interferometer 18 irradiates a moving mirror (or a mirror-finished reflecting surface (not shown)) provided at the end of the plate stage PST from the optical surface plate, and receives the reflected light from the moving mirror. As a result, the position of the plate stage PST is measured. The measurement result is supplied to the stage controller 51 (see FIG. 15). The stage controller 51 drives the plate stage PST via the stage drive system PSD according to the measurement result of the interferometer 18.

  FIG. 15 shows the configuration of a control system related to the stage control of the exposure apparatus 110. The control system shown in FIG. 15 is mainly configured by a stage control device 51 including a microcomputer.

  The exposure apparatus 110 performs alignment measurement (for example, EGA or the like) prior to exposure, and uses the result to expose the plate P according to the following procedure. First, according to an instruction from a main controller (not shown), the stage controller 51 scans the mask stage MST and the plate stage PST according to the measurement results of the mask interferometer system 16 and the plate interferometer system 18, respectively (start of acceleration). Position). Next, both stages MST and PST are synchronously driven at the same speed in the same X-axis direction. As a result, the pattern of the mask M is transferred to the first shot area on the plate P so as to overlap the pattern formed in the previous process layer. When the scanning exposure for the first shot area is completed, the stage control device 51 moves (steps) the plate stage PST to the scanning start position (acceleration start position) for the second shot area. At the same time, the driving direction of the mask stage MST is changed to the opposite direction. Then, scanning exposure is performed on the second shot area. Similarly, the stage controller 51 repeats the stepping between the shot areas of the plate P and the scanning exposure for the shot areas to transfer the pattern of the mask M to all the shot areas on the plate P.

  The configuration of the mask stage MST, the mask stage drive system MSD (69, 71A, 71B, 71C, 80), and the mask interferometer system 16 will be described in detail. FIG. 16 shows these configurations and arrangements.

  The mask stage MST includes a fine movement stage 32 that moves while holding the mask M, and a coarse movement stage 34 that moves while supporting the fine movement stage 32.

  Fine movement stage 32 has a rectangular shape in plan view (viewed from above). A rectangular opening (not shown) through which the illumination light IL passes is formed substantially at the center of the fine movement stage 32, and a plurality of (for example, four) vacuum chucks (not shown) are provided around the opening. Using these vacuum chucks, the mask M is held on the fine movement stage 32 by vacuum suction. Gas static pressure bearings (not shown) are provided at the four corners of the lower surface of fine movement stage 32, respectively. By these gas static pressure bearings, the fine movement stage 32 is levitated and supported on a support 74 (a guide surface formed on the upper surface thereof) via a clearance of about several μm. Moving mirrors 17X and 17Y are provided on the + X end surface and the −Y end surface of fine movement stage 32, respectively. Instead of the movable mirror, the end surface of the stage may be mirror-finished to form a reflecting surface.

  The coarse movement stage 34 has an L-shape in plan view, and is fixed in a cantilevered manner to the −Y end of the mover 72 constituting the linear motor 69.

  The mask stage drive system MSD includes a linear motor 69 that drives the coarse movement stage 34 in the Y-axis direction, voice coil motors 71A, 71B, and 71C that drive the fine movement stage 32 relative to the coarse movement stage 34, and the coarse movement stage 34. , And a gas flow device 80 that suppresses a reaction force acting on the fine movement stage 32.

Linear motor 69 is configured to include a stator 69 ₁ supported on the support 74, the movable element 69 ₂ along the stator 69 ₁ moves in the X-axis direction, from. The stator 69 ₁ has a prismatic shape that the X-axis direction is the longitudinal direction, is supported on the support 74 in a non-contact manner through an air bearing or the like. The stator 69 _1, a plurality of armature coils (not shown) are arranged in the X-axis direction. Mover 69 ₂ has a casing that the X-axis direction is the longitudinal, coarse movement stage 34 is fixed to the -Y end. The mover 69 _2, include magnetic member having a YZ cross-section of inverted U-shaped (U-shaped). On a pair of opposing surfaces inside the magnetic member, a plurality of permanent magnets (not shown) are arranged in the X-axis direction so as to oppose different magnetic pole surfaces and alternately face different magnetic pole surfaces.

Linear motor 69 includes a stator 69 ₁ and the mover 69 _2, configured as a moving magnet type linear motor for driving the coarse movement stage 34 in the X-axis direction. Stage controller 51 by using the linear motor 69, i.e., controls driving of the coarse movement stage 34 by controlling the exciting current for exciting the armature coils of the stator 69 in _one. Note that a plurality of permanent magnets in the stator 69 _1, provided with a plurality of armature coils to the movable member 69 within ₂ may constitute a moving coil type linear motor.

The voice coil motor 71A is composed of a stator 71A ₁ and mover 71A _2. The stator 71 </ b> A ₁ is fixed to the −Y end surface of the coarse movement stage 34. The stator 71A _1, magnetic member (not shown) having a YZ cross-section of a substantially U-shaped (U-shaped) are included. On a pair of opposing surfaces (upper and lower opposing surfaces) of the magnetic member, a plurality of permanent magnets (not shown) are arranged in the Y-axis direction so as to face different magnetic pole surfaces and alternately face different magnetic pole surfaces. Mover 71A ₂ is fixed to the + Y end surface of the fine movement stage 32. Mover 71A ₂ has a flat plate shape, a plurality of armature coils (not shown) are arranged in the Y-axis direction inside thereof. By the mover 71A ₂ is engaged with the stator 71A ₁ (by being positioned in the space substantially U), it is arranged in a magnetic field which armature coils emitted a plurality of permanent magnets. Thereby, the mover 71A ₂ (fine movement stage 32) can be finely driven in the Y-axis direction with respect to the stator 71A ₁ (coarse movement stage 34).

The voice coil motors 71B and 71C are configured similarly to the voice coil motor 71A. That is, the voice coil motor 71B (71C) includes a stator 71B ₁ (71C ₁ ) and a mover 71B ₂ (71C ₁ ). The stator 71B ₁ (71C ₁ ) is fixed to the + Y side (−Y side) of the + X end face of the coarse movement stage 34 (arm part). The mover 71B ₂ (71C ₂ ) is fixed to the + Y side (−Y side) of the −X end surface of the fine movement stage 32. When the movers 71B ₂ and 71C ₂ engage with the stators 71B ₁ and 71C ₁ respectively (by being arranged in a substantially U-shaped space), the movers 71B ₂ and 71C ₂ (fine movement stage 32) are moved. The stators 71B ₁ and 71C ₁ (coarse movement stage 34) can be finely driven in the X-axis direction and the θz direction.

Stage controller 51, a voice coil motor 71A, 71B, with 71C, i.e. mover _{71A 2,} 71B 2, drive control of the fine movement stage 32 by controlling the exciting current for exciting the armature coils 71C in ₂ To do.

Gas flow device 80 is composed of a fixed portion 80 ₁ and the movable portion 80 _2. Fixing unit 80 ₁ is secured between the stator _71B 1, 71C ₁ of + X end surface of the coarse movement stage 34 (arm portions). The movable portion 80 ₂ is fixed between the movable element _71B 2, 71C ₂ of the -X end surface of the fine movement stage 32.

Figure 17 shows details of the configuration of the gas flow device 80 (fixing unit 80 ₁ and the movable portion 80 _2). Fixing unit 80 ₁ has a rectangular parallelepiped housing 81, the housing 81 inside the -X inner surface and the + X inner surface a pair of which is fixed to the gas ejection mechanism 82A, and 82B. The movable portion 80 _2, an arm 83 projecting from the -X end surface of the fine movement stage 32, a plate-like member which is disposed fixed to the distal end of the arm 83 (-X end) of the gas ejection mechanism 82A, while the 82B 84. The arm 83 is inserted into the housing 81 through an opening 81a provided in the housing 81 and a gas ejection mechanism 82B.

  The gas ejection mechanisms 82A and 82B eject the same amount of pressurized gas such as air, nitrogen, and helium toward the −X plane and the + X plane of the plate-like member 84, respectively. Thereby, a thrust in the X-axis direction acts on the fine movement stage 32 with respect to the coarse movement stage 34. Here, the thrust in the + X direction and the −X direction acting on the plate member 84 from the gas ejection mechanisms 82A and 82B is proportional to the square of the separation distance between each of the gas ejection mechanisms 82A and 82B and the plate member 84. . Therefore, when the plate-like member 84 is located at the center of the gas ejection mechanisms 82A and 82B, the total thrust acting on the plate-like member 84 becomes zero, and the plate-like member 84 remains in that position. When the plate member 84 is located on the gas ejection mechanism 82A side, the + X direction thrust received from the gas ejection mechanism 82A is superior to the −X direction thrust received from the gas ejection mechanism 82B, that is, the total acting on the plate member 84. Since the thrust is directed in the + X direction, the plate member 84 is returned to the + X direction. On the other hand, when the plate member 84 is located on the gas ejection mechanism 82B side, the thrust in the −X direction received from the gas ejection mechanism 82B is superior to the thrust in the + X direction received from the gas ejection mechanism 82A, that is, acts on the plate member 84. Since the total thrust to be directed is in the −X direction, the plate member 84 is returned to the −X direction.

  When the fine movement stage 32 approaches the coarse movement stage 34 (when moved in the −X direction) by the gas flow device 80, a thrust is applied to the fine movement stage 32 in a direction away from the coarse movement stage 34 (+ X direction). When the stage 32 is separated from the coarse movement stage 34 (when moved in the + X direction), thrust is applied to the fine movement stage 32 in a direction approaching the coarse movement stage 34 (−X direction). As the gas flow device 80 functions as described above, the reaction force acting on the fine movement stage 32 by driving the coarse movement stage 34 can be suppressed. Therefore, it is not necessary to generate a large thrust by the voice coil motors 71A, 71B, 71C.

  The stage control device 51 may control the amount of pressurized gas ejected from the gas ejection mechanisms 82A and 82B in accordance with the position of the plate member 84, that is, the position of the fine movement stage 32 relative to the coarse movement stage 34. good.

The mask interferometer system 16 includes interferometer units 16X ₁ , 16X ₂ , and 16Y. The interferometer units 16X ₁ and 16X ₂ respectively irradiate the measurement beam to the + Y side and the −Y side of the movable mirror 17X of the fine movement stage 32, and receive the reflected light from the movable mirror 17X, thereby receiving the mask stage MST. The positions in the X-axis direction and the θz direction are measured. The interferometer unit 16Y measures the position of the mask stage MST in the Y-axis direction by irradiating the movable mirror 17Y of the fine movement stage 32 with a length measurement beam and receiving the reflected light from the movable mirror 17Y. As described above, the measurement result is supplied to the stage control device 51 (see FIG. 15).

  18A and 18B show the configuration of a control system 50 ″ for driving and controlling the mask stage MST according to the third embodiment. The configuration of the control system 50 ″ according to the first embodiment. Since it is similar to the configuration of the control system 50, the configuration will be described focusing on the differences.

  The control system 50 ″ is a feedback control system that drives and controls the fine movement stage 32 to be controlled by giving a control input U, and generates a drive target (having a driving force dimension) R of the fine movement stage 32. Then, the control input U is calculated using the result (and the correction amount dU described later), and the result is transmitted to the mask stage drive system MSD (voice coil motors 71A, 71B, 71C); From the above-described mask interferometer system 16 that measures the position X, Y, and θz of the mask stage MST (fine movement stage 32), which is a controlled variable, and the output of the mask interferometer system 16 (measurement results of the positions X, Y, and θz). A correction unit 53 ′ for obtaining a correction amount dU for the control input U, and a switch 55 for connecting the output of the correction unit 53 ′ to the stage control device 51. Is done.

  The mask stage drive system MSD (voice coil motors 71A, 71B, 71C) generates a drive force equal to the drive amount in accordance with the received control input U (having a force dimension in this embodiment). Thereby, the fine movement stage 32 is driven and controlled with respect to the coarse movement stage 34. In addition, d shown in the figure is a disturbance acting on the fine movement stage 32 from the outside, and is unknown. The control input U and the positions X, Y, and θz are input and output to the fine movement stage 32, respectively, and are both known. The actual input to fine movement stage 32 is the sum of control input U (equal to the driving force generated by voice coil motors 71A, 71B, 71C) output from stage control device 51 and disturbance d.

Stage controller 51 includes a target generator 51 ₀ and the switch 51 ₁ and the subtractor 51 _2. Drive target of the fine movement stage 32 target generator 51 ₀ generated (target with the dimensions of the driving force in this case) through the switch 51 ₁ is supplied to the subtracter 51 _2.

The correction unit 53 ′ has the same configuration as that of the correction unit 53 ′ according to the first embodiment. The output (correction amount dU) of the correction unit 53 ′ is output to the stage control device 51 (subtractor 51 ₂ ) via the switch 55.

  The coarse movement stage 34 that is driven with a relatively large driving force on the mask stage MST having the above-described structure, that is, the support 74 (the guide surface formed on the upper surface thereof), and the coarse movement stage 34 that is driven precisely. In the coarse / fine movement type mask stage MST constituted by the fine movement stage 32, the fine movement stage 34 (linear motor 69) is driven by the fine movement stage in order to drive the fine movement stage 32 holding the mask M at high speed. A catapult type driving method applied to 32 can be employed. Here, as long as the gas flow device 80 functions, the reaction force of the fine movement stage 32 received from the coarse movement stage 34 is suppressed. Therefore, the fine movement stage 32 does not contact the coarse movement stage 34 (force F (= 0)), and keeps the separated state. However, by applying a strong driving force of the coarse movement stage 34 (linear motor 69) to the fine movement stage 32, the limit of the function of the gas flow device 80 is exceeded, and the fine movement stage 32 comes into contact with the coarse movement stage 34 (force F caused by the collision F (Note 0), chattering (mechanical vibration) may occur.

When the mask M (mask stage MST) is driven at a constant speed to expose the plate P, the gas flow device 80 functions sufficiently, the fine movement stage 32 does not contact the coarse movement stage 34, and the fine movement stage is coarse. Since the force (that is, the external force d) F caused by the collision received from the moving stage is zero, the correction of the control input U by the correction unit 53 ′ is not necessary. Therefore, the control system 50 ', at a constant speed movement of the mask stage MST, 18 turns on the switch 51 ₁ in the stage controller 51 (A), a fine movement stage that target generator 51 ₀ generated 32 drive target R is supplied to the subtractor 51 _2, turn off the switch 55, blocking so that the output from the correction unit 53 '(the correction amount dU) is not output to the stage controller 51 (subtractor 51 ₂₎ To do.

On the other hand, if the mask M (mask stage MST) is accelerated or decelerated by the catapult driving method to step the plate P and change the driving direction of the mask M in the opposite direction, the limit of the function of the gas flow device 80 is exceeded, and the fine movement Since the stage 32 may come into contact with the coarse movement stage 34 and receive a force (external force d) F caused by a collision from the coarse movement stage (external), it is necessary to correct the control input U by the correction unit 53 ′. However, the drive target R of the fine movement stage 32 is not necessary. Therefore, the control system 50 ', during acceleration and deceleration of the mask stage MST, as shown in FIG. 18 (B), the fine movement stage by turning off the switch 51 ₁ in the stage control unit 51, target generator 51 ₀ generated 32 drive target R is cut off so as not to be supplied to the subtracter 51 _2, turn on the switch 55, inputs the output from the correcting unit 53 '(the correction amount dU) to the stage control unit 51 (the subtractor 51 ₂₎ To do.

  FIGS. 19A and 19B show simulation results of reaction force and position error received by the fine movement stage 32 when the coarse movement stage 34 collides with the fine movement stage 32, respectively. When the control system 50 ″ of the third embodiment is not adopted (w / o DDOB), the fine movement stage 32 does not have virtual viscoelasticity characteristics. By colliding with the fine movement stage 32, a large reaction force of −150 N is generated in the fine movement stage 32, and a strong vibration (chattering) occurs until time 0.2s, and the reaction force is gradually attenuated by time 0.5s. The position error flies to about 500 μm due to a collision, shows an overshoot (hunting) until time 0.15 s, and remains almost constant after time 0.2 s.

  On the other hand, when the control system 50 ″ of the third embodiment is employed (with DDOB), the fine movement stage 32 has virtual viscoelastic characteristics, so that the coarse movement stage 34 finely moves at time 0.05s. By colliding with the stage 32, a large reaction force of -150N is generated on the fine movement stage 32, and a strong vibration (chattering) occurs until time 0.1s, and the reaction force increases gently at time 0.1-0.25s. The position error jumps to about 500 μm due to a collision, shows an overshoot (hunting) until time 0.10 s, and remains almost constant after time 0.1 s. By adopting the control system 50 ″, the reaction force received by the fine movement stage 32 is reduced, that is, the chattering amplitude is reduced and the period thereof is shortened. The amplitude of the hunting that shows the position error is small and the period is short.

  As described above in detail, according to the control system 50 ″ of the third embodiment, the fine input stage 32 is corrected from the control input U when the fine movement stage 32 comes into contact with the coarse movement stage 34 and accelerates (when contact is possible). The output (correction amount dU) of the unit 53 ′ is reduced, so that the fine movement stage 32 is driven so that the fine movement stage 32 virtually functions viscoelasticity with respect to the force (external force) caused by the collision received from the coarse movement stage 34. As a result, chattering due to the collision is suppressed, and the fine movement tee 32 can be driven accurately and stably.

When the mask M (mask stage MST) is driven at a constant speed, the output of the first disturbance observer 531 is output in a stage without the second disturbance observer 533 instead of turning off the switch 55 in FIG. You may comprise so that it can output to the control apparatus 51 (subtractor 512). Thus, since the control input U to the first output of the disturbance observer 53 ₁ (disturbance) is subtracted is corrected from the control input U, to offset the external force fine movement stage 32 receives from the other coarse movement stage 34 virtually Thus, the fine movement stage 32 can be driven. As a result, the fine movement stage 32 can be driven accurately and stably.

  In addition, since the exposure apparatus 110 according to the present embodiment includes the control system 50 ″ for the mask stage MST designed as described above, the mask stage MST can be driven accurately and stably, and exposure accuracy, that is, The overlay accuracy can be improved.

  In the third embodiment, the control system 50 ″ is supported by the coarse movement stage 34 driven by a large driving force on the support body 74 (the guide surface formed on the upper surface thereof), and the coarse movement stage 34. Although applied to the drive control of the coarse / fine movement type mask stage MST composed of the fine movement stage 32 that is precisely driven, the fine movement stage that holds and moves the plate P, and moves while supporting the fine movement stage. The present invention may be applied to drive control of a coarse / fine movement type plate stage, whereby the fine movement stage can be driven accurately and stably, and exposure accuracy, that is, overlay accuracy can be improved.

  In the third embodiment, as the control system 50 ″, the control system (deformed configuration) 50 ′ in the first embodiment is adopted. However, the control system 50 in the second embodiment may be adopted.

  In addition to the mask interferometer system 16 and the plate interferometer system 18, for example, using a head provided on the optical surface plate 73, the measurement light is applied to the scale provided on the mask stage MST or the plate stage PST, It is also possible to use an encoder that receives the return light.

  Further, the configurations of the mask interferometer system 16 and the plate interferometer system 18 are not limited to the above-described configurations, and a configuration in which an interferometer is further added as appropriate can be adopted depending on the purpose. Further, instead of or together with the mask interferometer system 16 and the plate interferometer system 18, an encoder (or an encoder system composed of a plurality of encoders) may be used.

In the third embodiment, the illumination light is ultraviolet light such as ArF excimer laser light (wavelength 193 nm) or KrF excimer laser light (wavelength 248 nm), or vacuum ultraviolet light such as F ₂ laser light (wavelength 157 nm). There may be. As the illumination light, for example, a single wavelength laser beam oscillated from a DFB semiconductor laser or a fiber laser is amplified by a fiber amplifier doped with, for example, erbium (or both erbium and ytterbium). In addition, harmonics converted into ultraviolet light using a nonlinear optical crystal may be used. A solid laser (wavelength: 355 nm, 266 nm) or the like may be used.

  In the third embodiment, the case where the projection optical system PL is a multi-lens projection optical system including a plurality of optical systems has been described. However, the number of projection optical systems is not limited to this, and 1 There should be more than one. The projection optical system is not limited to a multi-lens type projection optical system, and may be a projection optical system using an Offner type large mirror, for example. In the above embodiment, the case where the projection optical system PL has the same magnification as the projection magnification has been described. However, the present invention is not limited to this, and the projection optical system may be either an enlargement system or a reduction system.

  Further, the third embodiment (stage drive system) can be applied to both a scanning exposure apparatus such as a batch exposure type or a scanning stepper and a stationary exposure apparatus such as a stepper. The above embodiments can also be applied to a step-and-stitch projection exposure apparatus that combines a shot area and a shot area. In addition, each of the above embodiments can be applied to a proximity type exposure apparatus that does not use a projection optical system, and is also applicable to an immersion type exposure apparatus that exposes a substrate via an optical system and a liquid. can do. In addition, in each of the above embodiments, an exposure apparatus (US) that combines two patterns on a substrate via a projection optical system and performs double exposure of one shot area on the substrate almost simultaneously by one scan exposure. (Patent No. 6,611,316) and the like.

  Further, the use of the exposure apparatus is not limited to a liquid crystal exposure apparatus that transfers a liquid crystal display element pattern onto a square glass plate. For example, an exposure apparatus for semiconductor manufacturing, a thin film magnetic head, a micromachine, and a DNA chip The present invention can also be widely applied to an exposure apparatus for manufacturing the above. Moreover, in order to manufacture not only microdevices such as semiconductor elements but also masks or reticles used in light exposure apparatuses, EUV exposure apparatuses, X-ray exposure apparatuses, electron beam exposure apparatuses, etc., glass substrates, silicon wafers, etc. The embodiments described above can also be applied to an exposure apparatus that transfers a circuit pattern. The object to be exposed is not limited to the glass plate, but may be another object such as a wafer, a ceramic substrate, or a mask blank.

  For electronic devices such as liquid crystal display elements (or semiconductor elements), the step of designing the function and performance of the device, the step of manufacturing a mask (or reticle) based on this design step, and the step of manufacturing a glass plate (or wafer) Other than the lithography step of transferring the mask (reticle) pattern to the glass plate by the exposure apparatus and the exposure method of each embodiment described above, the developing step of developing the exposed glass plate, and the portion where the resist remains It is manufactured through an etching step for removing the exposed member of the portion by etching, a resist removing step for removing a resist that has become unnecessary after etching, a device assembly step, an inspection step, and the like. In this case, in the lithography step, the exposure method described above is executed using the exposure apparatus of the above-described embodiment, and a device pattern is formed on the glass plate. Therefore, a highly integrated device can be manufactured with high productivity. .

  Although the exposure apparatus has been described as an example in the embodiment, the present invention can be applied to other apparatuses than the exposure apparatus. For example, if the present invention is applied to a vibration isolation device on which a measuring device or the like is mounted, it is possible to improve the vibration isolation efficiency.

16 (16X, 16Y) ... mask interferometer system (interferometer unit), 18 ... plate interferometer system (interferometer), 32 ... fine movement stage, 34 ... coarse movement stage, 50, 50 ', 50 "... control system, 51 ... stage controller, 51 0 _... target generator, 51 1 _... switch, 51 2 _... subtractor, 52 ... control object, 53, 53 '... correcting section, ₅₃ 1, 53 3 _... disturbance observer, 53a, 53b ... Controller, 55 ... switch, 69 ... linear motor, 71A, 71B, 71C ... voice coil motor, 110 ... exposure apparatus.

Claims (17)

A control system that generates an operation amount of the control target from a controlled amount of the control target,
An input unit for inputting the controlled variable;
The generated amount and the first model are different from the first amount obtained from the controlled amount or the generated amount generated by the calculation using the controlled amount and the first model corresponding to the controlled object. A control unit that generates the manipulated variable using a difference from the second amount obtained from the second model,
The second model is a control system in which a disturbance applied to the control target and the controlled amount are represented by a predetermined function.   The control system according to claim 1, wherein the function represents a translational motion of a rigid body having mass, viscosity, and elasticity.   The control unit includes a disturbance observer including the second model, and the disturbance observer inputs a third amount obtained from the first amount and the operation amount and the controlled amount, and performs the operation. The control system according to claim 1, wherein a correction amount for correcting the amount is output.   The control system according to claim 1, wherein the first model represents a free translational motion of a rigid body having a mass.   The control unit includes a second disturbance observer including the first model, and the second disturbance observer obtains the first amount based on the controlled amount and the operation amount. Or the control system of 4.   The control system according to any one of claims 1 to 5, further comprising a target unit that generates a target amount related to the position of the control target and outputs a difference between the target amount and the controlled amount as the generation amount. .   A target unit that generates a target amount related to the position of the control target, outputs a sum of the target amount and the controlled amount, a controller that generates the manipulated variable using the output of the target unit, and The control system according to claim 1, further comprising an adder that adds the difference between the amount of 1 and the second amount and the manipulated variable. When the control object is in contact with another member, the operation amount is generated using a difference between the first amount and the second amount,
The control system according to claim 6 or 7, wherein when the control object is separated from the other member, the operation amount is generated using the output of the target unit. An exposure apparatus that exposes an object with an energy beam to form a pattern on the object,
A moving mechanism having a first moving body that moves while holding the object, and a second moving body that supports the first moving body and is movable with respect to the first moving body;
The control system according to any one of claims 1 to 8, wherein the first moving body is the control target.
An exposure apparatus comprising: A control method for controlling a controlled object that inputs an operation amount and outputs a controlled amount,
Obtaining information on the position of the controlled object as the controlled amount;
Obtaining a first quantity input to the controlled object from the controlled quantity or a generated quantity generated by a calculation using the controlled quantity and a first model corresponding to the controlled object;
Obtaining a second amount input to the control object from the generated amount and a second model different from the first model;
Generating the manipulated variable using the first quantity and the second quantity,
The second model is a control method in which a disturbance applied to the controlled object and the controlled amount are represented by a predetermined function.   The control method according to claim 10, wherein the second model represents a translational motion of a rigid body having mass, viscosity, and elasticity.   The control method according to claim 10 or 11, wherein the first model represents a free translational motion of a rigid body having a mass.   The control method according to claim 10, further comprising: generating a target value related to the position of the control target, and generating the generation amount from a difference between the target value and the controlled amount. Generating a target value related to the position of the controlled object, obtaining a sum of the target value and the controlled quantity, and generating the manipulated variable;
The control method according to claim 10, further comprising: adding a difference between the first amount and the second amount and the operation amount. When the control object is in contact with another member, the operation amount is generated using a difference between the first amount and the second amount,
The control method according to claim 13 or 14, wherein when the control target is separated from the other member, the operation amount is generated using the output of the target unit. An exposure method for exposing an object with an energy beam to form a pattern on the object,
Among the moving mechanisms having a first moving body that moves while holding the object, and a second moving body that is movable relative to the first moving body while supporting the first moving body, the first moving body. The exposure method using the drive method as described in any one of Claims 10-15 which makes a moving body the said control object. Using the exposure method of claim 16 to form a pattern on an object;
Developing the object on which the pattern is formed;
A device manufacturing method including: