JPWO2005008754A1 - Flare measurement method, exposure method, and mask for flare measurement - Google Patents

Abstract

This is a flare measurement technology that can accurately measure the flare of the projection optical system. Three pairs of opening patterns and line-and-space pattern images (38AP, 39AP) and (38BP, 39BP) that are gradually widened on a wafer (W) coated with a resist via a projection optical system to be measured. ) And (38CP, 39CP) are transferred, and then the wafer (W) is developed. Among the resist images obtained after development, the line width of the space pattern image (40A, 40B, 40C) closest to the aperture pattern image (38AP, 38BP, 38CP) is measured. Ask for flare.

Description

  The present invention relates to a flare measurement technique for measuring a flare of a projection optical system that projects an image of a pattern on a first surface onto a second surface, such as a semiconductor integrated circuit, a liquid crystal display element, or a thin film magnetic head. It is suitable for use in measuring a flare of a projection optical system of a projection exposure apparatus used for transferring a mask pattern onto a substrate in a lithography process for manufacturing various devices. The present invention further relates to an exposure technique using a flare measurement technique and a mask that can be used when measuring flare.

  For example, in a lithography process for manufacturing a semiconductor integrated circuit, a reticle (or a photomask or the like) pattern as a mask is coated with a resist as a photosensitive substrate (sensitive object) through a projection optical system (or glass). A projection exposure apparatus such as a stepper type or a scanning stepper type is used for transfer to each shot area of a plate or the like. With the further miniaturization of integrated circuits in recent years, the line width uniformity with respect to a pattern after transfer required for a projection exposure apparatus is also increasing. For this reason, degradation of line width uniformity due to flare in the projection optical system, which could not be ignored in the past, is becoming a problem. The flare generated in the optical system is roughly divided into stray light that is not desirable for imaging caused by forward scattered light generated in a small angle range on the surface of the optical member (lens, etc.) constituting the optical system and the coating film. There are so-called local flare (or fog light) that is (or fog light) and so-called long-range flare that is undesirable stray light for imaging caused by reflection on the coating film of the optical member. Since it is mainly local flare that contributes to the degradation of the line width uniformity of the pattern, the following description will be made taking local flare as an example of flare.

  FIG. 16 is a schematic diagram showing local flare associated with a pattern projected on the wafer. In FIG. 16, an elongated desired imaging pattern 31 is projected at the center of the light shielding portion, and is attached to the imaging pattern 31. Thus, the local flare 32 is projected. Local flare is “fogging light” generated within a range of several hundred nm to several tens of μm around the original image formation pattern on the wafer. Usually, the intensity of the local flare 32 is about 1% or less with respect to the intensity of the imaging light beam forming the imaging pattern 31. Such local flare deteriorates the line width uniformity as follows.

  FIG. 17 shows a pattern of a reticle R1 to be transferred. In FIG. 17, a first reticle window 34 having an opening (light transmitting portion) in a chromium light-shielding film 35 and a reticle window having an opening are shown. A second reticle window 37 that is considerably wider than the second reticle window 37 is formed. In the center of the reticle windows 34 and 37, elongated reticle patterns 33 and 36 made of a chromium light-shielding film having the same shape are formed. FIG. 18 shows a resist image R1P obtained by projecting the pattern of the reticle R1 of FIG. 17 onto a resist-coated wafer through a projection optical system and developing the resist. In FIG. The images 33P and 36P are concave and convex patterns formed corresponding to the reticle patterns 33 and 36 in FIG. 17, respectively. If the resist is positive, the resist images 33P and 36P are convex patterns, respectively. If the resist is negative, the resist images 33P and 36P are concave patterns.

  In an ideal state with no local flare, the resist images corresponding to the reticle patterns 33 and 36 having the same pattern have the same shape. On the other hand, when there is a local flare, the reticle window 37 that exists around the reticle pattern 36 is wider than the reticle window 34 that exists around the reticle pattern 33. Therefore, the local flare generated from the latter reticle window 37 is larger. The amount becomes larger than the local flare generated from the former reticle window 34. That is, the amount of local flare generated on the reticle window 37 covering the resist image 36P is larger than the amount of local flare generated on the reticle window 34 covering the resist image 33P in FIG. The line width is narrower than the line width of the resist image 33P. As described above, the resist image formed by projecting the reticle pattern having the same shape changes in shape (line width) due to local flare generated from the opening pattern around each pattern. This is a variation in line width due to local flare, which lowers the exposure performance and deteriorates the yield of the exposure process. In order to reduce the variation in line width caused by such local flare, it is necessary to reduce local flare. For this purpose, it is necessary to accurately measure local flare.

Conventionally, as a method for measuring the flare of a projection optical system, for example, a plurality of light shielding patterns having a substantially uniform distribution are formed on the entire illumination area of the test reticle, and the test reticle pattern is projected onto the projection optical system. A method of performing projection exposure on a resist-coated wafer via a system is known (see, for example, Patent Document 1). In this method, exposure is performed a plurality of times while changing the integrated exposure amount, and after exposure, the shape of each resist image obtained by development is measured, so that the portions corresponding to the light shielding patterns are not exposed (resist The amount of flare is determined from the relationship between the exposure amount when no image is formed and the exposure amount when a portion corresponding to the light shielding pattern is exposed (when a resist image is formed).
International Publication No. 02/09163

Generally, it is not only flare that changes the line width of the resist image after development, but also the line width of the resist image or the exposure amount when the resist image is formed depending on the aberration of the projection optical system.
However, the conventional flare measurement method cannot always accurately determine whether the change in the line width of the resist image is caused by flare or aberration. Therefore, there is a problem that it is difficult to accurately measure the amount of flare in the projection optical system, particularly the amount of local flare.

  In addition, in order to improve the resolution, the exposure light has been shortened. Recently, a pulsed excimer laser such as a KrF excimer laser having a wavelength of 248 nm and an ArF excimer laser having a shorter wavelength of 193 nm is used as the exposure light. Has been. As a photosensitive material for this excimer laser, a highly sensitive chemically amplified resist is used. However, in the case of a chemically amplified resist, the line width of the resist image after development may slightly change due to the volatilization of acid from the resist in the PEB (Post-Exposure Bake) process, which is a baking before development. For this reason, when the exposure is performed using an excimer laser, the change in the line width of the resist image includes a change due to volatilization of the acid in the resist in addition to the change due to the flare and aberration. However, conventionally, there is no known method for accurately obtaining only the change in the line width of the resist image due to the volatilization of the acid in the resist, and this is a problem for further improving flare measurement accuracy. It was.

In view of this point, the present invention has a first object to provide a flare measurement technique capable of accurately measuring the flare of a projection optical system.
It is a second object of the present invention to provide an exposure technique that can reduce flare in a projection optical system and obtain high line width uniformity.
A third object of the present invention is to provide a mask that can be used for flare measurement of a projection optical system.

  A flare measurement method according to the present invention is a flare measurement method for measuring flare of a projection optical system (PL) that projects an image of a pattern on a first surface onto a second surface, and is arranged on the first surface. An image of the opening pattern (38A; 38) and one or more linear patterns (39A; 39) arranged adjacent to the opening pattern is projected onto the second surface via the projection optical system. A first step and a second step of obtaining a flare of the projection optical system based on the line width of the image (39AP) of the linear pattern projected on the second surface.

  According to the present invention, due to the combination of the opening pattern and the linear pattern, the change in the line width of the projected linear pattern image is small with respect to the occurrence of aberration, and the occurrence of flare is The line width change of the image of the linear pattern becomes large. Therefore, the flare of the projection optical system can be accurately measured by measuring the line width of the image of the linear pattern or its change. The line width of the image of the linear pattern may be measured as a concavo-convex pattern of the photosensitive material after development, or may be measured at the stage of the latent image or aerial image of the photosensitive material. When an image is projected onto the photosensitive material, the opening pattern and the linear pattern do not need to be exposed at the same time, and may be individually exposed by multiple exposure.

  In this case, in the first step, a plurality of sets of opening patterns (38A, 38B, 38C) and linear patterns (39A, 39B, 39C) having the same shape as the opening pattern and the linear pattern and different from each other. ) Is projected onto the second surface via the projection optical system, and the flare of the projection optical system is obtained based on the line widths of the images of the plurality of linear patterns in the second step. Good.

Since the influence of the flare on the image of the linear pattern differs depending on the interval between the opening pattern and the linear pattern, the flare can be more accurately obtained by using the measurement values of the line widths of the images of the linear pattern. It can be measured.
In the first step, the images of the opening pattern (38) and the linear pattern (39) may be exposed a plurality of times so as not to overlap each other and at different intervals. Thus, by performing multiple exposure of the same pattern, it is possible to reduce a mask pattern manufacturing error and a flare measurement error due to a difference in the light beam passage position in the projection optical system.

  In the first step, an image of another linear pattern (47B) is superimposed on the image of the linear pattern (47A) at a predetermined small crossing angle and exposed, and in the second step, the line is exposed. Measure the width (h1, hn) in the longitudinal direction of the part (47AP, 48BP) where the image of the linear pattern and the image of the other linear pattern overlap, and the flare of the projection optical system is calculated based on this measurement result. You may make it ask.

  By exposing the images of two linear patterns so as to overlap with a small crossing angle, the width in the longitudinal direction of the obtained pattern image is a value obtained by greatly expanding the line width of the image of the linear pattern. . Therefore, since the line width of the image of the linear pattern can be substantially enlarged and measured with high accuracy, the flare can be measured with higher accuracy, or the flare can be measured using a simpler measuring device. It can be measured.

  In the present invention, as an example, the size of the opening pattern is approximately several μm to several 100 μm square, the width of the linear pattern is approximately several 100 nm to several μm, and the opening pattern and the linear pattern Is approximately 1 μm to several tens of μm. Thereby, for example, when a projection optical system with a reduction magnification is used, a local flare spreading on the second surface around the aperture pattern image in a width of about several hundred nm to several tens of μm can be accurately measured.

  In the present invention, the first process includes an application process (steps 101 and 102) in which a photosensitive material (PR) is applied on the substrate (W) and then an upper layer film (TC) is applied on the photosensitive material. The surface of the substrate is substantially aligned with the second surface, and the opening pattern and an image of one or more linear patterns arranged adjacent to the opening pattern are passed through the projection optical system. A projection process (step 103) for projecting onto the surface of the substrate, a pre-development process (step 104) for performing pre-development baking of the photosensitive material on the substrate, and developing the photosensitive material on the substrate to develop the line And a development step (step 105) for forming an uneven image of the pattern.

In this way, by applying the upper layer film on the photosensitive material, volatilization of the substance that affects the line width from the photosensitive material during baking before development is suppressed, so that the lines of the image of the linear pattern after development are suppressed. The width is substantially a value corresponding to only the flare. Therefore, the flare can be measured with high accuracy by eliminating the error caused by the photosensitive material.
In this case, an example of the photosensitive material is a chemically amplified resist. Further, the upper layer film desirably suppresses volatilization of acid from the photosensitive material in the preliminary development step. By using a chemically amplified resist, an exposure beam having a short wavelength such as an excimer laser can be used, so that the line width of the linear pattern to be transferred can be further miniaturized and flare measurement accuracy can be improved. At that time, the deterioration of measurement accuracy can be prevented by suppressing the volatilization of the acid by the upper layer film.

Next, an exposure method according to the present invention uses the flare measurement method of the present invention in an exposure method in which a pattern of a first object (R) is projected and exposed onto two objects (W) via a projection optical system (PL). A measuring step for measuring the flare of the projection optical system, and a correcting step for correcting the flare of the projection optical system based on the measurement result in the measuring step.
According to this exposure method, the flare of the projection optical system can be accurately measured by using the flare measurement method of the present invention. Therefore, in the correction step, flare of the projection optical system can be reduced, for example, by reworking an optical member such as a lens in the projection optical system or replacing the optical member. Therefore, high line width uniformity can be obtained.
For example, when the line width uniformity is deteriorated, it is possible to accurately determine whether or not the cause is a flare of the projection optical system by performing the measurement process of the present invention. Furthermore, it is possible to accurately measure a change in flare over time. Accordingly, it is possible to appropriately deal with troubles occurring in the exposure process.

The mask according to the present invention is a flare measurement mask for a projection optical system, and includes an opening pattern (38A; 38) and one or a plurality of linear patterns (39A) arranged adjacent to the opening pattern. 39) is formed. By using this mask, the flare measuring method of the present invention can be used.
In this case, as an example, the opening pattern is a square pattern of about several μm to several 100 μm square, and the linear pattern is a line and space pattern whose width of the line pattern is about several hundred nm to several μm, The distance between the opening pattern and the linear pattern is approximately 1 μm to several tens of μm. As a result, local flare can be measured accurately.

According to the present invention, since the change in the line width of the projected image of the linear pattern is substantially caused by the flare of the projection optical system, substantially only the flare of the projection optical system can be measured accurately.
In addition, by applying an upper layer film on the photosensitive material, for example, when a highly sensitive photosensitive material is used, it is possible to prevent deterioration in flare measurement accuracy due to the photosensitive material.
Further, by correcting the flare of the projection optical system based on the measurement result, the line width uniformity of the transfer pattern can be improved.

FIG. 1 is a perspective view showing a projection exposure apparatus as an example of an embodiment of the present invention. FIG. 2 is a plan view showing an example of a test reticle pattern for flare measurement of the projection optical system. FIG. 3 is a plan view showing a resist image obtained by transferring the pattern of the test reticle of FIG. 2 onto a wafer via a projection optical system. FIG. 4 is an enlarged view showing a state of the wafer from when the photoresist and the top coat are applied on the wafer to when the flare measurement pattern is formed on the wafer. FIG. 5 is an enlarged view showing the state of the wafer from the application of the photoresist on the wafer to the formation of the flare measurement pattern on the wafer. FIG. 6 is a flowchart showing an example of the flare measurement operation in the embodiment of the present invention. FIG. 7 is a diagram showing a simulation result of the relationship between the pad-space distance of the resist image of FIG. 3 and the line width of the space pattern image in a state where there is no aberration of the projection optical system. FIG. 8 is a diagram showing a simulation result of the relationship between the pad-space distance of the resist image of FIG. 3 and the line width of the space pattern image in a state where there is an aberration of the projection optical system. FIG. 9 is a diagram showing a simulation result of the relationship between the pad-space distance of the resist image of FIG. 3 and the line width of the space pattern image in a state where there are more aberrations of the projection optical system. FIG. 10 is a plan view showing another test reticle used in an example of the embodiment. FIG. 11 is a plan view showing a resist image obtained by transferring a part of the pattern of the test reticle of FIG. 10 onto a wafer via a projection optical system. FIG. 12 is a plan view showing a resist image obtained by transferring a part of the pattern of the test reticle of FIG. 10 onto the wafer three times through the projection optical system. FIG. 13 is a plan view showing a resist image obtained by transferring the test reticle pattern of FIG. 10 onto the wafer six times via the projection optical system. FIG. 14 is a plan view showing a pattern of a test reticle used in another example of the embodiment of the present invention. FIG. 15 is a plan view showing a resist image obtained by superimposing and transferring the test reticle pattern of FIG. 14 onto a wafer via a projection optical system. FIG. 16 is a schematic diagram showing local flare associated with a pattern projected onto a wafer. FIG. 17 is a diagram showing an example of a reticle pattern used for explaining the flare. FIG. 18 is a view showing a resist image obtained by development after transferring the reticle pattern of FIG. 17 through the projection optical system.

  Hereinafter, an example of a preferred embodiment of the present invention will be described with reference to FIGS. In this example, the present invention is applied when measuring the flare of a projection optical system mounted on a scanning exposure type projection exposure apparatus. The flare of the projection optical system can be measured using an inspection apparatus that includes an illumination optical system and a simple stage mechanism that holds a mask and a wafer, for example, during assembly adjustment of the projection optical system. In the following description, an actual projection exposure apparatus is used as an inspection apparatus. Such flare measurement using an actual projection exposure apparatus can be performed in order to analyze the cause when, for example, the line width uniformity is reduced in the exposure process.

FIG. 1 shows a schematic configuration of a projection exposure apparatus equipped with a projection optical system PL for measuring flare. In FIG. 1, an ArF excimer laser light source (wavelength 193 nm) is used as the exposure light source 6. As an exposure light source, an ultraviolet pulse laser light source such as a KrF excimer laser light source (wavelength 247 nm), an F ₂ laser light source (wavelength 157 nm), a Kr ₂ laser light source (wavelength 146 nm), an Ar ₂ laser light source (wavelength 126 nm), or YAG A laser harmonic generation light source, a solid-state laser (semiconductor laser, etc.) harmonic generator, a mercury lamp (i-line, etc.), etc. can also be used.

  Exposure light (exposure illumination light) IL as an exposure beam pulsed from the exposure light source 6 during exposure is mirror 7, a beam shaping optical system (not shown), a first lens 8A, a mirror 9, and a second lens 8B. Then, the cross-sectional shape is shaped into a predetermined shape, and enters the fly-eye lens 10 as an optical integrator (a homogenizer or a homogenizer), and the illuminance distribution is made uniform. On the exit surface of the fly-eye lens 10 (the pupil plane of the illumination optical system), an aperture stop (σ stop) for determining the illumination condition by setting the light intensity distribution of the exposure light to a circular shape, a plurality of eccentric regions, an annular shape, or the like. ) The illumination system aperture stop member 11 having 13A, 13B, 13C, and 13D is rotatably arranged by the drive motor 12. The exposure light IL that has passed through the aperture stop in the illumination system aperture stop member 11 passes through a beam splitter 14 and a relay lens 17A having a low reflectivity, and passes through a fixed blind 18A as a fixed field stop and a movable blind 18B as a movable field stop. Pass sequentially. In this case, the movable blind 18B is disposed on a surface substantially conjugate with the pattern surface (reticle surface) of the reticle R as a mask, and the fixed blind 18A is a surface slightly defocused from the surface conjugate with the reticle surface. Has been placed.

  The fixed blind 18A is used to define the illumination area 21R on the reticle surface as a slit-like area elongated in the non-scanning direction orthogonal to the scanning direction of the reticle R. The movable blind 18B includes two pairs of blades that are relatively movable in directions corresponding to the scanning direction and the non-scanning direction of the reticle R, and is unnecessary at the start and end of the scanning exposure to each shot area to be exposed. It is used to close the illumination area in the scanning direction so that no exposure to is performed. The movable blind 18B is also used to define the center and width of the illumination area in the non-scanning direction. The exposure light IL that has passed through the blinds 18A and 18B passes through the sub-condenser lens 17B, the optical path bending mirror 19, and the main condenser lens 20, and the illumination area 21R of the pattern area of the reticle R as a mask has a uniform illuminance distribution. Illuminate.

  On the other hand, the exposure light reflected by the beam splitter 14 is received by the integrator sensor 16 composed of a photoelectric sensor via the condenser lens 15. The detection signal of the integrator sensor 16 is supplied to the exposure amount control system 3. The exposure amount control system 3 detects the detection signal and the optical system from the beam splitter 14 measured in advance to the wafer W as the substrate (photosensitive substrate). The exposure energy on the wafer W is indirectly calculated using the transmittance. The exposure amount control system 3 controls the exposure light source 6 so that an appropriate exposure amount is obtained on the wafer W based on the integrated value of the calculation result and the control information from the main control system 1 that controls the overall operation of the apparatus. Controls the light emission operation. Exposure light source 6, mirrors 7 and 9, lenses 8A and 8B, fly-eye lens 10, illumination system aperture stop member 11, beam splitter 14, relay lens 17A, blinds 18A and 18B, sub-condenser lens 17B, mirror 19 and main condenser The illumination optical system 5 includes the lens 20.

  Under the exposure light IL, the pattern in the illumination area 21R of the reticle R is coated with a resist at a projection magnification β (β is 1/4, 1/5, etc.) via a bilateral telecentric projection optical system PL. And projected onto a slit-like exposure area 21W elongated in the non-scanning direction on one shot area SA on the wafer W. The wafer W is a disk-shaped substrate having a diameter of about 200 to 300 mm, such as a semiconductor (silicon or the like) or SOI (silicon on insulator). The pattern surface (reticle surface) of the reticle R and the surface (wafer surface) of the wafer W correspond to the first surface (object surface) and the second surface (image surface) of the projection optical system, respectively. The reticle R and the wafer W can also be regarded as a first object and a second object (sensitive object), respectively. Hereinafter, in FIG. 1, the Z axis is taken in parallel to the optical axis AX of the projection optical system PL, and X in the non-scanning direction orthogonal to the scanning direction of the reticle R and the wafer W during scanning exposure in a plane perpendicular to the Z axis. A description will be given by taking the axis and taking the Y axis in the scanning direction.

  First, the reticle R is held on a reticle stage (movable body) 22, and the reticle stage 22 moves on the reticle base 23 at a constant speed in the Y direction, and corrects a synchronization error with a wafer stage 28 described later. The reticle R is scanned by slightly moving in the X direction, the Y direction, and the rotation direction. The position of the reticle stage 22 is measured by a movable mirror (not shown) and a laser interferometer (not shown) provided on the reticle stage 22, and based on the measured value and control information from the main control system 1, a stage drive system 2 controls the position and speed of the reticle stage 22 via a drive mechanism (not shown) such as a linear motor. In this example, a reticle stage system is configured by the reticle stage 22, the stage drive system 2, the drive mechanism, and the laser interferometer described above. A reticle alignment microscope (not shown) for reticle alignment is disposed above the periphery of the reticle R. Near the reticle stage 22, a reticle loader (not shown) for exchanging reticles on the reticle stage 22 and a reticle library storing a plurality of reticles are installed.

  On the other hand, the wafer W is held on the wafer stage 28 via the wafer holder 24. The wafer stage 28 moves on the wafer base 27 at a constant speed in the Y direction, and moves in steps in the X and Y directions. And a Z tilt stage 25. The Z tilt stage 25 performs focusing and leveling of the wafer W based on the measurement value of the position of the wafer W in the Z direction by an auto focus sensor (not shown). The position of the wafer stage 28 in the XY plane and the rotation angles around the X, Y, and Z axes are measured by a laser interferometer (not shown), and the measured values and control information from the main control system 1 are used. Based on this, the stage drive system 2 controls the operation of the wafer stage 28 via a drive mechanism (such as a linear motor) (not shown). In this example, the wafer stage system is constituted by the wafer holder 24, the wafer stage (movable body) 28, the stage drive system 2, the drive mechanism, and the laser interferometer.

  Further, a light amount sensor unit 29 including a dose monitor having a light receiving surface 30B larger than the exposure region 21W and an illuminance sensor having a pinhole-shaped light receiving surface 30A is fixed in the vicinity of the wafer W on the wafer stage 28. The two detection signals of the light quantity sensor unit 29 are supplied to the exposure amount control system 3. Further, an off-axis alignment sensor 36 for wafer alignment is disposed above the wafer stage 28, and the main control system 1 aligns the wafer W based on the detection result.

  At the time of exposure, the reticle stage 22 and the wafer stage 28 are driven so that the reticle R and one shot area on the wafer W are synchronously scanned in the Y direction while being irradiated with the exposure light IL, and the wafer stage 28 is driven. Then, the step of moving the wafer W stepwise in the X and Y directions is repeated. Thereby, the pattern image of the reticle R is exposed to each shot area on the wafer W by the step-and-scan method.

  Next, an example of the operation when measuring the flare of the projection optical system PL of this example will be described with reference to the flowchart of FIG. The flare measurement operation is roughly divided into a first process and a second process as follows. As described above, the flare of the projection optical system includes so-called local flare caused by forward scattered light generated in a small angle range on the surface of the optical member and the coating film, and so-called reflection caused by the coating on the surface of the optical member. There is a long-range flare. In the following, local flare is measured. For example, the long range flare can be measured in the same manner by changing the pattern of the test reticle described later.

[First step]
In this example, a test reticle R2 as a flare measurement mask shown in FIG. 2 is loaded on the reticle stage 22 shown in FIG. The X direction and Y direction in FIG. 2 (the same applies to FIGS. 3 and 10 to 13 below) correspond to the X direction (non-scanning direction) and Y direction (scanning direction) in FIG. 1, respectively.

  In FIG. 2, a pad pattern 38A is formed in the pattern area of the test reticle R2 with a square opening pattern having a width D in the X direction and a height H in the Y direction against a light shielding film 41 of chromium or the like. In addition, eight space patterns 40 each having a rectangular opening pattern having a width d in the X direction and a longitudinal direction in the Y direction are adjacent to the pad pattern 38A in the X direction in the light shielding film 41 in the X direction. A line and space pattern (hereinafter referred to as “L & S pattern”) 39A arranged at a pitch P is formed. The L & S pattern 39A corresponds to a linear pattern. Note that the number of space patterns 40 of the L & S pattern 39A is one or more, and when the L & S pattern 39A is composed of one space pattern 40 (isolated line), the space pattern 40 corresponds to a linear pattern. . Further, as the L & S pattern 39A, a linear pattern in which two or more rectangular light shielding patterns are arranged in the X direction may be used.

  In this example, the width D and the width H of the pad pattern 38A are equal to each other and are set to about several μm to several 100 μm, respectively. That is, the pad pattern 38A is a square opening pattern. However, the width D and the width H of the pad pattern 38A may be set to different values within a range of several μm to several 100 μm, respectively. Further, the width d in the X direction of the space pattern 40 constituting the L & S pattern 39A is about several hundred nm to several μm, and the length of the space pattern 40 in the Y direction is about several tens of times the width d, that is, several μm to It is about several tens of μm. In the example of FIG. 2, the length of the space pattern 40 in the Y direction is set to be approximately equal to the height H of the pad pattern 38A in the Y direction, but they may be different. The pitch P of the arrangement of the space patterns 40 is almost twice the width d.

  Further, in the light shielding film 41 of the test reticle R2, pad patterns 38B and L & S patterns 39B are formed at intervals of about several μm in the Y direction with respect to the pad patterns 38A and L & S patterns 39A. The pad patterns 38B and L & S patterns Pad patterns 38C and L & S patterns 39C are formed at intervals of about several μm in the Y direction with respect to 39B. Each of the pad patterns 38B and 38C is an opening pattern having the same shape as the pad pattern 38A, and each of the L & S patterns 39B and 39C has the same configuration as the L & S pattern 39A (8 space patterns 40 are arranged at a pitch P in the X direction). Is). However, the interval r2 in the X direction between the center pad pattern 38B and the L & S pattern 39B is set several times the interval r1 in the X direction between the upper pad pattern 38A and the L & S pattern 39A. The distance r3 in the X direction between the lower pad pattern 38C and the L & S pattern 39C is set to about 10 times. As an example, the interval r1 is about 1 μm, the interval r2 is about several μm, and the interval r3 is about several tens of μm.

  Next, in step 101 of FIG. 6, using a resist coater (not shown), as shown in FIG. 4A, a wafer as a substrate for flare evaluation (this is referred to as wafer W) has high flatness. A photoresist PR as a photosensitive material is applied on the upper surface. In this example, since ArF excimer laser light is used as the exposure beam, a chemically amplified resist that is a highly sensitive resist is used as the photoresist PR. In the next step 102, the acid from the photoresist at the time of PEB (Post-Exposure Bake) described later is formed on the photoresist PR (chemically amplified resist in this example) on the wafer W by using, for example, another coater. A top coat TC as an upper layer film for suppressing volatilization is applied in an overlapping manner. As an example, the thickness of the photoresist PR is about 100 to 200 nm, and the thickness of the topcoat TC is about 100 nm. 4A, 4B, 4C, and 4D show enlarged side views of a part of the wafer W, respectively, the magnification in the thickness direction of the part other than the wafer W is set to be large. Has been.

  Next, in step 103 in FIG. 6, the unexposed wafer W on which the photoresist PR and the top coat TC are applied in an overlapping manner is loaded onto the wafer holder 24 on the wafer stage 28 in FIG. Then, using the projection exposure apparatus of FIG. 1 as in normal exposure, the pattern of the test reticle R2 (flare evaluation pattern) of FIG. 2 is scanned on the wafer W by the scanning exposure system PL via the projection optical system PL. Transfer exposure on one shot area. At this time, the pattern of the test reticle R2 may be transferred and exposed onto a plurality of shot areas on the wafer W, respectively. For example, the exposure light IL including the imaging light beams 38A1 and 39A1 corresponding to the pad pattern 38A and the L & S pattern 39A of the test reticle R2 in FIG. The resist PR and the top coat TC are exposed. In FIGS. 4B and 4C, a portion with a large exposure amount is given a fine oblique line, and a portion with little exposure amount is given a rough oblique line.

  Then, after completion of the exposure, the process proceeds to step 104, and PEB which is a pre-development baking of the photoresist PR of the exposed wafer W is performed in a baking apparatus (not shown). Since the exposure beam in this example is a single-wavelength ArF excimer laser beam, PEB is required to reduce deformation of the resist pattern due to the standing wave effect. As shown in FIG. 4C, acid 50 is generated in a portion where the exposure amount of photoresist PR, which is a chemically amplified resist, is increased by the heat treatment with PEB. The volatilization of acid 50 is caused by topcoat TC. Is suppressed by. Accordingly, the acid 50 is prevented from adhering to the portion exposed by the exposure light 39A1 that has passed through the L & S pattern 39A. In the next step 105, the photoresist PR of the wafer W on which the PEB has been performed is developed in a developing device (not shown). As a result, an uneven resist image corresponding to the projected image of the pattern of the test reticle R2 in FIG. 2 is formed in one or a plurality of shot regions on the wafer W.

  Specifically, assuming that the photoresist PR is a positive type, a concave resist image 38AP corresponding to the image of the pad pattern 38A in FIG. 2 and the L & S pattern in FIG. 2 are formed on the wafer W in FIG. A periodic concave resist image 39AP corresponding to the image 39A is formed. As described later, in this example, the line width dP1 of the space pattern image 40A closest to the resist image 38AP in the resist image 39AP is measured. At this time, in this example, since the acid 50 generated in the photoresist PR of FIG. 4C does not adhere to other portions, the line width dP1 of the space pattern image 40A changes due to volatilization of the acid 50. Minutes have not occurred. Therefore, even if a chemically amplified resist is used, only a change in the line width of the resist pattern due to flare (particularly local flare) can be measured, and as a result, flare can be measured with high accuracy.

  On the other hand, for comparison, FIG. 5A (resist application) and FIG. 5 show changes in the state of the wafer W corresponding to FIG. (B) (exposure), FIG. 5 (C) (PEB), and FIG. 5 (D) (development). In this case, since there is no top coat, a PEB (Post-Exposure Bake) of a photoresist PR made of a chemically amplified resist is performed as shown in FIG. The acid 50 generated in the image portion is volatilized, and the volatilized acid 50 is mixed in the image portion of the L & S pattern adjacent to the pad pattern image. Therefore, after development of the photoresist PR, as shown in FIG. 5D, the line width dQ1 of the space pattern image 40A1 adjacent to the resist image 38AP becomes narrower than the original line width, and flare measurement accuracy is increased. Decreases.

  In this example, an excimer laser beam is used as the exposure beam, and a chemically amplified resist is used as the photosensitive material. Therefore, a top coat is used to suppress acid volatilization. However, for example, when using a photoresist with a small amount of acid volatilization, when the line width of the image of the L & S pattern is large and the influence of volatilization of the acid is small, or when using a bright line (i-line etc.) of a mercury lamp as the exposure beam For example, flare can be measured with high accuracy even if the top coat is omitted.

[Second step]
Next, in step 106 of FIG. 6, the line width of the resist pattern formed by development is measured. Below, it demonstrates from the principle which can measure a flare by the line width measurement.
FIG. 3 shows a resist image formed on the wafer W by transferring and developing the pattern image of the test reticle R2 of FIG. The projection optical system PL of FIG. 1 performs reverse projection, but for the sake of convenience of explanation, it is assumed that an erect image is projected in FIG. Further, the resist on the wafer W may be either a positive type or a negative type. In this case, if it is a positive type, the background portion 41P in FIG. 3 corresponding to the light shielding film 41 in FIG. It is a part of. In FIG. 3, rectangular resist images 38AP, 38BP, 38CP corresponding to the images of the pad patterns 38A, 38B, 38C of FIG. 2 are formed on the wafer W, and the L & S patterns 39A, 39B, 38C of FIG. Concave resist images 39AP, 39BP, and 39CP having an L & S pattern shape corresponding to the 39C image are formed. At this time, when the projection magnification β from the reticle of the projection optical system PL to the wafer is used, the intervals in the X direction of the resist images 38AP and 39AP, the resist images 38BP and 39BP, and the resist images 38CP and 39CP correspond to those in FIG. The intervals r1, r2, and r3 are approximately β times larger. If there is no flare in the projection optical system PL, the width in the X direction of each space pattern image formed by the concave portions constituting the resist images 39AP, 39BP, and 39CP is β of the width d of the space pattern 40 in FIG. Doubled.

  However, assuming that local flare occurs in the projection optical system PL, the local flare is closer to the resist images 38AP, 38BP, and 38CP in FIG. 3 corresponding to the images of the pad patterns 38A, 38B, and 38C in FIG. Become more. Accordingly, since the distance between the upper resist images 38AP and 39AP is the smallest and the distance between the lower resist images 38CP and 39CP is the widest, the amount of local flare is the upper resist image 39AP, the central resist image 39BP, and the lower resist image. The number decreases in the order of 39 CP. That is, among the resist images 39AP close to the upper resist image 38AP in FIG. 3 in the X direction, the space pattern image 40A closest to the resist image 38AP has the most local flare. Is considerably wider than the width β · d. On the other hand, in the space pattern image 40C closest to the resist image 38CP in the resist image 39CP close to the lower resist image 38CP in the X direction, the local flare hardly reaches, so the developed line width dP3 is almost the original width. It becomes the same as β · d. In the space pattern image 40B closest to the resist image 38BP in the resist image 39BP close to the center resist image 38BP in the X direction, the amount of local flare is almost in the middle between the upper and lower space pattern images 40A and 40C. The developed line width dP2 is approximately between the upper and lower line widths dP1 and dP3. Therefore, the following relationship is established.

dP1>dP2> dP3≈β · d (1)
Further, since the line width change of the space pattern images 40A, 40B, and 40C increases as the amount of local flare increases, the amount of local flare can be evaluated (determined) from the measured values of these line widths. Therefore, in this example, using a scanning electron microscope, for example, the line widths dP1, dP2, and dP3 in the X direction of the three space pattern images 40A, 40B, and 40C in the resist images after development shown in FIG. measure. At this time, when the pattern of the test reticle R2 in FIG. 2 is transferred to a plurality of shot areas on the wafer W, the X-direction lines of the space pattern images 40A, 40B, and 40C in the plurality of shot areas, respectively. The widths dP1, dP2, and dP3 may be measured, and the average value of these measurement results may be used as the line width of the space pattern images 40A, 40B, and 40C. As a result, the exposure error and the measurement error are averaged, and the measurement accuracy is improved. When the resist on the wafer W is a negative type, the relationship between the line widths of the equation (1) is the same as the unevenness of the resist image in FIG. 3 is reversed.

Next, in step 107 of FIG. 6, measurement (evaluation) of the local flare of the projection optical system PL is performed. For this purpose, the results obtained by the inventor's computer simulation regarding the relationship between the amount of local flare of the projection optical system PL and the line widths of the space pattern images 40A to 40C will be described with reference to FIGS. I will explain.
The horizontal axes of FIGS. 7, 8, and 9 are the pad patterns 38A, 38B, and 38C on the test reticle R2 of FIG. 2, and the space pattern 40 that is closest to the pad pattern in the corresponding L & S patterns 39A, 39B, and 39C. The X-direction spacing (pad-space distances) r1, r2, and r3 (nm), and their vertical axes are the spaces closest to the resist images 38AP, 38BP, and 38CP of the pad pattern in the resist image of FIG. Line widths dP1, dP2, and dP3 (nm) of the pattern images 40A, 40B, and 40C are shown.

  In the simulation, the wavelength λ of the exposure light IL in FIG. 1 is 193 nm (ArF excimer laser), and the numerical aperture NA of the projection optical system PL to be inspected is 0.78. The σ value that is the coherence factor of the illumination optical system 5 was set to 0.85. Further, the width D and the height H of the pad pattern 38A in FIG. 2 are both 1 μm, and the width d of the space pattern 40 of the L & S pattern 39A is 140 nm. Note that these values of the width D, the height H, and the width d are values in a projection image by the projection optical system PL. An n-th order (n is an integer of 1 or more) Zernike Polynomial (Zernike polynomial) is represented by Zn, and in this example, the aberration of the projection optical system PL is Zernike Polynomial up to the 37th order (Z1 to Z37) wavefront aberration. The local flare of the projection optical system PL is represented by the total RMS of higher-order wavefront aberrations of the 38th order (Z38) or higher of Zernike Polynomial. The unit of wavefront aberration is the exposure wavelength λ.

  Specifically, in this example, the aberration of the projection optical system PL, that is, the total RMS of wavefront aberrations up to the 37th order (Z1 to Z37) of Zernike Polynomial is 0 mλ, 13 mλ, and 26 mλ with respect to the pad-space distance. The line width of the space pattern image was calculated. FIG. 7 shows the calculation results when RMS = 0 mλ as aberration, FIG. 8 shows RMS = 13 mλ as aberration, and FIG. 9 shows RMS = 26 mλ as aberration. For each aberration state, the local flare of the projection optical system PL, that is, the pad-space in the case where the overall RMS of the Zernike Polynomial 38th order (Z38) or higher wavefront aberration is 0 mλ, 10 mλ, and 20 mλ. The line width of the space pattern image with respect to the distance was calculated. 7, 8, and 9, when solid line curves 42 </ b> A, 43 </ b> A, and 44 </ b> A are RMS = 0 mλ as local flare, and broken line curves 42 </ b> B, 43 </ b> B, and 44 </ b> B are RMS = 10 mλ as local flare, Curves 42C, 43C, and 44C are calculation results when RMS = 20 mλ as a local flare.

  From the calculation results of FIGS. 7 to 9, when the pad-space distance is shortened, the curves 42B, 43B, 44B when the local flare occurs are larger than the curves 42A, 43A, 44A when there is no local flare. It can be seen that the curves 42C, 43C, 44C when the local flare changes further upward (the line width becomes thicker) when the local flare further increases (the line width becomes thicker). That is, as the pad-space distance becomes shorter and the pad patterns 38A to 38C of FIG. 2 approach the L & S patterns 39A to 39C, the resist pattern lines of the space pattern closest to the pad patterns 38A to 38C of the L & S patterns 39A to 39C. It can be seen that the width becomes thicker. Furthermore, since the curves 42A to 42C in FIG. 7, the curves 43A to 43C in FIG. 8, and the curves 44A to 44C in FIG. 9 have almost the same tendency, the change in the line width of the resist image of the space pattern in this example is It can be seen that the aberration is not substantially affected by the aberration of the projection optical system PL (RMS of Z1 to Z37), and is mainly caused by the local flare (RMS of Z38 or higher) of the projection optical system PL. That is, the result of the line width measurement of the space pattern image substantially represents only the line width change caused by the local flare.

  Therefore, after measuring the line widths dP1, dP2, and dP3 of the space pattern images 40A, 40B, and 40C of FIG. 3 with the scanning electron microscope as described above, the corresponding pad-space distances r1, r2, and FIG. For example, r3 is plotted as a position on the horizontal axis in FIG. 7, and line widths dP1, dP2, and dP3 are plotted as positions on the vertical axis. Next, by interpolating these plotted points with respect to the curves 42A to 42C, local flare (RMS greater than Z38) LF1, LF2, LF3 of the projection optical system PL indicated by the corresponding line widths dP1, dP2, dP3 Can be determined quantitatively and accurately within a range of approximately 0 mλ to 20 mλ in this example. Thus, the process which calculates | requires local flare from the measured value of line | wire width is performed by the host computer (arithmetic unit for calculating | requiring flare) not shown, for example. In this case, the average value of LF1, LF2, and LF3 of the three local flares may be used as the local flare of the projection optical system PL. By averaging in this way, measurement accuracy may be improved. In addition, since the change in the line width of the space pattern image is the largest when the pad-space distance is the shortest (the detection sensitivity is high), the local flare only from the measurement result of the line width when the pad-space distance is the shortest. You may ask for.

  As described above, according to the present example, there is almost no change in the line width of the resist image with respect to the occurrence of aberration in the projection optical system PL, and the line of the resist image with respect to the occurrence of local flare in the projection optical system PL. Since a reticle pattern with a large change in width is used, the amount of local flare in the projection optical system PL can be accurately measured simply by measuring the line width (or change in line width) of the resist image using this reticle pattern. can do.

  In this example, since the pattern of the test reticle R2 of FIG. 2 is transferred onto the wafer by one exposure, the pattern line width error due to the drawing error of the test reticle R2 is the line width of the resist image. There is a risk that local flare measurement errors may occur when mixed in the measurement results. Therefore, in order to eliminate local flare measurement errors due to reticle pattern line width errors (drawing errors), overlay exposure as described with reference to FIGS. 10 to 13 below may be applied.

[Modification of the first step]
In this modification to the first step, a test reticle R3 as a flare measurement mask shown in FIG. 10 is loaded on the reticle stage 22 shown in FIG.
In FIG. 10, a pair of two-dimensional alignment marks 45A and 45B are formed so as to sandwich the pattern region of the test reticle R3 in the X direction. Further, a pad pattern 38 made of a square opening pattern having a width D in the X direction and a height H in the Y direction is formed in the pattern region of the test reticle R3 against the light shielding film 46 of chromium or the like. In addition, eight space patterns 40 each having a rectangular opening pattern having a width d in the X direction and a longitudinal direction in the Y direction are adjacent to the pad pattern 38 in the light shielding film 46 in the X direction. L & S patterns 39 arranged at a pitch P in the X direction are formed. The L & S pattern 39 corresponds to a linear pattern. The shapes of the pad pattern 38 and the L & S pattern 39 are the same as the pad pattern 38A and the L & S pattern 39A of FIG. 2, respectively, and the width D is set equal to the height H here. The positional relationship between the alignment marks 45A and 45B, the pad pattern 38, and the L & S pattern 39 is stored as exposure data in the main control system 1 of FIG.

  Next, an unexposed wafer coated with resist (referred to as wafer W1) is loaded onto the wafer holder 24 on the wafer stage 28 in FIG. First, the test reticle R3 is aligned by detecting the positions of the alignment marks 45A and 45B shown in FIG. 10 using a reticle alignment microscope (not shown) shown in FIG. Next, by adjusting the width of the movable blind 18B in FIG. 1 in the non-scanning direction, only the L & S pattern 39 is obtained during scanning exposure in the Y direction so that the illumination area of the exposure light IL is indicated by the illumination area 21RA in FIG. Set to illuminate. In this illumination state, using the projection exposure apparatus of FIG. 1 as in normal exposure, only the L & S pattern 39 of the test reticle R3 of FIG. 10 is scanned on the wafer W1 by the scanning exposure system PL via the projection optical system PL. Transfer exposure is performed on one shot area. At this time, the L & S pattern 39 may be transferred and exposed also on a plurality of other second, third, etc. shot areas on the wafer W1 (the same applies hereinafter). As a result, as shown in FIG. 11, the first resist image 39AP of the L & S pattern 39 is transferred to the first shot area on the wafer W1.

  Next, after the wafer W1 is moved stepwise in the Y direction by a width in the Y direction of the resist image 39AP and a width of several μm, only the L & S pattern 39 of the test reticle R3 in FIG. 10 is similarly passed through the projection optical system PL. Transfer exposure is performed on the first shot area on the wafer W1 by the scanning exposure method. Next, after the wafer W1 is again moved stepwise in the Y direction by the width of the resist image 39AP in the Y direction and a width of several μm, only the L & S pattern 39 of the test reticle R3 in FIG. 10 is similarly passed through the projection optical system PL. Then, transfer exposure is performed on the first shot area on the wafer W1 by the scanning exposure method. As a result, as shown in FIG. 12, the first, second, and third resist images 39AP, 39BP, and 39CP of the L & S pattern 39 are transferred to the first shot region on the wafer W1 at intervals of several μm in the Y direction. Is done.

  Next, by adjusting the width of the movable blind 18B in FIG. 1 in the non-scanning direction, only the pad pattern 38 is exposed during the scanning exposure in the Y direction so that the illumination area of the exposure light IL is indicated by the illumination area 21RB in FIG. Set to illuminate. 12 is moved stepwise in the X and Y directions so that the pad pattern 38 is projected in the positional relationship of the resist image 38AP in FIG. 3, and then the projection in FIG. 1 is performed in the same manner as in normal exposure. Using the exposure apparatus, only the pad pattern 38 of the test reticle R3 in FIG. 10 is transferred and exposed onto the first shot area on the wafer W1 by the scanning exposure system via the projection optical system PL. Similarly, the position of the wafer W1 is sequentially moved stepwise in the X and Y directions so that the pad pattern 38 is projected in the positional relationship between the resist images 38BP and 38CP in FIG. 3, and the test reticle R3 in FIG. Only the pad pattern 38 is transferred and exposed onto the first shot area on the wafer W1 by the scanning exposure system via the projection optical system PL. Accordingly, as shown in FIG. 13, the first, second, and third resist images of the pad pattern 38 are brought close to the resist images 39AP, 39BP, and 39CP in the X direction on the first shot region on the wafer W1. 38AP, 38BP, and 38CP are transferred. Further, the positional relationship between the resist images 38AP to 38CP and the resist images 39AP to 39CP in FIG. 13 is substantially the same as the positional relationship between the resist images 38AP to 38CP and the resist images 39AP to 39CP in FIG. That is, the distance between the upper resist images 38AP and 39AP in the X direction is the narrowest, and the distance between the central resist images 38BP and 39BP and the distance between the lower resist images 38CP and 39CP are gradually increased.

  Then, after the exposure, the exposed wafer W1 is developed in a developing device (not shown). Thereby, an uneven resist image corresponding to the resist image of FIG. 13 is formed in one or a plurality of shot regions on the wafer W1.

[Modification of the second step]
Here, as in the second step of the above-described embodiment, the resist of the pad pattern 38 is the most among the resist images 39AP, 39BP, and 39CP of the L & S pattern 39 on the wafer W1 in FIG. The line width in the X direction of the resist image corresponding to the space pattern images 40A, 40B, and 40C close to the images 38AP, 38BP, and 38CP is measured. Then, by applying this measurement result to the simulation of FIG. 7, for example, the amount of local flare of the projection optical system PL can be accurately obtained.

  In this modification, the same L & S pattern 39 of FIG. 10 is used as the original pattern of the resist images 39AP to 39CP of FIG. 13, and therefore the line width change of the space pattern image due to the change of the pad-space distance of FIG. The change due to the line width error (drawing error) of the reticle pattern is not included in. Therefore, it is possible to more accurately measure only the line width change due to the local flare of the projection optical system PL. Furthermore, for example, during double exposure, there may be a case where the local flare can be measured with higher accuracy because a portion with a large exposure amount occurs due to the influence of the local flare and the line width of the L & S pattern changes.

  In the above embodiment, the local flare of the projection optical system PL is measured. However, for example, in FIG. 2, by using a measurement pattern in which the distance between the pad pattern 38C and the L & S pattern 39C in the X direction is widened, by measuring the change in the line width of the resist image of the L & S pattern 39C, projection optics There is also a possibility that long range flare resulting from reflection on the coating film of the optical member of the system PL can be obtained.

In the above embodiment, the line width of the resist image after development is measured. However, the line width may be measured at the stage of the resist image (latent image) applied on the wafer, for example. . At this time, a photosensitive resin or the like may be used as the photosensitive material (photosensitive member).
Further, the reticle stage 22 of the projection exposure apparatus of FIG. 1 is stopped, and for example, an image of the pad pattern 38A and the L & S pattern 39A of the test reticle R2 of FIG. In the state projected onto the wafer stage 28 via the projection optical system PL, the wafer stage 28 is moved in the X direction, and the image of the L & S pattern 39A is captured by the pinhole-shaped light receiving surface 30A of the light quantity sensor unit 29 in the X direction. And the line width of the image (aerial image) of the L & S pattern 39A may be directly measured from the detection signal. Even if the measurement result of the line width is applied to the simulation result of FIG. 7, the amount of local flare of the projection optical system PL can be roughly determined. In addition to the light quantity sensor 29, an aerial image measurement system including a relay imaging system and an image sensor such as a line sensor (CCD or the like) is provided on the wafer stage 28. The line width of the image may be directly measured. Thus, the flare of the projection optical system PL can be measured in a very short time without performing the resist coating and developing steps.

  Next, another example of the embodiment of the present invention will be described with reference to FIGS. In this example, the line width of the resist image can be substantially enlarged and measured. In FIGS. 14 and 15, the same or similar reference numerals are given to portions corresponding to FIGS. Detailed description thereof will be omitted. In this example as well, the measurement operation is divided into the following first step and second step, assuming that the flare of the projection optical system PL is measured using the projection exposure apparatus of FIG.

[First step]
In this example, a test reticle R4 as a flare measurement mask shown in FIG. 14 is loaded on the reticle stage 22 shown in FIG.
In FIG. 14, a pair of two-dimensional alignment marks 45A and 45B are formed so as to sandwich the pattern region of the test reticle R4 in the X direction. Further, a pad pattern 38 made of a square opening pattern having a width D in the X direction and a height H in the Y direction is formed in the pattern region of the test reticle R3 against the light shielding film 46 of chromium or the like. Further, eight light-shielding films 46 are formed of rectangular opening patterns which are close to the pad pattern 38 in the X direction and have a distance r1 at the center in the Y direction and a width d in the X direction and a longitudinal direction substantially in the Y direction. A first L & S pattern 47A in which the space pattern 48 is arranged at a pitch P in the X direction is formed. However, in this example, each space pattern 48 of the L & S pattern 47A is inclined counterclockwise by the angle θ with respect to the Y direction.

  Further, eight space patterns formed of a rectangular opening pattern having a width d in the X direction and a length substantially in the Y direction at an interval of several μm in the Y direction with respect to the first L & S pattern 47A in the light shielding film 46. A second L & S pattern 47B in which 49 are arranged at a pitch P in the X direction is formed. However, each space pattern 49 of the second L & S pattern 47B is inclined clockwise by an angle θ with respect to the Y direction. That is, the space pattern 48 constituting the first L & S pattern 47A and the space pattern 49 constituting the second L & S pattern 47B are inclined so as to intersect at an angle 20. L & S patterns 47A and 47B (or space patterns 48 and 49) correspond to linear patterns, respectively. The shape of the pad pattern 38 is the same as the pad pattern 38A of FIG. 2, and the width D and the height H are set equal. The width d of the space patterns 48 and 49, the pitch P (= 2d) of the arrangement, and the interval r1 are the same as those of the space pattern 40 in FIG. The positional relationship between the alignment marks 45A and 45B, the pad pattern 38, and the L & S patterns 47A and 47B is stored as exposure data in the main control system 1 of FIG.

Assuming that the length of the space patterns 48 and 49 in the Y direction is equal to the height H, the inclination angle θ is, for example, the height H as follows, and the ends of the space patterns 48 and 49 are in the X direction. The angle is set so that the position is shifted substantially by the width P.
θ≈P / H = 2d / H (rad) (2)

  Next, an unexposed wafer coated with a photoresist is loaded onto the wafer holder 24 on the wafer stage 28 of FIG. Also in this example, when the photoresist is a chemically amplified resist, it is desirable to apply a top coat thereon. First, the alignment of the test reticle R4 is performed by detecting the positions of the alignment marks 45A and 45B in FIG. 14 using a reticle alignment microscope (not shown in FIG. 1). Next, using the projection exposure apparatus of FIG. 1 in the same manner as in normal exposure, only the pad pattern 38 and the first L & S pattern 47A of the test reticle R4 in FIG. 14 are scanned and exposed through the projection optical system PL. Transfer exposure is performed on the first shot area on the wafer. At this time, the pattern may be transferred and exposed also on a plurality of other second, third, etc. shot areas on the wafer (the same applies hereinafter).

  Next, after the wafer is stepped in the Y direction by a width in the Y direction and a width of several μm of the image of the first L & S pattern 47A, only the second L & S pattern 47B in FIG. 14 is passed through the projection optical system PL. Then, transfer exposure is performed on the first shot area on the wafer by the scanning exposure method. After the exposure, the exposed wafer is subjected to PEB and development in a developing device (not shown). As a result, as shown in FIG. 15, the resist pattern 38AP of the pad pattern 38 of FIG. 14 and the two images of FIG. Resist images 47AP and 47BP obtained by overlapping and exposing the L & S patterns 47A and 47B are formed. Note that the resist is a positive type. In FIG. 15, the wedge pattern (diamond-shaped) overlapping portion A of the space pattern images 48A and 49A in the portion of the resist images 47AP and 47BP that is closest to the resist image 38AP of the pad pattern 38 has a lot of local flare. The line width dP1 is increased, and the length h1 in the Y direction is increased accordingly. On the other hand, the wedge-shaped overlapping portion H of the space pattern images 48H and 49H farthest from the resist image 38AP has almost no local flare fogging light, so the line width dPn in the X direction has almost no local flare. The length hn in the Y direction is also shortened accordingly.

That is, when the inclination angles θ of the space patterns 48 and 49 are used, the lengths h1 and hn in the Y direction are greatly expanded as follows with respect to the widths dP1 and dPn in the X direction of the overlapping portions A and H.
h1≈dP1 / θ, hn≈dPn / θ (3)
In the equation (2), if the pitch P is several hundred nm and the height H is several μm, the inclination angle θ is approximately 0.1 (rad), and the width in the X direction is the height in the Y direction from the equation (3). It is magnified almost 10 times.

[Second step]
Here, instead of measuring the line width dP1 in the X direction of the space pattern image 48A closest to the resist image 38AP of the pad pattern in FIG. 15, the height h1 in the Y direction of the overlapping portion A of the space pattern images 48A and 49A is determined. measure. At this time, since the height h1 is enlarged by about 10 times the line width dP1, it can be measured with relatively high accuracy even using the alignment sensor 36 of FIG. 1, for example. Further, instead of measuring the line width dPn in the X direction of the space pattern image 48H farthest from the resist image 38AP in FIG. 15, the height hn in the Y direction of the overlapping portion H of the space pattern images 48H and 49H is also calculated, for example. Measurement is performed using the alignment sensor 36 of FIG. Thereafter, the line widths dP1 and dPn in the X direction of the space pattern images 48A and 48H are obtained from the equation (3), and this measurement result is applied to the simulation result of FIG. 7, for example, to determine the amount of local flare of the projection optical system PL. It can be determined accurately.

  In this embodiment, instead of exposing the image of the L & S pattern, the images of the two L & S patterns 47A and 47B intersecting each other are overlapped and exposed, and the measurement value of the length in the longitudinal direction of the overlappingly exposed portion is obtained. Since the line widths of the space pattern images 48A and 48H are converted, the line widths can be indirectly measured with high accuracy and efficiency with a measuring device having a simpler structure such as the alignment sensor 36, for example. it can. Therefore, flare measurement can be performed only by, for example, the projection exposure apparatus of FIG.

  It is also possible to measure the lengths h1 and hN of the overlapping portions A and H in FIG. 15 using a scanning electron microscope. In this case, the line width measurement accuracy can be further increased, and as a result, the flare measurement accuracy can be increased.

Next, an example of an exposure method using the above-described flare measurement method of the projection optical system PL will be described separately in the following measurement process and correction process.
[Measurement process]
In this measurement step, for example, the amount of flare (particularly local flare) of the projection optical system PL in FIG. 1 is measured by executing the first step and the second step described above.

[Correction process]
In this correction step, the flare (particularly local flare) of the projection optical system PL is corrected using the flare measurement result in the measurement step. Specifically, flare of the projection optical system PL is reduced by replacing a predetermined optical member (lens or the like) constituting the projection optical system PL. Further, for example, when the first step is executed during the assembly adjustment of the projection optical system PL, a predetermined optical member in the projection optical system PL may be reworked.

After that, by performing the exposure process using the projection optical system PL with corrected flare, the line width uniformity of the finally obtained device such as a semiconductor integrated circuit can be improved, and the yield of the device is improved. it can.
Further, for example, when the line width uniformity is deteriorated in the exposure process, it is possible to accurately determine whether the cause is a flare of the projection optical system by performing the measurement process. . Furthermore, it is possible to accurately measure a change in the flare of the projection optical system with time. Accordingly, it is possible to appropriately deal with troubles occurring in the exposure process.

  The projection exposure apparatus of the above embodiment includes an illumination optical system composed of a plurality of lenses, a projection optical system incorporated in the exposure apparatus main body, and optical adjustment, and a reticle stage and wafer stage composed of a large number of mechanical parts. Is attached to the exposure apparatus main body, wiring and piping are connected, and further comprehensive adjustment (electrical adjustment, operation check, etc.) is performed. The exposure apparatus is preferably manufactured in a clean room where the temperature, cleanliness, etc. are controlled.

  Further, when a semiconductor device is manufactured using the projection exposure apparatus of the above-described embodiment, the semiconductor device includes a step of designing a function / performance of the device, a step of manufacturing a reticle based on this step, and a silicon material. A step of forming a wafer, a step of performing alignment with the projection exposure apparatus of the above-described embodiment and exposing a reticle pattern onto the wafer, a step of forming a circuit pattern such as etching, a device assembly step (dicing process, bonding process, (Including a packaging process) and an inspection step.

  Note that the present invention can be applied not only to scanning exposure type projection exposure apparatuses but also to measuring flare in the projection optical system of a batch exposure type projection exposure apparatus. The present invention can also be applied to the case where the flare of the projection optical system is measured by an immersion type exposure apparatus disclosed in, for example, International Publication (WO) No. 99/49504. Further, the projection optical system in which flare is measured according to the present invention is not limited to application to an exposure apparatus for manufacturing a semiconductor device. For example, a liquid crystal display element or a plasma display formed on a square glass plate The present invention can be widely applied to exposure apparatuses for display devices such as display devices, and exposure devices for manufacturing various devices such as imaging devices (CCDs, etc.), micromachines, thin film magnetic heads, and DNA chips. Furthermore, the projection optical system in which flare is measured according to the present invention is an exposure process (exposure apparatus) for manufacturing a mask (photomask, reticle, etc.) on which a mask pattern of various devices is formed using a photolithography process. It can also be applied to.

  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. In addition, Japanese Patent Application No. 2003-277008 filed July 18, 2003, including specification, claims, drawings, and abstract, and Japanese Patent Application No. 2003 filed October 14, 2003, are filed. The entire disclosure of -353965 is incorporated herein by reference in its entirety.

  Since the flare of the projection optical system can be accurately measured by using the flare measurement method of the present invention, for example, by performing exposure using the projection optical system in which the flare is reduced based on the measurement result, various devices can be obtained. It becomes possible to manufacture with high accuracy with high line width uniformity.

Claims (11)

In a flare measurement method for measuring a flare of a projection optical system that projects an image of a pattern on a first surface onto a second surface,
A first projection that projects an image of an opening pattern arranged on the first surface and an image of one or a plurality of linear patterns arranged adjacent to the opening pattern onto the second surface via the projection optical system. Process,
And a second step of obtaining a flare of the projection optical system based on a line width of the image of the linear pattern projected on the second surface. In the first step, a plurality of sets of opening patterns and linear patterns having the same shape as the opening pattern and the linear pattern and different from each other are formed on the second surface via the projection optical system. Project,
The flare measurement method according to claim 1, wherein in the second step, the flare of the projection optical system is obtained based on line widths of images of the plurality of linear patterns. In the first step,
The flare measurement method according to claim 2, wherein the opening pattern and the image of the linear pattern are exposed a plurality of times so as not to overlap each other and at different intervals. In the first step, an image of another linear pattern is superimposed on the image of the linear pattern at a predetermined small crossing angle and exposed.
In the second step, a longitudinal width of a portion where the image of the linear pattern and the image of the other linear pattern overlap is measured, and a flare of the projection optical system is obtained based on the measurement result. The flare measuring method according to any one of claims 1 to 3, wherein The size of the opening pattern is about several μm to several hundreds μm square, the width of the linear pattern is about several hundred nm to several μm, and the distance between the opening pattern and the linear pattern is about 1 μm to several tens μm. The flare measuring method according to any one of claims 1 to 4, wherein The first step includes
An application step of applying an upper layer film on the photosensitive material after applying the photosensitive material on the substrate;
The surface of the substrate is substantially aligned with the second surface, and the opening pattern and an image of one or more linear patterns arranged adjacent to the opening pattern are passed through the projection optical system to the substrate. Projecting onto the surface of
A pre-development step of performing pre-development baking of the photosensitive material on the substrate;
6. The flare measurement method according to claim 1, further comprising: a development step of developing the photosensitive material on the substrate to form an image of unevenness of the linear pattern. The flare measuring method according to claim 6, wherein the photosensitive material is a chemically amplified resist. The flare measurement method according to claim 6 or 7, wherein the upper layer film suppresses volatilization of acid from the photosensitive material in the preliminary development step. In an exposure method for projecting and exposing a pattern of a first object onto two objects via a projection optical system,
A measurement step of measuring the flare of the projection optical system using the flare measurement method according to any one of claims 1 to 8,
An exposure method comprising: a correction step of correcting a flare of the projection optical system based on a measurement result in the measurement step. A mask for flare measurement of a projection optical system,
A flare measurement mask comprising an opening pattern and one or more linear patterns arranged adjacent to the opening pattern. The opening pattern is a square pattern of approximately several μm to several 100 μm square, and the linear pattern is a line-and-space pattern having a line pattern width of approximately several 100 nm to several μm, and the opening pattern and the line 11. The flare measuring mask according to claim 10, wherein the distance from the pattern is approximately 1 μm to several tens of μm.

Images