WO2001006549A1

WO2001006549A1 - Exposure method and device therefor

Info

Publication number: WO2001006549A1
Application number: PCT/JP2000/004707
Authority: WO
Inventors: Nobutaka Magome; Motokatsu Imai; Yukio Kakizaki
Original assignee: Nikon Corporation
Priority date: 1999-07-19
Filing date: 2000-07-13
Publication date: 2001-01-25
Also published as: AU6015200A; TW460933B

Abstract

An exposure method and device therefor, which can enhance a throughput at exposing by enabling patterns to be transferred onto a wide area on an object to be exposed without substantially upsizing a projection system. Electron beams (EB) shot from an electron gun (1) are shone on transfer-receiving character patterns on a mask (M) via a field-of-vision-selecting deflector (7), the electron beams passed through the character patterns are deflected back by a deflector (25) and then applied onto a wafer (W) via a projection system (PL). The projection system (PL) is supported so as to be displaceable by a drive system (34) in a direction perpendicular to an optical axis, and the projection system (PL) is mechanically displaced or vibrated when the reduced images of character patterns respectively corresponding to a plurality of positions on the wafer (W) are to be transferred.

Description

Description Exposure method and apparatus

The present invention relates to a lithographic apparatus for manufacturing a device such as a semiconductor device, an image pickup device (CCD or the like), a liquid crystal display device, a plasma display device, or a thin film magnetic head. The present invention relates to an exposure method and apparatus used for transferring onto a substrate, especially charged particles that transfer various patterns such as a mask pattern or a character pattern onto a substrate via a charged particle beam such as an electron beam or an ion beam. It is suitable for use in a line transfer device. Background art

In recent years, studies have been made on a charged particle beam transfer apparatus capable of improving both the resolution of a transfer pattern and the throughput (productivity). Conventionally, such a transfer device is equivalent to one die corresponding to one integrated circuit (semiconductor chip, etc.) formed on one wafer as a photosensitive substrate, or A batch transfer type apparatus for transferring a pattern of a plurality of dies from a mask onto a wafer coated with a resist sensitive to a charged particle beam has been studied. However, in the batch transfer method, it is difficult to manufacture a mask serving as a transfer master, and a charged particle projection system (hereinafter simply referred to as a “projection system”) in a large optical field equivalent to one die. It is difficult to keep the aberration within a predetermined range.

In this regard, in a semiconductor device or the like, the circuit pattern of each layer of one die on a substrate such as a wafer is, for example, a plurality of types of lines having a predetermined pitch • It can also be formed by combining and transferring images of a predetermined pattern (hereinafter referred to as “character pattern”) such as an and space pattern or a wiring pattern having a predetermined shape. it can. Therefore, a “character pattern transfer method” is used in which a large number of types of character patterns are formed in advance on a mask, and images of one character pattern selected from these are sequentially transferred onto a substrate in a predetermined arrangement. Development of a particle beam transfer device is being performed. For convenience of explanation, an area where a pattern image (for example, a reduced image) of one die is transferred on the substrate is called a shot area. The surface of the substrate is divided vertically and horizontally into a number of shot areas at predetermined pitches.

In such a character pattern transfer type apparatus, it is necessary to transfer a character pattern sequentially selected on the mask side into each shot area on the substrate in a predetermined arrangement as quickly as possible (with high throughput), so that the field of view of the projection system is limited. It is desirable to be as large as possible. However, simply increasing the size of the projection system increases various aberrations.Therefore, in order to substantially increase the field of view without increasing the size of the projection system, the distribution of the electromagnetic field in the projection system must be increased. The MOL (Moving Objective Lens) method, which electronically displaces the electron optical system of the projection system in a direction perpendicular to the optical axis by changing the optical system, is described in the literature (E. Goto and T. Soma: Optik 48, No. 255-270 (1977)).

In addition, the `` division transfer method '' is used to divide the circuit pattern to be transferred to the substrate into a plurality of sub-fields smaller than the size equivalent to one die, and transfer the reduced image of the pattern for each sub-field sequentially to the substrate. Development of a particle beam transfer device is also under consideration. Also in this division transfer method, it is desirable that the field of view of the projection system be as large as possible in order to transfer the pattern of each sub-field of view onto the substrate at high speed. Furthermore, even when a pattern (a character pattern or a sub-field pattern, etc.) at a position distant from the optical axis on the mask is transferred via the projection system, the charged particle beam must be Since it is desirable to pass through a position as close to the optical axis as possible, an MTP (Moving Trajectory Projection) optical system has been proposed. The MTP method is a method in which a deflector performs reduced projection using a projection system while the orbit of the charged particle beam is substantially aligned with the optical axis.

When a reduced image of a pattern sequentially selected from a mask is transferred to a different position on a substrate by the character pattern transfer method or the split transfer method as described above, when a narrow area is used, the deflector may use a charged particle beam. It is also possible to shift the position. However, the deflector alone cannot transfer the image of the pattern corresponding to each position in one wide shot area on the substrate sequentially, so that conventionally, the substrate is mechanically moved by the substrate stage. This changed the transfer position of the image on the substrate. Also, the transfer position on the substrate was corrected by a deflector provided on the substrate as needed.

In the conventional charged particle beam transfer apparatus as described above, in order to transfer images of a predetermined pattern to different positions in one shot area (one die area) on the substrate, respectively. The substrate stage was mechanically driven two-dimensionally. In this case, within one shot area on the substrate, the smallest exposed area exposed by the image of one minimum unit pattern on the mask is called a "sub-exposure area". By moving the stage for the substrate in the first direction, the images of the corresponding patterns are sequentially transferred into the elongated “main exposure region” having a width corresponding to approximately one sub-exposure region on the substrate, and then transferred. Moving the substrate stage in a second direction orthogonal to the first direction, and then moving the substrate stage again in the first direction. The operation of successively transferring the images of the corresponding patterns in the next main exposure area was repeated.

Therefore, conventionally, when exposing one shot area, it is necessary to repeatedly scan the stage for the substrate in the first direction many times, which increases the exposure time and reduces the throughput of the exposure process. There was an inconvenience.

In this regard, in order to increase the field of view of the projection system, a method of simply enlarging the projection system is conceivable, but enlarging the projection system without increasing the aberration significantly increases the manufacturing cost. There is fear. Furthermore, since the optical path of the charged particle beam of the charged particle beam transfer device needs to be evacuated, the projection system should be as small as possible in order to make the vacuum vessel as small as possible and reduce the manufacturing cost. desirable.

In addition, in the conventional electronic M〇L method, by changing the distribution of the electromagnetic field in the projection system, the projected image is transferred with low aberration within a width of one to two sub-exposure areas on the substrate. Since the position could be shifted, when the substrate stage was scanned in the first direction using this electronic MOL method, the transfer was performed, for example, to the main exposure area with the width of two sub-exposure areas. If we do, the throughput can be improved by a factor of two. However, even in this case, it is necessary to repeat scanning by the substrate stage many times in order to expose the entire one shot region, and therefore, there has been a demand for the development of a technology for greatly improving the throughput.

In view of the above, it is a first object of the present invention to provide an exposure method capable of transferring a pattern to a wide area on an object to be exposed (object to be exposed) without increasing the size of a projection system. .

It is a second object of the present invention to provide an exposure method that can increase the throughput when performing exposure on an object to be exposed. It is a third object of the present invention to provide an exposure method capable of increasing throughput when transferring an image of a small pattern sequentially selected from a mask onto an object to be exposed.

Further, a fourth object of the present invention is to provide an exposure apparatus capable of performing such an exposure method.

A fifth object of the present invention is to provide a device manufacturing method capable of manufacturing a high-performance device with high throughput by using the exposure method. Disclosure of the invention

The first exposure method according to the present invention comprises irradiating a first object (M) with a charged particle beam and projecting a charged particle beam having passed through the pattern of the first object (W) through a projection system (PL). In order to irradiate a plurality of different positions on the second object with charged particle beams having passed through corresponding patterns on the first object, at least a part of the The member (PL; 29) is displaced.

According to the present invention, for example, as shown in FIG. 10, at least some of the movable members of the projection system (PL) are moved in a direction perpendicular to the normal line of the second object (W), or When displaced by 1, Ay2 in the direction perpendicular to the optical axis, the transfer position on the second object moves by ΔΥ1, ΔY2 along the direction of displacement. The method of mechanically displacing at least a part of the movable member of the projection system in this manner is hereinafter referred to as a “mechanical moving objective lens (M〇L) method”. By using this mechanical MOL method, it is possible to reduce, for example, a reduced image of a pattern corresponding to a plurality of different positions on the second object, that is, to a wide area, without increasing the size of the projection system. Can be transferred. In this case, it is desirable to move the second object in a plane substantially perpendicular to the optical axis of the projection system and further with respect to the projection system. Thereby, a predetermined group of patterns can be transferred to a wider two-dimensional area on the second object.

Also, in synchronization with the operation of moving the second object or the operation of displacing the movable member, the electromagnetic field in the projection system is electrically changed, and the irradiation position of the charged particle beam on the second object is changed. Is desirably corrected. This means that the conventional electronic MOL method is used in combination with the mechanical M〇L method of the present invention. For example, as shown in Fig. 11, when at least a part of the projection system (PL) is mechanically displaced in a predetermined direction, the distribution of the electromagnetic field of the electromagnetic lens system constituting the projection system is changed. By changing the position (27 A, 27 B, 29 A, 29 B) shown by the original dotted line to the position (27, 29) shown by the solid line, the second object (W) The transfer position can be set to the same position (54Β). Therefore, the accuracy of the transfer position on the second object is improved, and the overlay accuracy is improved during the overlay exposure.

In addition, a plurality of different patterns are formed on the first object, and the charged particle beam passing through the pattern selected from the first object according to the irradiation position on the second object is obtained. It is desirable to lead to a projection system. This corresponds to a case where the exposure method of the present invention is applied to a character pattern transfer method, a division transfer method, or the like.

It is also desirable to detect rotation information of the movable member and correct the irradiation position of the charged particle beam on the second object based on the rotation information. Accordingly, even if the movable member rotates during the displacement, the transfer position on the second object can be accurately corrected. Further, the movable member is displaceable in a first direction, and the second object is moved in a second direction substantially orthogonal to the first direction. It is preferable that the pattern of the first object is transferred to an area on the second object that is longer than the moving distance of the movable member in the first direction. This further improves throughput.

Next, in the second exposure method of the present invention, the first object (M) is irradiated with a charged particle beam, and the charged particle beam passing through the pattern of the first object is projected through a projection system (PL). In the exposure method of irradiating two objects (W), a plurality of different positions on the second object are each irradiated with a charged particle beam having passed through a corresponding pattern on the first object. Vibrates at least a part of the movable members (PL; 29) in a predetermined direction.

In the second exposure method, at least a part of the movable members of the projection system is vibrated in a predetermined direction by applying the mechanical moving objective lens (M〇L) method of the present invention. Therefore, the pattern transfer position on the second object can be moved at high speed along the predetermined direction. At this time, at least some of the movable members of the projection system can be considerably reduced in weight compared to the stage for positioning the second object, so that the stage is mechanically moved in the predetermined direction. Exposure on the second object can be performed with greatly improved throughput compared to driving along.

In this case, in synchronization with oscillating the movable member in the first direction (Y direction), the second object is moved in a second direction (X direction) crossing the first direction, It is desirable to irradiate a two-dimensional area on the second object with a charged particle beam having passed through a corresponding pattern on the first object. Thus, by combining the vibration in the first direction by the mechanical MOL method and the mechanical driving of the second object in the second direction, for example, one of the When exposing the shot area (area for one die), the number of times the second object is mechanically driven in the second direction can be reduced. Therefore, the throughput of the exposure process is improved. Also in this case, it is desirable to correct the irradiation position of the charged particle beam on the second object by electrically changing the electromagnetic field in the projection system in synchronization with the operation of vibrating the movable member. . For example, as shown in Fig. 13 (A), simply moving the movable member in the first direction (Y direction) and moving the second object in the second direction (X direction), The transfer position of the pattern on the two objects deviates mainly from the target position (54A, 54B,...) In the second direction as shown by the locus (1 19). Therefore, the transfer position on the second object is corrected by correcting the transfer position of the pattern mainly in the second direction by combining the electronic MOL method so as to correct the displacement amount. For example, as shown by the locus (1 2 1) in Fig. 13 (D), the robot moves along the target position.

Furthermore, for example, at time points t1, t2,... On the locus (12 1) in FIG. 13D, the transfer positions have reached the target positions (54A, 54B,...), Respectively. Then, by irradiating the charged particle beam for a very short time at that time tl, t 2,..., The pattern corresponding to the target position can be transferred. However, more precisely, for example, when a photosensitive material is coated on the second object, it is necessary to irradiate the charged particle beam onto the second object for a predetermined exposure time Δt. In this case, for example, in order to keep the transfer position at the target position (54A) only during the exposure time Δt centered at the time point tl, the first direction (Y However, it is necessary to correct the transfer position.

In this case, the vibration frequency of the movable object may be controlled according to the sensitivity of the second object. Assuming that the intensity of the charged particle beam is constant, if the sensitivity of the second object is low, the exposure time Δt becomes long, so that the vibration frequency needs to be lowered. On the other hand, when the sensitivity of the second object is high, the exposure time Δt can be shortened, so that the vibration frequency The number can be higher.

On the other hand, for example, when the intensity of the charged particle beam source can be stably controlled, it is desirable to control the intensity of the charged particle beam according to the sensitivity of the second object. In this case, since the vibration frequency of the movable member can be kept constant, it is easy to control the mechanical drive system.

In addition, when a balancer (128) is connected to the movable member via a flexible member, and the movable member is vibrated, the position of the center of gravity of the mechanical system including the movable member and the balancer (G It is desirable that the balancer be vibrated in phase opposite to that of the movable member so that the movable member does not move. Thereby, the movable member can be more stably vibrated, and the accuracy of the transfer position on the second object is improved.

Further, also in the present invention, it is desirable to detect rotation information of the movable member and correct the irradiation position of the charged particle beam on the second object based on the rotation information. Thus, even if the movable member rotates during the displacement, the transfer position on the second object can be accurately corrected.

Next, the first exposure apparatus according to the present invention irradiates a first object (M) with a charged particle beam, and projects a charged particle beam having passed through the pattern of the first object through a projection system (PL). In the exposure apparatus for irradiating (W), in order to move the irradiation position of the charged particle beam on the second object, a mechanical drive system (at least one of the movable members of the projection system is displaced) 32 A, 32 B, 34) are provided.

In this case, it is desirable to provide a second object stage (46) for moving the second object with respect to the projection system in a plane substantially perpendicular to the optical axis of the projection system.

In addition, in order to correct the irradiation position of the charged particle beam on the second object, an electronic drive system for electrically changing an electromagnetic field in the projection system (35) And a control system (33) for controlling the operation of the electronic drive system according to the operation of the second object stage and the operation of the mechanical drive system.

Also, when a plurality of different patterns (15A, 15B, 15C, '...) are formed on the first object, the first object is moved to the optical axis of the projection system. A first object stage (20) for moving in a substantially vertical plane, a first deflector (7) for irradiating the charged particle beam to a pattern selected from the first object, and It is desirable to provide a second deflector (25) for turning back the charged particle beam passing through the selected pattern on the first object. According to the exposure apparatus of the present invention, the first exposure method of the present invention can be performed.

Further, the second exposure apparatus according to the present invention irradiates the first object (M) with a charged particle beam, and emits a charged particle beam having passed through the pattern of the first object via a projection system (PL). W), the mechanical drive system (3) that vibrates at least some of the movable members of the projection system in order to move the irradiation position of the charged particle beam on the second object. 2 A, 32 B, 34).

In this case, the mechanical drive system vibrates the movable member along the first direction (Y direction), and a corresponding pattern on the first object is stored in a two-dimensional area on the second object. In order to irradiate the charged particle beam, it is desirable to provide a second object stage (46) for moving the second object in a second direction (X direction) intersecting the first direction.

Further, in order to move the irradiation position of the charged particle beam on the second object, an electronic drive system (35) for electrically changing an electromagnetic field in the projection system is provided, and the movable member is provided. Of the charged particle beam via the electronic drive system in synchronism with the operation of oscillating the beam and the movement of the stage for the second object. It is desirable to correct the irradiation position on the second object.

The mechanical drive system includes, as an example, first and second leaf spring members (32A, 32B) arranged so as to sandwich the movable member and having mobility, and a support member for supporting the leaf spring member. (57A, 57B) and driving members (111A to 11D, 112A to 112D) for driving the movable member in a direction in which the two leaf spring members have flexibility. It has. Further, the mechanical drive system includes a balancer (128) arranged so as to surround the movable member, and a balancer holding member (32A to 32A) which holds the movable member and the balancer in a relatively displaceable state. 32D, 32E, 32F,) and a balancer driving member (111A, 1111C, 129A, 129C) for driving the balancer. It is desirable that the balancer be vibrated in a phase opposite to that of the movable member so that the center of gravity (G) of the mechanical system including the movable member and the balancer is not substantially displaced.

With the second exposure apparatus of the present invention, the second exposure method of the present invention can be performed.

At this time, a first position detector (40, 41) for measuring the position of the movable member is provided, and the operation of the mechanical drive system is controlled according to the detection result of the first position detector. It is desirable. As a result, the control accuracy of the position of the movable member and the control accuracy of the transfer position on the second object are improved.

In addition, at least a part of a path for carrying the first object and the second object into the above-described exposure apparatus has a spare chamber in which the degree of vacuum inside can be controlled independently of other parts. Is desirably provided. Thereby, the exposure apparatus and, for example, a mechanism for exchanging the first object or the second object can be operated in parallel, so that the throughput is improved. Next, the method of manufacturing an exposure apparatus according to the present invention includes irradiating a first object (M) with a charged particle beam, and projecting a charged particle beam having passed through the pattern of the first object through a projection system (PL). In the method of manufacturing an exposure apparatus for irradiating (W), at least a part of the movable member (PL; 29) of the projection system is moved to move the irradiation position of the charged particle beam on the second object. A mechanical drive system (32 A, 32 B, 11 A, 1 1 2) that is attached to the support member (57A, 57B) in a state where it can be displaced or vibrated, and displaces or vibrates the movable member. A) is to be attached. According to such a manufacturing method, the exposure apparatus of the present invention can be manufactured efficiently.

Next, a device manufacturing method according to the present invention includes a step of transferring a device pattern onto a workpiece using the above-described exposure method of the present invention. According to this device manufacturing method, an extremely fine pattern can be transferred with high precision onto a workpiece such as a substrate such as a wafer or a glass substrate for a mask by using a charged particle beam. Further, by applying the mechanical MOL method of the present invention, the throughput at the time of exposure is improved. BRIEF DESCRIPTION OF THE FIGURES

FIG. 1 is a configuration diagram showing a cross-sectional view of a part of an electron beam reduction transfer device according to a first embodiment of the present invention. In FIG. 2, (A) is a plan view showing the pattern arrangement of the mask M in FIG. 1, (B) is an enlarged plan view showing a portion B in FIG. 2 (A), and (C) is a plan view in FIG. It is sectional drawing which follows CC line. 3A is a plan view showing a pattern arrangement of the wafer W in FIG. 1, and FIG. 3B is an enlarged plan view of a portion B in FIG. 3A. FIG. 4 is a cross-sectional view of a part of a state where the electron beam reduction transfer device of FIG. FIG. 5 is a partially cutaway plan view showing the mask stage 20 of FIG. 4 and members related thereto. Figure 6 shows the wafer stage 4 in Figure 4. FIG. 6 is a partially cutaway plan view showing 6 and members related thereto. Figure 7 is an enlarged perspective view showing an example of a mechanical drive system of the projection system PL in FIG. 1 _£ 8 is an explanatory diagram of the operation of the mechanical drive system of FIG. FIG. 9 is a partially cutaway plan view showing another example of the mechanical drive system of the projection system PL. FIG. 10 is an explanatory diagram in the case where the position of the reduced image is changed by displacing the projection system PL by the mechanical MOL method in FIG. FIG. 11 is an explanatory diagram of the case where the position of the reduced image is changed by changing the electromagnetic field of the electron optical system of the projection system PL by the electronic MOL method in FIG. 11. FIG. 12 is an explanatory diagram in the case where exposure is performed by the character pattern transfer method in the first embodiment. FIG. 13 is an explanatory diagram in the case of combining the operation of the mechanical MOL method with the operation of the electronic MOL method in the operation of FIG. FIG. 14 is a plan view showing a locus of a reduced image on a wafer when the projection system PL is vibrated in a triangular waveform with respect to time by a mechanical MOL method. FIG. 15 is an explanatory diagram in the case where exposure is performed by the division transfer method in the second embodiment of the present invention. FIG. 16 is an explanatory diagram in the case where some members of the projection system PL are displaced or vibrated by the mechanical MOL method in the third embodiment of the present invention. FIG. 17 is a perspective view showing a drive system when the projection system PL is displaced or vibrated by the counterbalance method and the mechanical M〇L method in the fourth embodiment of the present invention. . FIG. 18 is an explanatory diagram of the arrangement and operation of the drive elements of the drive system in FIG. FIG. 19 is a diagram illustrating an example of a manufacturing process of a semiconductor device. BEST MODE FOR CARRYING OUT THE INVENTION

Hereinafter, a first preferred embodiment of the present invention will be described with reference to FIGS. In this example, the present invention is applied to a reduction transfer apparatus using an electron beam as a charged particle beam in a character pattern transfer method, that is, an electron beam exposure apparatus. Applied.

First, FIG. 1 shows a schematic configuration of an electron beam reduction transfer apparatus of the present example as an exposure apparatus. In FIG. 1, an electron beam EB emitted from an electron gun 1 is converted into a parallel beam by a capacitor lens 2, and The light passes between the blanking deflectors 3 and is incident on an aperture plate 4 having an opening of a predetermined shape (square in this example). The electron beam EB whose cross-sectional shape has been shaped by passing through the aperture of the aperture plate 4 during transfer passes through the condenser lens system 5 including the first condenser lens 6A and the second condenser lens 6B, and then becomes a parallel beam again. Are deflected two-dimensionally by a deflector 7 for selecting a field of view composed of a first deflector 8A and a second deflector 8B, and guided into one character pattern on a mask M as a first object. .

An illumination system is composed of an electron gun 1, a condenser lens 2, an aperture plate 4, and a condenser lens system 5, and the optical axis AX 1 of the illumination system is perpendicular to the pattern surface of the mask M. In this example, the pattern surface almost matches the horizontal plane. In this case, as the electron gun 1, for example, a thermionic emission type lanthanum hexaborite (L a B ₆ ) or tantalum (T a) can be used. In addition, the electron gun 1 is used in a so-called low acceleration method in which the acceleration voltage is, for example, about several hundred V to 5 kV. Furthermore, the optical path of the electron beam EB is set to, for example, a high vacuum of about 1 0- ⁷ T orr. The mask M is arbitrarily irradiated with the electron beam EB by operating the blanking deflector 3 to remove the electron beam EB from the opening of the aperture plate 4 as shown by the dotted optical path 11. Can be interrupted. Further, the electron beam EB can be irradiated almost vertically on one character pattern located at a certain distance in the X direction and the Y direction from the optical axis AX 1 on the mask M by the deflector 7 for selecting the visual field. It is configured as follows. The opening and closing operation of the electron beam EB by the deflector 3 and the amount of deflection in the deflector 7 are A main control system 10 that controls and controls the operation is controlled via a deflection amount setting unit 9. Hereinafter, the Z-axis is taken parallel to the optical axis AX1 of the illumination system, the X-axis is set in the direction perpendicular to the paper of Fig. 1 in the plane perpendicular to the Z-axis, and the Y-axis is set in the direction A description will be given taking the axis.

The electron beam EB that has passed through one character pattern on the mask M is returned to the optical axis of the illumination system again by the swingback deflector 25 composed of the first deflector 26 A and the second deflector 26 B. Continue along AX 1. The deflection amount of the electron beam EB in the deflector 25 is set by the main control system 10 via a deflection amount setting unit 36. Each of the deflectors 8A and 8B and the deflectors 26A and 26B is, for example, a four-pole electrostatic deflector, but instead includes a coil mounted inside a ferrite core. Alternatively, an electromagnetic deflector may be used.

After that, the electron beam EB passes through a projection system PL composed of an imaging system to reduce the electron beam EB as a second object at a reduction magnification β (for example, 3 is 1/5, 1/10, or 120). A reduced image of one character pattern is formed in the exposed area (sub-exposure area) on W8. That is, the pattern surface of the mask Μ as the first object is located on the object surface of the projection system PL, and the surface of the wafer W as the second object (or object to be exposed) is located on the image surface of the projection system PL. I have. An electron beam resist is applied to the surface of the wafer W. The wafer W is a disk-shaped substrate such as a semiconductor (silicon or the like) or S 等 I (silicon on insulator).

As an example, the projection system PL of this example includes a front group 27 including a first lens 28A and a second lens 28B, and a rear group 29 including a first lens 30A and a second lens 30B. The front group 27 and the rear group 29 are housed in a cylindrical lens barrel 31. The lenses 28A and 28B and the lenses 30A and 30B are each an eight-pole electrostatic lens as an example, but an electromagnetic lens may be used instead. Then, in the projection system PL, the electron beam EB is once focused by the front group 27 to form a crossover (image of the electron beam source) 42, and thereafter, the rear group 29 forms a projection image on the wafer W. . The optical axis AX2 of the projection system PL is parallel to the optical axis AX1 of the illumination system.In the initial state, the optical axis AX2 of the projection system PL matches the optical axis AX1 of the illumination system and the lens barrel. 3 The axis of symmetry (center axis) of 1 also matches.

The lens barrel 31 in which the projection system PL is housed is supported by a pair of leaf springs 32A and 32B as members having flexibility in the Y direction. Both ends in the X direction of A and 32B are fixed to stably held support members 57A and 57B (not shown) (see FIG. 7). Therefore, the lens barrel 31 (projection system PL) is supported so that it can be displaced and vibrated within a predetermined range in the Y direction. A drive system 34 for displacing and vibrating the leaf springs 32A and 328 in the directions is provided, and the operation of the drive system 34 is controlled by a synchronous drive control system 33 under the control of the main control system 10. Has been done. Further, a movable mirror 40 is fixed to a side surface of the lens barrel 31 in the Y direction, and the movable mirror 40 is irradiated with a laser beam of a plurality of axes for measurement from a laser interferometer 41, and the laser interferometer 41 is not shown. The displacement of the lens barrel 31 in the Y direction, the rotation angle around the X axis, and the rotation angle around the Z axis are measured based on the position of the reference mirror of 34 and the alignment control system 39 described later. The drive system 34 moves the lens barrel 31 (projection system PL) in a state where the optical axis AX 2 is parallel to the optical axis AX 1 based on the measurement data of the laser interferometer 41 and the control information of the synchronous drive control system 33. Displace in the Y direction. That is, the projection system PL of this example is a mechanical M〇L (Moving Objective Lens) system using a mechanical drive system including the leaf springs 32A and 32B, the laser interferometer 41, and the drive system 34. , So that it can be displaced and vibrated in the Y direction. Therefore, in this example, the projection system Ρ However, the whole is a movable member.

In this case, the projection system PL is, for example, a small electron optical system in which the distance from the object plane (the pattern plane of the mask M) to the image plane (the surface of the wafer W) is about 100 to 150 mm. The MOL method enables high-speed, high-precision displacement and vibration.

In the projection system PL of this example, the distribution of the internal electromagnetic field (in this example, the electric field) is indicated by dotted virtual lenses 44 and 45, using the electronic M〇L (Moving Objective Lens) method. Thus, the optical axis AX2 can be displaced within a predetermined range in the X direction and the Y direction with respect to the symmetry axis of the lens barrel 31. However, with respect to the displacement of the projection system PL (optical axis AX2) by the mechanical MOL method, the displacement of the optical axis AX2 by the electronic MOL method with the aberration kept within the allowable range is, for example, In this example, when the projection system PL is displaced or oscillated relatively large by the mechanical M〇L method, the transfer position on the wafer W is corrected. The optical axis AX2 is displaced by a typical MOL method. Therefore, the lenses 28 A and 28 B and the lenses 3 OA and 30 B as the electron optical systems constituting the projection system PL are respectively driven by the lens driving system 35 as the electronic driving system, and the lens driving system 35 is It is controlled by the synchronous drive control system 33.

Next, the mask M is held on the mask stage 20 in parallel with the XY plane by electrostatic attraction so as not to affect the electron beam, and the mask stage 20 is placed on the mask base 21 by, for example, a linear motor. It is configured to be able to move stepwise in the X and Y directions. Since the exposure unit of the transfer device of this example is placed in a vacuum, the mask stage 20 is an air-bearing type that has almost no effect on the vacuum space, or a magnetic levitation type bearing that strictly shields the electron beam. Almost non-contact state On the mask base 21.

On the mask M of this example, a group of a plurality of character patterns in each of which a plurality of character patterns are arranged close to each other (described later in detail), is first used by moving the mask stage 20. After the center of the character pattern group is positioned approximately in the vicinity of the optical axis AX1, a deflector 7 for selecting a visual field is used to select one character and one pattern from the character and pattern group.

A moving mirror 22 fixed to the mask stage 20 (actually composed of an X-axis and a Y-axis) is irradiated with a multi-axis laser beam from the laser interferometer 23 on the mask side. Measures the position of the mask stage 20 in the X and Y directions, the rotation angle around the X axis, the rotation angle around the Y axis, and the rotation angle around the Z axis (jowing amount). Alignment control system 3

9 and the mask stage control system 24. The alignment control system 39 supplies measured values and alignment information to the main control system 10, and the mask stage control system 24 supplies a mask stage based on the measured values and control information from the main control system 10. 20 (Mask M) is made parallel to the XY plane, and its position, speed, and jogging amount are controlled.

Further, a masked backscattered electron detector 37 for detecting backscattered electrons and the like from the mask M is disposed diagonally above the mask stage 20, and a detection signal of the backscattered electron detector 37 is supplied to an alignment control system 3. Supplied to 9. Since the state in which the electron beam EB is irradiated on the predetermined alignment mark on the mask M can be detected by the reflection electron detector 37, the mask M can be aligned based on the detection result.

On the other hand, the wafer W is held on the wafer stage 46 in parallel with the XY plane by an electrostatic chuck or the like that does not affect the electron beam via a wafer holder (not shown). The wafer stage 4 6 is the wafer base 4 7 Above, for example, it is configured to be able to continuously move in the X direction by a linear motion method, and to be able to move stepwise in the X direction and the Y direction. The wafer stage 46 is mounted on the wafer base 47 almost in a non-contact state by an air-bearing system that has almost no effect on the vacuum space or a magnetic levitation type bearing system that strictly shields the electron beam. Have been. Further, the wafer stage 46 also incorporates a Z leveling mechanism for controlling the position (focus position) of the surface of the wafer W in the Z direction and its two-dimensional inclination angle. Information on the focus position at multiple measurement points on the surface of the wafer W Force is measured by an optical, multipoint autofocus sensor (AF sensor) (not shown), and Z leveling in the wafer stage 46 during exposure. The mechanism focuses the surface of the wafer W on the image plane of the projection system PL by an autofocus method based on the measurement value of the AF sensor.

A movable mirror 48 fixed to the side of the wafer stage 46 (also composed of X-axis and Y-axis) is irradiated with a laser beam of multiple axes from a laser interferometer 49 on the wafer side. The interferometer 49 is used to determine the position of the wafer stage 46 in the X and Y directions, the rotation angle around the X axis (the amount of rolling), the rotation angle around the Y axis (the amount of pitching), and the rotation around the Z axis. The angle (jowing amount) is measured, and the measured value is supplied to the alignment control system 39 and the wafer stage control system 50. The wafer stage control system 50 controls the measured value and the synchronous drive control system 33. Based on the information, the wafer stage 46 (wafer W) is made parallel to the χγ plane, and its position and amount of movement are controlled. A reflected electron detector 38 for the wafer for detecting backscattered electrons and the like from the wafer W is disposed obliquely above the wafer stage 46, and a detection signal of the backscattered electron detector 38 is used for the alignment control system 3. Supplied to 9. The reflected electron detector 38 can detect the state in which the electron beam EB irradiates a predetermined alignment mark on the wafer W, and based on this detection result, The wafer W can be aligned.

In FIG. 1, the exposure data such as the pattern configuration of the mask M to be exposed and the arrangement information of a plurality of shot areas on the wafer W are supplied to the main control system 10 from an exposure data storage device (not shown). Is done. Based on the exposure data, the main control system 10 sequentially irradiates the electron beam EB onto a plurality of character patterns selected from the mask M via the mask stage drive system 24 and the deflection amount setting units 9 and 24. At the same time, by scanning the wafer stage 46 and driving the projection system PL mechanically and electronically by the MOL system via the synchronous drive control system 33, a reduced image of the selected character and one pattern is formed on the wafer. It is transferred to the shot area to be exposed on W. Thus, a reduced image of the target pattern is transferred to each shot area on the wafer W.

Here, an example of the pattern arrangement of the mask M of the present example and an example of the shot arrangement on the wafer W will be described.

2 (A) is a plan view showing an example of a mask M to be transferred, FIG. 2 (B) is an enlarged view of a portion B in FIG. 2 (A), and FIG. 2 (C) is a CC in FIG. 2 (B). It is sectional drawing which follows a line. In FIG. 2A, a character pattern group 13 is formed at a predetermined pitch in the X direction and the Y direction in the pattern area of the mask M, and a pair of two-dimensional patterns is sandwiched between the pattern areas in the Y direction. Alignment marks 14A and 14B made of a metal film are formed. The mask M has a large number of character patterns formed inside a silicon wafer, and a notch portion 12 for angle detection is formed on an outer peripheral portion thereof. As shown in FIG. 2 (B), each character pattern group 13 on the mask M has different character patterns 15 A, 15 B, 15 C,... At different pitches in the X and Y directions. , 15 Y are formed.

As shown in Fig. 2 (C), the character pattern group in the mask M A portion 17 where 13 is formed is formed thinner than other regions, and a portion corresponding to the character pattern (a portion that transmits an electron beam) in the portion 17 is a hole. The portion 17 may be a thin film of silicon (Si), but the portion 17 may be a thin film of, for example, silicon nitride (SiN). As described above, the mask M of this example is a so-called perforated stencil mask, and the mask M is a so-called so-called stencil mask provided with a thin film for scattering electron beams such as tungsten (W) on a film that transmits electron beams. A scattering mask may be used.

In FIG. 2B, each character pattern 15A to 15Y is formed in a square area having a width D, and the boundary between adjacent character patterns is an electron beam. This is a region that does not transmit light or a region that scatters electron beams (non-pattern region). Further, the irradiation area 14 in FIG. 2 (B) by the electron beam EB in FIG. 1 is a square having a width slightly larger than D, and one character selected from the character pattern group 13 is selected. It is configured such that only the pattern (for example, 15 A) can be completely covered by the irradiation area 14. Also, a plurality of different character patterns are formed in different character pattern groups 13 respectively. In the case of the division transfer method, each of the character patterns 15A to 15Y corresponds to a “sub-field of view” which is a minimum unit pattern obtained by dividing a large original pattern.

Further, at the four corners of the character pattern group 13, two-dimensional (here, cross-shaped) alignment marks 16 A to 16 D made of a metal film that reflects an electron beam are formed. In this case, for example, in FIG. 1, after driving the mask stage 20 to sequentially move the alignment marks 14 A and 14 B of FIG. 2A on the mask M substantially in the vicinity of the optical axis AX 1 By driving the deflector 7 and scanning the electron beam EB in the X and Y directions, By processing the detection signal, the coordinates of the alignment marks 14A and 14B are detected, and the center coordinate and the rotation error of the mask M are calculated from the detection result. Thereafter, for example, by rotating the mask stage 20 so as to cancel the rotation error, the array coordinates of each character and pattern group 13 on the mask M can be calculated.

Then, by driving the mask stage 20 based on the array coordinates calculated in this way, the center of the character pattern group 13 used for exposure in FIG. Almost positioned on the optical axis AX1. In this state, the deflector 7 is driven to scan the electron beam EB in the X and Y directions to process the detection signal from the backscattered electron detector 37, thereby obtaining the alignment mark 1 in FIG. 2 (B). Detect the coordinates of at least two alignment marks in 6A to 16D. Then, by processing the obtained coordinates of the two alignment marks, the optical axis AX1 of each of the character patterns 15A to 15Y in the character pattern group 13 is set as the origin. Can be calculated with high accuracy. By driving the deflector 7 of FIG. 1 based on the array coordinates calculated in this way, the electron beam irradiation area can be accurately placed on the desired character pattern in the character pattern 15A to 15Y. 14 can be moved, and the electron beam transmitted through the desired character pattern can be returned to the optical axis AX1 with high precision via the deflector 25 in FIG. That is, in this example, the arrangement coordinates of the character-to-character patterns 15A to 15Y in each character pattern group 13 are finally obtained based on the positions of the alignment marks 16A to 16D. Therefore, the positioning accuracy of the mask stage 20 in FIG. 1 does not need to be so high.

The width D of the outer shape of the character one pattern 15 A to 15 Y is, for example, about 100 m to l 00 / zm. In this example, each character one pattern group 1 3 Each character pattern of 5 rows x 5 columns is formed in each character pattern, but in each character pattern group 13, character patterns of about 5 rows x 5 columns to 20 rows x 20 columns are formed. be able to.

On the other hand, FIG. 3 (A) is a plan view showing the wafer W, FIG. 3 (B) is an enlarged view of a portion B in FIG. 3 (A), and FIG. A large number of shot areas 52 are arranged at predetermined pitches in the X and Y directions, and a circuit pattern for one die is transferred to each shot area 52. Note that a circuit pattern for a plurality of dies may be transferred to each shot area 52. In addition, a pair of two-dimensional metal film search alignment marks 53 A and 53 B are formed so as to sandwich the shot regions 52 in the Y direction. Has a notch 51 for angle detection.

As shown in FIG. 3 (B), each shot area 52 is a minimum unit exposure area to which a reduced image of one character pattern is transferred. Are arranged closely in the X and Y directions. Also, assuming that a reduced image of one character pattern by the projection system PL in Fig. 1 is a reduced image 43 in Fig. 3 (B), the width D of the outer shape of the character pattern and the projection magnification of the projection system PL are used. Each of the reduced image 43 and the sub-exposure area 54 is a square area of width 3D. Then, by moving the shot area 52 relative to the sequentially changing reduced image 43 in the X and Y directions, the character pattern corresponding to all the sub-exposure areas 54 of the shot area 52 is obtained. Is transferred.

Assuming that the size of each character pattern on the mask M is 500 / xm square and the projection magnification / 3 is 110, each sub-exposure area 54 in the shot area 52 on the wafer W is 50 The shot area 52 is, for example, a rectangular or square area having a side length of about 20 to 30 mm. In this example, The projection system PL is mechanically displaced or vibrated as described later in order to transfer a reduced image of the character pattern corresponding to all the sub-exposure regions 54 in the projection region 52 at a high throughput.

Here, the overall mechanical configuration of the electron beam reduction transfer apparatus of the present embodiment will be described with reference to FIGS.

FIG. 4 is a cross-sectional view showing the entirety of the transfer device of the present example. In FIG. 4, the transfer device of the present example has a box-shaped solid frame 60 capable of maintaining the inside in an airtight state. Are located in The frame 60 is housed in a larger chamber (not shown). Around the frame 60 in the chamber, for example, a dry air or the like, which is temperature-controlled at a substantially atmospheric pressure and is highly dust-proof, is provided. Is supplied. The frame 60 is mounted on the base member 65 via the vibration isolating tables 64 A and 64 B (actually arranged at three or four places). A particularly airtight exposure chamber 66, a mask-side decompression chamber 67 used when replacing the mask M, and a mask chamber communicating with outside air (gas inside the chamber surrounding the frame 60). It is divided into a storage room 68, a wafer-side decompression room 69 used when exchanging the wafer W, and a Jehachi storage room 70 communicating with the outside air. The decompression chambers 67 and 69 correspond to the spare chamber of the present invention. A plurality of exhaust holes are formed from the top to the bottom on the side wall surrounding the exposure chamber 66, and these exhaust holes are connected to the vacuum pump 63 via exhaust pipes 62A to 62E, respectively. internal exposure chamber 6 6 is maintained at a high vacuum of about 1 0- ⁷ T orr example by a vacuum pump 6 3 when.

Further, a partition plate 93 (see FIG. 5) is provided so as to surround the decompression chamber 67 on the mask side, and a boundary between the exposure chamber 66 and the decompression chamber 67 is opened and closed by a drive unit 77A. At the boundary between the decompression chamber 67 and the mask storage chamber 68, there is a shirt member that is opened and closed by the drive unit 77B. 7 6 B is installed. Similarly, a partition plate 94 (see FIG. 6) is provided so as to surround the decompression chamber 69 on the wafer side, and the boundary between the exposure chamber 66 and the decompression chamber 69 is opened and closed by a drive unit 87A. At the boundary between the decompression chamber 69 and the wafer storage room 70, a shirt evening member 86B that is opened and closed by a drive unit 87B is installed. ing. At the two openings of the partition walls of the decompression chambers 67 and 69, there are exhaust pipes 62F and 62G, respectively, which lead to a vacuum pump 63, and a solenoid valve that can be opened and closed to take in ambient gas as needed. A supply line with 788 and 88 is connected. Further, tables 79 and 89 are provided in the decompression chambers 67 and 69 for transferring a mask and a wafer, respectively.

Then, in the exposure room 66, the electron gun 1 is arranged in the sealed cover 61 above the frame 60, and the condenser lens 2, the condenser lens system 5, and the field selection The deflector 7 and the mask M are arranged. In addition, members such as the aperture plate 4 in FIG. 1 are not shown in FIG. 4 for easy understanding of the drawing. Further, the mask base 21 is fixed to the frame 60, and a coarse movement stage 73 for driving the mask stage 20 holding the mask M is arranged. The mask base 21 is driven by a pair of linear motors 74 A and 74 B in the Y direction with respect to the mask base 21. In FIG. 4 as well, the measurement system for the mask stage including the movable mirror 22 on the mask side and the laser interferometer 23 is actually composed of elements for two axes as shown in FIG. FIG. 5 is a plan view in which a part of a drive system and a measurement system of the mask stage 20 in FIG. 4 is shown in cross section. In FIG. 5, the side surfaces of the mask stage 20 in the X direction and the Y direction are shown. X-axis movable mirror 2 2 X and Y-axis movable mirror 2 2 Y are fixed, and reference mirrors 2 2 RX and 2 are placed on mask base 2 1 in parallel with movable mirrors 2 2 X and 2 2 Y, respectively. 2 RY is fixed. Then, from the laser interferometer 23 X on the X axis, through the window member 71 X provided on the frame 60, the moving mirror 22 X and the reference mirror 22 RX are moved along the X axis with respect to a plurality of axes, respectively. A single-axis laser beam is irradiated, and the Y-axis laser interferometer 23 Y is moved to the moving mirror 22 Y and the reference mirror 22 RY via the window member 71 Y provided on the frame 60. The laser interferometers 23 X and 23 Y are irradiated with laser beams of multiple axes and one axis, respectively, along with, and the moving mirrors 22 X and 22 Y are referenced with respect to the reference mirrors 22 RX and 22 RY. The X coordinate and Y coordinate of Y (mask stage 20) and the rotation angle around three axes are measured. Further, the mask stage 20 is disposed in the U-shaped coarse movement stage 73, and the coarse movement stage 73 is substantially parallel to the Y axis with respect to the mask base 21. It is driven in the Y direction by linear motors 74A and 74B along the simple guide surface 21a. A guide surface 73a substantially parallel to the X-axis is formed on the inner surface of the coarse movement stage 73, and a guide member 96 is slidably disposed along the guide surface 73a. The mask stage 20 is disposed via two actuating sections 95A and 95B that can be extended and contracted in the Y direction with respect to the guide member 96, respectively. Examples of the actuary 95A and 95B include, for example, a so-called EI core type non-contact actuator having an electromagnet having an E-shaped core and an I-shaped core. One-touch contact or a small linear motor can be used.

A mover fixed to the end in the + Y direction of the mask stage 20 and a stator fixed to one side of the coarse movement stage 73 constitute a linear motor 74 D. The linear motor 74 C is composed of a mover fixed to the guide member 96 and a stator fixed to the other side of the coarse movement stage 73. The guide member 96 and the mask stage 20 are integrally driven in the X direction with respect to the coarse movement stage 73 by 4C and 74D. Be moved. Further, by controlling the amount of expansion and contraction of the two-axis factories 95A and 95B, the rotation angle (jowing amount) of the mask stage 20 with respect to the coarse movement stage 73 (mask base 21) is controlled. Can be controlled. The positional relationship between the mask and the table 79 in the decompression chamber 67 is maintained close to the boundary between the decompression chamber 67 and the mask stage 20 in the exposure chamber 66. A robot arm 75 for transferring the mask up to the mask is provided.

In the mask storage chamber 68 on the right side of the decompression chamber 67, a rotary table 81 for temporarily mounting a mask is installed, and an imaging type position detecting device 82A around the rotary table 81. ~ 82 C has been installed. Further, a mask library 97 is installed near the opening communicating with the outside air of the mask storage room 68, and a plurality of types of character patterns different from the plurality of character patterns formed on the mask M are provided in the mask library 97. A plurality of masks on which a character pattern group is formed (the uppermost mask MN appears) are stored. A robot arm 80 for transferring a mask between the mask library 97, the rotary table 81, and the table 79 in the decompression chamber 67 while maintaining the positional relationship of the mask in a predetermined state is installed. Have been. A mask loader system is composed of the robot arm 75, the table 79, the mouth pot 80, and the like.

In this case, the position of the notch at the outer periphery of the mask placed at the position P2 on the rotary table 81 taken out of the mask library 97 and the other two edges are detected by the position detecting device 82A. By detecting at up to 82 C, it is possible to detect a rotation error and a two-dimensional position error based on the notch of the mask. Then, the rotary table 81 is rotated so as to cancel the rotation error, and the mask is moved through the robot arm 80 so as to cancel the two-dimensional position error. one After placing the mask on the table 7 9 on the table 7 9 via the robot arm 75 in the exposure chamber 6 6 so that the rotation angle and the like do not change, By mounting the mask on the substrate, the mask to be exposed can be aligned.

Returning to FIG. 4, when carrying the mask from the position P 2 on the rotary table 8 1 in the mask storage chamber 6 8 to the position P 1 on the table 7 9 in the decompression chamber 67, the shirt evening member 7 6A is closed, the shirt member 76B is opened, and the shirt member 76B and the electromagnetic valve 78 are closed with the mask placed at the position P1. Then, after evacuation is performed through the exhaust pipe 62F and the inside of the decompression chamber 67 is evacuated, the shirt member 76A is opened and the mask is placed on the mask stage 20 from the position P1. Is carried in. On the other hand, after the exposed mask M of the mask stage 20 is carried out onto the table 79 in the decompression chamber 67, the shirt members 76A and 76B are closed and the electromagnetic valve 78 is turned on. After being opened, the inside of the decompression chamber 67 is brought to the atmospheric pressure, the shirt member 76B is opened, and the mask on the table 79 is returned to the mask library 97 in the mask storage chamber 68 (see FIG. 5). Thus, in this example, since the decompression chamber 67 is provided, the mask can be quickly exchanged between the vacuum exposure chamber 66 and the atmospheric pressure mask storage chamber 68.

Next, a deflector 25 for rewinding, a projection system PL and a wafer W are sequentially arranged below the mask base 21 in the exposure chamber 66, and the projection system PL is connected to the lens barrel 31 and the leaf spring 3 2. The supporting members 57 A and 57 B, which are held via A and 32 B (see FIG. 1), are stably fixed to the frame 60, and the wafer W is placed on the wafer stage 46 via the wafer holder 98. The wafer stage 46 is mounted on a wafer base 47 so as to be movable two-dimensionally. The wafer base 47 is fixed on the bottom surface of the frame 60, and a coarse movement stage 83 for driving the wafer stage 46 is mounted on the wafer base 47. The coarse movement stage 83 is arranged and driven by a pair of linear motors 84 A and 84 B in the Y direction with respect to the wafer base 47. In FIG. 4, the measurement system for the wafer stage, which includes the moving mirror 48 on the wafer side and the laser interferometer 49, etc., is actually composed of elements for two axes as shown in FIG. ing.

FIG. 6 is a plan view in which a part of a drive system and a measurement system of the wafer stage 46 in FIG. 4 is shown in cross section. In FIG. 6, the side surfaces of the wafer stage 46 in the X direction and the Y direction are shown. The X-axis movable mirror 48 and the X- and Y-axis movable mirrors 48 Y are fixed, respectively, and the reference mirrors 48 RX and 48 RY are mounted on the wafer base 47 in parallel with the movable mirrors 48 X and 48 Y, respectively. Has been fixed. Then, from the laser interferometer 49 X on the X axis, via the window member 72 X provided on the frame 60, the moving mirror 48 X and the reference mirror 48 RX are respectively moved along the X axis to the plural axes and A single-axis laser beam is irradiated, and a Y-axis laser interferometer 49 Y is applied to a moving mirror 48 Y and a reference mirror 48 RY via a window member 72 Y provided in a frame 60. The laser interferometers 49 X and 49 Y are illuminated by a multi-axis and a single-axis laser beam, respectively.

The X- and Y-coordinates of the movable mirror 48 X, 48 Y (wafer stage 46) and the rotation angles around three axes are measured with reference to 8 RX and 48 RY. The wafer stage 46 is disposed in a U-shaped coarse movement stage 83 laterally, and the coarse movement stage 83 is a guide substantially parallel to the Y axis with respect to the wafer base 47. Driven in the Y direction by linear motors 84 A and 84 B along the drive surface 47 a. A guide surface 83 a substantially parallel to the X-axis is formed on the inner surface of the coarse movement stage 83, and a guide member 100 is slidably disposed along the guide surface 83 a. , Each of which is non-contact and extendable and contractible in the Y direction with respect to the guide member 100.

A wafer stage 46 is arranged via 9 9 B. As in the case of the mask stage 20 of FIG. 5, the wafer stage 46 and the guide member 100 of FIG. 6 are integrally moved by a pair of linear motors 84 C and 84 D. Driven in the X direction along guide surface 83a with respect to 83. Further, by controlling the amount of expansion and contraction of the two-axis actuators 99A and 99B, the rotation angle (jewing amount) of the wafer stage 46 with respect to the coarse movement stage 83 (the wafer base 47) is controlled. Can be controlled. Further, the positional relationship of the wafer is maintained in a predetermined state in the exposure chamber 66 near the boundary with the decompression chamber 69, and between the table 89 and the wafer stage 46 in the decompression chamber 69. A robot arm 85 that transfers wafers to the robot is installed.

Then, in the wafer storage room 70 on the right side of the decompression room 69, a rotary table 91 for temporarily mounting a wafer is installed, and around the rotary table 91, three image-sensing type position detecting devices are provided. 9 2 A to 92 C are installed. Further, a wafer cassette 101 is installed near the opening communicating with the outside air of the wafer storage room 70, and the exposed or unexposed wafer (the uppermost wafer WN appears in the wafer cassette 101). Is stored. Then, a wafer is transferred between the wafer cassette 101, the rotary table 91, and the table 89 in the decompression chamber 69 while maintaining a predetermined positional relationship of the wafer. Arm 90 is installed. The wafer loader system is composed of the mouth pot arm 85, the table 89, the robot arm 90, and the like.

And, like the mask storage room 68, also in the wafer storage room 70, the notch portion of the outer peripheral portion of the wafer taken out of the wafer cassette 101 and placed on the position P4 on the rotary table 91, and The positions of the other two edge portions are detected by the position detection devices 92A to 92C, and the rotation error and the two-dimensional position error are calculated. After that, the rotation error and position error are cancelled. After placing the wafer at a position P3 on the table 89 in the decompression chamber 69 via the robot arm 90 so that the wafer can be moved, the wafer on the table 89 is loaded into the opening in the exposure chamber 66. By placing the wafer on the wafer holder 98 via the pot arm 85, the wafer to be exposed can be realigned.

Referring back to FIG. 4, when carrying the wafer from the position P4 in the wafer storage chamber 70 to the position P3 in the decompression chamber 69, the shirt evening member 86A is closed and the shirt evening member 86B is moved. The shutter member 86B and the electromagnetic valve 88 are also closed in a state where the wafer is opened and the wafer is placed at the position P3. Then, after the inside of the decompression chamber 69 becomes a vacuum state, the shirt member 86 A is opened, and the wafer is loaded onto the wafer holder 98 from the position P 3. On the other hand, after the exposed wafer W on the wafer stage 46 is carried out onto the table 89 in the decompression chamber 69, the shirt members 86A and 86B are closed and the electromagnetic valve 88 is turned on. After opening the vacuum chamber 69 to atmospheric pressure, the shirt member 86B is opened and the wafer on the table 89 is returned to the wafer cassette 101 in the wafer storage chamber 70 (see Fig. 6). It is. As described above, in this example, since the decompression chamber 69 is provided, the wafer can be quickly exchanged between the vacuum exposure chamber 66 and the atmospheric pressure wafer storage chamber 70.

Note that the wafer opening system (the same applies to the mask loader system) may be divided into a load system and an unload system so as to shorten the wafer (mask) replacement time. Further, for example, in FIG. 6, at least two wafer stages 46 may be arranged to execute various operations such as load alignment of the next wafer in parallel with exposure of one wafer. . In order to transfer the mask pattern to the second and subsequent layers of each shot area on the wafer, a plurality of alignment marks on the wafer are detected by a mark detection system, and a pattern is detected based on the detection result. Sub-dew in each shot area on Yagami The light region and the corresponding reduced image of the sub-field on the mask are accurately aligned. In this embodiment, the mark detection system detects the mark to be inspected by an electron beam, but an optical mark detection system may be used instead.

Next, a specific configuration example of the drive system 34 that mechanically drives the projection system PL in the Y direction in FIG. 1 and an example of the operation of the mechanical MOL method and the electronic MOL method of this example will be described. This will be described with reference to FIGS.

FIG. 7 is an enlarged perspective view, partially cut away, showing an example of members from the field-of-view selection deflector 7 to the wafer W in FIG. 1. In FIG. 7, the character pattern to be exposed on the mask M is shown. The center of group 13 almost coincides with the optical axis AX1 of the illumination system. In addition, by detecting the positions of at least two alignment marks in the alignment marks 16A to 16D in FIG. 2 (B), the positions of the squares each having a width D in the character pattern group 13 are detected. It is assumed that the arrangement coordinates with the optical axis AX1 of the character pattern 15A to 15Y as the origin are determined with high precision.

In FIG. 7, for example, when a reduced image of the character pattern 15 N on the mask M is transferred into one shot area 52 on the wafer W, the irradiation area 14 of the electron beam EB is The electron beam EB that has moved on the character pattern 15 N and has passed through the character pattern 15 N is accurately moved on the optical axis AX 1 by the deflector 25 for swingback. In the state shown in FIG. 7, the optical axis AX1 of the illumination system coincides with the optical axis AX2 of the projection system PL. Then, the electron beam EB moved on the optical axis AX 1 is transformed into a plurality of widths d (= β · D) of the shot region 52 by the projection system PL including the lenses 28 A and 28 B. On one sub-exposure area of the square sub-exposure area 54, a reduced image 43 obtained by reducing the character pattern 15N at a projection magnification of 3 is formed.

Then, the projection system PL is housed in the lens barrel 31, and the lens barrel 31 is a pair of plates. The leaf springs 32A and 32B are held so as to be sandwiched in the Y direction, and both ends of the leaf springs 32A and 32B in the X direction are fixed to support members 57A and 57B. Therefore, the projection system PL is supported by the leaf springs 32A and 32B so as to be able to be displaced and vibrated in the Y direction. The extensible drive elements 111 A, 111 B, 112 A, and 112 B, each composed of a piezo element, are fixed to one surface of the leaf springs 32 A and 32 B in the Y direction by bonding or the like. The leaf springs 32A and 32B also have expandable and contractible piezo elements so as to face the drive elements 11A, 1118 and 1128, and 112B on the + Y direction surfaces, respectively. Driving elements 1 1 1C, 1 1 10 and 1 12 (:, 1 1 2D are fixed. Note that driving elements 1 1 1A to: L 1 1D, 1 12A to 1 12D are piezo elements. It is also possible to use an electrostrictive element or a magnetostrictive element other than the above. In this example, a control unit (not shown) separates one set of drive elements 11A, 11B, 12A, and 12B. By expanding and contracting a set of the driving elements 1 1 1 C, 1 1 1 D, 1 1 2 C, 1 1 2 D at a constant period while the phases are inverted, the projection system PL as a whole Y Vibrates in the direction parallel to the axis (Y direction). The mechanical vibration direction MV of the projection system PL is in the Y direction The drive elements 111 A to 111 D, 112 A to 112 D, and the control system are the drive system shown in Fig. 1. 34 is supported.

Fig. 8 shows the relationship between the expansion and contraction state of the driving elements 111A and L11D and 112A to 112D and the displacement of the projection system PL (barrel 31). As shown in), when the expansion and contraction amount of those driving elements is 0, the projection system PL is not displaced. On the other hand, as shown in FIG. 8 (B), one set of drive elements 1 1 1 A, 1 1 1 B, 1 1 2 A, 1 1 2 B shrinks and another set of drive elements 1 1 When 1 C, 1 1 1 D, 1 1 2 C and 1 1 2 D are extended, the leaf springs 32 A and 32 B are bent in the + Y direction, and the projection system PL is also displaced in the + Y direction (direction MVA). You. On the other hand, as shown in Fig. 8 (C), one set of driving elements 1 1 1A and 1 1 1 When B, 1 12 A, 1 12 B expands and another set of driving elements 1 1 1 C, 1 1 1 D, 1 1 2 C, 1 1 2 D contracts, leaf springs 32 A, 32 When B moves in the Y direction, the projection system PL is also displaced in one Y direction (direction MVB). In this way, the projection system PL can be oscillated in the Y direction by expanding and contracting the two driving elements at a constant period (constant frequency) with the phases inverted. Further, by controlling the maximum value of the amount of expansion and contraction of the two drive elements, the amplitude of the projection system PL during vibration can be controlled.

Further, in FIG. 7, a movable mirror 40 composed of a mirror having a reflecting surface substantially parallel to the ZX plane is fixed to a side surface in the Y direction of the lens barrel 31, and a reference mirror 40 R substantially parallel to the movable mirror 40. Is arranged. The reference mirror 4OR is fixed to a support member (not shown) so as not to be displaced relative to the support members 57A and 57B. Then, the laser interferometer 41 fixed to the supporting member irradiates the movable mirror 40 with a three-axis laser beam, and irradiates the reference mirror 40R with, for example, a one-axis laser beam. The displacement amount of the moving mirror 40 and the lens barrel 31 (projection system PL) in the Y direction, the rotation angle around the Z axis, and the rotation angle around the X axis with respect to the reference mirror 40R as a predetermined sample. It is measured with the grating. These measured values are continuously supplied to the control system of the driving elements 111A to 111D and 112A to 112D, and the control system controls the driving elements 111 based on the measured values. A ~: By controlling the amount of expansion and contraction of L11D, 112A ~ 112D, the lens barrel 31 and the projection system PL are rotated at a predetermined amplitude and at a predetermined frequency without rotation. Vibrates along the mechanical vibration direction MV.

Thus, in this example, the displacement and rotation angle of the projection system PL are actually measured by the laser interferometer 41, and the measured values are fed back to the drive system. Can be stably vibrated in the Y direction as a whole. The displacement of the projection system PL is monitored. For this purpose, an expansion / contraction measuring element such as a strain gauge is attached to at least one surface of each of the leaf springs 32A and 32B, and the expansion / contraction measuring element is used as the leaf spring 3A.

The amount of expansion and contraction of 2 A and 32 B may be directly measured. Further, in order to monitor the amount of displacement of the projection system PL, a non-contact capacitance sensor or an optical gap sensor may be used.

In the embodiment shown in FIG. 7, two pairs of drive elements 11 1A to 11 ID are attached to one leaf spring 32A (the same applies to 32B). Elements, for example, four pairs of drive elements in two rows in the Z direction

It may be attached to 32 A. Thus, the amplitude of projection system PL can be increased. For example, when the expansion and contraction characteristics of the driving element 111A are improved, a pair of driving elements or only one driving element is attached to one leaf spring 32A (the same applies to 32B). Just do it.

As described above, the mechanical drive system of the projection system PL of this example uses a telescopic drive element. In addition, for example, the projection system PL is configured using a voice coil motor or linear motor. It may be mechanically displaced.

Fig. 9 shows an example of the configuration of a mechanical drive system that drives the projection system PL using a voice coil motor system. In Fig. 9, the lens barrel 31 in which the projection system PL is housed is formed by leaf springs 32A and 32B. The leaf springs 32A and 32B are fixed to support members 57A and 57B. A cylindrical member 1 13 A, 1 13 B made of a non-magnetic material is fixed to the outer surface of each of the leaf springs 32 A, 32 B, and a coil is attached to the tip of the cylindrical member 1 13 A, 1 13 B. 1 1

4 A and 1 14 B are wound. In addition, permanent magnets 115A and 115B as magnets are inserted into the cylindrical members 113A and 113B in a non-contact state, and inserted into the permanent magnets 115A and 115B. Is a yoke with a cylindrical tip made of a magnetic material so as to sandwich the cylindrical members 113A and 113B in a non-contact state. A, 1 16 B are fixed. The permanent magnets 1 15 A, 1 15 B and the yoke 1 16 A, 1 16 B are fixed to a support member (not shown) so as not to be displaced relative to the support members 57 A, 57 B. .

The voice coil motor (VCM) is composed of coils 114A and 114B, permanent magnets 115A and 115B, and yokes 1116A and 116B. In this case, by passing a current through the coils 114A and 114B, the coils 114A and 114B move along the direction in which the leaf springs 32A and 32B have flexibility, that is, the mechanical vibration direction MV. A thrust consisting of Lorentz force is generated in the same direction. As a result, the leaf springs 32A and 32B are radiused in that direction, and the lens barrel 31 (projection system PL) is also displaced. At this time, for example, by supplying an alternating current having a predetermined frequency and a predetermined amplitude to the coils 114A and 114B, the projection system PL is vibrated at the predetermined frequency and the predetermined amplitude along the vibration direction MV. Can be done.

Also in this case, in order to monitor the displacement of the lens barrel 31 (projection system PL), the inner surfaces of the leaf springs 32A and 32B are covered with expansion and contraction measuring elements 127A and 127B, respectively, each composed of a strain gauge. By driving the voice coil motor using the measured values, the projection system PL can be vibrated with high accuracy. In the example shown in FIG. 9, a laser interferometer or the like may be used to monitor the displacement of the lens barrel 31. In particular, in the case of the voice coil motor system, the projection system PL can be stably displaced along the vibration direction MV by a desired amount with high accuracy and stopped.

Returning to FIG. 7, the basic operation of sequentially transferring one or a plurality of character patterns on the mask M into the shot area 52 on the wafer W in this example will be described. In this example, the projection system PL was moved along the Y direction by expanding and contracting the drive elements 111A to 111D and 112A to 112D to deflect the leaf springs 32A and 32B in the Y direction. Vibration direction Predetermined vibration at predetermined frequency in MV Vibrate by width. As a result, the characters on the mask M corresponding to the sub-exposure areas 54 in the row in the main exposure area 110 A having the width L in the Y direction in the shot area 52 on the wafer W, respectively. Transfer a reduced image of the pattern. The width L1 is, for example, about 10 to 30 times the width d of the sub-exposure area 54, but for simplicity in FIG. 7, the width L1 is set to 9 times the width d. I have.

FIG. 10 shows a state in which the projection system PL is vibrated in the Y direction. In FIG. 10, in FIG. 10A, the optical axis AX 2 of the projection system P is compared with the optical axis AX 1 of the illumination system. The reduced image of the character pattern 15 B on the mask M by the projection system PL is exposed on the sub-exposure area 54 A on the wafer W. Next, as shown in FIG. 10 (B), with the electron beam EB passing through the character-to-pattern 15 C on the mask M turned back onto the optical axis AX1, the projection system PL is moved in the direction MVA ( + Y direction) by Δγΐ.

In this case, assuming that the projection system PL of this example is an inverted projection system composed of the front group 27 and the rear group 29, the projection image / 3 (] 3 <1) and the + Y direction of the reduced image are used. The displacement amount ΔΥ 1 is as follows.

ΔΥ 1 = (1 + β) Δ γ ΐ (1)

Therefore, by setting the displacement amount ΔΥ1 to the width d (arrangement pitch) of the sub-exposure area 54 on the wafer W, the character is added to the sub-exposure area 54B adjacent to the sub-exposure area 54A on the wafer W. A reduced image of the pattern 15 C can be exposed.

Similarly, as shown in FIG. 10 (C), a reduced image of the character pattern 15A on the mask M is exposed on the sub-exposure area 54C adjacent to the sub-exposure area 54B on the wafer W. The displacement of the reduced image due to the displacement Δy2 of the projection system PL in the direction MVA with the electron beam EB transmitted through the character pattern 15 A turned back onto the optical axis AX1 (= (1 + 3) Δ y 2) Should be 2 · d.

However, as shown in FIGS. 10 (A), (B), and (C), the time points at which the displacement of the reduced image by the projection system PL becomes 0, ΔY1, ΔY2 are t1, t2, Assuming that the minimum value of the interval between t 3 and time points tl, t 2, and t 3 is Τ, when the sensitivity of the electron beam resist on the wafer W is high (the required exposure amount is small), it is almost at the time point tl. The deflector 3 for blanking shown in FIG. 1 is deactivated for the exposure time △ t 1 (where m 1 is considerably shorter than the spacing plate) around t, t 2, and t 3. By irradiating the EB, a reduced image of one character pattern corresponding to each of the different sub-exposure areas 54A, 54B, and 54C on the wafer W can be exposed. In this case, for example, when the sensitivity of the electron beam resist changes to some extent by the layer on the wafer W, the output of the electron gun 1 in FIG. 1, that is, the intensity of the electron beam EB is controlled according to the sensitivity. You may do so. That is, when the sensitivity of the electron beam resist is slightly low (appropriate exposure amount is slightly large), the intensity of the electron beam EB is increased, and when the sensitivity of the electron beam resist is slightly high (appropriate exposure amount is slightly small), the electron beam is increased. The strength of the EB may be reduced. This exposure amount control method is effective when the intensity of the electron beam EB can be controlled accurately and stably within a predetermined range, and thus the exposure amount can be easily controlled.

On the other hand, for example, when the sensitivity of the electron beam resist on the wafer W is low (the appropriate exposure amount is large), the exposure time 2 around the time points t1, t2, and t3 is the interval Shorter than Τ, but may approach the interval Τ. In this case, simply irradiating the electron beam EB onto the mask M for the exposure time △ t2, the reduced image on the wafer W moves in the Y direction, and is transferred to the sub-exposure areas 54A to 54C. Image resolution may be degraded. Thus, in this example, when the projection system PL is moving in the Y direction, the reduced images on the wafer W are stopped on the respective sub-exposure areas 54A to 54C, so that an electronic P 04707

39

The optical axis of the projection system PL is electronically displaced by applying the MOL method. Here, for convenience of explanation, the optical axis in the initial state of the projection system PL (before the optical axis is displaced by the electronic MOL method) is set to the optical axis AX2, and after being displaced by the electronic MOL method. The optical axis is called the optical axis AX3 (see Fig. 11).

Then, for example, assuming that a reduced image of the character pattern 15C is to be exposed on the sub-exposure area 54B on the wafer W for an exposure time Δt2 around the time point t2 in FIG. 10 (B). In (t 2-Δt 2 2), the electromagnetic fields of the front group 27 and the rear group 29 are indicated by dotted lines 27 A and 29 A in the electronic MOL method as shown in Fig. 11 (A). The optical axis AX3 is displaced in the + Y direction with respect to the optical axis AX2 in the initial state by changing from the initial state. As a result, the reduced image moves from the dotted line locus 117A to the solid line locus, and the reduced image is transferred to the sub-exposure area 54B. At this time, the relationship between the displacement <5 y of the optical axis AX3 and the displacement δY of the reduced image on the wafer W is also expressed by the following equation similar to the equation (1).

δ Υ = (1 + β) 6 y (2)

Therefore, if the displacement between the optical axis AX2 and the center of the sub-exposure area 54 (exposure target area) is <5Ya, the displacement of the optical axis AX3 by the electronic M〇L method 6 y can be expressed by the following equation.

6 y = 6 Ya / (1 + ^ 3) (3)

Thereafter, the displacement of the optical axis AX3 with respect to the optical axis AX2 is gradually reduced, and in FIG. 11B showing the state at the time point t2, the optical axis AX3 is matched with the optical axis AX2. After that, since the optical axis AX2 shifts in the + Y direction with respect to the sub-exposure area 54, the front group 27 and the rear group 27 are formed by an electronic MOL method as shown in Fig. 11 (C). The electromagnetic field (optical axis AX3) of group 29 is displaced in one Y direction with respect to the initial state (optical axis AX2) indicated by dotted lines 27B and 29B. As a result, the reduced image moves from the dotted locus 1 17 B to the solid locus, The reduced image is transferred to the sub-exposure area 54B.

Thus, the optical axis AX3 determined by the electromagnetic field distribution of the projection system PL is displaced by the electronic M〇L method at the exposure time Δt2 around each time point t1, t2, t3, ... In this way, the reduced images of the corresponding character patterns are transferred with high resolution to the sub-exposure areas 54 A, 54 B, 54 C,... In a row along the mechanical vibration direction MV on the wafer W. Can be. If the exposure time Δt 2 becomes longer than the minimum interval T at each point due to a further decrease in the resist sensitivity, the mechanical oscillation frequency of the projection system PL is lowered to increase the interval T. You may. Conversely, when the resist sensitivity is high and the exposure time Δt 2 is short, the mechanical vibration frequency of the projection system PL may be increased. Further, the exposure amount control may be performed by combining the control of the mechanical vibration frequency of the projection system PL and the control of the intensity of the electron beam EB.

In this example, the position of the reduced image on the wafer W is also corrected in the X direction by an electronic MOL method as described later.

Returning to FIG. 7, as described above, in this example, the projection system PL is oscillated in the Y direction by the mechanical MOL method, and the position of the reduced image 43 is corrected by the electronic MOL method, thereby obtaining the width in the Y direction. A reduced image of the character pattern corresponding to the sub-exposure area 54 having a width d of one row arranged in the Y direction in the main exposure area 11 1 OA of L1 can be transferred. Then, in order to expose a reduced image of one character pattern corresponding to one row of the sub-exposure area 54 sequentially in the main exposure area 110A, in this example, the projection system PL is synchronized with the oscillation operation in the Y direction in the Y direction. Then, the wafer W is continuously moved at a predetermined speed in the + X direction orthogonal to the Y direction. That is, the wafer W is scanned in the + X direction relative to the projection system PL. Therefore, the scanning direction MSA of the wafer W is in the + X direction, and the scanning direction MSA is orthogonal to the mechanical vibration direction MV of the projection system PL.

In this case, the frequency at which the projection system PL is vibrated along the vibration direction MV Is _M0L and the period is T OL —— 丄 ^ / 丄 MOL). In this example, two rows of sub-exposure areas 54 are exposed during one period T _MOL. It is necessary to move the wafer W along the scanning direction MSA by 2 · d which is twice the width of the sub-exposure area 54. Accordingly, the scanning speed Vw in the running direction MSA of Jehwa W is as follows.

MOL (4)

Then, after the exposure of the main exposure area 11 OA is completed, the wafer W is moved stepwise in the Y direction by an interval L1, and then synchronized with the mechanical MOL method of oscillating the projection system PL in the Y direction. Then, by moving the wafer W in the −X direction, the reduced images of the corresponding character patterns are sequentially exposed in the adjacent main exposure area 110B. Hereinafter, the image of the target circuit pattern is transferred onto the entire surface of the shot area 52 by sequentially exposing the adjacent main exposure areas 110C,... While reversing the scanning direction of the wafer W. There is a monkey.

In this case, the width D of the outer shape of the character

Assuming that the projection magnification of the projection system PL is about 1 m and the projection magnification / 3 of the projection system PL is 1 to 10, the width d of the sub-exposure area 54 on the bottom surface is about 50 m. Furthermore, if the width L 1 of the main exposure area 11 OA in the Y direction is 20 times the width .d of the sub-exposure area 54, the width L 1 is about 2 mm (± lmm). Therefore, using equation (1), the minimum value of the amplitude when the projection system PL is vibrated in the Y direction by the mechanical MOL method is approximately L IZ (1 +) 3), that is, approximately 1.82 mm. . In this case, the width of the effective visual field on the object plane side of the projection system PL only needs to be approximately (D + L1Z (1 + β)) or more. That is, the effective visual field may be a circular area having a diameter of about 2.3 mm or more. In addition, in order to provide a margin, the amplitude of the projection system PL is set to about 1.2 to 1.6 times the minimum value, and is set within the range of the width L1 on the surface of the wafer. Only one character pattern reduction Small images may be transferred.

The frequency f _MOL when the projection system PL is vibrated in the Y direction by a mechanical MOL method is, for example, about 40 to 60 Hz. As a vibration method at this time, it is excellent in terms of stability to vibrate the projection system PL in a sinusoidal waveform with respect to time. However, from the viewpoint of facilitating control when the transfer position in the X direction is corrected by the electronic MOL method as described later, the projection system PL may be vibrated in a substantially triangular waveform with respect to time. Specifically, _assuming that the frequency f _M0L is 50 Hz (period T _MOL is 20 msec) and the width d of the sub-exposure area 54 is 50 _im, the scanning speed Vw of the wafer W in the + X direction is given by the following equation (4). SmmZs ec.

Next, FIGS. 12 to 13 show an example of an operation of transferring a predetermined pattern to one shot area on a wafer by the character-to-pattern transfer method using the reduction transfer apparatus of FIG. 1 of the present embodiment. It will be described with reference to FIG.

FIG. 12 (A) shows an example of a pattern arrangement of a mask M to be exposed. In FIG. 12 (A), a character pattern 15 A to 1 A in a character pattern group 13 A on the mask M is shown. Patterns represented by symbols A1 to Y1 are formed as 5Y, and symbols {2,...,... , A3,... And Α4,... Are formed. In this case, the electron beam irradiation area 14 is relatively moved along the trajectory 118 on the mask Μ, so that the reduced images of the patterns Β1, C1,... The main exposure area 11 having a width L1 and a length L5 in the shot area 52 on the wafer W of FIG. 12 (B) 11 is sequentially transferred to the sub-exposure areas 54A, 54B, 54C,. I do. Therefore, as described with reference to FIG. 7, as the first step, the projection system PL is vibrated in the vibration direction MV (Y direction) by the mechanical MOL method, and the reduced image 43 is vibrated. Synchronously, the wafer W is scanned in the scanning direction MSA along the + X direction. Assuming that the vibration of the projection system PL is sinusoidal with respect to time at this time, the relative locus of the center of the reduced image 43 with respect to the main exposure area 110 A on the wafer W is shown in FIG. The trajectory of the solid line of FIG. In FIGS. 12 to 14, for convenience of explanation, it is assumed that the width L1 of the main exposure area 110A in the Y direction is 10 times the width d of the sub-exposure areas 54A, 54B,. I have.

In FIG. 13 (A), the reduced image 43 by the projection system PL moves relative to the main exposure area 11 OA in a sinusoidal shape at a pitch 2 At tl, t2, t3,..., t20, the Y coordinate of the center of the reduced image 43 matches the Y coordinate of the center of the sub-exposure areas 54A, 54B, 54C,…, 54U, respectively. . At this time, if the reduced image 43 of the corresponding character pattern is exposed simply by irradiating an electron beam at the time points tl, t 2,…, the sub-exposure areas 54 Α, 54 Β,… in the X direction The position of the reduced image 43 is shifted. Therefore, in this example, as the second step, the projection system PL is changed by changing the electromagnetic field of the electron optical system of the projection system P using the electronic MOL method in synchronization with the operation of oscillating the projection system PL in the Υ direction. The optical axis (corresponding to the optical axis AX 3 in Fig. 11) is periodically displaced in the X direction (the direction orthogonal to the vibration direction), and the reduced image 43 is transferred as shown in Fig. 13 (B). Displace the position.

In FIG. 13 (B), the position of the lens barrel 31 of the projection system PL at the time points t1, t5, t10, and til is shown corresponding to FIG. 13 (A). In this case, at the time tl, the reduced image 43 is displaced in the X direction, and at the time t5, the reduced image 43 does not need to be displaced substantially, and at the time t10, the reduced image 43 needs to be displaced in the + X direction. There is. At time t11 following time t10, the contraction is performed to expose the sub-exposure area 54L adjacent in the X direction. It is necessary to displace small image 43 in one X direction. To summarize, as shown in Fig. 13 (C), the reduced image 43 is converted to the amplitude L 1 (= 1) by the mechanical MOL method that vibrates the lens barrel 31 of the projection system PL in the Y direction with the amplitude L4. Synchronized with the operation of oscillating at (1 + 3) L4), the reduced image 43 is periodically displaced in the X direction with almost the amplitude d in the X direction as shown by the locus 120 by the electronic M〇L method. That is all we need to do. The frequency of the vibration in the Y direction of the projection system P in the mechanical MOL method is equal to the frequency F _M0L of the vibration of the reduced image 43 in the X direction in the electronic MOL method, but the electronic MOL In the movement of the reduced image 43 by the method, as shown in Fig. 13 (C), when the projection system PL is located at the end, the reduced image 43 is almost instantaneously moved from the position 43B or from the position 43B. A must move in the X direction by width d in A. In other words, due to the electronic MOL method, the reduced image 43 has a sawtooth-like waveform with respect to time at a frequency of 2 • f MOL with almost the amplitude d in the X direction (direction perpendicular to the mechanical vibration direction of the projection system PL). It will vibrate. When the mechanical MOL vibration of the projection system PL and the electronic MOL vibration are combined, the reduced image 43 follows an approximately “8-shaped” locus on the image plane (wafer W). To move.

As described above, the projection system PL in the Y direction by the mechanical MOL method, the scanning of the main exposure area 11 OA on the wafer W in the X direction, and the X direction of the reduced image 43 by the electronic MOL method When the displacement is performed in synchronization with the main exposure area 11 OA, the relative locus of the center of the reduced image 43 with respect to the main exposure area 11 OA is

3 It becomes a rectangular wave like the locus 1 2 1 of the solid line in (D).

In Fig. 13 (D), the reduced image 43 by the projection system PL moves relative to the main exposure area 11 OA in the form of a rectangular wave at a pitch of 2 Reduced images at t 1, t 2, t 3,-, t 20 respectively

The center of 43 is in the center of the sub-exposure area 54 A, 54 B, 54 C,…, 54U Agree. At this time, by irradiating the electron beam at the points in time tl, t 2,..., The reduced images of the corresponding character patterns can be exposed in the sub-exposure areas 54 領域 to 54U. Also, if the sensitivity of the electron beam resist on the wafer W is low and the required exposure time Δt2 approaches the minimum value T of the interval between the time points tl, t2, ..., see Fig. 11 As described above, the transfer position of the reduced image 43 is also corrected in the Y direction by the electronic MOL method during the exposure time Δt 2 around the time points tl, t 2,. You can do it.

Note that when the projection system PL is vibrated in the Y direction in a triangular waveform with respect to time by the mechanical MOL method and the wafer W is scanned in the X direction, the main exposure area is not driven when the electronic MOL method is not driven. The relative locus of the center of the reduced image 43 with respect to 11 OA has a triangular wave shape as shown by the solid line locus 122 in FIG. In this case, since the correction amount of the reduced image 43 in the X direction by the electronic MOL method is expressed by a linear function with respect to time, the drive control of the electronic MOL method is easy.

Returning to FIG. 12, exposure is continued in the manner described with reference to FIG. 13, and in FIG. 12 (A), after the transfer of the character pattern group 13 A of the character pattern group 13 A of the mask M after the transfer of the character pattern V 1. When transferring the character pattern A 3 of the next character-pattern group 13 C, the mask stage 20 of FIG. 1 is step-moved in the X direction, so that the character-pattern group 13 C is transferred. Move the center to near the optical axis AX1 of the illumination system. Then, the electron beam EB is irradiated onto the character pattern A3 via the deflector 7 in FIG. Similarly, when transferring the character-pattern W1 after transferring the character-pattern A3, transferring the character-pattern U2 after transferring the character-pattern X1, and character-pattern U When transferring character-to-pattern A4 after transfer of 2, etc. 707

At step 46, the mask stage 20 moves the mask M stepwise.

As described above, in this example, when selecting a character pattern to be transferred from a large number of character patterns on the mask M, the character stage pattern group 13 A, 13 B,… is selected by the step movement of the mask stage 20. After that, since the deflector 7 for selecting the field of view is used, a character pattern to be transferred can be selected quickly and efficiently from a large number of characters and patterns.

As described above, when the exposure to the main exposure area 11 OA in FIG. 12B is completed, the wafer stage 46 in FIG. By moving the wafer W in the X direction (ie, the scanning direction MSB of the adjacent main exposure area 110B) in synchronization with the operation of vibrating the PL in the Y direction, the main exposure area 110 Exposure to B is performed, and thereafter, the wafer W is scanned in the opposite scanning directions MSA, MSB, and MSA alternately for each of the adjacent main exposure areas 110C, 110D, and 110E. Finally, an image of a circuit pattern of a predetermined layer is transferred onto the entire surface of the shot area 52. Similarly, the same or another circuit pattern image is transferred to other shot areas on the wafer W, respectively. Subsequently, the circuit pattern of the layer is formed through a development process of a resist on the surface of the wafer W and a pattern formation process such as etching and ion implantation.

At this time, in this example, the width L1 in the Y direction of the main exposure areas 110A, 110B, ... is about 10 to 30 times the width of one sub-exposure area 54A, 54B, ... Therefore, for example, the throughput of the exposure process is greatly improved to about 10 to 30 times as compared with the case where the wafer W is mechanically scanned and exposed for each row of the sub-exposure areas 54.

In this regard, for example, the projection system PL is vibrated by the mechanical MOL method. Without exposure, as shown in Fig. 12 (C), exposure can be performed by vibrating the reduced image 43 in the Y direction with the width d of one sub-exposure area by the electronic MOL method. Conceivable. In this case, each time the wafer W is scanned once in the X direction, it is possible to expose about two rows of sub-exposure areas. However, even in this example described with reference to FIG. The throughput is about 15 to about;

In the above-described embodiment, the projection system PL is vibrated by the mechanical MOL method, and at the time of scanning the wedge W, for example, the locus 11 1 of the center of the reduced image 43 in FIG. The optical axis of the projection system PL is displaced by an electronic MOL method in order to align 9 with the center of a series of sub-exposure areas 54 A, 54 B, 54 C,. An irradiation position correction deflector similar to the deflectors 7 and 25 is arranged between the projection system PL and the wafer W, and the position of the reduced image 43 is corrected by the irradiation position correction deflector. You may. Further, instead of separately providing a deflector for correcting the irradiation position, the position of the electron beam EB incident on the projection system PL may be displaced from the optical axis AX1 by, for example, a deflector 25 for rewinding.

Further, in the above embodiment, the projection system PL is vibrated. For example, when a test pattern is transferred onto the wafer W, the corresponding patterns are sequentially shifted while gradually displacing the projection system PL in the Y direction. It may be transcribed. Further, in the above embodiment, the projection system PL is displaced or vibrated one-dimensionally. However, a support mechanism that supports the projection system PL so as to be freely displaceable in two directions orthogonal to a plane perpendicular to the optical axis is provided. The projection system PL may be displaced or vibrated two-dimensionally in the X direction and the Y direction. When the projection system PL is displaced or vibrated two-dimensionally in this way, a measurement system (for example, a laser interferometer) that can detect the position of the projection system PL in the X and Y directions and the rotation angle around three axes ) Is desirable. Further, in order to support the projection system PL as a movable member so as to be displaceable, four portions of the projection system PL may be supported by leaf springs, and these leaf springs may be vibrated by telescopic drive elements. Alternatively, instead of the drive element, the actuator may be driven by an EI core type 3-axis or 4-axis actuator having an E-shaped electromagnet section and an I-shaped core section.

In the above embodiment, the position of the wafer stage 46 in FIG. 1 is measured with reference to the predetermined reference mirrors 22 RX and 22 RY, but the position of the wafer stage 46 is fixed to the projection system PL. The position of wafer stage 46 may be controlled based on the difference between the measured value of the position of projection system PL and the measured value of the position of wafer stage 46, with reference to the reference mirror obtained as a reference. Next, a second embodiment of the present invention will be described with reference to FIG. In the above-described first embodiment, the present invention is applied to the case where the exposure is performed by the character-to-pattern transfer method. In the second embodiment, the present invention is applied to the exposure by the split transfer method. This is applied when performing. Also in this embodiment, the exposure is performed by using the same electron beam reduction transfer device as the first embodiment in FIG.

FIG. 15 shows the mask Ml and the wafer W when performing the exposure by the division transfer method using the electron beam reduction transfer apparatus of FIG. 1, and as shown in FIG. 15 (A), the pattern of the mask M1 is shown. On the surface, the main fields of view 13A1, 13B1, 13C1, 13D1, ... are arranged at predetermined pitches in the X and Y directions, and each of the main fields of view 13A1, 13B1, ... In the sub-field of view 15 formed at predetermined pitches in the X and Y directions inside the pattern, a pattern (a code A, B, C, D is shown) in which a predetermined circuit pattern is enlarged. ) Is divided into 5 rows and 5 columns (the number of divisions in this case is, for example, about 5 rows and 5 columns to 20 rows and 20 columns) to form an original pattern. Also, as shown in FIG. 15 (B), an image of a circuit pattern for one or more dies on the wafer W is formed. The shot area 52A, which is an area to be exposed, is divided into a plurality of main exposure areas 125 A to 125 F in the Y direction, and each main exposure area 125 A to 125 F The exposure units 124 A, 124 B,… are divided in the X direction, and these exposure units 124 A, 124 B,… are each composed of 5 rows × 5 columns of sub-exposure areas 54 A 1. I have. A reduced image 43 of the original pattern in the corresponding sub-field of view 15 on the mask M1 is transferred to the sub-exposure area 54A1.

In this example, the mask Ml is placed on the mask stage 20 in FIG. 1, and the wafer W is placed on the wafer stage 46 in FIG. Then, by driving the mask stage 20 and the deflector 7 for selecting the field of view, in FIG. 15 (A), the irradiation area 14 of the electron beam EB is changed to the first sub-field of the main field of view 13 A1. , The second sub-field of view,..., And the 25th sub-field of view are sequentially moved, and each is irradiated with the electron beam EB for a predetermined exposure time.

Corresponding to this, in Fig. 1, the projection system PL is vibrated in the Y direction (mechanical vibration direction MV) by the mechanical MOL method, and the wafer W is moved in the + X direction (mechanical scanning direction MS A). As shown in Fig. 15 (B), 5 lines X in the main exposure area 125A on wafer W are scanned by correcting the irradiation position of the reduced image using the electronic MOL method. A reduced image of a circuit pattern represented by a symbol A is transferred to a first exposure unit 124A composed of five rows of sub-exposure areas 54A1.

Then, in FIG. 15 (A), the electron beam irradiation area 14 is regularly moved as the first sub-field of view, the second sub-field of the next main field of view 13 B 1 on the mask Ml, and so on. Then, the operation of oscillating the projection system PL in the Y direction by the mechanical MOL method and the operation of scanning the wafer W in the scanning direction MSA are performed as shown in Fig. 15 (B). , The second, third,... Exposed units 124 B, 124 C,... In the main exposure area 125 A are denoted by B, C,. The reduced image of the circuit pattern represented by is transferred. After the exposure to the main exposure area 125A is completed, the wafer W is alternately moved in the opposite scanning direction MSB, MSA to the adjacent main exposure areas 125B, 125C,. Exposure is performed by the division transfer method by scanning in the order of.

Thus, in this example, when performing the transfer by the division transfer method, the reduced image 43 (projection system PL) is vibrated in the Y direction by the mechanical M〇L method, and each main exposure area 125 A , 125 B,... In the Y direction can be increased, thereby improving the throughput of the exposure process. In the example of FIG. 15, the number of divisions of the enlarged pattern in the main field of view 13 A 1, 13 B 1,... On the mask Ml and the corresponding exposed units 124 A, 124 B on the wafer W ,… Have the same number of divisions in the sub-exposure area (5 rows x 5 columns), but the main field of view 13A1, 13B1,… , 124 B,... May be different from the number of divisions of the sub-exposure area. Furthermore, in the division transfer method, an overlapping portion (joint portion) is provided at the boundary of the adjacent sub-exposure area 54A1 on the wafer W, and a reduced image of the adjacent original pattern is overlapped at the overlapping portion. Exposure may be performed together. This reduces splice errors.

When the electron beam irradiation area 14 on the mask M 1 is switched from the main field of view 13 A 1 to the adjacent main field of view 13 B 1, for example, instead of or in addition to deflecting the electron beam, a mask stage 20 is used. May be moved. In this example, in particular, when the mask pattern to be transferred to each shot area on the wafer is simply divided into a matrix and divided patterns are formed in a number of sub-areas, the mask stage 20 and the wafer stage 46 are used. It will be moved similarly.

Next, a third embodiment of the present invention will be described with reference to FIG. In the first embodiment, the entire projection system PL is mechanically MOL-based. In the third embodiment, some members of the projection system PL are mechanically displaced or vibrated.

FIG. 16 is a schematic diagram showing members from the mask M to the wafer W of the present example corresponding to FIG. 10 and, in FIG. 16 (A), the character pattern 15 on the mask M is shown. The reduced image by the projection system PL of B is projected on the sub-exposure area 54A on the wafer W. The projection system PL in this example also includes the front group 27 and the rear group 29. For convenience of explanation, if the optical axis of the front group 27 is the optical axis AX2, the optical axis AX2 is the optical axis AX1 of the illumination system. Is matched. In this example, only the rear group 29 as a movable member is arranged so as to be able to displace and vibrate in a direction perpendicular to the optical axis AX2. That is, the sub lens barrel 126 holding the rear group 29 is held by, for example, the leaf springs 32A and 32B in FIG. 1, and the leaf springs 32A and 32B are displaced in the Y direction by the drive system 34 in FIG. And it can be vibrated.

In this case, as shown in FIG. 16 (B), when the reduced image of the character 15 C on the mask M is transferred onto the adjacent sub-exposure area 54 B on the wafer W, The electron beam transmitted through the character 1 pattern 15 C is returned to the optical axis AX 1 by the deflector 25, and the rear group 29 of the projection system PL is moved together with the sub-barrel 126 in the + Y direction (vibration direction MVA). Displace by Ay. As a result, the optical axis AX 28 of the rear group 29 is displaced by Ay in the + Y direction with respect to the optical axis AX 2 of the front group 27, so that the position of the reduced image is displaced by ΔY in the + Y direction. However, in this example, Δy and △ Y are equal as follows.

Δ Υ = Δγ (5)

Therefore, the reduced image of the character pattern 15C can be transferred to the sub-exposure area 54B by matching the displacement ΔΥ with the interval (d) between the sub-exposure areas 54 ° and 54 °.

Instead of using the rear group 29 as a movable member, the front group 27 is used as a movable member. It is also possible. However, if the projection system PL is a reduction system and the projection magnification 3. (/ 3 × 1) is used, the displacement AYF of the reduced image with respect to the displacement Ay F of the front group 27 is as follows.

Therefore, in order to displace the reduced image by AYF, it is necessary to displace the front group 27 more than AYFZβ, that is, AYF, so that the displacement amount increases and the configuration of the mechanical drive system becomes complicated. Further, in a reduced system such as the projection system PL, the rear group 29 is more likely to be smaller and lighter than the front group 27, so the mechanical drive system is designed by displacing the rear group 29. Also, there is an advantage that the manufacturing becomes easy.

Next, a fourth embodiment of the present invention will be described with reference to FIGS. In the first embodiment described above, the entire projection system PL is displaced or vibrated by the mechanical MOL method, but in the fourth embodiment, the displacement of the center of gravity during displacement or vibration is prevented. In order to carry out the council, components for the balance of the evening and evening are provided.

FIG. 17 shows a mechanical drive system of the projection system PL of the present example corresponding to FIG. 7. In FIG. 17, the projection system PL is housed in the lens barrel 31 and the lens barrel 3 1 Are held so as to be sandwiched in the Y direction by two pairs of leaf springs 32A, 328, and 32 (:, 32D. Both ends of the leaf springs 32A to 32D in the X direction are supported by supporting members 57A1, 57B1. A ring-shaped balance member 128 is arranged so as to surround the center of the lens barrel 31 in a non-contact state, and the balance member 128 is also moved in the Y direction by a pair of leaf springs 32E and 32F. And both ends of the leaf springs 32E and 32F in the X direction are also fixed to the support members 57A1 and 57B 1. Therefore, the projection system PL and the balance member 128 are formed by the leaf springs. It is supported so that it can perform relative displacement and relative vibration in the Y direction via 32 A to 32 F. In this case, the projection system PL and the lens barrel 3 1, and the weight of the member including the leaf springs 32A to 32D is substantially equal to the weight of the member including the balance member (counterhead) 128 and the leaf springs 32E and 32F.

Further, the displacement amount of the lens barrel 31 (projection system PL) in the Y direction is determined by the movable mirror 40 fixed to one Y-direction side surface of the lens barrel 31 (leaf spring 32A) and the laser interferometer 41. A rotation angle around the Z axis and a rotation angle around the X axis are measured at a predetermined sampling rate, and a movable mirror 40 G fixed to a side surface in the Y direction of the balance member 128 (leaf spring 32 E); With the laser interferometer 41G, the displacement amount of the balance member 128 in the Y direction, the rotation angle around the Z axis, and the rotation angle around the X axis are measured at a predetermined sampling rate. In this example, when the projection system PL is vibrated in the Y direction (mechanical vibration direction MV) based on these measured values, the balance member 128 is moved in the Y direction with the same amplitude and the opposite phase. Vibrate along. In other words, when the projection system PL is displaced by ΔYG1 in the + Y direction, the balance member 128 is displaced by AYG1 along the opposite Y direction. This means that the movable member (the projection system PL itself in this example) is vibrated and displaced together with the balance member (counter-mass) 128 while substantially satisfying the law of conservation of momentum. As a result, since the position of the center of gravity of the mechanical system including the projection system PL and the balance member 128 does not change, the projection system PL can be vibrated and displaced stably, and the accuracy of the transfer position on the wafer, etc. Is improved.

Fig. 18 shows an example of the mechanical drive member of Fig. 17. In Fig. 18 (A), the leaf spring 32A that holds the lens barrel 31 can be extended and contracted on the surface in the Y direction. The drive elements 11 A and 11 B are fixed, and the drive elements 11 C and 11 ID are also fixed to the + Y direction surface of the leaf spring 32A. Similarly, the driving elements are also provided in the leaf spring 32B and the leaf springs 32C and 32D in FIG. Fixed. The extensible drive elements 12 9 A and 12 9 B are also fixed to the surface of the leaf spring 3 2 E holding the balance member 1 2 8 in the —Y direction, and the + Y direction of the leaf spring 3 2 E is fixed. The expandable and contractible drive elements 129 / C and 1 / 29D are also fixed to the surface of each. A driving element is similarly fixed to the leaf spring 32F. Also in this example, as shown in FIG. 18 (B), the leaf springs 32A and 32B are bent in the vibration direction MV along the Y direction by controlling the amount of expansion and contraction of the drive elements. By bending the leaf springs 32E and 32F in the opposite vibration direction MG by the same amount, the position of the center of gravity G of the mechanical system including the projection system PL and the balance member 128 does not change.

In the embodiment shown in FIG. 17 as well, a voice filter, an EI core type actuator, a linear motor, etc. may be used as the mechanical drive system.

In each of the above embodiments, for example, as shown in FIGS. 12 (B) and 15 (B), while moving the wafer W in the X direction, a predetermined width (for example, After the reduced image of each sub-region of the mask is transferred to the region of L 1), the wafer W is transferred to the region on the wafer W adjacent to the region in the Y direction to transfer the reduced image of each sub-region to Y. When stepping in the direction, the velocity components of the wafer stage 46 in the X direction and the Y direction are not simultaneously set to 0, that is, the wafer W sequentially follows the U-shaped and inverted U-shaped trajectories. It is desirable to move the wafer so that it moves. By driving the wafer stage 46 in this way, it is possible to improve the throughput and reduce the vibration.

In particular, in the second embodiment described above, when the mask stage 20 is moved in order to move the irradiation area of the electron beam between the main visual fields of the mask M1, the mask stage 20 is also substantially moved. So that the velocity components of the mask stage 20 in the X and Y directions do not become 0 at the same time. By doing so, it is possible to improve the throughput and reduce the vibration.

Further, in each of the above embodiments, at least a part of the movable member of the projection system PL (barrel 31) is vibrated or moved, so that the movable member is inclined or rotated with the movement or the like. Sometimes. Therefore, during the scanning exposure of the wafer W, the rotation amount of the projection system PL around the X axis and the rotation amount around the Y axis are respectively measured by the measurement system including the laser interferometer 41 shown in FIG. 17, for example. It is desirable to measure constantly and continuously correct the relative positional relationship between the electron beam from the projection system PL and the wafer W based on the measurement result. This makes it possible to further improve the alignment (alignment) accuracy between the reduced image of the sub-field of view on the mask and the sub-exposure area on the wafer. The rotation amount of the projection system PL about the Z axis may be measured during the scanning exposure, and at least one of the mask and the wafer may be rotated based on the measurement result.

Further, in each of the above embodiments, the projection system PL has one axis, but in FIG. 1, a plurality of projection systems PL that can be driven by a mechanical MOL method are arranged at variable intervals in the Y direction. Is also good. In this case, a control mechanism that adjusts the interval between the projection axes PL of the plurality of axes to be, for example, an integral multiple of the pitch of the shot array on the wafer to be exposed is provided. The offset of the electron beam irradiation position of each projection system PL is input so as to cancel the difference (residual error), and exposure is performed in parallel using the projection systems PL using the same control system. This further improves the throughput. In this case, a large fiducial mark member may be used to measure the distance between the exposure centers of the multi-axis projection system PL, and one fiducial mark may be used relative to the plurality of projection systems PL. You may move.

Also, at least a part of the movable members of the projection system is driven by the mechanical MOL method. A movable stage (hereinafter referred to as a “MOL stage”) is provided, and a plurality of movable members of the projection system are installed on the MOL stage so as to be drivable, and these movable members are moved according to the pitch of the shot arrangement on the wafer. The distance between the members may be adjusted. In this mechanism, a plurality of movable members can be displaced or vibrated in parallel using one MOL stage, so that the mechanism of the drive system can be simplified.

Next, in the transfer device using a low-acceleration voltage electron beam (low-acceleration electron beam transfer device) according to the above embodiment, the shape of the electron beam on the wafer stage 46 and the mechanical MOL method are used. In order to measure the amplitude of the electron beam when oscillating the electron beam, the first part that generates a secondary electron beam by irradiating the electron beam onto the wafer stage 46 and the second part that does not generate a secondary electron beam Reference marks provided with the portions may be formed at predetermined intervals. In this case, the output of the secondary electron beam when the electron beam is displaced by the mechanical M〇L method is detected, and for example, the output is differentiated with respect to the position of the electron beam (differential calculation in the case of a digital signal). From the obtained signals, the position and shape of the electron beam can be calibrated.

Also, in a low-acceleration electron beam transfer apparatus, if a magnetic field exists near the mask stage or the wafer stage, the magnetic field may bend the electron beam and deteriorate the transfer accuracy. Therefore, for example, a reference plate for detecting the deflection error of the electron beam according to the magnetic field strength is installed on the wafer stage, and the deflection error of the electron beam is detected. The speed at which the electron beam is vibrated by the MOL method, the position of the wafer stage, and the like may be controlled based on a predetermined table.

Next, an example of a manufacturing process of a semiconductor device using the electron beam reduction transfer device of the above embodiment as an exposure device will be described with reference to FIG. FIG. 19 shows an example of a semiconductor device manufacturing process. In this FIG. 19, when manufacturing a semiconductor device, first, for example, a single crystal silicon ingot is sliced and polished to manufacture a wafer W. I do. At this time, a notch (notch or the like) serving as a reference for wafer alignment is provided on the outer periphery of the wafer W. In the next step ST1, for example, a metal film or an insulating film is deposited on the wafer W, and an electron beam resist is applied. In a subsequent step ST2, a large number of characters selected in a predetermined order from a plurality of character-one pattern groups 13 on the mask M by the character pattern transfer method using the electron beam reduction transfer device of FIG. The reduced images of the patterns are sequentially joined and transferred into one shot area (area where a circuit pattern for one or a plurality of dies is transferred) 52 on the wafer W. Then, similarly, the reduced images of many character patterns are sequentially transferred to other shot areas 52 on the wafer W in the same manner. At this time, since the projection system PL is vibrated using a mechanical MOL method, the throughput of the exposure process is extremely high. Thereafter, in step ST3, a pattern F is formed in each shot region 52 on the wafer W by performing development, etching (or ion implantation), and the like.

When exposing the next layer, first, in step ST4, for example, a metal film or an insulating film is deposited on the wafer W and an electron beam resist is applied. In step ST5, the electron beam shown in FIG. Using a reduction transfer device, reduced images of a number of character patterns selected from the mask M in the character pattern transfer method from the mask M in a different order from step ST2 are sequentially placed in one shot area 52 on the wafer W. Transcribe. Similarly, reduced images of a large number of character patterns are sequentially transferred to other shot areas 52 on the wafer W. At this time, the throughput of the exposure process is extremely high because the projection system PL is vibrated by using the mechanical MOL method. afterwards, In step ST6, a pattern G is formed in each shot region 52 on the wafer W by performing development, etching, and the like.

The above-described electron beam resist coating process to pattern forming process (steps ST1 to ST3 or steps ST4 to ST6) are repeated as many times as necessary to manufacture a desired semiconductor device (step ST7). Then, a semiconductor device SP as a product is manufactured through a dicing step (step ST8) for separating each chip CP on the wafer W (step ST8), a bonding step, a packaging step, and the like (step ST9). In this case, in this example, since the throughput of the exposure step using the electron beam transfer device is extremely high, a high-performance device can be produced with a high throughput as a whole.

Further, in the above embodiment, a mechanical M〇L method is used when manufacturing a semiconductor device. However, other devices such as an image pickup device (CCD, etc.), a liquid crystal display device, a plasma display, etc. The present invention can also be applied to the manufacture of devices, thin-film magnetic heads, micro machines, and the like. Further, the present invention is also applicable to the production of photomasks (reticles) for optical projection exposure apparatuses or other X-ray exposure apparatuses (including EUV exposure apparatuses), masks for electron beam transfer apparatuses, and the like. Can be applied. That is, in the case of manufacturing a photomask, for example, as a first step, a pattern obtained by enlarging an original pattern is drawn on a glass substrate by the exposure method of the above-described embodiment, and development and etching are performed. Manufacture reticles. Then, as a second step, the pattern of the master reticle may be transferred onto a glass substrate for a working reticle using an optical projection exposure apparatus.

Further, when manufacturing a mask for an X-ray exposure apparatus or an electron beam transfer apparatus, a wafer as a mask substrate is formed using the exposure method of the above embodiment. Then, a predetermined original pattern may be transferred.

Further, the electron beam reduction transfer apparatus as the exposure apparatus of the above-described embodiment includes an illumination system composed of a plurality of electron lenses and deflectors, and a projection system as a movable member incorporated in the transfer apparatus main body. Manufacturing, by performing symmetrical adjustments, attaching a reticle stage or wafer stage consisting of many mechanical parts to the transfer device body, connecting wiring and piping, and performing comprehensive adjustments (electrical adjustment, operation confirmation, etc.) can do. It is desirable to manufacture the transfer device in a clean room where the temperature and cleanliness are controlled.

In the above-described embodiment, the exposure is performed by the character-pattern transfer method or the split transfer method. However, the present invention also includes an electron beam transfer device using a variable shaped beam or a Gaussian beam type electron beam. It goes without saying that the present invention can be applied to a transfer device and the like. In this regard, in the above embodiment, the electron gun 1 is used in a low-acceleration system, but in addition to this, the electron gun is moderately accelerated (acceleration voltage of about 15 kV to 30 kV) or high. The present invention can also be applied to a case where acceleration is used (acceleration voltage is about 50 kV to about 100 kV).

Further, in the above embodiment, an electron beam is used as a charged particle beam, but the present invention can also be applied to a charged particle beam transfer device using an ion beam or the like as a charged particle beam.

It should be noted that the present invention is not limited to the above-described embodiment, and can take various configurations without departing from the gist of the present invention. In addition, all disclosures in Japanese Patent Application No. 111-210, filed on July 19, 1999, including the specification, claims, drawings, and abstract, are as follows: , Which are hereby incorporated by reference in their entirety. Industrial applicability

According to the first exposure method of the present invention, since the projection system can be displaced by the mechanical MOL method, the projection system can be formed on the object (exposure target) without increasing the size of the projection system. This has the advantage that the pattern can be transferred to a wide area.

Further, according to the second exposure method of the present invention, since the projection system can be vibrated by the mechanical MOL method, there is an advantage that the throughput when performing exposure on an object to be exposed can be increased. is there.

Further, when the present invention is applied to the case of transferring an image of a small pattern sequentially selected from a mask onto an object to be exposed, the transfer speed can be increased, so that the throughput can be increased.

According to the exposure apparatus of the present invention, such an exposure method can be performed. Further, according to the device manufacturing method of the present invention, a high-performance device can be manufactured with high throughput.

Claims

The scope of the claims

1. An exposure method for irradiating a first object with a charged particle beam and irradiating a second object via a projection system with a charged particle beam having passed through the pattern of the first object,

Displacing at least some of the movable members in the projection system to irradiate a plurality of different positions on the second object with charged particle beams having passed through corresponding patterns on the first object, respectively. Characteristic exposure method.

2. The exposure method according to claim 1, wherein the second object is moved within a plane substantially perpendicular to the optical axis of the projection system, and further with respect to the projection system.

3. The irradiation position of the charged particle beam on the second object by electrically changing an electromagnetic field in the projection system in synchronization with an operation of moving the second object or an operation of displacing the movable member. 3. The exposure method according to claim 2, wherein the correction is performed.

4. A plurality of different patterns are formed on the first object, and the charged particle beam having passed through the pattern selected from the first object according to the irradiation position on the second object is projected. 4. The exposure method according to claim 1, 2 or 3, wherein the exposure method is directed to a system.

5. The rotation information of the movable member is detected, and an irradiation position of the charged particle beam on the second object is corrected based on the rotation information. Exposure method according to the item.

6. The movable member is displaceable in a first direction, and the second object is moved in a second direction substantially orthogonal to the first direction, and is larger than a moving distance of the movable member in the first direction. The exposure method according to any one of claims 1 to 5, wherein a pattern of the first object is transferred to an area on a second object.

7. An exposure method for irradiating a first object with a charged particle beam, and irradiating a second object with a charged particle beam having passed through the pattern of the first object via a projection system, comprising a plurality of different positions on the second object. An exposure method characterized by vibrating at least some of the movable members in the projection system in a predetermined direction in order to irradiate a charged particle beam having passed through a corresponding pattern on the first object.

8. Synchronize with vibrating the movable member along the first direction, move the second object in a second direction intersecting the first direction, and move the second object to a two-dimensional area on the second object. 8. The exposure method according to claim 7, further comprising irradiating a charged particle beam having passed through a corresponding pattern on the first object.

9. The exposure method according to claim 7, wherein the movable member is vibrated in a sine wave shape with respect to time in the first direction.

10. In synchronization with the operation of vibrating the movable member, the electromagnetic field in the projection system is electrically changed to correct the irradiation position of the charged particle beam on the second object. 10. The exposure method according to claim 8 or 9, wherein

11. The exposure method according to any one of claims 7 to 10, wherein a vibration frequency of the movable member is controlled according to a sensitivity of the second object.

12. The exposure method according to any one of claims 7 to 10, wherein the intensity of the charged particle beam is controlled according to the sensitivity of the second object.

13. A balancer is connected to the movable member via a flexible member, and when the movable member is vibrated, the center of gravity of a mechanical system including the movable member and the balancer is not displaced. The exposure method according to any one of claims 7 to 12, wherein the balancer is vibrated in a phase opposite to that of the movable member.

14. An exposure apparatus that irradiates a first object with a charged particle beam and irradiates a second object via a projection system with a charged particle beam having passed through the pattern of the first object, An exposure apparatus, comprising: a mechanical drive system for displacing at least a part of a movable member of the projection system in order to move an irradiation position of the charged particle beam on the second object.

15. A second object stage for moving the second object with respect to the projection system in a plane substantially perpendicular to the optical axis of the projection system. Exposure apparatus according to the above.

16. An electronic drive system for electrically changing an electromagnetic field in the projection system to correct an irradiation position of the charged particle beam on the second object;

16. The exposure according to claim 15, further comprising: a control system that controls an operation of the electronic drive system according to an operation of the second object stage and an operation of the mechanical drive system. apparatus.

17. A first object stage in which a plurality of different patterns are formed on the first object, and the first object stage moves the first object in a plane substantially perpendicular to the optical axis of the projection system.

A first deflector that irradiates the charged particle beam onto a pattern selected from the first object,

17. The exposure apparatus according to claim 14, further comprising: a second deflector for turning back a charged particle beam having passed through a pattern selected on the first object.

1 8. Provide a measurement system to detect the rotation information of the movable member,

The exposure according to any one of claims 14 to 17, wherein an irradiation position of the charged particle beam on the second object is corrected based on rotation information detected by the measurement system. apparatus.

19. The movable member is displaceable in a first direction, the second object is moved in a second direction substantially orthogonal to the first direction, and is larger than a moving distance of the movable member in the first direction. The area of the first object is The exposure apparatus according to any one of claims 14 to 18, wherein the pattern is transferred.

20. An exposure apparatus that irradiates a first object with a charged particle beam and irradiates a second object via a projection system with a charged particle beam having passed through the pattern of the first object, wherein the second object on the second object is An exposure apparatus comprising: a mechanical drive system that vibrates at least a part of a movable member of the projection system in order to move an irradiation position of a charged particle beam.

21. The mechanical drive system vibrates the movable member along a first direction.

Moving a second object in a second direction intersecting with the first direction to irradiate a two-dimensional area on the second object with a charged particle beam having passed through a corresponding pattern on the first object; 20. The exposure apparatus according to claim 20, further comprising an object stage.

22. In order to move the irradiation position of the charged particle beam on the second object, an electronic drive system for electrically changing an electromagnetic field in the projection system is provided, and an operation of vibrating the movable member is provided. And correcting the irradiation position of the charged particle beam on the second object via the electronic drive system in synchronization with the movement operation of the second object stage. 21 Exposure device as described in 1.

2 3. The mechanical drive system is

First and second leaf spring members arranged so as to sandwich the movable member and having mobility, respectively;

A support member for supporting the leaf spring member;

The exposure apparatus according to any one of claims 20 to 22, further comprising: a driving member that drives the movable member in a direction in which the two leaf spring members have flexibility.

24. The driving member includes a pair of expandable and contractible elements fixed on both surfaces of the first and second leaf spring members.

24. The exposure apparatus according to claim 23, wherein said pair of expandable and contractible elements expand and contract in opposite phases, respectively.

25. The projection system is a reduction projection system,

The exposure apparatus according to any one of claims 20 to 24, wherein the movable member is a part of the electron optical system on the side closer to the second object in the projection system.

26. The exposure apparatus according to any one of claims 20 to 24, wherein the movable member is the entirety of the projection system.

2 7. The mechanical drive system is

A balancer arranged to surround the movable member,

A balancer holding member for holding the movable member and the balancer in a state where they can be relatively displaced;

And a balancer driving member for driving the balancer.

The mechanical drive system vibrates the balancer in a phase opposite to that of the movable member so that the center of gravity of the mechanical system including the movable member and the balancer does not substantially displace. The exposure apparatus according to any one of 20 to 26.

28. A first position detector for measuring the position of the movable member,

The exposure according to any one of claims 20 to 27, wherein the operation of the mechanical drive system is controlled according to a detection result of the first position detector.

2 9. A first and a second position detector for measuring the positions of the movable member and the balancer, respectively.

The operation of the mechanical drive system is controlled according to the detection results of the two position detectors. 28. The exposure apparatus according to claim 27, wherein the exposure apparatus is controlled.

30. A spare chamber capable of controlling the degree of vacuum inside independently of other parts is provided at least in a part of a path for carrying the first object and the second object into the exposure apparatus. The exposure apparatus according to any one of claims 14 to 29, wherein:

31. A method for manufacturing an exposure apparatus, which irradiates a first object with a charged particle beam and irradiates a second object via a projection system with a charged particle beam having passed through the pattern of the first object,

In order to move the irradiation position of the charged particle beam on the second object, at least a part of the movable member of the projection system is attached to a support member in a displaceable or vibrable state,

A method for manufacturing an exposure apparatus, comprising: attaching a mechanical drive system for displacing or vibrating the movable member.

32. A second object stage for moving the second object with respect to the projection system in a plane substantially perpendicular to the optical axis of the projection system,

An electronic drive system for electrically changing an electromagnetic field in the projection system for moving an irradiation position of the charged particle beam on the second object is further provided. 2. The method for manufacturing the exposure apparatus according to 1.

33. A device manufacturing method including a step of transferring a device pattern onto a work piece using the exposure method according to any one of claims 1 to 13.