JP2007027350A - Exposure apparatus and method, and electronic device manufacturing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an exposure technique efficiently and inexpensively forming a fine pattern of having a wavelength not more than that of an illumination light on a region wider than an irradiation region of an illumination light in a still state. <P>SOLUTION: An interference fringe pattern 92 is formed in a narrow illumination region 42 by irradiating a wafer W with an illumination light via two diffraction gratings. By repeating scanning exposure in which the wafer W is moved in a Y direction for the illumination region 42 and an operation in which the wafer is moved stepwise in an X direction, the interference fringe pattern 92 is exposed on the entire surface of the wafer W. The cycle of the interference fringe pattern 92 is previously measured, and a moving quantity SX used when the wafer W is moved stepwise is set at the natural-number multiple of the cycle. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an exposure technique used in a fine pattern forming process in an electronic device manufacturing process such as a semiconductor integrated circuit, a flat panel display device, a thin film magnetic head, and a micromachine, and an electronic device manufacturing technique using the exposure technique.

  In forming a fine pattern in a manufacturing process of an electronic device such as a semiconductor integrated circuit, a photolithography technique is generally used. This is a coating process in which a photoresist (photosensitive thin film) is applied to the surface of a processed substrate such as a wafer, and illumination light for exposure having a light amount distribution corresponding to the shape of a pattern to be formed is irradiated onto the processed substrate. A desired pattern is formed on the substrate to be processed by an exposure process, a development process, an etching process, and the like.

  In an exposure process for manufacturing a state-of-the-art electronic device at present, a projection exposure method is mainly used as an exposure method. In this method, a pattern to be formed on a mask such as a reticle is enlarged by 4 times or 5 times, irradiated with illumination light for exposure, and the transmitted light is reduced using a reduction projection optical system. By exposing the substrate to be processed, the pattern is transferred onto the substrate to be processed.

The fineness of the pattern that can be formed by the projection exposure method is determined by the resolution of the reduction projection optical system, which is approximately equal to the value obtained by dividing the exposure wavelength by the numerical aperture (NA) of the projection optical system. Therefore, in order to form a finer circuit pattern, it is necessary to shorten the exposure wavelength or increase the NA of the projection optical system.
On the other hand, for example, as disclosed in Non-Patent Document 1 and Non-Patent Document 2, a plurality of diffracted lights generated by disposing a diffraction grating between a light source and a substrate to be processed and irradiating illumination light on the diffraction grating. Has also been proposed (hereinafter referred to as an interference exposure method) in which a fine pattern is formed on a substrate using a light / dark pattern formed by the interference fringes.
JM Carter et al: "Interference Lithography" http://snl.mit.edu/project_document/SNL-8.pdf Mark L. Schattenburg et al: "Grating Production Methods" http://snl.mit.edu/papers/presentations/2002/MLS-Con-X-2002-07-03.pdf

Of the conventional exposure methods described above, the projection exposure method requires a shorter wavelength exposure light source or a higher NA projection optical system in order to obtain higher resolution.
However, in the current state-of-the-art exposure apparatus, the exposure wavelength is shortened to 193 nm, and further shortening in the future is difficult from the viewpoint of usable lens materials. Further, the NA of the most advanced projection optical system has reached about 0.92, and it is difficult to increase the NA further, and this causes a significant increase in the manufacturing cost of the exposure apparatus.

  On the other hand, the interference exposure method of the conventional exposure methods has the advantage that, in principle, exposure of fine patterns below the exposure wavelength can be realized with a relatively inexpensive apparatus. However, in order to form interference fringes with a fine period with high accuracy, the irradiation area of the interference fringes may be considerably narrower than a substrate to be processed such as a wafer. Thus, even when the irradiation area of the interference fringe is narrow, it may be required to efficiently expose the interference fringe pattern on almost the entire surface of the substrate to be processed.

In addition, for example, in the case where an existing pattern formed on the substrate to be processed is overlaid with an interference fringe pattern, it is necessary to overlay the existing pattern and the interference fringe pattern with high accuracy. There is.
The present invention has been made in view of such problems, and a fine pattern, for example, a pattern having a line width less than or equal to the wavelength of illumination light, is cheaper and wider than the irradiation area of illumination light in a stationary state. It is a first object to provide an exposure technique that can be efficiently formed in a region.

It is a second object of the present invention to provide an exposure technique that can form such a fine pattern with high overlay accuracy.
Another object of the present invention is to provide an electronic device manufacturing technique using the exposure technique.

The following reference numerals with parentheses attached to some elements of the present invention correspond to the configuration of the embodiment of the present invention. However, each symbol is merely an example of the element, and the element of the present invention is not limited to the configuration of the embodiment.
The exposure method according to the present invention is an exposure method in which a pattern is exposed on a photosensitive substrate with illumination light from a light source (1), and the first diffraction grating (G11) and the second diffraction grating ( G21) is irradiated with the illumination light from the first diffraction grating side, and a predetermined illumination field (42) having a predetermined width in the first direction on the opposite side of the second diffraction grating from the first diffraction grating. ) To form an interference fringe (92) having periodicity in the first direction, and the interference fringes at two positions on the interference fringe separated by a first interval in the first direction. The second step of measuring the position in the first direction and determining the period of the interference fringe, and the relative fringe and the substrate are relatively scanned in the second direction orthogonal to the first direction, First exposing an exposure region on the substrate having the predetermined width in the first direction; The interference fringes and the substrate relative to each other in the first direction by a second interval that is a natural number multiple of the period of the interference fringes determined in the step and the second step and is narrower than the predetermined width. The interference fringes so as to overlap the exposure pattern formed by the interference fringes exposed on the substrate in the third step at least at the end in the first direction of the exposure region. And a fifth step of exposing the substrate with the interference fringes while relatively moving the substrate in the second direction.

  According to the present invention, by exposing the substrate a plurality of times such that the exposure pattern by the interference fringe and the end of the interference fringe overlap each other, the area is wider than the illumination light irradiation area in a stationary state. An interference fringe pattern can be formed efficiently while suppressing splicing errors.

In the present invention, as an example, the shape of the predetermined illumination field is a shape elongated in the first direction, and both end portions in the first direction are gradually narrowed toward the outside. Thereby, a change in the exposure amount at the joint can be suppressed.
As another example, the shape of the predetermined illumination field is an elongated shape in the first direction, and the amount of light at both ends in the first direction gradually decreases outward.

Moreover, you may further have the 6th process of correct | amending the period of the interference fringe calculated | required at the 2nd process before the 3rd process. Thereby, the overlay accuracy can be improved.
The sixth step includes a step of controlling the telecentricity of the illumination light irradiated on the first diffraction grating and a step of controlling the height of the substrate in order to correct the period of the interference fringes. At least one of them.

The second step uses the change in the amount of transmitted light from the measurement grating accompanying the relative movement of the interference fringe and the periodic measurement grating in the first direction in the first direction. A step of measuring the position of the.
In the third step and the fifth step, a liquid is applied to at least one of the gap between the second diffraction grating and the substrate and the gap between the first diffraction grating and the second diffraction grating. The interference fringe pattern may be exposed on the substrate in a filled state.

The method further includes a step of measuring pattern position information of a pattern formed in advance on the photosensitive substrate. In the sixth step, the pattern position information is used to correct the period of the interference fringes. Good.
Moreover, the manufacturing method of the electronic device by this invention is a manufacturing method of the electronic device which has a pattern formation process, Comprising: In the pattern formation process, the exposure method of this invention is used.

  Next, the exposure apparatus according to the present invention is generated by irradiating illumination light from the light source (1) to the first diffraction grating (G11) and the second diffraction grating (G21) arranged to face each other. An exposure apparatus for exposing a pattern of interference fringes having periodicity in one direction onto a photosensitive substrate, and an illumination optical system for irradiating the first diffraction grating with the illumination light from the light source (IS), a substrate holding mechanism (38) for holding the substrate, a period measuring mechanism (41) for measuring the period of the interference fringes in the first direction, and the interference fringes and the substrate holding mechanism. A stage mechanism (36A, 37A, 80A, 80B, 38) that relatively moves in a second direction orthogonal to the first direction and relatively moves the interference fringes and the substrate holding mechanism in the first direction. And by its periodic measurement mechanism And a control device (70) that drives the stage mechanism to move the interference fringe and the substrate holding mechanism relative to each other in the first direction based on the period of the interference fringe to be measured. is there. According to the present invention, the exposure method of the present invention can be used.

In the present invention, as an example, the periodic measurement mechanism is provided in the substrate holding mechanism and is periodically measured in the first direction (95A), and a photoelectric detector that detects transmitted light from the measurement grating ( 101).
Moreover, you may further provide the correction apparatus which correct | amends the period of the 1st direction of the interference fringe. The correction device includes a member (31a, 31b, 33a, 33b, 34a, 34b) that controls the telecentricity of the illumination light with respect to the first diffraction grating, and a height control device that controls the height of the substrate ( 38Z).

Further, as an example, it has a position measurement mechanism (43) that measures pattern position information of a pattern formed in advance on the substrate, and based on the measurement result of the position measurement device, the correction device uses the period of the interference fringes. May be corrected.
Further, a liquid filling mechanism (54, 55) for filling at least one of the gap between the first diffraction grating and the second diffraction grating and the gap between the second diffraction grating and the substrate is provided. You may have.

  According to the present invention, by exposing a substrate (substrate to be processed) with an interference fringe pattern formed through two diffraction gratings, a fine pattern such as a line width pattern less than or equal to the wavelength of illumination light is formed. It can be formed at low cost. In addition, by exposing the substrate a plurality of times so that the edge of the interference fringe and the end of the interference fringe overlap each other, the pattern of the interference fringe is wider than the illumination light irradiation area in a stationary state. Can be formed efficiently while suppressing splicing errors.

  Further, by correcting the period of the interference fringes, it is possible to improve the overlay accuracy when performing overlay exposure.

Hereinafter, an example of a preferred embodiment of the present invention will be described with reference to the drawings.
FIG. 1 is a view showing the schematic arrangement of the exposure apparatus of this example. Note that the XYZ coordinate system shown in FIG. 1 is the same as the coordinate systems shown in the following drawings. In FIG. 1, an ArF (argon / fluorine) excimer laser that outputs illumination light IL <b> 1 composed of a pulsed laser beam having a wavelength of 193 nm is used as the light source 1. As the light source 1, a KrF (krypton / fluorine) excimer laser with an oscillation wavelength of 248 nm, an F ₂ (fluorine dimer) laser with an oscillation wavelength of 157 nm, a harmonic laser using a wavelength conversion element, or the like can also be used.

The illumination light IL1 emitted from the light source 1 is converted into a parallel light beam (having a predetermined beam size) by the lenses 2, 3, 4, and 6 constituting the first lens group arranged along the first optical axis AX1. The illumination light IL2 is a parallel beam.
The illumination light IL2 becomes illumination light IL3 set in a predetermined polarization state by the polarization control element 9, and enters the condensing optical system 10 constituting a part of the illumination light uniformizing means. The illumination light IL5 emitted from the condensing optical system 10 is incident on a fly-eye lens 13 as an optical integrator (a homogenizer or a homogenizer) that constitutes a part of the illumination light uniformizing means. An aperture stop 17 is disposed near the exit side surface of the fly-eye lens 13 as necessary.

The details of the illumination light uniformizing means including the condensing optical system 10, the fly-eye lens 13, the aperture stop 17 and the like will be described later.
The illumination light IL7 emitted from the fly-eye lens 13 is incident on the lenses 19, 20, and 21 constituting the second lens group arranged along the second optical axis AX2, and is refracted by these lenses to illuminate. It becomes the light IL8 and reaches the field stop 22 (details will be described later).

  The illumination light transmitted through the field stop 22 is further refracted by the lenses 25, 26, and 27 constituting the third lens group disposed along the second optical axis AX2, and reaches the condensing point. The condensing point 28 is conjugated with one point on the exit-side surface of the fly-eye lens 13 via the second lens group (19, 20, 21) and the third lens group (25, 26, 27). (Imaging relationship).

  The illumination light IL9 that has passed through the condensing point 28 is further refracted by the lenses 29, 30, 32, and 35 constituting the fourth lens group to become the illumination light IL10 and is incident on the first translucent flat plate P1. . In addition, the optical member on the optical path of illumination light IL1-IL10 from the above 1st lens group (2,3,4,5) to 4th lens group (29,30,32,35) is included. The optical system is hereinafter referred to as illumination optical system IS. This illumination optical system IS can also be regarded as an illumination optical device in which the surface on which the first translucent flat plate P1 is disposed is a predetermined irradiation surface.

Hereinafter, the Z-axis is taken along the optical axis AX2 on the emission side of the illumination optical system IS so that the light source 1 side is in the + direction, and the X-axis is set in a direction parallel to the paper surface of FIG. A description will be given taking the Y axis in a direction perpendicular to the paper surface of FIG. In this example, the XY plane including the X axis and the Y axis is substantially parallel to the horizontal plane.
A second translucent flat plate P2 is provided below the first translucent flat plate P1 (−Z direction). The second translucent flat plate P2 is disposed so as to face a substrate W (hereinafter also referred to as a wafer as appropriate) such as a semiconductor wafer to be processed. A first diffraction grating, which will be described later, is formed on the first light-transmitting flat plate P1, and the diffracted light generated by irradiating the first diffraction grating with the illumination light IL10 is the second light-transmitting light. Is irradiated to the conductive flat plate P2. A second diffraction grating, which will be described later, is formed on the second light transmitting flat plate P2, and the second diffraction grating is irradiated with the diffracted light. Then, the diffracted light generated by the second diffraction grating is applied to the wafer W, and a light / dark pattern is formed on the wafer W by interference fringes composed of a plurality of diffracted lights.

  A photosensitive member PR such as a photoresist for exposing and recording the light and dark pattern is formed on the surface of the wafer W. That is, the wafer W can be regarded as a photosensitive substrate. The wafer W has a disk shape with a diameter of 300 mm as an example, and the second light-transmitting flat plate P2 has a disk shape with a diameter that covers the entire surface of the wafer W as an example. Similarly, the first light-transmitting flat plate P1 has a disk shape with a diameter that covers the entire surface of the second light-transmitting flat plate P2, as an example. However, as will be described later, the diameter of the second translucent flat plate P2 is desirably about 30 mm or more larger than the diameter of the wafer W. In this example, the light-transmitting flat plates P1 and P2 are disk-shaped, but the light-transmitting flat plates P1 and P2 may be rectangular flat plates that are large enough to cover the surface of the wafer W.

  The wafer W is held on a wafer stage 38 which is a substrate holding mechanism movable on the wafer base 50 in the X direction and the Y direction, and is movable in the X direction and the Y direction. A wafer stage system is configured to include a drive mechanism such as a wafer stage 38 and a linear motor (not shown). Further, the positions of the wafer W in the X direction and the Y direction are measured by the laser interferometer 40 through the position of the movable mirror 39 provided on the wafer stage 38. That is, the laser beam from the laser interferometer 40 is divided into a measurement beam and a reference beam by a beam splitter (for example, a polarization beam splitter) 90, the measurement beam is irradiated to the movable mirror 39, and the reference beam is column-structured via the mirror. The reference mirror 91 fixed to the body 75 (a reference frame serving as a position reference) is irradiated. Then, the interference light between the measurement beam from the movable mirror 39 and the reference beam from the reference mirror 91 is detected by the laser interferometer 40, so that the position of the wafer stage 38 in the X direction and Y direction with respect to the column structure 75. Can be measured with a resolution of about 1 to 0.1 nm, for example. The laser interferometer 40 actually represents a biaxial X-axis laser interferometer 40X1, 40X2 and a triaxial Y-axis laser interferometer 40Y1, 40Y2, 40Y3, as shown in FIG. The movable mirror 39 also represents an X-axis movable mirror 39X and a Y-axis movable mirror 39Y.

  In FIG. 1, a wafer mark detection mechanism 43 is disposed above the wafer base 50 in the + X direction. The wafer mark detection mechanism 43 is formed of, for example, an optical microscope and detects the position of an existing circuit pattern or alignment mark formed on the wafer W. The wafer W held on the wafer stage 38 is moved immediately below the wafer mark detection mechanism 43 before exposure as necessary, and the position of the pattern or mark on the wafer W is detected. The measurement beam of the Y-axis laser interferometer 40Y3 in FIG. 7 is set so as to pass through the detection center of the wafer mark detection mechanism 43 in parallel with the Y-axis, and the laser interferometer 40Y3 is, for example, when the wafer W is aligned. It is used to measure the Y coordinate of the wafer stage 38.

  In order to measure the height distribution of the surface of the wafer W, for example, a projection optical system that projects a slit image obliquely to a plurality of measurement points on the surface of the wafer W, and reflected light from the surface of the wafer W are received. And a light receiving optical system that re-images the plurality of slit images, and a predetermined reference plane (hereinafter referred to as a first reference surface) of the plurality of measurement points based on the lateral shift amount of the re-imaged slit images. An oblique incidence type wafer surface position detection system (not shown) that measures the amount of positional deviation in the Z direction from the surface) is disposed in the vicinity of the wafer mark detection mechanism 43, for example. A specific configuration of the oblique incidence type multi-point wafer surface position detection system is disclosed in, for example, Japanese Patent Application Laid-Open No. 05-129182. In the present example, the first reference plane is parallel to the XY plane.

  The wafer stage 38 of this example also incorporates a Z leveling mechanism 38Z that controls the position of the wafer W in the Z direction and the tilt angles (rotation angles) around the X and Y axes. Based on the measurement result, the Z leveling mechanism 38Z causes the average surface of the surface of the wafer W to coincide with the first reference surface when the wafer W is exposed. The Z leveling mechanism 38Z has, for example, a table portion for holding the wafer W, and the height of the table portion in the Z direction with respect to the base portion of the wafer stage 38 at three locations arranged near the apex of the regular triangle. And a drive member to be controlled.

  The translucent flat plates P1 and P2 have a substantially U-shaped cross section when viewed in the Y direction, respectively, and a first holder 36A and a second holder 37A each having a circular opening through which illumination light passes in the center. It is adsorbed and held by vacuum adsorption or electromagnetic adsorption. The translucent flat plate P2 is arranged parallel to the XY plane and facing the wafer W with a predetermined interval by the second holding drive mechanism including the holder 37A. Further, the translucent flat plate P1 is arranged in parallel to the XY plane and facing the second translucent flat plate P2 with a predetermined interval by the first holding drive mechanism including the holder 36A. Further, in this example, the portion that holds the translucent flat plate P1 of the first holder 36A is disposed so as to be housed in the portion that holds the translucent flat plate P2 of the second holder 37A in the X direction. ing. For this reason, for example, when it is necessary to replace the translucent flat plates P1 and P2, the holders 36A and 37A can be pulled out in the Y direction independently of each other.

  In FIG. 1, on the + X direction side of the optical axis AX2, a pair of elongated flat plate-like base members 75A and 75B that are parallel to the XY plane, parallel to the Y axis, and at a predetermined interval in the Z direction are disposed, and the optical axis AX2 A pair of base members 75C and 75D are arranged symmetrically with respect to the base members 75A and 75B and parallel to the Y axis. The upper surfaces of the base members 75A to 75D are each processed into a guide surface with good flatness. The column structure 75 is configured by fixing the ends in the + Y direction and the −Y direction of the base members 75A to 75D with connecting members (not shown).

  Further, both end portions in the X direction of the holder 36A for holding the first light transmitting flat plate P1 are movably mounted on the base members 75A and 75C via flat sliders 36B and 36C, respectively. The holder 36A and the sliders 36B and 36C are connected via three Z-axis actuators 81A capable of controlling the position of the holder 36A in the Z direction and two pins (not shown) for preventing relative rotation. . Accordingly, the holder 36A and the sliders 36B and 36C can be moved integrally in the X direction, the Y direction, and the rotational direction around the Z axis on the upper surfaces of the base members 75A and 75C, and the three Z-axis actuators 81A can By controlling the height, the position in the Z direction of the holder 36A (first translucent flat plate P1) with respect to the base members 75A and 75C and the rotation angles around the X axis and the Y axis can be adjusted.

  Further, the + X direction slider 36B is connected to the slider 77A via two X axis actuators 80A separated in the Y direction, and the slider 77A is driven along a Y axis guide fixed on the base member 75A. It is driven in the Y direction by a motor. By adjusting the drive amount of the two X-axis actuators 80A, the position in the X direction of the holder 36A (first translucent flat plate P1) relative to the base member 75A and the rotation angle around the Z axis are adjusted. Can do. Further, by driving the slider 77A in the Y direction, the holder 36A (first translucent flat plate P1) can be moved greatly in the Y direction in conjunction with it, and the position in the Y direction can be adjusted. .

As a result, the position of the holder 36 </ b> A (first translucent flat plate P <b> 1) in the X direction, Y direction, and Z direction with respect to the column structure 75, and the rotation angle of 6 around the X axis, Y axis, and Z axis are 6. The displacement of the degree of freedom can be adjusted, and the holder 36A (first translucent flat plate P1) is pulled out from the lower side of the illumination optical system IS in the + Y direction as necessary to replace the translucent flat plate P1. Can do.
Further, the position of the holder 36A (first translucent flat plate P1) in the X direction and the Y direction, and the rotation angle around the Z axis are determined by laser interference via the position of the movable mirror 84 provided on the holder 36A. It is measured by a total 82. That is, the laser beam from the laser interferometer 82 is split into a measurement beam and a reference beam by the beam splitter 83, the measurement beam is irradiated to the movable mirror 84, and the reference beam is passed through the mirror to the base 75C (column structure 75). The reference mirror 85 fixed to the beam is irradiated. Then, the interference light between the measurement beam from the movable mirror 84 and the reference beam from the reference mirror 85 is detected by the laser interferometer 82, whereby the positions of the holder 36A in the X direction and the Y direction are determined with the column structure 75 as a reference. For example, the measurement can be performed with a resolution of about 1 to 0.1 nm, and the rotation angle around the Z axis of the holder 36A can be measured from the difference between the positions in two Y directions (or in the X direction). For this reason, the laser interferometer 82 is actually composed of one axis in the X direction and two axes in the Y direction, and the movable mirror 84 is also one axis in the X direction and two axes in the Y direction. It is composed of

  Further, if necessary, a plurality of, for example, three laser beams are irradiated obliquely on the upper surface of the translucent flat plate P1 from a laser light source of the first Z position detection system (not shown) and reflected by the upper surface. By detecting the position of the laser beam with a photoelectric detector, the position of the upper surface of the translucent flat plate P1 in the Z direction with respect to a predetermined reference plane (hereinafter referred to as a second reference plane), The rotation angle around the X axis and the Y axis is measured. The second reference plane is parallel to the XY plane.

  As a result, the displacement of the holder 36A (first translucent flat plate P1) in the X direction, the Y direction, and the Z direction, and the six-degree-of-freedom displacement of the rotation angle around the X axis, the Y axis, and the Z axis are measured. By driving the three Z-axis actuators 81A, the two X-axis actuators 80A, and the Y-axis drive motor for the slider 77A based on the measured values, the holder 36A (first translucent flat plate P1) is driven. ) Is controlled to a target state.

  In FIG. 1, as in the case of the first holder 36A, both end portions in the X direction of the second holder 37A that holds the second translucent flat plate P2 are respectively base members via flat sliders 37B and 37C. It is movably mounted on 75B and 75D. The holder 37A and the sliders 37B and 37C are connected via three Z-axis actuators 81B that can control the position of the holder 37A in the Z direction and two pins (not shown) for preventing relative rotation. . The + X direction slider 37B is connected to the slider 77B via two X axis actuators 80B separated in the Y direction, and the slider 77B is driven along a Y axis guide fixed on the base member 75B. It is driven in the Y direction by a motor.

As a result, the position of the holder 37 </ b> A (second translucent flat plate P <b> 2) in the X direction, the Y direction, and the Z direction with respect to the column structure 75 and the rotation angle of 6 around the X axis, Y axis, and Z axis are 6. The displacement of the degree of freedom can be adjusted, and the holder 37A (second translucent flat plate P2) is pulled out from the lower side of the illumination optical system IS in the + Y direction as necessary to replace the translucent flat plate P2. Can do.
Further, the position of the holder 37A (second translucent flat plate P2) in the X and Y directions and the rotation angle around the Z axis are fixed to the base 75D (column structure 75) in the same manner as the holder 36A. The measurement is performed by the laser interferometer 86 through the position of the movable mirror 88 provided on the holder 37A with the reference mirror 89 as a reference.

  In addition to this configuration, for example, by fixing the reference mirror 89 of the laser interferometer 86 to the holder 36A, the X direction between the holder 36A and the holder 37A can be directly adjusted by the laser interferometer 86 without using the laser interferometer 82. The relative positional relationship in the Y direction may be measured. Further, for example, by fixing the reference mirror 91 of the laser interferometer 40 to the holder 37A, the X- and Y-direction relative of the wafer stage 38 and the holder 37A can be directly adjusted by the laser interferometer 86 without using the laser interferometer 86. A general positional relationship may be measured.

  Further, if necessary, a plurality of, for example, three laser beams are obliquely applied to the upper surface of the translucent flat plate P2 from a laser light source of a second Z position detection system (not shown) and reflected by the upper surface. By detecting the position of the laser beam with a photoelectric detector, the position of the upper surface of the translucent flat plate P2 in the Z direction with respect to a predetermined reference surface (hereinafter referred to as a third reference surface), The rotation angle around the X axis and the Y axis is measured. The third reference plane is parallel to the XY plane. In place of the first and second Z position detection systems, a detection system having the same configuration as the oblique incidence type multi-point wafer surface position detection system for measuring the Z position of the wafer W, or Z It is also possible to use a laser interferometer that measures the position in the direction.

  As a result, the displacement of the holder 37A (second translucent flat plate P2) in the X direction, the Y direction, and the Z direction, and the six-degree-of-freedom displacement of the rotation angles around the X, Y, and Z axes are measured. The Then, based on the measured values, the holder 37A (second translucent flat plate) is driven by driving the three Z-axis actuators 81B, the two X-axis actuators 80B, and the Y-axis drive motor for the slider 77B. The displacement of 6 degrees of freedom of P2) is controlled to a target state.

  In this example, each of the first, second, and third reference planes is parallel to the XY plane, and the interval between the first and second reference planes in the Z direction, and the third and first reference planes. The distance between the reference planes in the Z direction is stored in advance in a main control system that controls the operation of the entire apparatus. In this example, the surface of the wafer W is controlled to be parallel to the first reference plane at the time of exposure, and the position of the surface of the wafer W and the second translucent flat plate P2 (second diffraction grating). The relationship and the positional relationship between the first translucent flat plate P1 (first diffraction grating) and the second translucent flat plate P2 are controlled to be a predetermined relationship. Furthermore, as necessary, the surface of the wafer W is controlled so as to be shifted from the first reference plane by a predetermined distance in the Z direction.

Next, a bright and dark pattern of interference fringes formed on the wafer W by the exposure apparatus of this example will be described with reference to FIGS.
A one-dimensional phase modulation type diffraction grating G11 having periodicity in the X direction is formed on the + Z side of the first translucent flat plate P1 in FIG. A one-dimensional phase modulation type diffraction grating G21 having periodicity in the X direction is also formed on the surface of the second translucent flat plate P2 on the + Z side, that is, the first translucent flat plate P1 side. Yes.

First, these diffraction gratings G11 and G21 will be described with reference to FIG.
FIG. 2A is a view of the first light-transmitting flat plate P1 of FIG. 1 as viewed from the + Z side. The surface has a longitudinal direction in the Y direction and is one-dimensional in the X direction perpendicular thereto. A phase-modulation type first diffraction grating G11 having a long period T1 is formed.
The first diffraction grating G11 includes a surface portion of the first light-transmitting flat plate P1 and a dug portion in which the flat plate surface is dug by etching or the like as in a so-called chromeless phase shift reticle (in FIG. 2A). (Shaded part). The depth of the digging portion is set so that a phase difference of approximately 180 degrees is formed between the illumination light transmitted through the surface portion and the illumination light transmitted through the digging portion. When a phase difference of 180 degrees is formed in both illumination lights, the digging depth is set for the wavelength λ of the exposure light, the refractive index n of the first light-transmissive plate P1, and an arbitrary natural number m.
(2m−1) λ / (2 (n−1)) (1)
And it is sufficient.

Moreover, it is preferable that the ratio (duty ratio) of the width between the surface portion and the dug portion is approximately 1: 1. However, values different from 180 degrees and 1: 1 can be adopted for both the phase difference and the duty ratio.
FIG. 2B is a view of the second light-transmitting flat plate P2 of FIG. 1 viewed from the + Z side, and the surface (the surface on the first light-transmitting flat plate P1 side) is long in the Y direction. A second diffraction grating G21 having a direction and a one-dimensional period T2 in the X direction is formed. The structure of the second diffraction grating G21 is the same as that of the first diffraction grating G11 described above.

The first light-transmitting flat plate P1 and the second light-transmitting flat plate P2 are made of synthetic quartz or the like and have a high transmittance (translucency) to ultraviolet rays and a small thermal expansion coefficient (linear expansion coefficient). It is made of a material that has a small thermal deformation accompanying absorption. The thickness is preferably 5 mm or more, for example, in order to prevent deformation such as self-weight deformation. However, in order to further prevent the self-weight deformation and the like, the thickness may be 10 mm or more. In particular, when an F ₂ laser is used as the light source 1, it is preferable to use synthetic quartz to which fluorine is added.

  2A and 2B, the period T1 is represented as about 10% of the diameter (for example, 300 mm or more) of the first translucent flat plate P1 for convenience of explanation. T1 is about 240 nm, for example, and the period T2 is about 120 nm, for example, which is overwhelmingly smaller than the diameter of the first translucent flat plate P1. The same applies to each figure other than FIGS. 2 (A) and 2 (B).

Hereinafter, the principle of forming a bright / dark pattern of interference fringes on the wafer W by irradiating the first diffraction grating G11 and the second diffraction grating G21 with the illumination light IL10 will be described with reference to FIG.
FIG. 3 is a cross-sectional view of the first translucent flat plate P1, the second translucent flat plate P2, and the wafer W arranged to face each other. In FIG. 3, when the illumination light IL10 is irradiated. From the first diffraction grating G11, diffracted light corresponding to the period T1 is generated. If the first diffraction grating G11 is a phase modulation type grating having a duty ratio of 1: 1 and a phase difference of 180 degrees, the generated diffracted light is mainly + 1st order diffracted light LP and −1st order diffracted light LM. However, other orders of diffracted light may be generated.

The diffraction angle θ of the ± first-order diffracted light LP and LM is relative to the wavelength λ of the exposure light (illumination light IL10).
sin θ = λ / T1 (2)
Is the angle represented by
However, this is because ± 1st-order diffracted light LP, LM passes through the first light-transmitting flat plate P1 and is in the air (it may be nitrogen or a rare gas instead. The same applies hereinafter). It is the diffraction angle after ejection. That is, the diffraction angle θ ′ of the ± first-order diffracted light LP, LM in the first light-transmitting flat plate P1 is obtained by using the refractive index n of the first light-transmitting flat plate P1.
sin θ ′ = λ / (n × T1) (3)
Is the angle represented by

Subsequently, the ± first-order diffracted lights LP and LM are incident on the second diffraction grating G21 on the second translucent flat plate P2. Here, since the second diffraction grating G21 is also a phase modulation type diffraction grating as described above, ± first-order diffracted light is mainly generated from the second diffraction grating G21.
In this example, the period T2 of the second diffraction grating G21 satisfies the condition of half the period T1 of the first diffraction grating G11, that is, T1 = 2 × T2. In this case, the −1st order diffracted light LP1 generated by irradiating the second diffraction grating G21 with the + 1st order diffracted light LP is generated with an inclination angle θ with respect to the Z direction and tilted in the −X direction. Further, the + 1st order diffracted light LM1 generated by irradiating the second diffraction grating G21 with the −1st order diffracted light LM is generated tilted in the + X direction with an inclination angle θ with respect to the Z direction. Note that a second diffraction grating with T2 = T1 may be used on the assumption that the second-order diffracted light generated by the second diffraction grating G21 is used.
Furthermore, assuming that higher order K-order (K is a natural number of 3 or more) diffracted light generated from the second diffraction grating is used, the second diffraction grating T2 = K × T1 / 2. Can also be used.

As shown in FIG. 4, the two diffracted lights are irradiated onto the wafer W while maintaining the tilt angle θ with respect to the vertical direction (normal direction) VW of the wafer W, and the wafer W is subjected to light and dark as interference fringes. A pattern IF is formed. At this time, the period (intensity distribution period) T3 of the light / dark pattern IF of the interference fringes formed on the wafer W is:
T3 = λ / (2 × sin θ) (4)
It becomes. This is half of the period T1 of the first diffraction grating G11 and is equal to the period T2 of the second diffraction grating G21.

The light / dark pattern IF exposes a photosensitive member such as a photoresist PR formed on the surface of the wafer W according to the light / dark, and the light / dark pattern IF is exposed and transferred onto the wafer W.
Accordingly, a bright and dark pattern parallel to the Y direction having a period T3 in the X direction is formed on the entire surface of the wafer W. The light and dark pattern is irradiated and exposed on the photoresist PR formed on the wafer W.

  In this example, in FIG. 3, the first distance L1 between the first diffraction grating G11 and the second diffraction grating G21, and the second distance L2 between the second diffraction grating G21 and the wafer W. As an example, the first-order diffracted light LP, LM emitted from an arbitrary point on the first diffraction grating G11 is set so as to irradiate substantially the same point on the wafer W. As a result, the diffracted light irradiated to each point on the wafer W is ± 1st order diffracted light LP and LM emitted from substantially the same point on the first diffraction grating G11. Interference is high, and interference fringes can be formed with good contrast.

Even in this case, the interference fringes formed via the diffraction gratings G11 and G21 are formed in the vicinity of the second diffraction grating G21 on the opposite side to the first diffraction grating G11, for example, about several mm apart. I can say.
Next, an example of the illumination light uniformizing means of this example will be described with reference to FIG. The illumination light uniformizing means of this example comprises the condensing optical system 10 and the fly eye lens 13 in FIG.

  FIG. 5A shows a fly-eye lens 13, an illumination system rear group lens 35 a that collectively represents the lens system from the lens 19 to the lens 35 in FIG. 1, and translucent plates P 1 and P 2. It is the figure seen from the direction, FIG.5 (B) is the figure seen from the -Y direction. 5A, the fly-eye lens 13 is configured by arranging lens elements F1, F2, F3, F4, F5, F6, F7, and F8 in a line along the Y direction on a light-shielding member. ing. Illumination light IL5 from the condensing optical system 10 of FIG. 1 is applied to the fly-eye lens 13, and the illumination light IL7 emitted from the fly-eye lens 13 is refracted by the illumination system rear group lens 35a and becomes the illumination light IL10. 1 is incident on a translucent flat plate P1.

In this case, you may provide the input optical system which condenses a light beam to several lens element F1-F8 arrange | positioned at the line of the fly eye lens 13. FIG.
At this time, since the fly-eye lens 13 is composed of a plurality of lens elements F1 to F8 arranged in a line along the Y direction, the incident angle characteristic of the illumination light IL10 to the first translucent flat plate P1 is as follows. The X direction and the Y direction are different.

The illumination system rear group lens 35a has an incident-side focal plane that coincides with the exit surface of the fly-eye lens 13, and an exit-side focal plane that coincides with the upper surface (the surface in the + Z direction) of the first translucent flat plate P1. Arranged. Accordingly, the illumination system rear group lens 35a constitutes a so-called Fourier transform lens.
The illumination light IL7 emitted from each lens element of the fly-eye lens 13 is refracted by the illumination system rear group lens 35a, and is irradiated with the illumination light IL10 superimposed on the first translucent flat plate P1. Therefore, the intensity distribution of the illumination light on the first translucent flat plate P1 is made uniform by the averaging effect due to the superposition.

An incident angle range φ in the Y direction of the illumination light IL10 to an arbitrary point IP on the first translucent flat plate P1 is shown in FIG. 5A according to the arrangement of the fly-eye lenses 13 in the Y direction. It becomes such a predetermined value.
On the other hand, since the fly-eye lens 13 has only one row in the X direction, the incident angle range in the X direction can be set to a predetermined value or less. Therefore, the illumination light IL10 for one point IP can be a plurality of illumination lights that have a traveling direction in a plane that includes the Y direction and includes the one point IP, and whose traveling directions are not parallel to each other.

  Further, if necessary, an aperture stop 17 having a slit-shaped opening 18 that is long in the Y direction and narrow in the X direction is provided on the exit surface of the fly-eye lens 13 as shown in FIG. The traveling direction in the X direction can be further limited to a plane parallel to the plane including the Y direction and including one point IP. Further, the aperture stop 17 can be provided at the position of the condensing point 28 in FIG. 1 which is conjugate with the exit surface of the fly-eye lens 13.

  By the way, when the interference fringe light / dark pattern IF having a one-dimensional period is formed, the polarization direction (electric field direction) of the illumination light IL10 used for the formation is parallel to the longitudinal direction of the light / dark pattern IF. The linearly polarized light is preferably parallel to the longitudinal direction of the diffraction grating G11 and the second diffraction grating G21. This is because the contrast of the interference fringe IF can be maximized in this case.

However, even if the illumination light IL10 is not perfect linearly polarized light, it may be illumination light in which the electric field component in the longitudinal direction (Y direction) of the first diffraction grating G11 is larger than the electric field component in the periodic direction (X direction). Thus, the above-described contrast improvement effect can be obtained.
Such polarization characteristics of the illumination light IL10 are realized by the polarization control element 9 provided in the illumination optical system of FIG. The polarization control element 9 is, for example, a polarization filter (polaroid plate) or a polarization beam splitter that is rotatably provided with the optical axis AX1 as the rotation axis direction, and the polarization direction of the illumination light IL3 is changed to a predetermined linear polarization by the rotation. be able to.

  When the light source 1 is a light source that emits illumination light IL1 polarized in substantially linear polarization, such as a laser, a half-wave plate that is also rotatably provided can be used as the polarization control element 9. It is also possible to employ two quarter-wave plates provided in series so as to be independently rotatable. In this case, the polarization states of the illumination lights IL3 to IL10 can be not only substantially linearly polarized light but also circularly polarized light and elliptically polarized light.

Note that a pattern has already been formed in the previous exposure process (photolithography process) on the wafer W to be exposed by the exposure apparatus. In the new exposure process, the pattern is kept in a predetermined positional relationship. Generally, it is necessary to form.
In many cases, the existing pattern on the wafer W has a certain degree of expansion and contraction compared to the design value due to thermal deformation and stress deformation accompanying the film forming process and etching process on the wafer W. Therefore, the exposure apparatus is required to be applied to such expansion / contraction of the wafer W to form a new pattern on the wafer W by correcting the expansion / contraction to some extent.

In the exposure apparatus of this example, either the Z position of the wafer W set by the Z leveling mechanism 38Z (height control apparatus) or telecentricity as the convergence / divergence state of the illumination light IL10, or both are changed. By doing so, the expansion / contraction correction of the bright and dark pattern formed on the wafer W can be performed.
First, the convergence / divergence state of the illumination light IL10 in FIG. 1 will be described with reference to FIG.

  FIG. 6A is a diagram illustrating a state where the illumination light IL10 is a parallel light beam, that is, a state in which neither the convergence nor the divergence occurs. At this time, outer edges LEa and LEb in the X direction of the illumination area of the illumination light IL10 are perpendicular to the first light-transmissive plate P1, and the illumination light IL10c, illumination light IL10d, and illumination light IL10e between them are Regardless of the location within one translucent flat plate P1, the light enters perpendicularly to the first translucent flat plate P1.

  On the other hand, FIG. 6B is a diagram showing a case where the illumination light IL10 is a divergent light beam, and the illumination light IL10 defined by the outer edges LEa1 and LEb1 is a diverging light path as a whole. At this time, the outer edges LEa1 and LEb1 are inclined (diverged) by ψe from the vertical directions LEa and LEb, respectively. Accordingly, the incident angle of the illumination light IL10 to the first light-transmissive plate P1 changes according to the position.

  That is, the illumination light IL10f irradiated through the optical path portion close to the outer edge LEa1 is inclined slightly outward and enters the first translucent flat plate P1. When the tilt angle is ψf, the position of the bright and dark pattern of the interference fringes formed on the wafer W by the illumination light IL10f is shifted in the −X direction by ΔZ × tan ψf in which the angle φ is replaced by the angle ψf in the following equation 6. Formed in position.

δp = ΔZ × tanφ (6)
In FIG. 3, the angle φ in Expression 6 is the X direction (diffraction grating G11) from the normal direction to the first diffraction grating G11 in the direction of the center of gravity of the illumination light IL10 incident on each position of the first diffraction grating G11 in FIG. Of the illumination light IL10 incident on the respective positions of the first diffraction grating G11. According to Equation 6, in FIG. 3, from the state where the distances L1 and L2 are set so that the diffracted lights LP and LM emitted from the first diffraction grating G11 are focused on the same point on the wafer W. By changing the Z position of W by ΔZ and controlling the telecentricity of the illumination light IL10 incident on a certain point on the diffraction grating G11 by the angle φ, the interference on the wafer W corresponding to that point The position of the stripe changes by δp in the X direction. Further, in the illumination optical system IS of FIG. 1 of this example, the angle φ of the illumination light IL10 is 0 on the optical axis AX2, and the position where the light and dark pattern of the interference fringes is formed on the optical axis AX2 6 is a reference position in the X direction.

  On the other hand, the position of the bright and dark pattern of the interference fringes formed on the wafer W by the illumination light IL10h irradiated through the optical path portion close to the outer edge LEb1 is ΔZ × according to Expression 6 by the outward inclination angle ψh of the illumination light IL10h. It is formed at a position shifted in the + X direction by tan ψh. Further, the position of the bright and dark pattern of the interference fringes formed on the wafer W by the illumination light IL10g irradiated through the optical path portion close to the center is not displaced because the illumination light IL10g is substantially perpendicularly incident.

Therefore, the relationship of the size of the interference fringe pattern IF exposed to the wafer W with respect to the first diffraction grating G11 is expanded by using the illumination light IL10 as a divergent light beam when ΔZ is positive. Therefore, it is possible to reduce the interference fringe pattern IF exposed on the wafer W.
Specifically, assuming that the angle φ (X1) [rad] is the angle of the illumination light IL10 incident on the position separated from the optical axis AX2 in FIG. Can be approximated as follows:

δp = ΔZ × φ (X1) (7)
Further, when the period in the X direction of the interference fringe pattern IF is to be expanded or contracted by k (k is a real number near 1) times, the following equation should be established for the positional deviation amount δp of the interference fringe pattern IF at the interval X1.
(X1 + δp) / X1 = k (8)
From Expression 7 and Expression 8, the angle φ (X1) is as follows.

φ (X1) = X1 (k−1) / ΔZ (9)
That is, the telecentricity of the illumination light IL10 in FIG. 1 is controlled, and the angle φ (X1) in the X direction of the illumination light IL10 at a position separated from the optical axis AX2 by the interval X1 in the X direction satisfies Expression 9. By doing so, the period of the interference fringe pattern IF in the X direction can be expanded and contracted by k times.

  As described above, in this example, the telecentricity of the illumination light IL10 is controlled, that is, the light quantity center of gravity direction of the illumination light IL10 and the normal direction of the diffraction grating G11 according to the position in the first diffraction grating G11. The period of interference fringes formed on the wafer W is controlled by changing the relationship (by changing the incident angle of the illumination light IL10 according to the position in the diffraction grating G11).

  In the exposure apparatus of the present example, as shown in FIG. 1, among the lenses 29, 30, 32, and 35 constituting the fourth lens group in the illumination optical system IS, the negative lens 30 has a lens driving mechanism. Reference numerals 31a and 31b are attached, and lens drive mechanisms 33a and 33b are attached to the positive lens 32. These lens driving mechanisms 31a, 31b, 33a, and 33b are movable in the Z direction on the fixed shafts 34a and 34b, whereby the lens 30 and the lens 32 are also independently movable in the Z direction.

  That is, the fourth lens group 29, 30, 32, 35 constitutes a so-called inner focus lens as a whole, and its focal length or focal position is variable. The lens driving mechanisms 31a, 31b, 33a, and 33b (members that control the telecentricity of the illumination light) control the focal lengths or focal positions of the fourth lens groups 29, 30, 32, and 35, thereby illuminating. The convergence / divergence state of the light IL10 can be made variable. In addition to this, a Z position adjusting mechanism is also provided for the first lens groups 2, 3, 4, and 6 of the illumination optical system IS, and together with the fourth lens groups 29, 30, 32, and 35, The convergence / divergence state of the illumination light IL10 may be variable.

  For this purpose, as shown in FIG. 1, lens driving mechanisms 5a and 5b are attached to the negative lens 4 in the first lens group 2, 3, 4 and 6, and the lens driving mechanisms 7a and 5a are attached to the positive lens 6. 7b is attached. These lens driving mechanisms 5a, 5b, 7a and 7b are movable in the Z direction on the fixed shafts 8a and 8b, whereby the lens 4 and the lens 6 can be independently moved in the Z direction. it can.

Further, from Expression 6, the convergence / divergence state of the illumination light IL10 is not changed as described above, but is fixed to a predetermined convergence state or a divergent state, and the interference is changed by changing the Z position where the wafer W is disposed. It is also possible to perform expansion / contraction correction of the fringe period.
These expansion / contraction corrections are performed in advance by measuring the positions of existing circuit patterns or alignment marks formed at a plurality of locations on the wafer W by the wafer mark detection mechanism 43 prior to exposure of the wafer W. It is desirable to carry out based on the amount of expansion and contraction (detailed later).

  Prior to measurement of the amount of expansion / contraction of the wafer W, the position of the detection reference 44 of the wafer mark detection mechanism 43 may be measured using a predetermined reference mark on the interference fringe measurement system 41 on the wafer stage. desirable. The exposure apparatus also includes a Y-axis laser interference that is a detection mechanism that enables the position measurement of the wafer stage 38 at the position of the wafer mark detection mechanism 43 in order to improve the position measurement accuracy by the wafer mark detection mechanism 43. It is desirable to provide a total of 40Y3 (see FIG. 7).

Next, an example of the exposure method of this example will be described. In this example, the area of the illumination light IL10 irradiated on the first light-transmissive plate P1 in FIG. 1, that is, the predetermined illumination field is an illumination area elongated in the X direction smaller than the wafer W as shown in FIG. In the exposure, an interference fringe pattern 92 is formed in the illumination region 42 at a predetermined cycle in the X direction.
That is, FIG. 7 is a view of the wafer stage 38 of FIG. 1 as viewed from the lens 35 side, and the exposure to the wafer W is performed by the light source 1, the illumination optical system IS, the first translucent flat plate P1, FIG. Further, as shown in FIG. 7, the wafer W is scanned by the wafer stage 38 in the Y direction with respect to the second translucent flat plate P2. As described above, the illumination light IL10, the first diffraction grating G11 on the first translucent flat plate P1, and the second diffraction grating G21 on the second translucent flat plate P2 are placed on the wafer W in the X direction. Since the interference fringe pattern 92 (corresponding to the light / dark pattern IF in FIG. 4) having the periodic direction in the Y direction and the longitudinal direction in the Y direction is formed, the scanning in the Y direction is performed in the longitudinal direction of the interference fringe pattern 92. Will be performed along.

  During the scanning exposure, the position and rotation angle of the wafer W (wafer stage 38) in the X direction and the Y direction are changed via the X axis moving mirror 39X and the Y axis moving mirror 39Y provided on the wafer stage 38, respectively. Measurement is performed using the X-axis laser interferometers 40X1 and 40X2 and the Y-axis laser interferometers 40Y1 and 40Y2, and the measured positions and rotation angles are controlled by a stage control mechanism (not shown). By such scanning exposure, since the interference fringe pattern 92 formed by the first diffraction grating G11 and the second diffraction grating G21 is integrated and exposed in the Y direction on the wafer W, these diffraction patterns are exposed. The effects of lattice defects and foreign matter are alleviated, and a good pattern without defects is exposed on the wafer W.

Note that the positional deviation and rotational deviation in the X and Y directions between the interference fringe pattern 92 and the wafer W generated during scanning exposure are the measurement results of the laser interferometer at the positions of the transmissive plates P1 and P2 and the wafer stage 38 described above. Is corrected by movement of at least one of the first and second transparent flat plates P1 and P2 and the wafer W.
In addition, the illuminance non-uniformity that may remain in the illumination area 42 is also accumulated and averaged in the Y direction, so that substantially higher uniformity can be realized. Further, the shape of the illumination area 42 may be such that the width in the Y direction changes depending on the position in the X direction. This is because the integrated value in the Y direction of the illumination light illuminance distribution in the illumination area 42 can be made more uniform by changing the shape of the illumination area 42 itself.

The shape of the illumination area 42 can be determined by the shape of the opening provided in the field stop 22 in the illumination optical system IS of FIG. The field stop 22 may be disposed in the vicinity of the light source side of the first light-transmitting substrate P1.
The illumination area 42 in this example is a fixed rectangular area whose width in the Y direction is narrower than D except for both end portions 42L and 42R in the X direction. However, the width d in the X direction (the width d is an example of the width D). As an example, the end portions 42L and 42R are symmetrically outward and the width in the Y direction is linearly reduced. Further, the width D in the X direction of the illumination area 42 is smaller than the diameter of the wafer W. Therefore, in order to expose the entire surface of the wafer W in the illumination area 42, as shown by a locus 45 at the center of the illumination area 42 in FIG. 7, for example, the wafer W and the illumination area 42 are relatively moved in the Y direction. Thus, it is necessary to repeat the exposure operation (scanning exposure) and the operation of relatively moving the wafer W and the illumination area 42 in the X direction (step movement operation) between them. At this time, in the adjacent rows, the direction of relative movement between the wafer W and the illumination area 42 is reversed. In this example, the illumination area 42 is actually fixed, and the wafer W moves in the X direction and the Y direction relative to the illumination area 42 by driving the wafer stage 38. However, FIG. For the sake of convenience, the illumination area 42 is shown to move with respect to the wafer W.

  In this case, when the illumination area 42 scans and exposes the wafer W in the + Y direction along a certain row 45a in the trajectory 45, an illumination area 42P (the same as the illumination area 42) is located along the row 45b adjacent thereto. The exposure is performed so that one end 42R of the width d of the illumination area 42 and the other end 42LP of the width d of the illumination area 42P overlap in the X direction when the wafer W is scanned and exposed in the −Y direction. Is called. For this purpose, the distance SX in the X direction between the rows 45a and 45b may be narrower than the width D by the width d as follows.

SX = D−d (10)
As a result, the exposure amount of the portion SB exposed by overlapping the end portions 42R and 42LP becomes uniform and equal to the exposure amount of the other exposure regions, so that the uniformity of the exposure distribution over the entire surface of the wafer W is improved. .
Further, if the period of the interference fringe pattern 92 in the X direction is T3, the end portion 42R and the end portion 42LP do not shift the positional relationship between the bright and dark portions of the interference fringe pattern 92, so that no joint error occurs. In order to achieve this, N may be a predetermined natural number, and the interval SX may be N times the period T3 as follows.

SX = N × T3 (11)
More precisely, the interval SX is set to a value closest to Expression 10 and satisfying Expression 11. As a result, the interval SX may be a value that differs by about ± T3 from the value calculated by Equation 10, but since the period T3 is a minute value that is about the exposure wavelength, the exposure amount distribution is uniform. There is no substantial effect on sex.

The point is that the end portions 42R and 42LP (which have the same shape as the end portion 42L) have the following shapes because the exposure amount after the end portion 42R and the end portion 42LP are overlapped and exposed only to be uniform. Needless to say, the shape is not limited to the symmetric and linear shape exemplified above.
That is, when the width in the Y direction of the end portion 42R is R (X) and the width in the Y direction of the end portion 42LP is L (X), R (X) + L (X) has a constant value. Any shape is possible.
For example, even if the shape is a rectangular shape, a neutral density filter or the like that reduces the amount of illumination light toward the outside may be disposed at both ends of the field stop 22 in the X direction.

  If the exposure on the wafer W is a superposition exposure, it is necessary to previously align the bright and dark pattern of the interference fringes with the circuit pattern formed on the wafer W in the steps so far. Hereinafter, an example of a mechanism and an alignment method for performing alignment in this example will be described. At this time, it is assumed that the rotation angles are adjusted so that the periodic directions of the diffraction gratings of the two light-transmitting flat plates P1 and P2 in FIG. 1 are parallel and in the X direction.

  FIG. 8 is a plan view showing the wafer stage 38 of FIG. 2. In FIG. 8, as an example, the upper surface of the wafer W is divided into a number of shot areas SA (partition areas) each having a predetermined width in the X and Y directions. In each shot area SA, a predetermined circuit pattern is formed by the device manufacturing process so far, and wafer marks WMx and WMy as alignment marks (alignment marks) indicating positions in the X direction and the Y direction are formed. Is also formed. The alignment may be performed using a predetermined circuit pattern formed in each shot area SA instead of the wafer mark.

  In the exposure using the exposure apparatus of this example, the interference fringe pattern 92 and the circuit pattern in each shot area SA on the wafer W must satisfy a predetermined positional relationship particularly in the X direction. When the interference fringe pattern 92 is a two-dimensional lattice pattern having a predetermined pitch in the X and Y directions, the lattice pattern and the circuit pattern in each shot area SA are predetermined in the X and Y directions. It is necessary to satisfy the positional relationship.

In order to perform alignment between the interference fringe pattern and the shot area on the wafer W, an interference fringe measurement system 41 (period measurement mechanism) is fixed to the wafer stage 38, and the upper surface of the interference fringe measurement system 41 is on the wafer W. It is set to the same height as the surface.
In FIG. 8, two two-dimensional reference marks 93A and 93B are formed at a predetermined interval in the Y direction on the upper surface of the interference fringe measurement system 41, and a transmissive substrate 94 is embedded between these reference marks 93A and 93B. ing. In addition, the transmissive substrate 94 may be enlarged and the reference marks 93A and 93B may be formed thereon. As an example, the rotation angle around the Z axis of the wafer stage 38 is set so that a straight line passing through the centers of the reference marks 93A and 93B is parallel to the Y axis.

  On the surface of the transmissive substrate 94, measurement gratings 95A and 95B are formed in which light transmissive portions and light shielding portions are arranged in different directions in the X direction for measuring the position of the interference fringe pattern. The measurement grating 95A is used to detect the position of the interference fringe pattern 92 in the X direction, while the other measurement grating 95B detects the position of another interference fringe pattern having a period different from that of the interference fringe pattern 92. Used for. Therefore, the number of measurement gratings 95A and 95B is set according to the type of the period of the interference fringe pattern to be detected.

  When detecting the position of the interference fringe pattern 92 in FIG. 8, the wafer stage 38 is driven, and the center of the measurement grating 95A is the position to be measured in the interference fringe pattern 92 (not yet irradiated with illumination light). Move to match (measurement point). Thereafter, the field stop 22 of FIG. 1 is driven to limit the illumination area so that the illumination light is not irradiated onto the wafer W, and the illumination light IL10 is emitted to form the interference fringe pattern 92 on the measurement grating 95A. To do.

  FIG. 9 shows a configuration example of the interference fringe measurement system 41 of FIG. 8. In FIG. 9, the light transmitted through the measurement grating 95A is sequentially collected on the bottom surface of the region where the measurement grating 95A of the transmission substrate 94 is formed. And a photoelectric detector 101 that detects the collected light. Assuming that the period of the designed intensity distribution in the X direction of the interference fringe pattern 92 is T3, the period PA in the X direction of the measurement grating 95A is set to be substantially equal to the period T3 as an example. The ratio of the width to the part is set to approximately 1: 1. The lens 100 transmits the measurement grating 95A through the lens 100, with the period PA set to substantially an integral multiple (eg, twice) of the period T3 and the width of the light transmission portion in the measurement grating 95A is set to about the period T3. A configuration in which only the 0th-order light is condensed is also possible. A first position measurement member is configured including the measurement grid 95 </ b> A, the lens 100, and the photoelectric detector 101. Similarly, a lens and a photoelectric detector are also arranged on the bottom surface of the measurement grid 95B in FIG. 8, thereby constituting a second position measurement member. An interference fringe measurement system 41 is configured including the plurality of position measurement members.

In addition, the structure which uses the lens 100 and the photoelectric detector 101 in common with respect to several measurement grating | lattices 95A and 95B is also possible. Also, a plurality of interference fringe measurement systems similar to the interference fringe measurement system 41 of FIG. 8 are arranged on the wafer stage 38 apart in the X direction, and the positions of the interference fringe pattern at a plurality of measurement points separated in the X direction. May be measured simultaneously.
In FIG. 9, the detection signal of the photoelectric detector 101 is supplied to the alignment information processing system 73. In the storage unit in the alignment information processing system 73, the positional relationship between the reference marks 93A and 93B and the measurement grids 95A and 95B in FIG. 8 (for example, in the measurement grids 95A and 95B with respect to a straight line passing through the centers of the reference marks 93A and 93B). Information of offset amounts ΔIAX, ΔIBX) in the X direction at the center of the predetermined light transmitting portion is stored. The alignment information processing system 73 is also supplied with information on the position in the X and Y directions of the wafer stage 38 measured by the laser interferometer 40 of FIG.

  In this case, the wafer stage 38 in FIG. 8 is moved in the + X direction within a range of, for example, several times the period of the interference fringe pattern 92 under the control of the main control system 70 that controls the operation of the entire exposure apparatus. When the alignment information processing system 73 captures the detection signal of the photoelectric detector 101 in correspondence with the coordinates of the wafer stage 38, the detection signal changes in a sine wave shape with a period T3 with respect to the X coordinate of the wafer stage 38. Therefore, as an example, the X coordinate of the wafer stage 38 when the detection signal first peaks after the movement of the wafer stage 38 is started can be set as the position of the interference fringe pattern 92 at the measurement point. In the following, after the measurement grating 95A is moved to the measurement point of the interference fringe pattern 92, the detection signal of the photoelectric detector 101 first peaks when the wafer stage 38 is driven and the measurement grating 95A is moved in the + X direction. The X coordinate and Y coordinate of the wafer stage 38 at that time are set as the positions of the interference fringe pattern 92 detected by using the interference fringe measurement system 41 at the measurement point.

  In FIG. 9, the alignment information processing system 73 supplies information on the position of the interference fringe pattern 92 thus detected to the main control system 70. The main control system 70 (control device) calculates the period of the interference fringe pattern 92 based on position information at a plurality of measurement points of the interference fringe pattern 92 as described later. Next, in order to correct the period of the interference fringe pattern 92, the main control system 70 uses the lens driving mechanisms 31a, 31b, 33a, 33b of FIG. 1 as an example and the fourth lens groups 29, 30, 32, 35. By controlling the focal length or the focal position, the convergence / divergence state of the illumination light IL10 is made variable, and the Z position of the wafer W is controlled via the Z leveling mechanism 38Z as necessary.

  In alignment and exposure, it is necessary to measure and store the distance (baseline amount) between the detection center 44 of the wafer mark detection mechanism 43 in FIG. 8 and the exposure center that is the center of the interference fringe pattern 92 in advance. is there. Therefore, the values of the X and Y coordinates (AX1, AY1) of the wafer stage 38 when the centers of the reference marks 93A and 93B coincide with the detection center 44 of the wafer mark detection mechanism 43 are measured in advance, and the alignment shown in FIG. It is stored in a storage unit in the information processing system 73. As an example, the wafer stage 38 is driven to move the center of the measurement grating 95A to the vicinity of the optical axis AX2 in FIG. 1 (equal to the center of the illumination area 42 in FIG. 7 in this example), and then the interference fringe pattern in FIG. The X and Y coordinates of the wafer stage 38 when the measurement grating 95A is moved in the + X direction with respect to 92 and the detection signal of the photoelectric detector 101 first peaks are the coordinates of the exposure center of the interference fringe pattern 92 ( EX1, EY1).

  In this case, the difference between the coordinates (EX1, EY1) and the coordinates (AX1, AY1) is corrected by the offset amount ΔIAX in the X direction of the measurement grid 95A with respect to the reference marks 93A, 93B, thereby the wafer mark detection mechanism. 43 baseline amounts (BEX, BEY) can be obtained. This baseline amount (BEX, BEY) is supplied to the main control system 70. Each shot area on the wafer W is accurately moved to the exposure area of the interference fringe pattern 92 by correcting the coordinates of each shot area on the wafer W measured via the wafer mark detection mechanism 43 with the baseline amount. can do.

Next, an example of the alignment operation and the exposure operation of the exposure apparatus of this example will be described with reference to the flowchart of FIG. In the following description, the shape of the illumination region 42 when measuring the period of the interference fringe pattern 92 is the same elongated shape as that during exposure, but the shape may be a circle or the like as long as the wafer is not exposed. .
First, in a state where the illumination light for forming the interference fringe pattern 92 is not irradiated, a plurality of marks on the wafer W, for example, in step 200 of FIG. 10 using the wafer mark detection mechanism 43 of FIG. Then, the coordinates of the X-axis and Y-axis wafer marks (alignment marks) attached to a plurality of shot areas SA including three shot areas that are not on the same straight line selected from the wafer W are measured. Specifically, the wafer mark detection mechanism 43 detects the amount of positional deviation in the X and Y directions from the detection center 44 of the wafer mark to be measured, and supplies the detection result to the alignment information processing system 73 in FIG. To do. In the alignment information processing system 73, the X coordinate and Y coordinate of the wafer mark are obtained by adding the position in the X direction and Y direction of the wafer stage 38 at that time to the positional deviation amount.

  Thereafter, the alignment information processing system 73 uses, for example, an enhanced global alignment (EGA) method disclosed in Japanese Patent Publication No. 4-47968 based on the measurement values of the X and Y coordinates of the wafer marks. Offset Xoff in the X direction, offset Yoff in the Y direction, shot array rotation (rotation angle) Θ, scaling in the X direction (linear expansion / contraction amount) to calculate the array coordinates of all shot areas on W The parameters of the shot array including kx and scaling (linear expansion / contraction amount) ky in the Y direction are obtained. These parameters are supplied to the main control system 70, and the offsets Xoff and Yoff are corrected using the above-described baseline amounts (BEX, BEY).

  In this example, as an example, in order to correct the rotation Θ of the shot arrangement, the wafer W is rotated by −Θ through the wafer stage 38. Instead, the translucent flat plates P1 and P2 in FIG. 1 are rotated to rotate the interference fringe pattern 92 by an angle Θ, and the scanning direction of the wafer W during exposure is inclined by an angle Θ with respect to the Y axis. It is also possible to use a method of setting the direction.

  Note that the scalings kx and ky are 1 when there is no linear expansion / contraction, and when the interference fringe pattern with the period T20 is exposed on the wafer W in FIG. 8 in a state where there is no linear expansion / contraction in the X direction (kx = 1). It is assumed that the overlay accuracy is set to be the best. Thus, in the state where there is no linear expansion and contraction in the X direction, the arrangement period in the X direction of the shot areas SA on the wafer W is an integral multiple of the period T20.

  In the next step 201, for example, the illumination light IL10 is irradiated to a region other than the wafer W using the field stop 22 of FIG. 1, for example, and as shown in FIG. The interference fringe pattern 92 formed in the period T3 is formed in the illumination area 42. Although the period T3 of the interference fringe pattern 92 is about the exposure wavelength, it is displayed large for convenience of explanation in FIG.

  In the next step 202, as shown in FIG. 11A, the interference fringe measurement of FIG. 8 is performed at two measurement points Q4 and Q5 that are separated by an interval LX1 on the interference fringe pattern 92 in the illumination area 42 in the X direction. The measurement grid 95A of the system 41 is sequentially moved to measure the position of the interference fringe pattern 92 in the X direction, and the measurement result is supplied to the main control system 70 of FIG. The interval LX1 is M1 times the design period of the interference fringe pattern 92 (M1 is an integer), and the value of M1 is the actual period T3 of the interference fringe pattern 92 between the measurement points Q4 and Q5. (= LX1 / T3) is set so as to fall between (M1−1 / 2) and (M1 + 1/2), that is, not so large. Then, the main control system 70 obtains a fraction m by dividing the difference in position in the X direction at the two measurement points by the design cycle T3, and then interferes by dividing the interval LX1 by the number of cycles (M + m). The period T3 of the stripe pattern 92 is obtained.

  In the next step 203, the period T3 of the interference fringe pattern 92 is corrected to T3A in accordance with the scaling kx of the shot arrangement of the wafer W obtained in step 200. In this case, since the period of the interference fringe pattern 92 should be T20 when the scaling kx is 1, the period T3A of the interference fringe pattern 92 for optimizing the overlay accuracy in the X direction is as follows: It becomes kx1 times the original period T3.

kx1 = T3A / T3 = kx · T20 / T3 (12)
Specifically, the main control system 70 substitutes the coefficient kx1 of Expression 12 instead of the coefficient k of Expression 9, and corrects the telecentricity of the illumination light IL10 irradiated on the translucent flat plate P1 of FIG. By correcting the height of the wafer via the wafer stage 38, the period of the interference fringe pattern 92 in the X direction is expanded and contracted by k × 1. As an example, when kx1 is smaller than 1, the interference fringe pattern 92 is contracted in the X direction as shown in FIG.

  In the next step 204, as shown in FIG. 11B, the interference fringe measurement system of FIG. 8 is placed at two measurement points Q4 and Q6 separated by an interval LX2 (> LX1) in the X direction on the interference fringe pattern 92. The measurement grating 95A of 41 is sequentially moved to measure the position of the interference fringe pattern 92 in the X direction, and the measurement result is supplied to the main control system 70 of FIG. Note that the measurement points Q4 and Q6 may be set in the vicinity of both ends 42L and 42R of the illumination area 42 in order to increase the interval LX2 and increase the measurement accuracy of the period of the interference fringe pattern 92. However, the interval LX2 may be equal to or less than the interval LX1. The interval LX2 is also M2 times the design period of the interference fringe pattern 92 (M2 is an integer), and in the same manner as in Step 202, the main control system 70 determines the difference in position in the X direction at the two measurement points. Then, the fraction m ′ is obtained by dividing by the period (here, equal to T3A), and then the period T3A of the interference fringe pattern 92 is obtained with higher accuracy by dividing the interval LX2 by the number of periods (M2 + m ′).

  In the next step 205, the period T 3 A of the interference fringe pattern 92 is corrected in the same manner as in step 203 in accordance with the shot array scaling kx obtained in step 200. In this case as well, the period of the interference fringe pattern 92 for optimizing the overlay accuracy in the X direction is kx1 times the original period T3 as in Expression 12, but the accuracy of the coefficient kx1 is improved.

  In the next step 206, using the period of the interference fringe pattern 92 corrected in step 205, the exposure start position (here, the locus 45) that provides the best overlay accuracy between the interference fringe pattern 92 and the shot area on the wafer W. 7 is moved to the starting point), the illumination area 42 is set by the field stop 22 of FIG. 1, and the irradiation of the illumination light IL10 is started. Then, the movement of the wafer W in the Y direction is started, and the exposure region of the first column on the wafer W is scanned and exposed with the interference fringe pattern 92 in the illumination region 42, and then the irradiation of the illumination light IL10 is stopped.

  In the next step 207, when the exposure is not completed, the process proceeds to step 208, and the wafer stage 38 in FIG. 7 is driven to move the wafer W stepwise in the X direction by an interval SX. This interval SX is calculated by substituting the period T3A after being corrected in step 205 as the period T3 of Equation 11. Thereafter, the operation returns to step 206 again, and scanning exposure is performed by the illumination area 42 on the exposure area of the second row on the wafer W. Steps 206 to 208 are repeated until the entire surface of the wafer W is exposed.

  According to the present example after completion of exposure, since the period of the interference fringe pattern 92 in the illumination area 42 is corrected according to the shot arrangement on the wafer W as described above, it is already formed in each shot area on the wafer W. The overlay accuracy between the circuit pattern and the interference fringe pattern 92 is improved. Further, since the exposure is performed by repeatedly scanning the wafer W in the Y direction with respect to the illumination area 42, even when the illumination area 42 is small, a pattern with a fine cycle is efficiently exposed on the entire surface of the wafer W. it can. Further, since the width SX of the step movement at the time of exposing the adjacent row on the wafer W is set to a natural number multiple of the period of the interference fringe pattern 92, no joint error occurs and the shape of the end portion of the illumination region 42 ( (Alternatively, it may be a light amount distribution), so that a uniform exposure amount distribution can be obtained on the entire surface of the wafer W.

In the operation of FIG. 10, the second cycle correction operation (steps 204 and 205) during the steps 202 to 205 can be omitted. For example, when the interference fringe pattern 92 is exposed on a wafer on which a circuit pattern is not yet formed, the operation of step 200 can be omitted.
In the example of FIG. 1, the first diffraction grating G11 and the second diffraction grating G21 are each a phase modulation type diffraction grating, but the configuration of both diffraction gratings is not limited to this. For example, any of the diffraction gratings may be a diffraction grating that modulates both the phase and intensity of transmitted light, such as a halftone phase shift reticle. Further, the first diffraction grating may be a phase modulation type diffraction grating, and the second diffraction grating may be an intensity modulation type diffraction grating.

FIG. 12 shows the first modulation gratings G11 and G12 having a period T1 in the X direction formed on the surface on the + Z direction side (light source side) of the first translucent flat plate P1 in FIG. This shows a case where an intensity-modulated second diffraction grating G21A having a period T2 (here, equal to T1 / 2) is formed on the −Z direction side (wafer W side) of the second translucent flat plate P2.
Hereinafter, the principle that the bright and dark patterns of interference fringes are formed on the wafer W by irradiating the first diffraction gratings G11 and G12 and the second diffraction grating G21A with the illumination light IL10 will be described with reference to FIG.

  In FIG. 12, when the illumination light IL10 is irradiated, diffracted light corresponding to the period T1 is generated from the first diffraction gratings G11 and G12. If the first diffraction gratings G11 and G12 are phase modulation type gratings having a duty ratio of 1: 1 and a phase difference of 180 degrees, the 0th-order diffracted light disappears and does not occur. In this case, two diffracted lights of ± 1st order light are mainly generated, but there is a possibility that higher order diffracted lights such as ± 2nd order light are also generated.

  However, when the period T1 is shorter than three times the effective wavelength λ of the illumination light, the third-order or higher-order diffracted light cannot be generated. Further, if the phase modulation type grating has a duty ratio of 1: 1 and a phase difference of 180 degrees as described above, no second-order diffracted light can be generated. Therefore, in this case, only two of the + 1st order diffracted light LP and the −1st order diffracted light LM are generated from the first diffraction gratings G11 and G12, and pass through the first translucent flat plate P1 to obtain the second. 2 is incident on the translucent flat plate P2. Subsequently, the + 1st order diffracted light LP and the −1st order diffracted light LM are applied to the second diffraction grating G21A provided on the wafer W side surface of the second translucent flat plate P2. Since both diffracted lights are symmetrical, only the + 1st order diffracted light LP will be described below.

  The + 1st-order diffracted light LP is inclined by a predetermined angle from a direction perpendicular to the second diffraction grating G21A (normal direction) by the period T1 of the first diffraction gratings G11 and G12, and enters the second diffraction grating G21A. Incident. When the + 1st order diffracted light LP is irradiated onto the second diffraction grating G21A, diffracted light is also generated from the second diffraction grating G21A. Since the second diffraction grating G21A is an intensity modulation type diffraction grating, the diffracted light is diffracted light including zeroth-order light.

Here, the angle direction in which each diffracted light is generated is inclined according to the inclination of the incident angle of the illuminating illumination light (+ 1st order diffracted light LP). That is, the second diffraction grating G21A is diffracted according to the zero-order diffracted light LP0 traveling in the direction parallel to the irradiated + 1st-order diffracted light LP and the X-direction period T2 of the second diffraction grating G21A. First-order diffracted light LP1 is generated.
If the period T2 of the second diffraction grating G21A is larger than a predetermined value determined by the relationship between the period T1 and the effective wavelength, + 1st order diffracted light (not shown) may also be generated. However, by setting the period T2 to be equal to or less than the effective wavelength of the illumination light, it is possible to substantially prevent the generation of + 1st order diffracted light (not shown).

As a result, two diffracted lights of 0th-order diffracted light LP0 and −1st-order diffracted light LP1 are irradiated on the wafer W, and a light / dark pattern of interference fringes is formed by interference of these diffracted lights.
In the configuration of FIG. 12, as an example, the distance D2 between the second diffraction grating G21A and the surface of the wafer W is different from the distance D1 between the first diffraction grating G11, G12 and the second diffraction grating G21A. Set small.

In the above example, the first diffraction grating G11 (or G11, G12) and the second diffraction grating G21 (or G21A) are formed on different translucent flat plates, respectively. Both diffraction gratings can also be formed on the same translucent flat plate.
FIG. 13 is a diagram showing an example in which the first diffraction grating G13 and the second diffraction grating G14 are formed on the light source side and the wafer W side of one translucent flat plate P3, respectively. Also in this example, the structure and manufacturing method of each diffraction grating are the same as those in the above example. The lens 35 and the illumination optical system upstream thereof are the same as in the above example.

  The translucent flat plate P4 in FIG. 13 is provided to prevent contamination of the second diffraction grating G14 and to make the first distance L1 and the second distance L2 substantially equal. Further, as an example, the light transmitting flat plate P3 and the light transmitting flat plate P4 can be integrally held by the holder 37A of FIG. 1, and in this case, the holder 36A of FIG. It can be omitted.

Further, the first diffraction grating and the second diffraction grating are not limited to those provided only on the surface of the translucent flat plate.
For example, as shown in FIG. 14, although the second diffraction grating G16 is formed on the surface of the second light-transmitting flat plate P6, a thin third light-transmitting flat plate P7 is bonded to the second diffraction grating G16. The grating G16 can also be effectively formed inside the translucent flat plate. Note that the second translucent flat plate P5 and the first diffraction grating G15 in FIG. 14 are the same as those shown in FIG.

Even when these translucent flat plates P7 and the like are used, the first distance L1 and the second distance L2 are, for example, a pair of diffracted light generated from one point on the translucent flat plate P5. Are set so as to intersect on substantially the same point on the wafer W.
In the above example, air is present between the second translucent flat plate P2 and the wafer W, but instead, a predetermined dielectric may be filled. Thereby, the substantial wavelength of the illumination light (diffracted light) irradiated to the wafer W can be reduced by the refractive index of the dielectric, and the period of the bright and dark pattern of the interference fringes formed on the wafer W. It becomes possible to further reduce T3. For this purpose, it goes without saying that the period T2 of the second diffraction grating G21 and the period T1 of the first diffraction grating G11 also need to be reduced in proportion thereto.

  FIG. 15A shows an example of a wafer stage 38a and the like suitable for this. Continuous side walls 38b and 38c are provided around the wafer stage 38a, and a liquid 56 such as water can be held in a portion surrounded by the side walls 38b and 38c. As a result, the space between the wafer W and the second translucent flat plate P2 held by the holder 37A is filled with water, and the wavelength of the illumination light has a refractive index of water (1.46 for light with a wavelength of 193 nm). ) Is reduced.

A water supply mechanism 54 and a drainage mechanism 55 are also provided so that a clean liquid without contamination is supplied to and discharged from a portion surrounded by the side walls 38b and 38c.
Further, as shown in FIG. 15B, the uppermost surfaces of the side walls 38d and 38e of the wafer stage 38a are made higher than the lower surface of the first light transmitting flat plate P1, and the first light transmitting flat plate held by the holder 36A. The space between P1 and the second translucent flat plate P2 held by the holder 37A can also be filled with liquid. The functions of the liquid supply mechanism 54a and the drainage mechanism 55b are the same as described above.

As a result, the entire optical path from the first diffraction grating G11 to the wafer W can be covered with a dielectric other than air, and the effective wavelength λ of the illumination light can be reduced by the amount of refraction of the liquid. It becomes. As a result, a pattern having a finer period can be exposed.
Needless to say, the dielectric filled between the second translucent flat plate P2 and the wafer W is not limited to water but may be another dielectric liquid. In that case, the refractive index of the dielectric liquid is preferably 1.2 or more from the viewpoint of reducing the period of the bright and dark pattern of interference fringes.

  15A and 15B, the wafer W is disposed in the liquid, but includes at least an interference fringe pattern formation region between the second transparent flat plate P2 and the wafer W. The supply and discharge may be performed so that the predetermined area is filled with the liquid. At this time, particularly in the scanning exposure apparatus, the liquid may flow along the scanning direction at the time of scanning exposure, and the height of the surface of a predetermined region surrounding the wafer W on the wafer stage 38 is set to the surface of the wafer W. It is preferable to make them substantially coincide. Further, a predetermined area including at least the passage area of the illumination light IL10 may be filled with the liquid between the first and second transmissive flat plates P1 and P2, and in the scanning exposure apparatus, in particular, along the scanning direction. The liquid may flow. At this time, the liquid may be supplied and discharged between the first and second permeable flat plates P1 and P2 independently of the space between the second permeable flat plate P2 and the wafer W.

  In the exposure apparatus of this example, various translucent flat plates are arranged close to each other along the optical path, and there is a risk of adverse effects due to multiple interference accompanying surface reflection on each surface. Therefore, in this example, as the illumination lights IL1 to IL10 from the light source 1 in FIG. 1, light whose temporal coherence distance (coherence distance in the light traveling direction) is 100 [μm] or less is used. It is preferable to do. Thereby, it is possible to avoid generation of unnecessary interference fringes due to multiple interference.

The temporal coherence distance of light is a distance approximately represented by λ ² / Δλ, where λ is the wavelength of the light and Δλ is the half-value width of the wavelength distribution of the light. Therefore, when the exposure wavelength λ is 193 nm from the ArF laser, it is desirable to use the illumination light IL1 to IL10 whose wavelength half width Δλ is about 370 pm or more.
Moreover, it is desirable to use illumination light of 200 [nm] or less as the wavelengths of the illumination lights IL1 to IL10 in order to obtain a finer interference fringe pattern IF.

  In the above embodiment, for example, the movable mirrors 39, 89, 84 such as the laser interferometers 40, 86, 82 in FIG. 1 are fixed to the wafer stage 38 and the holders 36A, 37A as independent members. Instead of using the movable mirror, the side surface of the upper end portion of the wafer stage 38 and the side surfaces of the holders 36A and 37A themselves may be mirror-finished, and these mirror-finished reflecting surfaces may be used as the movable mirror. The same applies to the reference mirrors 91, 89, and 85.

  In the above embodiment, the position of the interference fringe pattern in the XY plane is measured. For example, based on the detection signal of the photoelectric detector 101 obtained by moving the measurement grating 95A at different positions in the Z direction. Alternatively, the position of the interference fringe pattern having the highest contrast in the Z direction may be determined, and the wafer W may be placed at the determined position to transfer the interference fringe pattern. In the above embodiment, the position and inclination of the measurement grating 95A and the wafer W in the Z direction are adjusted using a wafer surface position detection system (not shown). A sensor (for example, a laser interferometer) that measures the relative position in the Z direction between the transmissive flat plate P2 (or the holder 37A) and the wafer stage 38 may be used.

In addition, the configuration of the exposure apparatus (in particular, the illumination optical system IS and the transmissive flat plates P1 and P2) for forming the interference fringe pattern is not limited to the above embodiment, and the exposure apparatus of the interference exposure system for forming the interference fringe pattern. Then, the present invention can be applied.
The wafer W having the light and dark pattern exposed by the interference fringes as described above is transferred to the outside of the exposure apparatus by a wafer loader (not shown) and is transferred to the developing apparatus. By development, a resist pattern corresponding to the exposed light and dark pattern is formed on the photoresist on the wafer W. In the etching apparatus, a predetermined pattern is formed on the wafer W by etching the wafer W or a predetermined film formed on the wafer W using the resist pattern as an etching mask.

A manufacturing process of an electronic device such as a semiconductor integrated circuit, a flat panel display, a thin film magnetic head, or a micromachine includes a process of forming the fine pattern as described above over a plurality of layers. An electronic device can be manufactured by using the above-described exposure method by the exposure apparatus of the present invention in at least one of such multiple pattern forming steps.
In addition, in the at least one step, a pattern having a predetermined shape is formed by a general projection exposure apparatus on the photoresist PR on the wafer W that has been exposed to the bright and dark pattern by the interference fringes using the exposure method by the exposure apparatus of the present invention. And the pattern formation can be performed by developing the photoresist PR subjected to the synthetic exposure.

  In addition, this invention is not limited to the above-mentioned embodiment, Of course, a various structure can be taken in the range which does not deviate from the summary of this invention.

  The exposure method and apparatus of the present invention can be used in the manufacturing process of electronic devices such as semiconductor integrated circuits, flat panel displays, thin film magnetic heads, and micromachines.

1 is a partially cutaway view showing a schematic configuration of an exposure apparatus as an example of an embodiment of the present invention. (A) shows the first diffraction grating G11 formed on the first translucent flat plate P1, and (B) shows the second diffraction grating G21 formed on the second translucent flat plate P2. FIG. It is sectional drawing which shows the positional relationship of the 1st diffraction grating G11, the 2nd diffraction grating G21, and the wafer W, and the diffracted light LP, LM, LM1, LP1. 3 is a cross-sectional view showing an intensity distribution of interference fringes formed on a wafer W. FIG. (A) is a side view of the incident angle range of illumination light on the first light-transmitting flat plate viewed from the + X direction, (B) is a side view of the incident light range viewed from the −Y direction, and (C) is the aperture stop 28. FIG. It is a figure explaining the convergence divergence state of the illumination light to the 1st translucent flat plate P1 of FIG. It is a top view which shows an example of the illumination area | region of the interference fringe pattern in embodiment of this invention. FIG. 2 is a plan view showing a relationship between a wafer stage 38 and a wafer mark detection mechanism 43 in FIG. 1. It is a figure which shows the structural example of the interference fringe measurement system 41 in FIG. It is a flowchart which shows an example of the operation | movement at the time of alignment of the embodiment of this invention, and exposure. It is explanatory drawing of the measurement of the period of an interference fringe pattern, and correction | amendment operation | movement. It is sectional drawing which shows the positional relationship of 1st diffraction grating G11, G12, 2nd diffraction grating G21A, and wafer W in the other example of embodiment of this invention. It is a figure which shows the state which provided the 1st diffraction grating G13 and the 2nd diffraction grating G14 on each of both surfaces of the translucent flat plate P3. It is a figure which shows the example which provides the 2nd diffraction grating G16 substantially inside the 2nd translucent flat plate P6. (A) is explanatory drawing of the mechanism with which a liquid is filled only between the wafer W and the 2nd translucent flat plate P2, (B) is also liquid also between the translucent flat plate P2 and the translucent flat plate P1. It is explanatory drawing of the mechanism to satisfy | fill.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Light source, 2, 3, 4, 6 ... 1st lens group lens, 10 ... Condensing optical system, 13 ... Fly eye lens, 17 ... Illumination aperture stop, 29, 30, 32, 35 ... 4th lens Group lens P1, first translucent flat plate, P2 second translucent flat plate, 36A ... first holder, 37A ... second holder, W ... substrate (wafer), 38 ... wafer stage, 40 ... laser Interferometer, 41 ... interference fringe measurement system, G11 ... first diffraction grating, G21 ... second diffraction grating, IL1 to IL10 ... illumination light

Claims (20)

An exposure method for exposing a pattern on a photosensitive substrate with illumination light from a light source,
The illumination light is irradiated from the first diffraction grating side to the first diffraction grating and the second diffraction grating that are arranged to face each other, and the first diffraction grating is opposite to the first diffraction grating. A first step of forming interference fringes having periodicity in the first direction in a predetermined illumination field having a predetermined width in the direction;
A second step of measuring a position of the interference fringe by measuring a position of the interference fringe in the first direction at two positions separated by a first interval in the first direction on the interference fringe;
Thirdly exposing the exposure region on the substrate having the predetermined width in the first direction by the interference fringes while relatively scanning the interference fringes and the substrate in a second direction orthogonal to the first direction. Process,
The interference fringes and the substrate are relatively moved in the first direction by a second interval that is a natural number multiple of the period of the interference fringes obtained in the second step and is narrower than the predetermined width. And a fourth step to
At least at the edge of the exposure region in the first direction, the interference fringes and the substrate are placed in the second direction so as to overlap with the exposure pattern formed by the interference fringes exposed on the substrate in the third step. And a fifth step of exposing the substrate with the interference fringes while relatively moving.   The shape of the predetermined illumination field is an elongated shape in the first direction, and both end portions in the first direction are gradually narrowed toward the outside. Exposure method.   2. The exposure according to claim 1, wherein the predetermined illumination field has a shape elongated in the first direction, and light amounts at both ends in the first direction gradually decrease toward the outside. Method.   4. The exposure method according to claim 1, further comprising a sixth step of correcting a period of the interference fringes obtained in the second step before the third step. 5. .   The sixth step includes at least a step of controlling a telecentricity of the illumination light irradiated on the first diffraction grating and a step of controlling the height of the substrate in order to correct the period of the interference fringes. The exposure method according to claim 4, comprising one of them.   The second step uses the change in the amount of transmitted light from the measurement grating due to the relative movement in the first direction between the interference fringe and the measurement grating periodic in the first direction. The exposure method according to any one of claims 1 to 5, further comprising a step of measuring.   The said 1st diffraction grating is a diffraction grating which has a period twice as long as the period of the said 1st direction of the said interference fringe in the said 1st direction, It is any one of Claim 1 to 6 characterized by the above-mentioned. Exposure method.   The exposure method according to claim 1, wherein the second diffraction grating is a diffraction grating having a period in the first direction equal to a period of the interference fringes in the first direction. .   The exposure method according to claim 1, wherein the wavelength of the illumination light is 200 nm or less.   In the third step and the fifth step, at least one of a gap between the second diffraction grating and the substrate and a gap between the first diffraction grating and the second diffraction grating is filled with a liquid. The exposure method according to claim 1, wherein the interference fringe pattern is exposed on the substrate in a state. Further comprising measuring pattern position information of a pattern previously formed on the photosensitive substrate;
6. The exposure method according to claim 4, wherein in the sixth step, a period of the interference fringes is corrected using the pattern position information. A method of manufacturing an electronic device having a pattern forming step,
12. A device manufacturing method using the exposure method according to claim 1 in the pattern forming step. A pattern of interference fringes having periodicity in the first direction generated by irradiating the first diffraction grating and the second diffraction grating arranged opposite to each other with illumination light from the light source is formed on the photosensitive substrate. An exposure apparatus for exposing,
An illumination optical system for irradiating the first diffraction grating with the illumination light from the light source;
A substrate holding mechanism for holding the substrate;
A period measurement mechanism for measuring a period of the interference fringes in the first direction;
A stage mechanism that relatively moves the interference fringes and the substrate holding mechanism in a second direction orthogonal to the first direction, and relatively moves the interference fringes and the substrate holding mechanism in the first direction. When,
And a control device that drives the stage mechanism to move the interference fringe and the substrate holding mechanism relative to each other in the first direction based on the period of the interference fringe measured by the period measurement mechanism. An exposure apparatus characterized by that.   The periodic measurement mechanism includes a measurement grating that is provided in the substrate holding mechanism and that is periodic in the first direction, and a photoelectric detector that detects transmitted light from the measurement grating. The exposure apparatus described in 1.   The position of the interference fringe is measured based on a detection signal of the photoelectric detector when the interference fringe and the measurement grating are relatively moved in the first direction by the stage mechanism. The exposure apparatus described.   The exposure apparatus according to claim 13, further comprising a correction device that corrects the period of the interference fringes in the first direction.   The correction device includes at least one of a member that controls the telecentricity of the illumination light with respect to the first diffraction grating and a height control device that controls the height of the substrate. The exposure apparatus described in 1. The first diffraction grating is a diffraction grating having a period twice as long as the period of the interference fringes in the first direction in the first direction;
18. The exposure apparatus according to claim 13, wherein the second diffraction grating is a diffraction grating having a period equal to a period of the interference fringes in the first direction in the first direction. . While having a position measuring mechanism for measuring pattern position information of a pattern previously formed on the substrate,
The exposure apparatus according to claim 16 or 17, wherein the correction device corrects the period of the interference fringes based on a measurement result of the position measurement device. The liquid filling mechanism for filling at least one of a gap between the first diffraction grating and the second diffraction grating and a gap between the second diffraction grating and the substrate. Item 20. The exposure apparatus according to any one of Items 13 to 19.

Images