JP6745647B2 - Scanning exposure apparatus and article manufacturing method - Google Patents

Description

The present invention relates to a scanning exposure apparatus and an article manufacturing method.

In the manufacturing process of devices such as semiconductor devices and display devices, a scanning exposure apparatus that exposes a substrate while scanning the original plate and the substrate may be used. In the scanning exposure apparatus, the original plate and the substrate are scanned with exposure light having an elongated (for example, rectangular) cross-sectional shape in a direction orthogonal to the scanning direction of the original plate and the substrate. When a light source that generates pulsed light is used as a light source that generates exposure light, it is necessary to form a light intensity distribution having a trapezoidal shape along the scanning direction on the substrate in order to reduce unevenness in exposure amount. is there. Therefore, the light shielding member can be arranged at a position slightly deviated from the surface conjugate with the original plate. The light shielding member forms an inclined portion in the trapezoidal light intensity distribution. Here, the inclined portion means a right triangle in which a trapezoidal leg (two sides other than the upper base and the lower base) is a hypotenuse and a part of the lower base of the trapezoid is an adjacent side.

In Patent Document 1, a light blocking member is arranged in the vicinity of a surface conjugate with the mask, and the position of this light blocking member is changed to change the width of a portion (inclined portion) corresponding to the hypotenuse in the trapezoidal light intensity distribution. It is described to do. Further, Patent Document 1 describes that the width of the inclined portion is changed according to the pattern of the device to be manufactured. Patent Document 2 describes that the illuminance distribution in the scanning direction is measured and the position of the aperture stop is adjusted according to the measurement result. In Patent Document 3, the relationship between the number of pulses received by the substrate and the exposure amount error (exposure spot) is calculated according to the shape of the light intensity distribution, and the exposure is performed while the substrate moves in the scanning direction by a unit amount. A method is disclosed in which the number of received light pulses is controlled so that the size and slope of the spots are below a threshold value.

JP, 2011-40716, A JP, 2011-29672, A JP, 2010-21211, A

The light rays forming the inclined portion in the trapezoidal light intensity distribution have asymmetry with respect to the optical axis, and thus can cause positional deviation of the pattern formed on the substrate. In the past, the width of the inclined portion was simply changed according to the pattern of the device to be manufactured. However, in such a method, it is necessary to optimize the width of the inclined portion by repeating the trial exposure many times.

An object of the present invention is to provide an advantageous technique for predicting the positional deviation of the pattern formed on the substrate.

One aspect of the present invention relates to a scanning exposure apparatus that exposes a substrate with exposure light that forms a light intensity distribution having a trapezoidal shape along a scanning direction on the substrate. A detecting unit for detecting an image forming position error of a light ray forming the inclined portion in the shape , wherein the detecting unit detects a pupil plane light intensity distribution observed from a position where the light ray forming the inclined portion is incident. A detector and a processing unit that obtains an imaging position error based on the pupil plane light intensity distribution detected by the photodetector .

According to the present invention, an advantageous technique for predicting the positional deviation of the pattern formed on the substrate is provided.

The figure which shows the structure of the scanning exposure apparatus of one Embodiment of this invention. The figure which shows the state which the original plate is illuminated by exposure light. The figure which shows the method of determining the width|variety of the inclination part in trapezoidal light intensity distribution. The figure which shows the image formation state at the time of detecting a ring-shaped exposure light with a photodetector. The figure which shows the scanning exposure apparatus at the time of measuring a pupil surface intensity distribution. FIG. 6 is a diagram illustrating an image forming position error when the width of the inclined portion in the trapezoidal light intensity distribution is large. FIG. 6 is a diagram illustrating an image forming position error when the width of the inclined portion in the trapezoidal light intensity distribution is small. The figure which shows the exposure amount distribution by one irradiation of pulsed light. The figure which illustrates the trapezoidal light intensity distribution. A diagram showing the relationship between the number of received light pulses and the exposure amount error, The figure explaining measurement of the width of the inclination part in trapezoidal light intensity distribution. The figure explaining operation|movement of the scanning exposure apparatus of one Embodiment of this invention. The figure which shows the principle which changes the width|variety of the inclination part in trapezoidal light intensity distribution. The figure explaining the relationship between exposure amount error, irradiation pulse number, and required accuracy. The figure explaining a quadrupole illumination mode. The figure which shows the relationship between a trapezoidal light intensity distribution and a pupil surface light intensity distribution. The figure which shows the relationship between the deviation (telecentric characteristic) of a light amount gravity center, and an imaging position error. FIG. 6 is a diagram showing a relationship between a width of an inclined portion in a trapezoidal light intensity distribution and an image forming position error.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings.

FIG. 1 shows the configuration of a scanning exposure apparatus SS according to an embodiment of the present invention. The scanning exposure apparatus SS exposes the substrate with exposure light that forms a light intensity distribution having a trapezoidal shape along the scanning direction (Y direction) on the substrate. The scanning exposure apparatus SS scans the original 113 and the substrate 115 with exposure light having a slit shape on the original 113 and the substrate 115, and exposes each shot area of the substrate 115 via the original 113. For example, a gas such as KrF is enclosed, and the pulsed laser light source 101 generates exposure light having a wavelength of 248 nm in the far ultraviolet region, for example. A gas exchange operation of the pulsed laser light source 101, an operation for stabilizing the wavelength, a discharge applied voltage, and the like are controlled by the laser control device 102. The laser control device 102 may be configured to control the pulsed laser light source 101 according to an instruction from the main control device 103, for example.

The exposure light emitted from the pulsed laser light source 101 passes through a shaping optical system (not shown) of the illumination optical system 104 into a predetermined shape (circular shape, ring shape, quadrupole shape, dipole shape, etc.). After being shaped, the light enters the integrator lens 105. The exposure light forms a plurality of secondary light sources on the exit surface of the integrator lens 105. The condenser lens 107 may have a function of changing the illuminance distribution of the original plate (reticle) 113. The condenser lens 107 illuminates the variable slit mechanism 110 whose exposure width is changeable with exposure light from the exit surface (secondary light source) of the integrator lens 105. The variable slit mechanism 110 is arranged on the conjugate plane MB which is a plane conjugate with the original plate RP or the imaging plane WP. The aperture of the aperture stop 106 defines the numerical aperture (NA) of the illumination optical system. The aperture of the aperture stop 106 can be substantially circular. The diameter of the aperture of the aperture stop 106, that is, the numerical aperture (NA) of the illumination optical system is set by the illumination system control device 108. The illumination system control device 108 can set the σ value by controlling the aperture of the aperture stop 106 of the illumination optical system 104.

A half mirror 111 is arranged on the optical path of the illumination optical system 104, and a part of the exposure light illuminating the original 113 is reflected by the half mirror 111 and taken out. A photo light detector 109 is arranged on the optical path of the reflected light of the half mirror 111, and the photo sensor 109 generates an output corresponding to the intensity of the exposure light (exposure energy amount). The output of the photo sensor 109 is converted into an exposure energy amount per pulse by an integration circuit (not shown) that performs integration for each pulse emission of the pulse laser light source 101, and is sent to the main controller 103 via the illumination system controller 108. Provided.

A pattern corresponding to the circuit pattern of the semiconductor element is formed on the original 113 and is illuminated by the illumination optical system 104. The two-dimensional variable blade 112 has a plurality of movable light shielding members in a plane orthogonal to the optical axis LA, and by adjusting the positions thereof, the exposure light irradiation area with respect to the pattern area of the original 113 can be arbitrarily set. It is possible to set. FIG. 2 shows a state in which the original 113 is illuminated by the exposure light 203. The slit-shaped exposure light (illumination light) 203 formed by the variable slit mechanism 110 and the variable blade 112 illuminates a part of the pattern area 202 indicated by diagonal lines. The variable slit mechanism 110 has a pair of light blocking members arranged in the vicinity of the conjugate plane MB, and has a function of changing the opening width which is the interval between the pair of light blocking members. The exposure amount distribution of the illumination light 203 in the non-scanning direction (long side) is controlled by controlling the opening width. The variable slit mechanism 110 has, at a position near the conjugate plane MB, a mechanism that moves a pair of light blocking members in a direction along the optical axis LA at the position. The exposure light 203 forms a light intensity distribution having a trapezoidal shape along the scanning direction (Y direction) on the substrate 115 by moving the pair of light shielding members away from the conjugate plane MB. By changing the distance between the conjugate plane MB and the pair of light shielding members, the width of the inclined portion in the trapezoidal light intensity distribution in the scanning direction can be changed. Therefore, the variable slit mechanism 110 can be understood as a changing mechanism that changes the width of the inclined portion in the trapezoidal light intensity distribution in the scanning direction.

The projection optical system 114 reduces and projects a part of the pattern area 202 on the substrate 115 coated with the photoresist at a reduction magnification β (β is, for example, ¼). In this state, the original stage 123 and the substrate stage 116 scan the exposure light 203 in opposite directions (Y: scanning direction) at the same speed ratio as the reduction ratio β of the projection optical system 114. The pulsed laser light source 101 repeats pulsed light emission. As a result, the entire pattern area 202 of the original 113 is transferred to the shot area (the shot area includes one or a plurality of chip areas) on the substrate 115. In FIG. 2, when the axis parallel to the optical axis of the projection optical system 114 is the Z axis, the substrate 115 or a substrate stage 116 described later is scanned during exposure among two axes that are orthogonal to each other and orthogonal to each other. The axis parallel to the scanning direction, which is the direction, is the Y axis, and the remaining axis is the X axis.

The projection optical system 114 has a movable optical element 127. The movable optical element 127 is held by the lens barrel 130 of the projection optical system 114, and can be driven by the drive mechanism 128 in the optical axis direction of the projection optical system 114. The drive mechanism 128 may be configured to drive the movable optical element 127 by, for example, pneumatic pressure or a piezoelectric element. By adjusting the position of the movable optical element 127 in the optical axis direction of the projection optical system 114, the projection magnification and/or the distortion error of the projection optical system 114 can be adjusted.

The substrate stage 116 holds the substrate 115 and is driven by the drive mechanism 119 to move in the optical axis direction (Z direction) of the projection optical system 114 and in the plane (XY plane) orthogonal to the optical axis direction. be able to. The substrate stage 116 can also be driven for rotation about the Z axis, the X axis, and the Y axis. A movable mirror 117 is provided on the substrate stage 116, and the position or displacement of the movable mirror 117 is measured by the laser interferometer 118. As a result, the position of the substrate stage 116 in the X and Y directions is detected. The substrate stage controller 120 controlled by the main controller 103 controls the drive mechanism 119 based on the position of the substrate stage 116 detected by using the laser interferometer 118, so that the position and movement of the substrate stage 116 are controlled. Control. A photodetector 135 is mounted on the substrate stage 116.

FIG. 3 illustrates a method of determining the width of the inclined portion IP in the light intensity distribution having a trapezoidal shape along the scanning direction. The width of the inclined portion IP is changed by moving the pair of light shielding members 110a and 110b of the variable slit mechanism 110 in the direction along the optical axis LA. In FIG. 3A, the exposure light emitted from the condenser lens 107 passes through the opening between the pair of light blocking members 110a and 110b, and is trapezoidal on the conjugate plane MB conjugate to the original plate RP or the image forming plane WP. A light intensity distribution 301 having a shape is formed. In FIG. 3A, the pair of light blocking members 110a and 110b are arranged at positions at a distance s2 from the conjugate plane MB. The upper light ray 302a emitted from the condenser lens 107 is blocked by the light blocking member 110a, and the lower light ray 302b is blocked by the light blocking member 110b. FIG. 3B shows a light intensity distribution 304 formed on the image plane WP surface in the arrangement of FIG. The light intensity distribution 304 has a trapezoidal shape along the scanning direction. The horizontal axis of FIG. 3B shows the position in the scanning direction, and the vertical axis of FIG. 3B shows the energy amount E of the exposure light. The width 305 of the inclined portion IP in the light intensity distribution 304 can be controlled by the distance s2 of the light shielding members 110a and 110b from the position MB conjugate with the original plate surface RP or the image plane WP.

FIG. 3C shows a light intensity distribution 306 when the pair of light shielding members 110a and 110b of the variable slit mechanism 110 are brought closer to the conjugate plane MB from the state of FIG. 3A. In FIG. 3C, the pair of light blocking members 110a and 110b are arranged at a position of a distance s1 from the conjugate surface placement MB. The upper light beam 307a emitted from the condenser lens 107 is blocked by the light blocking member 110a, and the lower light beam 307b is blocked by the light blocking member 110b to form a trapezoidal light intensity distribution 306. FIG. 3D shows a light intensity distribution 309 formed on the image plane WP surface in the arrangement of FIG. 3C. The light intensity distribution 309 has a trapezoidal shape along the scanning direction. The width of the inclined portion IP in the trapezoidal shape of the light intensity distribution 309 of FIG. 3D is smaller than the width of the inclined portion IP in the trapezoidal shape of the light intensity distribution 309 of FIG. 3B.

Next, the configuration of the photodetector 135 that measures the pupil plane light intensity distribution of the exposure light (illumination light) and the method of measuring the pupil plane light intensity distribution will be described. FIG. 5 shows the scanning exposure apparatus SS when measuring the pupil plane intensity distribution of the illumination light 501. In FIG. 5, the horizontal direction is the scanning direction, and the pupil plane intensity distribution corresponding to the position Ph in the light intensity distribution 501 formed in the illumination optical system 104 is detected by the photodetector 135 mounted on the substrate stage 116. It The photodetector 135 has a pinhole on the imaging plane WP, and the laser interferometer 118 is arranged so that the pinhole of the photodetector 135 is arranged at the position −Yh corresponding to the position Ph in the light intensity distribution 501. The substrate stage 116 is driven by the drive mechanism 119 of the substrate stage 116. Further, the original stage 123 is provided with a dedicated plate 136 having a pinhole, and the original stage 123 is arranged so that the pinhole of the dedicated plate 136 is arranged at a position rh corresponding to the position Ph in the light intensity distribution 501. Is driven.

FIG. 4A shows an image formation state when the ring-shaped exposure light is detected by the photodetector 135. The photodetector 135 has a pinhole 403. The pinhole 403 has a diameter of, for example, several tens of μm. The pinhole 403 is arranged on the image plane WP of the projection optical system 114. A pupil plane light intensity distribution 401 according to the illumination mode set in the illumination optical system 104 is formed on the pupil plane of the projection optical system 114 by the exposure light. In the example of FIG. 4A, the illumination mode is annular illumination, and the pupil surface light intensity distribution 401 has an annular shape. The exposure light emitted from the pupil plane of the projection optical system 114 passes through the pinhole 403 of the photodetector 135, and a light intensity distribution 404 equivalent to the pupil plane light intensity distribution 401 is transmitted to the light receiving section 405 of the photodetector 135. Form. Therefore, the photodetector 135 can observe the pupil plane light intensity distribution 401 from an arbitrary position on the imaging plane WP (the position where the light rays from the pupil plane are incident).

The light receiving unit 405 includes, for example, a two-dimensional image light detector (for example, a CCD sensor), and outputs an image obtained by capturing a light intensity distribution 404 equivalent to the pupil plane intensity distribution 401. That is, the image output from the light receiving unit 405 is an image showing the pupil plane intensity distribution 401. FIG. 4B illustrates the pixel array 406 of the light receiving unit 405. The XY directions in FIG. 4B match the XY directions presented in the scanning exposure apparatus SS. The light intensity distribution 404 in FIG. 4A is shown as the light intensity distribution 407 in FIG. 4B. Pixel values of a plurality of pixels forming the pixel array 406 are represented by A(x, y). Coordinate values in the x and X directions, and y are coordinate values in the Y direction.

In the present embodiment, the image forming position error of the light ray forming the inclined portion in the trapezoidal light intensity distribution is detected. The image forming position error can be detected based on the pupil plane light intensity distribution observed from the position where the light rays forming the inclined portion in the trapezoidal light intensity distribution are incident. More specifically, the imaging position error can be detected based on the asymmetry of the pupil plane light intensity distribution (for example, the deviation of the center of gravity). The image forming position error detected in this manner can be used as an index value indicating the amount of positional deviation of the pattern formed on the substrate. Therefore, the width of the inclined portion can be determined so as to satisfy the required specifications (allowable positional deviation of the pattern) based on the image forming position error of the light beam forming the inclined portion in the trapezoidal light intensity distribution.

In the present embodiment, the photodetector 135 and the processing unit 200 that processes the signal provided from the photodetector 135 constitute the detection unit DP. The detection unit DP is configured to detect an image forming position error of a light ray forming an inclined portion in a light intensity distribution having a trapezoidal shape along the scanning direction. The function of the processing unit 200 is provided by the main control unit 103 in this embodiment. However, the processing unit 200 may be provided separately from the main control unit 103, or may be incorporated in the photodetector 135.

FIG. 6A illustrates a light intensity distribution 601 formed on the image plane WP when the width of the inclined portion is relatively large. The horizontal axis represents the position in the scanning direction (Y direction), and the vertical axis represents the energy amount E of the exposure light. The inclined portions 602 and 603 in the light intensity distribution 601 are formed by blocking a part of the light beam by the light blocking members 110a and 110b of the variable slit mechanism 110, as described with reference to FIG. A part of the exposure light to be incident on the positions P1, P2, P3 in the inclined portion 602 is blocked by one light shielding member of the variable slit mechanism 110, and the exposure light to be incident on the positions P4, P5, P6 in the inclined portion 603. Is partially blocked by the other light blocking member of the variable slit mechanism 110. As a result, a trapezoidal light intensity distribution 601 is formed on the image plane WP. The exposure light to be incident on the central portion (for example, the position Pa on the optical axis) where the intensity distribution between the inclined portion 602 and the inclined portion 603 is flat passes through the opening between the light blocking members 110a and 110b of the variable slit. So you will not be blocked.

FIG. 6B illustrates the pupil plane light intensity distribution observed by the photodetector 135 from the positions P1, P2, P3, P4, Pa, P5, and P6 in the scanning direction in the circular illumination mode (normal illumination mode). Has been done. Here, the positions P1, P2, P3, P4, P5, and P6 are positions where the light rays forming the inclined portion are incident. The horizontal direction indicates the scanning direction (Y direction), and the vertical direction indicates the direction (X direction) orthogonal to the scanning direction. The pupil plane light intensity distribution observed from each of the above positions is shown in gray.

The pupil plane light intensity distribution observed from the position Pa on the optical axis (center O) is circular because the exposure light is not blocked by the light blocking members 110a and 110b of the variable slit mechanism 110, and the light amount center of gravity Ga Coincides with the optical axis LA, and the center of gravity is not biased. The pupil plane light intensity distribution observed from the positions P1, P2, and P3 in the inclined portion 602 has an asymmetrical shape (a biased shape) because part of the exposure light is blocked by the light blocking members 110a and 110b of the variable slit mechanism 110. ), the light amount centroids G1, G2, and G3 deviate from the center O. In the pupil plane light intensity distribution observed at the position P3 closer to the center of the inclined portion 602, the exposure light blocking amount is small, and the light amount center of gravity G3 is biased by +g3 in the scanning direction. In the pupil plane light intensity distribution observed at the position P1 near the end of the inclined portion 602, the amount of blocking of the exposure light amount is large, and the light amount centroid G1 is biased by +g1 in the scanning direction. As the blocking amount of the exposure light increases from the position P3 of the inclined portion 602 toward the position P1, the deviation amount of the light amount gravity center G also increases, and the telecentric characteristic of the exposure light deteriorates. As for the pupil plane intensity distribution at the positions P4, P5, and P6 of the inclined portion 603, similarly to the inclined portion 602, the blocking amount of the exposure light increases from the position P4 to P6, and the deviation amount of the light amount gravity center G increases. , The telecentric characteristic of exposure light deteriorates. Since the inclined portions 602 and 603 of the light intensity distribution 601 are relatively large, the region where the deviation of the light amount centroid is large is large.

FIG. 6C schematically shows an image forming position error caused by deterioration of the telecentric characteristic of exposure light (ray inclination). In FIG. 6C, the horizontal axis represents the scanning direction and the vertical axis represents the Z direction (focus direction). FC indicated by an arrow indicates a running surface of the substrate when scanning exposure is performed on the substrate while continuously irradiating the exposure light forming the light intensity distribution 601 as shown in FIG. Move along). When a certain point on the substrate moves from the position P6 (exposure start position) to the position P1 (exposure end position), it is indicated by z1 to z6 between the running surface FC (substrate surface) and the imaging surface WP. A focus difference dz occurs. Due to the deviations g1 to g6 of the light amount gravity center G shown in FIG. 6B, ray tilts L1 to L6 due to the telecentric characteristic of the exposure light are generated, and a coupling position error dy (shift) indicated by y1 to y6 occurs. The pupil plane light intensity distribution is observed from the position where the light rays forming the inclined portions 602 and 603 of the light intensity distribution 601 are incident, the deviation dg (g1 to g6) of the light amount centroid of the pupil surface light intensity distribution is obtained, and the deviation The telecentric characteristics (light ray inclinations L1 to L6) of the exposure light can be obtained based on Then, the imaging position error dy when the focus difference dz occurs can be obtained based on the telecentric characteristic. In the example of FIG. 6, since the widths of the inclined portions 602 and 603 of the light intensity distribution 601 are relatively large, the region in which the deviation of the light amount centroid occurs is large. Therefore, the area in which the telecentric characteristic of the exposure light deteriorates becomes large, and the imaging position error easily deteriorates.

FIG. 7A illustrates a light intensity distribution 701 formed on the image plane WP when the width of the inclined portion is relatively small. The horizontal axis represents the position in the scanning direction (Y direction), and the vertical axis represents the energy amount E of the exposure light. The inclined portions 702 and 703 in the light intensity distribution 701 are formed by blocking a part of the light beam by the light blocking members 110a and 110b of the variable slit mechanism 110, as described with reference to FIG. A part of the exposure light that should enter the positions P1 and P2 in the inclined portion 702 is blocked by the light blocking members 110a and 110b of the variable slit mechanism 110. Further, a part of the exposure light to be incident on the positions P5 and P6 in the inclined portion 603 is blocked by the other light blocking member of the variable slit mechanism 110. As a result, a trapezoidal light intensity distribution 701 is formed on the image plane WP. The exposure light to be incident on the central portion (for example, the position P3, the position Pa, the position P4) where the intensity distribution between the inclined portion 602 and the inclined portion 603 is flat is between the light shielding members 110a and 110b of the variable slit mechanism 110. Since it passes through the opening, it is not blocked.

FIG. 7B illustrates the pupil plane light intensity distribution observed by the photodetector 135 from the positions P1, P2, P3, P4, Pa, P5, and P6 in the scanning direction in the circular illumination mode (normal illumination mode). Has been done. Here, the positions P1, P2, P5, and P6 are positions where the light rays forming the inclined portion are incident. The horizontal direction indicates the scanning direction (Y direction), and the vertical direction indicates the direction (X direction) orthogonal to the scanning direction. The pupil plane light intensity distribution observed from each of the above positions is shown in gray.

The pupil plane light intensity distribution observed from the position P3, the position Pa (position on the optical axis), and the position P4 is circular because the exposure light is not blocked by the light blocking members 110a and 110b of the variable slit mechanism 110. The light amount center of gravity Ga and the optical axis LA coincide with each other, and the center of gravity is not biased. The pupil plane light intensity distribution at the positions P1 and P2 in the inclined portion 702 has an asymmetrical shape (biased shape) because a part of the exposure light is blocked by the light blocking members 110a and 110b of the variable slit mechanism 110, and the light amount center of gravity is obtained. G1 and G2 are deviated from the center O. In the pupil plane light intensity distribution observed at the position P2 closer to the central portion of the inclined portion 602, the exposure light blocking amount is small, and the light amount center of gravity G2 is biased by +g2 in the scanning direction. In the pupil plane light intensity distribution observed at the position P1 near the end of the inclined portion 602, the amount of blocking of the exposure light amount is large, and the light amount centroid G1 is biased by +g1 in the scanning direction. As the blocking amount of the exposure light increases from the position P2 of the inclined portion 602 toward the position P1, the deviation amount of the light amount gravity center G also increases, and the telecentric characteristic of the exposure light deteriorates. As for the pupil plane intensity distribution at the positions P5 and P6 of the tilted portion 603, similarly to the tilted portion 702, the blocking amount of the exposure light increases from the position P5 to the position P6, and the deviation amount of the light amount gravity center G also increases, resulting in the exposure. The telecentricity of light deteriorates. Since the inclined portions 702 and 703 of the light intensity distribution 701 are relatively small, the region where the deviation of the light amount center of gravity occurs is small.

FIG. 7C schematically shows an image forming position error caused by deterioration of the telecentric characteristic of exposure light (light ray inclination). In FIG. 7C, the horizontal axis represents the scanning direction and the vertical axis represents the Z direction (focus direction). FC shown by an arrow represents the running surface of the substrate when scanning exposure is performed on the substrate while continuously irradiating the exposure light forming the light intensity distribution 701 of FIG. When a certain point on the substrate moves from the position P6 (exposure start position) to the position P1 (exposure end position), it is indicated by z1 to z6 between the running surface FC (substrate surface) and the imaging surface WP. A focus difference dz occurs. Due to the deviations g1 to g6 of the light amount gravity center G shown in FIG. 7B, ray tilts L1 to L6 due to the telecentric characteristic of the exposure light are generated, and a coupling position error dy (shift) indicated by y1 to y6 occurs. Due to the deviations g1 to g6 of the light amount gravity center G shown in FIG. 7B, ray tilts L1 to L6 due to the telecentric characteristic of the exposure light are generated, and a coupling position error dy (shift) indicated by y1 to y6 occurs. The pupil plane light intensity distribution is observed from the position where the light rays forming the inclined portions 702 and 703 of the light intensity distribution 701 are incident, and the deviation dg (g1 to g6) of the light quantity center of gravity of the pupil surface light intensity distribution is obtained, and the deviation The telecentric characteristics (light ray inclinations L1 to L6) of the exposure light can be obtained based on Then, the imaging position error dy when the focus difference dz occurs can be obtained based on the telecentric characteristic. In the example of FIG. 7, since the widths of the inclined portions 702 and 703 of the light intensity distribution 701 are relatively small, the region where the deviation of the light amount centroid is generated is small. Therefore, the region where the telecentric characteristic of the exposure light is deteriorated is small, The imaging position error becomes small.

Next, a method for obtaining the exposure amount error (unevenness of exposure) in the scanning direction from the trapezoidal light intensity distribution will be described. FIG. 8A shows a shot area when pulsed light is intermittently irradiated while the substrate 115 is moving in the scanning direction Y. In scanning exposure, the pulsed light is integrated while the substrate 115 is displaced by ΔY in the scanning direction for each pulse. When the oscillation frequency of the pulsed laser light source 101 that determines the period of the pulsed light is f and the scanning speed of the substrate 115 is v, the displacement amount ΔY is expressed by (Equation 1).

ΔY=v/f (1)
FIG. 8B shows the light intensity distribution 901 in the scanning direction, and shows the relationship with the displacement amount ΔY for each pulse. In FIG. 8B, the horizontal axis represents the coordinate value of the substrate 115 in the scanning direction Y, and the vertical axis represents the energy amount E (exposure amount) of the exposure light. The energy amount E (exposure amount) is proportional to the light intensity. On the substrate 115, the exposure amounts e1 to e9 are integrated for each pulse for each ΔY section. The result of integrating the exposure amounts of e1 to e9 for each section of ΔY indicates the difference in exposure amount (exposure unevenness) between the sections of ΔY.

The relationship between the exposure amount error in the scanning direction, the shape of the light intensity distribution, and the number of received light pulses will be described with reference to FIGS. 9 and 10. FIG. 9 shows a trapezoidal light intensity distribution in which the width of the inclined portion is L1 and the width of the central portion having a constant light intensity is L2. FIG. 10 illustrates the relationship between the number of light receiving pulses per unit length (1/ΔY [pulse/mm]) and the exposure amount error (exposure amount unevenness). As described in Patent Document 3, when the number of light-receiving pulses (1/ΔY) on the substrate is short, the exposure error is small at 1/(L1+L2), and the exposure error is small at long period 1/L1. However, there exists. The exposure amount error changes according to the shape of the light intensity distribution (L1, L2) and the number of light receiving pulses per unit length (1/ΔY).

FIG. 12 exemplifies the operation of the scanning exposure apparatus SS. The operation shown in FIG. 12 can be executed or controlled by the main control unit 103. In step S01, the main control unit 103 sets an illumination condition, an exposure amount, and an exposure mode as conditions for exposing the substrate 115 based on information input by an operator. The exposure mode may include an image formation position priority mode that prioritizes reduction of the image formation position error and an exposure amount distribution priority mode that has reduction of the exposure amount distribution error.

In step S02, the main control unit 103 forms an illumination shape based on the illumination condition set in step S01. Selectable illumination shapes can include, for example, circular shapes, ring shapes, and multipole shapes.

In steps S03 to S05, the main control unit 103 (processing unit 200) detects the imaging position error while changing the width of the inclined portion in the light intensity distribution having a trapezoidal shape along the scanning direction, and the inclined portion is thereby detected. And the relationship between the image forming position error and the image forming position error. Here, the main control unit 103 can detect the image formation position measurement error based on the pupil plane intensity distribution observed from the position where the light rays forming the inclined portion are incident. Hereinafter, an example in which the width of the inclined portion is changed in three steps will be described.

In step S03, the main control unit 103 sets or changes the width of the inclined portion. In this example, as illustrated in FIGS. 13A1 to 13A3, the width of the inclined portion is changed by moving the light blocking members 110a and 110b of the variable slit mechanism 110 in the direction along the optical axis LA. .. FIG. 13(a1) shows a state in which the light blocking members 110a and 110b are controlled to the position of the distance s1 from the conjugate plane MB, and at this time, the width of the inclined portion becomes the minimum value. FIG. 13A2 shows a state in which the light blocking members 110a and 110b are controlled to the position of a distance s2 from the conjugate plane MB surface, and at this time, the width of the inclined portion becomes the maximum value. FIG. 13(a3) shows a state in which the light blocking members 110a and 110b are controlled to a position of a distance s3 from the conjugate plane MB, and at this time, the width of the inclined portion has an intermediate value between the minimum value and the maximum value. Become.

In step S04, the main control unit 103 uses the photodetector 135 to measure the width (range) of the inclined portion. FIG. 11 schematically shows how the width of the inclined portion is measured using the photodetector 135. A trapezoidal light intensity distribution 1101 is formed on the original plate surface RP, and a trapezoidal light intensity distribution 1102 is formed on the imaging plane WP so as to correspond to the light intensity distribution 1101. 11A and 11C show a state in which measurement of the width of the inclined portion is started, and in this example, from one end Ps of the light intensity distribution 1101 (one end of the light intensity distribution 1102) from outside. Measurement starts. FIGS. 11B and 11D show a state in which the measurement of the width of the inclined portion is completed. In this example, the other end Pe of the light intensity distribution 1101 (the other end of the light intensity distribution 1102) is shown. The measurement ends outside. From the start of measurement in FIG. 11A to the end of measurement in FIG. 11B, the photodetector 135 of the substrate stage 116 is scan-driven from −Ys to +Ye, and the exposure light is pulsed to the substrate stage 116. Is intermittently irradiated and the exposure light is detected by the photodetector 135. The exposure light can be detected as a pupil plane light intensity distribution by the photodetector 135, but the information to be obtained in step S04 indicates that the exposure light having a predetermined light amount or brightness has entered the photodetector 135. It is only necessary to use information, and it is not necessary to use the pupil plane light intensity distribution.

As described above, in step S04, the main control unit 103 controls the position of the photodetector 135 so that the photodetector 135 is sequentially arranged at a plurality of positions in the scanning direction, and controls the photodetector 135 at each of the plurality of positions. The range in which the inclined portion is formed is specified based on the output of the photodetector 135. The range in which the width of the inclined portion is measured (the range from the position where the measurement starts to the position where the measurement ends) is determined so that the range includes the entire area of the inclined portion. Such a range can be determined by, for example, adding a margin to the width of the designed inclined portion determined by the positions of the light blocking members 110a and 110b of the variable slit mechanism 110.

FIG. 11E shows a light intensity distribution 1101 on the image formation plane W detected by the photodetector 135. Based on the position of the substrate stage 116 and the light intensity detected by the light detection unit 135, the width Ls of the inclined portion on one end side of the light intensity distribution 1102 and the width Le of the inclined portion on the other end side of the light intensity distribution 1102. And can be detected or measured.

FIG. 13B1 shows a light intensity distribution 1301 formed on the image plane WP in the state of FIG. 13A1. The widths (minimum widths) of the two inclined portions of the light intensity distribution 1301 are Ls1 and Le1. FIG. 13(a2) shows a light intensity distribution 1302 formed on the image plane WP in the state of FIG. 13(a2). The widths (maximum widths) of the two inclined portions of the light intensity distribution 1302 are Ls2 and Le2. A light intensity distribution 1303 formed on the image plane WP in the state of FIG. 13A3 is shown. The widths of the two inclined portions of the light intensity distribution 1303 are Ls3 and Le3.

In step S05, the main controller 103 arranges the photodetector 135 at a predetermined position within the range of the width L of the inclined portion measured in step S04, causes the photodetector 135 to detect the pupil plane light intensity distribution, and An image forming position error is obtained based on the pupil plane light intensity distribution. Here, for example, the deviation (telecentric characteristic) of the light amount centroid can be obtained based on the pupil plane light intensity distribution, and the imaging position error can be obtained based on this deviation.

In FIG. 16A, light intensity distributions 1301, 1302, 1303 obtained by changing the width of the inclined portion in three steps are shown in an overlapping manner. On the one end Ps side, the photodetector unit 135 may be arranged at a predetermined position in a common range within each of the widths Ls1, Ls2, and Ls3, and the photodetector unit 135 may detect the pupil surface light intensity distribution. Similarly, on the other end Pe side, the photodetector unit 135 may be arranged at a predetermined position in a common range within each of the widths Le1, Le2, and Le3, and the photodetector unit 135 may detect the pupil surface light intensity distribution. ..

Here, in order to measure the difference in the deviation of the center of gravity of the light amount due to the difference in the width of the inclined portion with higher sensitivity, on the one end Ps side, a predetermined position in the common range within each range of the widths Ls1, Ls2, and Ls3 (Position for detecting the pupil surface light intensity distribution) is preferably Ps1. Similarly, on the one end Pe side, it is preferable that the predetermined position (the position for detecting the pupil surface light intensity distribution) in the common range within the widths Le1, Le2, and Le3 is Pe1.

FIG. 16B shows a pupil plane light intensity distribution detected by (a pinhole of) the photodetector 135 at the positions Ps1 and Pe1. FIG. 16B shows a pupil plane light intensity distribution detected by (a pinhole of) the photodetector 135 at the positions Ps1 and Pe1. FIG. 16C shows a pupil plane light intensity distribution detected by the photodetector 135 (pinholes thereof) arranged at the positions Ps1 and Pe1.

In FIG. 16B, the exposure light is hardly blocked by the pair of light blocking members of the variable slit mechanism 110. Therefore, the light amount centroids Gs1 and Ge1 of the pupil surface light intensity distribution coincide with the center O, and the light amount centroid is the same. Almost no bias occurs. In FIG. 16C, as indicated by Es2 and Ee2 in FIG. 16A, the light intensity of the trapezoidal light intensity distribution 1302 is a value less than half the maximum light intensity, and the variable slit mechanism 110 has A large amount of exposure light is blocked by the pair of light blocking members. Therefore, in FIG. 16C, the light amount centroids Gs2 and Ge2 of the pupil plane light intensity distribution have deviation amounts from the center O of gs2 and ge2, and the telecentric characteristic of the exposure light is in a bad state.

In FIG. 16D, the light intensity of the trapezoidal light intensity distribution 1303 is half the maximum light intensity, as indicated by Es3 and Ee3 in FIG. 16A, and the telecentric characteristic of the exposure light is as shown in FIG. Better than in case (c). Comparing the amounts from the mold in the case of FIG. 16C and the case of FIG. 16D, gs2>gs3 and ge2>ge3. Here, when obtaining the deviation of the light quantity center of gravity from the pupil surface light intensity distribution, the deviation of the light quantity center of gravity on one end Ps side of the trapezoidal light intensity distribution and the deviation of the light quantity center of gravity on the other end Pe side of the trapezoidal light intensity distribution The average value of and may be used, or either one of them may be used as a representative value. Alternatively, another method may be used.

Next, with reference to FIG. 17, a method of obtaining the image forming position error from the deviation of the light amount centroid (telecentric characteristic) will be described. FIG. 17A shows the same state as the pupil plane light intensity distribution (width of inclined portion=Ls2, Le2) of FIG. 16C, and shows a state in which the telecentric characteristic of the exposure light is large (bad). In FIG. 17B, the horizontal axis represents the scanning direction and the vertical axis represents the Z direction (focus direction). The left side of FIG. 17B corresponds to a state in which the exposure light whose inclined portion width is Ls2 is observed at the position Ps1, and the exposure light is equivalent to the light ray 1702. The ray 1702 has an angle θ with respect to the ray 1701 having good telecentricity (a ray parallel to the optical axis), which indicates that the telecentricity is poor. The light ray 1702 is incident on the image plane WP at an inclination angle θ with respect to the light ray 1701. gs2 is the bias of the light amount centroid obtained based on the pupil surface light intensity distribution. By obtaining the deviation Gs2 of the light amount centroid based on the pupil surface light intensity distribution, the inclination angle θ of the light ray 1702 can be obtained based on the deviation Gs2. The tilt angle θ is an index value indicating the telecentric characteristic.

Since the optical path length C of the light ray 1701 between the pupil plane PP and the image plane WP is known, the tilt angle θ of the light ray 1702 with respect to the light ray 1701 is obtained by obtaining the deviation gs2 of the light amount center of gravity in the pupil plane light intensity distribution. You can ask. The imaging position error dys2 at the distance dz between the running surface FC of the substrate and the imaging surface WP at the position Es2 can be calculated based on the distance dz and the tilt angle θ.

The right side of FIG. 17B corresponds to a state in which the exposure light having the inclined portion width Le2 is observed at the position Pe1, and the exposure light is equivalent to the light ray 1703. Also for the right side of FIG. 17B, similarly to the left side of FIG. 17B, the tilt angle θ of the light ray 1703 with respect to the light ray 1701 can be obtained by obtaining the deviation ge2 of the light amount center of gravity in the pupil plane light intensity distribution. it can. The imaging position error dye2 at the distance dz between the running surface FC of the substrate and the imaging surface WP at the position Pe1 can be calculated based on the distance dz and the tilt angle θ.

Here, the method of obtaining the telecentric characteristic (light ray inclination θ) based on the deviation of the light amount centroid has been described. Instead of this method, exposure is performed to transfer the pattern of the original onto the substrate while changing the width of the inclined portion, and the width of the inclined portion and the image forming position error are measured by measuring the image forming position error of the transferred pattern. May be specified.

The main control unit 103 uses information obtained by repeating steps S03 to S05, that is, information indicating the relationship between the width of the inclined portion and the image forming position error (hereinafter, characteristic information) inside or outside the main control unit 103. Save to memory.

In steps S06 to S08, the main control unit 103 determines the width of the inclined portion of the trapezoidal light intensity distribution according to the exposure mode set in step S01. Specifically, in step S06, the main control unit 103 determines the exposure mode. If the exposure mode is the imaging position priority mode, the process proceeds to step S07. If the exposure mode is the exposure amount distribution priority mode, Go to step S08.

When the exposure mode is the image formation position priority mode, in step S07, the main control unit 103 sets the determination threshold of the image formation position error to a low threshold value. In this case, in step S09, the main control unit 103 ends the connection. The width of the inclined portion is determined under the condition that the image position error is severe. On the other hand, when the exposure mode is the exposure amount distribution priority mode, in step S08, the main control unit 103 sets the determination threshold of the imaging position error to a high threshold value. In this case, in step S09, the main control unit 103 The width of the inclined portion is determined under the condition that the image forming position error is gentle. Here, the determination threshold means an allowable imaging position error.

FIG. 18A shows an example in which a plurality of threshold values are set according to the required accuracy of the image forming position error. In this example, a low threshold value (for example, 8 [nm/um] per unit length) is set as the determination threshold value of the imaging position error in the imaging position priority mode, and the imaging position error is set in the exposure amount distribution priority mode. A high threshold value (for example, 15 [nm/um] per unit length) is set as the determination threshold value. FIG. 18B shows a method of determining the width of the inclined portion based on the threshold value of the image formation position error. In FIG. 18B, the horizontal axis represents the width (L) of the inclined portion and the vertical axis represents the image forming position error (dy), and the information indicated by 1801 is the relationship between the width of the inclined portion and the image forming position error. Is characteristic information indicating.

When the imaging position priority mode is set, in step S09, the main control unit 103, based on the characteristic information 1801, the width Ls1 that satisfies the imaging position error of 8 [nm/um] or less, which is the low threshold value. Is determined as the width of the inclined portion. When the exposure amount distribution priority mode is set, in step S09, the main control unit 103, based on the characteristic information 1801, has a high threshold of 15 [nm/um] and a low threshold of 8 [nm/um]. um] is determined as the width of the inclined portion. In step S10, the main controller 103 controls the positions (distance from the conjugate plane MB) of the pair of light blocking members 110a and 110b of the variable slit mechanism 110 according to the width of the inclined portion determined in step S09.

In step S11, the main control unit 103 obtains the exposure amount error in the scanning direction according to the method described with reference to FIGS. FIG. 14A illustrates the exposure amount error obtained from the width of the inclined portion determined in step S09. In FIG. 14A, the relationship between the number of received light pulses per unit length (1/ΔY [pulse/mm]) and the exposure amount error is shown by an exposure error characteristic 1401. In order to achieve the exposure amount set in step S01, the exposure amount range 1402 is determined based on the pulse energy (exposure amount per pulse) obtained from the pupil plane light intensity distribution detected in step S04. In the example of FIG. 14, the exposure amount range 1402 indicates that the lower limit value of the irradiation pulse number is 4 pulses and the upper limit value is 5.1 pulses.

Next, in steps S12 to S15, the main control unit 103 determines the number of irradiation pulses according to the exposure mode set in step S01. In step S12, the main control unit 103 determines the exposure mode, and if the exposure mode is the imaging position priority mode, the process proceeds to step S13, and if the exposure mode is the exposure amount distribution priority mode, the process proceeds to step S14.

When the exposure mode is the imaging position priority mode, in step S13, the main control unit 103 sets the exposure amount error determination threshold value to a high threshold value, and determines the irradiation pulse number under the condition that the exposure amount error is gentle. On the other hand, when the exposure mode is the exposure amount distribution priority mode, in step S14, the main control unit 103 sets the determination threshold value of the exposure amount error to a low threshold value, and determines the irradiation pulse number under a severe exposure amount error condition. .. Here, the determination threshold regarding the exposure amount error means an allowable exposure amount error.

FIG. 14B shows an example in which a plurality of threshold values are set according to the required accuracy of the exposure amount error. In this example, a high threshold value (for example, exposure amount error≦0.6[%]) is set as the exposure amount error determination threshold value in the imaging position priority mode, and the exposure amount error determination is performed in the exposure amount distribution priority mode. A low threshold value (for example, exposure amount error≦0.3 [%]) is set as the threshold value.

When the imaging position priority mode is set, in step S13, the main control unit 103 determines, based on the exposure error characteristic 1401, that the exposure amount error is equal to or less than the high threshold value 0.6 [%], and The search is performed in the direction 1404 in which the number of irradiation pulses becomes small (exposure in a short time). Then, the main control unit 103 determines the irradiation pulse number from the intersection 1405 that matches the condition. When the exposure amount distribution priority mode is set, in step S14, the main control unit 103 determines, on the basis of the exposure amount error characteristic 1401, that the exposure amount error is a low threshold value of 0.3 [%] or less. A search is performed in the direction 1407 in which the number of irradiation pulses increases (exposure for a long time). Then, the main control unit 103 determines the irradiation pulse number from the intersection 1408 that matches the condition. When the exposure amount error characteristic 1401 and the exposure amount range 1402 have the relationship illustrated in FIG. 14, the irradiation pulse number may be simply determined as 4 [pulse] in any exposure mode. ..

In step S15, the main control unit 103 calculates the scanning speed and the laser oscillation frequency that satisfy the irradiation pulse number per unit length (1/ΔY [pulse/mm]). In the example of FIG. 14A, when 4 [pulse] is selected as the irradiation pulse number for the imaging position priority mode, the oscillation frequency f of the pulse laser light source 101 is set to 3000 [Hz] from the equation (1). Then, the scanning speed becomes 750 [mm/sec]. When 5 [pulse] is selected as the irradiation pulse number for the exposure amount distribution priority mode, the scanning speed is 600 [mm/sec] when the oscillation frequency f of the pulse laser light source 101 is 3000 [Hz].

In step S16, the main control unit 103 exposes the substrate 114 with the set width of the inclined portion and the set irradiation pulse number.

Steps S03 to S05 may be executed in the calibration mode. Further, in this case, steps S03 to S05 may be executed for each illumination mode. Illumination modes may include, for example, circular illumination modes, ring illumination modes, multipole illumination modes, and the like.

FIG. 15A shows the pupil plane light intensity distribution in the quadrupole illumination mode. FIG. 15B shows a light intensity distribution 1501 on the substrate in the quadrupole illumination mode. In such a case, since the light intensity distribution changes stepwise in the inclined portion L1, the pupil plane light intensity distribution is changed at a plurality of positions P1, P2, P3, P1′, P2′, P3′ in the inclined portion. It is also possible to detect and obtain the imaging position error based on them.

Hereinafter, an article manufacturing method using the above scanning exposure apparatus will be described. The article manufacturing method may include a coating step, an exposure device, a developing step, and a processing step. In the coating process, a photoresist is coated on the substrate. In the exposure process, the photoresist on the substrate that has been subjected to the coating process can be exposed by the above scanning exposure apparatus. In the developing process, the photoresist on the substrate that has undergone the exposing process may be developed to form a resist pattern. In the processing step, the substrate that has undergone the developing step can be processed. This treatment may include, for example, etching, ion implantation or oxidation.

SS: Scanning exposure device, 135: Photodetector, 114: Projection optical system

Claims (11)

A scanning exposure apparatus that exposes a substrate with exposure light that forms a light intensity distribution having a trapezoidal shape along a scanning direction on the substrate,
A detection unit for detecting an image forming position error of a light ray forming an inclined portion in the trapezoidal shape ,
The detection unit, based on the photodetector for detecting the pupil plane light intensity distribution observed from the position where the light rays forming the inclined portion are incident, and the pupil plane light intensity distribution detected by the photodetector. And a processing unit that obtains an imaging position error,
A scanning exposure apparatus characterized by the above. The processing unit obtains an inclination of a light ray forming the inclined portion based on the pupil plane light intensity distribution, and obtains an image forming position error based on the inclination.
The scanning exposure apparatus according to claim 1 , wherein: The processing unit obtains a tilt of a light beam forming the tilted portion based on a deviation of a center of gravity of the pupil surface light intensity distribution,
The scanning exposure apparatus according to claim 2 , wherein: Further comprising a changing mechanism for changing the width of the inclined portion in the scanning direction,
The processing unit controls the changing mechanism and the detection unit so that the image forming position error is detected for each of the plurality of widths of the inclined portion, and determines the relationship between the width of the inclined portion and the image forming position error. get,
The scanning exposure apparatus according to any one of claims 1 to 3 , characterized in that. The processing unit acquires the relationship for each lighting mode,
The scanning exposure apparatus according to claim 4 , wherein. The processing unit controls the position of the photodetector so that the photodetector is sequentially arranged at a plurality of positions in the scanning direction, and based on the output of the photodetector at each of the plurality of positions. By specifying the range in which the inclined portion is formed and disposing the photodetector at a predetermined position in the range, the photodetector is caused to detect the pupil plane light intensity distribution,
Scanning exposure apparatus according to any one of claims 1 to 5, characterized in that. The scanning exposure apparatus according to claim 4 or 5 , further comprising a control unit that determines a width of the inclined portion based on an allowable imaging position error and the relationship. The control unit determines the number of exposure light irradiation pulses per unit length of scanning of the substrate based on the determined width of the inclined portion and the allowable exposure amount error.
The scanning exposure apparatus according to claim 7 , wherein: The control unit determines the allowable image forming position error depending on which of the reduction of the image forming position error and the reduction of the exposure amount error should be prioritized.
Scanning exposure apparatus according to claim 7 or 8, characterized in that. The control unit sets the exposure amount error that is allowed when giving priority to reduction of the imaging position error to a value larger than the exposure amount error that is allowed when giving priority to reduction of the exposure amount error.
The scanning exposure apparatus according to claim 9 , wherein: A coating step of coating a photoresist on the substrate,
An exposure step of exposing by scanning exposure apparatus according to the photoresist on the substrate after the coating step to any one of claims 1 to 10,
A developing step of forming the resist pattern by developing the photoresist on the substrate after the exposure step,
A processing step of processing the substrate that has undergone the developing step,
A method for manufacturing an article, comprising: