JP2014178389A - Measurement method and position adjustment method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To relatively simply and quickly measure the defocus amounts of the measurement position of a surface to be inspected.SOLUTION: A measurement method for measuring defocus amounts as deviation amounts in an optical axial direction from the focal position of an imaging optical system 104 at the measurement position of a surface to be inspected includes: acquiring an evaluation value on the basis of the image of a mask 103 formed by the imaging optical system; and acquiring the defocus amounts corresponding to the evaluation value by using preliminarily acquired relation between the defocus amounts and evaluation value. The mask includes: first regions 35 and 36 having a first phase; and second regions 37 and 38 having a second phase different from the first phase. The evaluation value is acquired on the basis of a first value indicative of the imaging state of an image corresponding to the first regions and a second value indicative of the imaging state of the image corresponding to the second regions.

Description

  The present invention relates to a measurement method for measuring a defocus amount and a position adjustment method using the same.

  In a microscope system for pathological diagnosis called a virtual slide, an enlarged image of a sample that has passed through a microscope optical system is captured by an image sensor. With virtual slides, it is necessary to correct the amount of defocus that occurs due to deformation of the specimen, mechanical errors in the microscope system, and deformation and alteration due to temperature fluctuations, and measurement of the defocus amount distribution is required. It becomes. Patent Document 1 proposes a method of irradiating a sample with a measurement beam and detecting a focused state based on which reflected light forms a spot on a plurality of light receiving portions.

  Further, in the semiconductor exposure apparatus, in order to correct defocus (focus nonuniformity) at a plurality of locations in the exposure shot area, it is necessary to measure each defocus amount (field curvature). As a measuring method, there are a method in which a wafer coated with a resist is exposed and observed with an electron microscope, or a method in which a photoelectric conversion element is provided on a wafer stage and an intensity distribution on an image plane is directly measured. As a mask used for this measurement, Patent Document 2 proposes a mask in which an asymmetric diffraction grating and a reference pattern are juxtaposed, and Patent Document 3 discloses a mask in which an isolated linear light-shielding region and a micro phase shifter are juxtaposed. is suggesting.

JP 2001-305420 A JP 2002-055435 A JP 2010-282115 A

  However, the method of Patent Document 1 cannot calculate the defocus amount by one measurement, and in order to correct the defocus, it is necessary to continuously perform the focusing operation while continuously monitoring the light receiving spot. , Correction takes time. Further, in this method, one set of light receiving portions is required for measurement of one point on the sample, and when this method is applied to a bright field microscope for observing transmitted light from the sample, illumination optics for measurement is used. Since a separate system must be provided, the cost and size of the measuring device increase.

  On the other hand, in the measurement of the defocus amount (field curvature) of the semiconductor exposure apparatus, if the masks disclosed in Patent Documents 2 and 3 are used, two measurement patterns that move according to the defocus amount may approach each other. In this case, the peak or edge of the measurement pattern becomes unclear, and there is a risk that the measurement accuracy will decrease. In addition, in the conventional mask, if there is no prior information measured in advance in order to associate the measurement result with the defocus amount, it is determined whether or not the in-focus position is in front of or behind the wafer. Can not. Furthermore, performance is exhibited only under specific lighting conditions, and it is necessary to switch the lighting conditions from normal use.

  An object of the present invention is to provide a measurement method capable of measuring a defocus amount of a measurement position on a surface to be measured in a relatively simple time and a position adjustment method using the same.

  The measurement method of the present invention is a measurement method for measuring a defocus amount, which is a deviation amount in the optical axis direction from the focal position of the imaging optical system, at the measurement position of the surface to be measured. The defocus amount corresponding to the evaluation value is acquired using the step of acquiring the evaluation value based on the image of the mask to be formed and the relationship between the defocus amount and the evaluation value acquired in advance. And the mask includes a first region having a first phase and a second region having a second phase different from the first phase, and the evaluation value is , Obtained based on a first value representing the imaging state of the image corresponding to the first region and a second value representing the imaging state of the image corresponding to the second region. It is characterized by.

  ADVANTAGE OF THE INVENTION According to this invention, the measuring method which can measure the defocus amount of the measurement position of a to-be-measured surface comparatively easily in a short time, and the position adjustment method using it can be provided.

It is a block diagram of a virtual slide to which the present invention is applicable. 1 is a block diagram of a projection exposure apparatus to which the present invention can be applied. It is a figure which shows the transmittance | permeability distribution and phase distribution of the mask for a measurement used with the measuring method of this invention. Example 1 FIG. 4 is an output image of the image sensor with respect to the measurement mask shown in FIG. 3. Example 1 4 is a graph illustrating a relationship between an image sensor defocus amount and an evaluation value with respect to the measurement mask illustrated in FIG. 3. Example 1 It is a figure which shows the transmittance | permeability distribution and phase distribution of the mask for a measurement used with the measuring method of this invention. (Example 2) FIG. 7 is an output image of the image sensor with respect to the measurement mask shown in FIG. 6. FIG. (Example 2) It is a graph which shows the relationship between the defocus amount of an image sensor with respect to the measurement mask shown in FIG. 6, and an evaluation value. (Example 2) It is a figure which shows the structure of the mask for a measurement used with the measuring method of this invention. (Example 3) FIG. 10 is an output image of the image sensor with respect to the measurement mask shown in FIG. 9. FIG. (Example 3) 10 is a graph showing a relationship between an image sensor defocus amount and an evaluation value with respect to the measurement mask shown in FIG. 9. (Example 3) It is a figure which shows the transmittance | permeability distribution and phase distribution of the mask for a measurement used with the measuring method of this invention. Example 4 13 is an output image of the image sensor with respect to the measurement mask shown in FIG. 12. Example 4 13 is a graph showing a relationship between an image sensor defocus amount and an evaluation value with respect to the measurement mask shown in FIG. 12. Example 4 It is a figure which shows the transmittance | permeability distribution and phase distribution of the mask for a measurement used with the measuring method of this invention. (Example 5) It is an output image of the image pick-up element with respect to the mask for measurement shown in FIG. (Example 5) 16 is a graph illustrating a relationship between an image sensor defocus amount and an evaluation value with respect to the measurement mask illustrated in FIG. 15. (Example 5) It is a figure which shows the transmittance | permeability distribution and phase distribution of the mask for a measurement used with the measuring method of this invention. (Example 6) It is an output image of the image pick-up element with respect to the measurement mask shown in FIG. (Example 6) It is a graph which shows the relationship between the defocus amount of an image sensor with respect to the mask for measurement shown in FIG. 18, and an evaluation value. (Example 6) It is a flowchart of the measuring method and position adjustment method of this embodiment. (Examples 1-6)

  FIG. 1 is a block diagram of a microscope system for pathological diagnosis called a virtual slide to which the present invention can be applied. The virtual slide includes a microscope 100, a transport unit 201, a control unit 202, an image processing unit 203, and a personal computer (PC) 204. The microscope 100 includes an illumination optical system 101, a stage 102, an imaging optical system 104, and an imaging element 105.

  The illumination optical system 101 illuminates a sample such as a measurement mask (hereinafter simply referred to as “mask”) 103 mounted on the stage 102 or a human tissue piece (not shown) under appropriate illumination conditions. The stage 102 supports the mask 103 and the sample, and moves in the optical axis direction of the illumination optical system 101 and two directions orthogonal thereto. The imaging optical system 104 is an imaging optical system that forms an optical image of a mask 103 that is an object or a sample. The image sensor 105 photoelectrically converts the optical image formed by the imaging optical system 104.

  The transport unit 201 mounts the mask 103 and the sample on the stage 102 based on a command from the control unit 202. The control unit 202 controls the operation of each part of the virtual slide. The image processing unit 203 performs various processes on a digital signal obtained by A / D converting the output (analog signal) of the image sensor 105.

  The PC 204 includes a CPU (arithmetic unit) that performs arithmetic processing and a storage unit that stores an imaging result of the sample. The PC 204 is connected to a network such as the Internet and provides an image corresponding to a read request from the outside. The storage unit also stores a relationship between a defocus amount (described later) and an evaluation value acquired in advance. The PC 204 acquires the defocus amount at the measurement position of the surface to be measured from the measurement result described later and the relationship. In addition, the controller 202 is instructed to adjust the position of the image sensor 105 or the stage 102 so that the defocus amount of the measurement position on the surface to be measured is reduced.

  In the virtual slide, when the visual field (observation region of the sample) is larger than one image sensor 105, a configuration including a plurality of image sensors 105 may be employed. The mask 103 or the sample and the imaging surface of the imaging element 105 must have a conjugate relationship (that is, the relationship between the object and the image) by the imaging optical system 104, but the imaging surface of the imaging element 105 may be defocused. . For this reason, it is necessary to measure and correct the defocus amount on the imaging surface (test surface) of each imaging element 105. Here, the “defocus amount” is a shift amount in the optical axis direction from the focal position of the imaging optical system at the measurement position of the surface to be measured such as the imaging surface of the image sensor 105. If the in-focus state is 0, a positive value is obtained when the light source (object) is moved away from the light source (object), and a negative value is obtained when the light is moved away from the light source (object).

  FIG. 21 is a flowchart for explaining a measurement method and a position adjustment method according to this embodiment.

  Before the sample is observed and imaged, a measurement step (S10) for measuring the defocus amount (or distribution of the defocus amount) at one or a plurality of measurement positions on the test surface using the mask 103 is performed. Then, a position adjustment step (S20) for adjusting the position of the image sensor 105 or the stage 102 so as to correct the defocus amount measured in the measurement step is performed.

  In the measurement step (measurement method) of the present embodiment, the output of the image sensor 105 is acquired using the mask 103, and the measurement position of the test surface is determined based on the acquired output of the image sensor 105 and previously acquired information. Get the defocus amount. Specifically, the illumination optical system 101 illuminates the mask 103 mounted on the stage 102 by the transport unit 201 under measurement conditions, and the imaging element 105 is imaged on the surface to be measured by the imaging optical system 104. The optical image of the mask 103 is photoelectrically converted (exposure and imaging). The image processing unit 203 performs data processing on the image of the mask imaged by the plurality of image sensors 105, and obtains an evaluation value to be described later for evaluating the in-focus state of the measurement position on the test surface (S12). . The PC 204 acquires the defocus amount at the measurement position of the test surface corresponding to the evaluation value from the acquired evaluation value and information acquired in advance (data related to the relationship between the defocus amount and the evaluation value) (S14). ). Then, the PC 204 adjusts the position of the image sensor 105 or the stage 102 based on this information, and then enters a process of observing the sample. S12 and S14 may be embodied as a program executed by a computer.

  The mask 103 has a light passing portion through which light passes. In the following embodiments, a first region that is a light passing portion having a first phase and a second phase different from the first phase are used. The light passage part having a phase will be described with respect to the second region. The light from the illumination optical system 101 passes through the first region and the second region of the mask 103, and the imaging optical system 104 forms an image corresponding to each of the first region and the second region.

  The manner in which the imaging state of the image corresponding to the first region changes according to the defocus amount is different from the manner in which the imaging state of the image corresponding to the second region changes according to the defocus amount. If only the image formation state of one image is used as an index (for example, using only an index indicating the image formation state of an image corresponding to the first region), there is no comparison reference, so that the measurement accuracy decreases, It takes a long time to measure the number of times. Therefore, in the present embodiment, a combination of the image formation states of two types of regions having different ways of changing the defocus amount is used as the evaluation value. The evaluation value is acquired based on the first value that represents the imaging state of the image corresponding to the first region and the second value that represents the imaging state of the image corresponding to the second region. . That is, it is generated by combining the first value and the second value, and uniquely corresponds to the defocus amount (one to one). In this embodiment, by using such an evaluation value, the measurement accuracy is increased and the measurement time is shortened.

  In one embodiment, the mask 103 has an effect of shifting the position of a feature point (for example, a point at which the luminance takes a maximum value) in the acquired image in different directions according to the defocus amount. In another embodiment, the mask 103 has an effect of changing the magnitude relationship between the image intensities of a plurality of feature points in the acquired image according to the defocus amount.

  The first light passage portion corresponding to the first region and the second light passage portion corresponding to the second region may be separated by a light shielding portion, or within one continuous light passage portion. It may be provided. The shape of the first region and the second region may be the same or different. The shapes of the first light passage part and the second light passage part may be the same or different. The transmittance of the first light passage section and the transmittance of the second light passage section may be the same or different. When the first light passing portion and the second light passing portion are provided in pairs, the pattern image of the mask 103 tends to be clear. That is, the mask 103 may include a plurality of each of the first region and the second region. The first phase and the second phase may be constant in the first region and the second region, respectively, or may have a distribution. When forming one pattern including the first region and the second region, the mask 103 may have a plurality of patterns. Thereby, the defocus amount of the several measurement position of a to-be-examined surface can be acquired by one measurement.

  FIG. 2 is a block diagram of an exposure apparatus to which the present invention can be applied. As in FIG. 1, before the exposure of the wafer or liquid crystal substrate as the object to be processed, in order to obtain the position adjustment amount of the stage 301 for correcting the defocus amount in advance, the image intensity distribution of the image using the mask 303 is obtained. Acquisition or exposure to a measurement substrate coated with a resist is performed. The mask 303 is replaced with a mask (reticle) on which a pattern to be transferred is formed after the measurement is completed. First, the transport unit 401 moves the mask 303 to the projection exposure apparatus 300 based on a command from the control unit 402. Next, the illumination optical system 304 illuminates the mask 303 under the same conditions as normal exposure or measurement conditions, and the projection optical system 302 as an imaging optical system applies a photoelectric conversion element or resist on the stage 301. An image is formed on a measurement substrate (not shown). When a photoelectric conversion element is used, a signal related to the image intensity distribution at a plurality of positions on the stage 301 is transmitted to the image processing unit 403, and predetermined data processing to be described later is performed, whereby defocus amounts at the plurality of positions are obtained. Is calculated. On the other hand, when a measurement wafer is used, the measurement wafer is developed, the resist shape is observed with an electron microscope to obtain image information, and a predetermined data process is performed on the image in the same manner. The evaluation value at the measurement position of the surface inspection is acquired. As in FIG. 1, the PC 404 calculates the defocus amount from the acquired evaluation value and information acquired in advance (data relating to the relationship between the defocus amount and the evaluation value). Based on this information, the position of the stage 301 or the mask 303 is corrected or the optical system 302 is adjusted before exposing the object to be processed, or the position of the stage 301 is corrected during the exposure of the object to be processed.

  Hereinafter, in each embodiment, a method for measuring the defocus amount at the measurement position on the test surface will be described using the virtual slide in FIG. 1 as an example.

  In Example 1, the image-side numerical aperture of the imaging optical system 104 is 0.7, the imaging magnification is 20 times, the pixel size of each imaging element 105 is 1.5 μm, and the area aperture ratio of the pixel is 80%. The wavelength was assumed to be white light. The relative intensities at wavelengths of 400 nm, 500 nm, 550 nm, 600 nm, and 700 nm are 0.1, 1.0, 1.5, 1.0, and 0.1, respectively.

  In the first embodiment, the (measurement) mask 103a shown in FIG. 3 is used as the mask 103 shown in FIG. 3A shows the transmittance distribution of the mask 103a, and FIG. 3B shows the phase distribution of the mask 103a. In the following, the unit of phase in all the diagrams showing the phase distribution is radians. In FIG. 3, the plurality of first regions and the plurality of second regions are arranged so as to intersect with each other, and the intersecting portion functions as a light shielding portion.

  In FIG. 3A, a light shielding portion 30 and four openings 31 to 34 are provided. The four openings form a cross shape, and the crossing portion (center portion in FIG. 3) is a light shielding portion. Each opening has a square shape with a side of 0.25 μm. A square with a side of 0.25 μm centered at (x, y) = (0, 0) is a light shielding portion. The upper and lower openings 31, 32 arranged in the y direction (first direction) of the light shielding part at the center of the cross form a pair and function as a first opening (first light passing part). The left and right openings 33 and 34 arranged in the x direction (second direction) of the light shielding part at the center of the cross form a pair and function as a second opening (second light passing part). The transmittances of the first opening and the second opening are both 1, and the other is 0.

  In FIG.3 (b), the same area | region as each opening part is provided in the same position. The phase of the pair of first regions 35 and 36 corresponding to the pair of first openings is 1.02 radians, and the phase of the pair of second regions corresponding to the pair of second openings 37 and 38 is Both phases are -1.02 radians. The phases of the areas other than the first areas 35 and 36 and the second areas 37 and 38 are zero. The phase of the first region is constant, and the phase of the second region is constant. The first regions 35 and 36 and the second regions 37 and 38 are in contact with each other at the corners (in a diagonal direction), but may be separated from each other. The first regions 35 and 36 and the second regions 37 and 38 have a phase difference other than 0 radians, in this embodiment, a phase difference of 2.04 radians with respect to a wavelength of 550 nm.

  This phase difference can be realized, for example, by providing different amounts of digging for each opening in the transparent substrate of the mask 103a. Here, even when it is assumed that a manufacturing error of 100 nm is added to the digging amount, the phase difference error is about 0.29 radians or less, and the same effect as described below can be obtained. In consideration of manufacturing errors, the phase difference is preferably 0.61 radians or more and 3.24 radians or less. The image shape to be described later depends on this phase difference, and if the phase difference is less than 0.61 radians, the difference in shape for each image corresponding to each defocus amount becomes small, and it is difficult to obtain the defocus amount. Become. On the other hand, if it is larger than 3.24, the mask pattern image obtained from the image sensor 105 becomes difficult to be clear.

  FIG. 4 shows an output image of the image sensor 105 with respect to the defocus amount of the test surface when the mask 103a is illuminated with vertically incident coherent light. In FIG. 5, the output image is binarized with a threshold value of 35% of the maximum value, and the horizontal width (second value) of the obtained elliptical region is changed to the vertical width (second value). It is a graph which shows the result of having plotted the evaluation value which subtracted 1st value) with respect to the said defocus amount. The vertical direction of the elliptical area corresponds to the arrangement direction of the plurality of first areas 35 and 36, and the horizontal direction of the elliptical area corresponds to the arrangement direction of the plurality of second areas 37 and 38.

  FIG. 5 is a graph showing the relationship between the defocus amount and the evaluation value. As described above, the evaluation value is an index indicating how long an ellipse is (here, how long it is horizontally long). The ellipse is set by approximating each image (white image) shown in FIG. For example, a vertically long ellipse is set for an image with a defocus amount of −200 μm in FIG. 4, an almost perfect ellipse is set for an image with a defocus amount of 0 μm, and an image with a defocus amount of 200 μm is set. The horizontal ellipse is set. However, the evaluation value need not be limited to the difference in the width of the ellipse, and may be an index that quantitatively reflects the shape of the mask pattern image.

  If the horizontal and vertical end points of the elliptical area are feature points, the position of the feature point is moved according to defocus. If the measurement for a known defocus amount is performed in advance, and the information shown in FIG. 5 is acquired and stored in a storage means (not shown), the corresponding measurement can be performed by acquiring the evaluation value by the measurement method of this embodiment. The defocus amount of the position can be acquired. For example, the defocus amount at the position calculated as the evaluation value of the test surface of −6.0 μm is identified as −100 μm. The relationship between the defocus amount and the evaluation value is not limited to the graph, and may be a table or an expression.

  The advantage of using the mask 103a over the conventional method is that it is possible to determine whether the in-focus state or the defocus direction without any prior information. For example, in this embodiment, if the evaluation value is negative, it can be determined that the image sensor 105 is defocused in a direction away from the light source. Needless to say, if the evaluation value is positive, the evaluation value may be defined so as to determine that the focus is away from the light source. Furthermore, by using a mask provided with a plurality of patterns shown in FIG. 3 and performing the above-described series of procedures, defocus amounts at a plurality of positions on the test surface can be simultaneously measured.

  The number of times of imaging does not need to be limited to one, and a plurality of images are acquired, and the final measurement is performed by performing statistical processing such as averaging on the defocus amount acquired from the images using the above-described method. It may be a value.

  In the above measurement, it is not always necessary to switch the illumination condition in accordance with the mask 103a, and the measurement may be performed under the same illumination condition as in normal sample observation. Further, the field curvature of the projection exposure apparatus can be measured by the same method.

  The second embodiment uses the same conditions as the first embodiment except that the mask 103B is used instead of the mask 103a. The mask 103b has a transmittance distribution shown in FIG. 6A and a phase distribution shown in FIG. This is the result of Fourier transform of the complex number distribution whose phase is the fifth term of the Zernike polynomial representing astigmatism, the amplitude is 50% of the maximum value, and the phase is ± 1.58 for the wavelength of 550 nm. It is obtained by ternizing with radians as a threshold value. The result of this Fourier transform is analytically represented by the sum of the Bessel functions. When the phase distribution of the mask 103b is represented by a Bessel function, the pattern image of the mask 103b tends to be clear. The means for realizing the phase difference is the same as in the first embodiment.

  In FIG. 6A, one rhombus opening (light passage portion) 41 having a transmittance of 1 is provided in the mask 103 b, the surrounding black portion is a light shielding portion 40, and the transmittance is zero. The first light passing portion and the second light passing portion are provided inside one opening.

  As shown in FIG. 6B, in the mask 103b, the first regions 45 and 46 and the second regions 47 and 48 have the same spade shape, and a pair of spades arranged in the Y direction are the first. A pair of spades arranged in the X direction is a second area. The pair of first regions 45 and 46 are provided symmetrically with respect to the line y = 0, and the pair of second regions 47 and 48 are provided symmetrically with respect to the line x = 0. The phase of the pair of first regions 45 and 46 is 1.58 radians, and the phase of the pair of second regions 47 and 48 is −1.58 radians. The phases of the areas other than the first area and the second area are zero. The phases of the first regions 45 and 46 are constant, the phases of the second regions 47 and 48 are constant, and the first phase and the second phase are different. The first regions 45 and 46 and the second regions 47 and 48 are spaced apart.

  FIG. 7 shows an output image of the image sensor 105 for each defocus amount when the mask 103b is illuminated with vertically incident coherent light. In FIG. 8, the output image is binarized with a value of 30% of the maximum value as a threshold value, and the horizontal width (second value) of the obtained elliptical region is changed to the vertical width (second value). It is a graph which shows the result of having plotted the evaluation value which subtracted 1st value) with respect to the defocus amount. The vertical direction of the elliptical area corresponds to the arrangement direction of the plurality of first areas 45 and 46, and the horizontal direction of the elliptical area corresponds to the arrangement direction of the plurality of second areas 47 and 48.

  If the horizontal and vertical end points of the elliptical area are feature points, the position of the feature point is moved according to defocus. If the measurement for a known defocus amount is performed in advance and the information shown in FIG. 8 is acquired and stored in a storage means (not shown), the corresponding measurement can be performed by acquiring the evaluation value by the measurement method of this embodiment. The position defocus amount can be identified. For example, the defocus amount at the position calculated as the evaluation value of the test surface of −6.0 μm is identified as −100 μm.

  The evaluation value is not limited to the difference in length, and may be an index that quantitatively reflects the image shape. The advantages over the conventional method, simultaneous measurement at a plurality of positions, imaging of a plurality of sheets, illumination conditions, and application to a projection exposure apparatus are the same as in the first embodiment.

  The third embodiment uses the same conditions as the first embodiment except that the illumination conditions and the mask 103c are used instead of the mask 103a. The illumination light has a single wavelength of 550 nm, and the mask 103c has the structure shown in FIG. FIG. 9A is an enlarged perspective view of the mask 103c. FIG. 9B is a top view of the mask 103c. FIG. 9C is a cross-sectional view of the mask 103c.

  As shown in FIG. 9C, the mask 103 c includes a light shielding film 51 and a transparent substrate 52. The light shielding film 51 is made of, for example, a material that hardly transmits illumination light, such as chromium, and the transparent substrate 52 is made of, for example, a material having a constant refractive index and low light absorption, such as optical glass. . In FIG. 9, the transparent substrate 52 is optical glass having no absorption and a refractive index of 1.4, and the light shielding film 51 is chromium having a complex refractive index of 2.38 + 2.97i.

  The transparent substrate 52 has two kinds of intersecting inclined surfaces when viewed from the side as shown in FIG. 9C, in the y coordinate range of −2 μm to 2 μm. As shown in FIG. 9C, the first inclined surface is an inclined surface having an angle θ1 with respect to the horizontal plane, and the second inclined surface is an inclined surface having an inclination angle angle θ2 different from θ1 with respect to the horizontal plane.

  As shown in FIGS. 9A and 9C, the first inclined surface is formed in a region where the x coordinate is positive, and the second inclined surface is formed in a region where the x coordinate is negative. Yes. It is desirable that the angles θ1 and θ2 have opposite directions and the same size. In this embodiment, θ1 = 0.79 radians (45 °) and θ2 = −0.79 radians (−45 °), but are not limited thereto.

  As shown in FIG. 9A, a pair of rectangular light-shielding films 51 are formed on the first inclined surface, and a rectangular slit (first slit) 53 is formed therebetween. Similarly, a pair of rectangular light shielding films 51 are formed on the second inclined surface, and a rectangular slit (second slit) 53 is formed therebetween. The two slits 53 are openings (light passage portions) through which light passes. The pair of slits 53 shown in FIG. 9B has different phase distributions in the y direction. That is, one of the two slits 53 functions as a first region and the other functions as a second region, and the phase distribution inside the first slit and the phase distribution inside the second slit are different.

  FIG. 10 shows an output image of the image sensor 105 for each defocus amount when the mask 103c is illuminated with circular illumination having a coherence factor of 0.7. The reason why the peak positions of the image intensities of the two slits 53 change is that the y coordinate of the slit 53 in the focused z coordinate is different between the two slits. Using this phenomenon, the defocus amount can be determined by calculating the difference (distance) between the y coordinates (first value and second value) of two image intensity peaks as an evaluation value.

  As shown in FIG. 11, as the defocus amount of the test surface increases positively, the difference in the y coordinate of the peak also increases monotonously. If the peak is a feature point, the position of the feature point is moved according to the defocus amount. If the measurement for a known defocus amount is performed in advance, and the data shown in FIG. 11 is acquired and stored in a storage means (not shown), the corresponding measurement is performed by acquiring the evaluation value by the measurement method of this embodiment. The position defocus amount can be identified. For example, the defocus amount at the position where the evaluation value of the test surface is calculated to be −9.0 μm is identified as −100 μm.

  The advantages over the conventional method, simultaneous measurement at a plurality of positions, imaging of a plurality of sheets, illumination conditions, and application to a projection exposure apparatus are the same as in the first embodiment.

  In the present embodiment, the absolute values of θ1 and θ2 are set to 0.79 radians (45 °). However, if these values are reduced, more accurate defocus measurement can be performed, and if the values are increased, the measurement is performed. The range of possible defocus amounts is expanded. Furthermore, the same effect can be obtained even if the slope of the transparent substrate 52 has a stepped shape approximating it.

  The fourth embodiment uses the same conditions as in the first embodiment except that the mask 103d is used instead of the mask 103a. The mask 103d has a transmittance distribution shown in FIG. 12A and a phase distribution shown in FIG.

  As shown in FIG. 12A, three openings 61, 62, 63 as light passing portions are arranged in the y direction, and the other portions are light shielding portions 60. Each opening is a square having a side of 1.0 μm, the interval is 2.0 μm, and if the central opening 61 is a first opening (first light passage), the openings 62 and 63 at both ends are second. This is an opening (second light passing portion).

  As shown in FIG. 12B, a first region 65 having a first phase is formed at the same position as the central opening 61 in the xy cross section. The first region 65 and the first opening have the same shape and the same position, and the first phase has a distribution that decreases in the positive direction of the x-axis. The phase difference between both ends in the x direction of the first region 65 is 4.71 radians with respect to a wavelength of 550 nm, and is a linear inclination. A second region 67 having a second phase is formed at the same position as the openings on both sides in the y direction of the central opening, but the second region 67 excludes the first region 65. This is the entire surface of the mask 103. Further, the phase of the second region is constant. The first phase and the second phase are different.

  In the present embodiment, such a phase gradient can be realized, for example, by digging the transparent substrate of the mask 103d so as to have a slope in the opening. Here, even when a manufacturing error is added to the phase inclination or the amount of digging, the same effect as described below can be obtained.

  FIG. 13 shows an output image of the image sensor 105 for each defocus amount when the mask 103D is illuminated with vertically incident coherent light. For this output image, three peaks (points where the brightness is maximum) are detected. FIG. 14 shows the result of plotting the defocus amount with the relative position (which may be a signed shortest distance or a relative x coordinate) between the central peak and the line connecting the upper and lower peaks as an evaluation value. It is a graph. Either one of the upper and lower openings may be provided, and the evaluation value in this case is that the brightness of the image corresponding to the first area is maximum and the brightness of the image corresponding to the second area is maximum. Relative position in a predetermined direction. In any case, the evaluation value is a relative position between the point where the brightness of the image corresponding to the first region is maximum and the point where the brightness of the image corresponding to the second region is maximum.

  If the three peaks are feature points, the position of the feature point moves according to the defocus amount. The advantage of using the mask 103d over the conventional method is that the three patterns are too close to any defocus amount, the image is not deteriorated and the measurement accuracy is not lowered, and the defocus amount is reduced. The measurable range is expanded.

  If the measurement for a known defocus amount is performed in advance, and the information shown in FIG. 14 is acquired and stored in a storage means (not shown), the corresponding measurement is performed by acquiring the evaluation value by the measurement method of this embodiment. The position defocus amount can be identified. For example, the defocus amount at the position calculated as the evaluation value of the test surface of −6.0 μm is identified as −200 μm.

  The evaluation value is not limited to the positional deviation amount, and may be an index that quantitatively reflects the relative position of the image intensity peak.

  The advantages over the conventional method, simultaneous measurement at a plurality of positions, imaging of a plurality of sheets, illumination conditions, and application to a projection exposure apparatus are the same as in the first embodiment. Furthermore, the phase gradient of the mask 103d may be a stepwise phase distribution that approximates this, and the same effect can be obtained in this case.

  The fifth embodiment uses the same conditions as the first embodiment except that the mask 103e is used instead of the mask 103a. The mask 103e has a transmittance distribution shown in FIG. 15A and a phase distribution shown in FIG. As shown in FIG. 15A, the pattern of the mask 103 e has a light shielding portion 70 and four openings 71 to 74. The four openings are provided at positions corresponding to the four corners of the rectangle, and a pair of openings arranged in a diagonal direction form a pair. That is, the upper left opening 71 and the lower right opening 72 function as a first opening (first light passing portion), and the upper right opening 73 and the lower left opening 74 are the second opening ( 2nd light passage part). As shown in FIG. 15B, first regions 75 and 76 having the same shape are provided at the same position as the first opening, and the first shape having the same shape is provided at the same position as the second opening. Two regions 77 and 78 are provided. Note that the phase distribution of the paired regions only needs to have the same inclination direction and does not necessarily need to be completely coincident.

  Each opening is a square having a side of 1.0 μm, and the interval is 3.0 μm. The first phase has a distribution that decreases in the x direction, the second phase has an inverse distribution that increases in the x direction, and the first phase and the second phase are different. The phase difference between both ends of the first region and the second region in the x direction is 4.71 radians with respect to the wavelength of 550 nm, which is a linear inclination. The means for realizing the phase gradient is the same as in the fourth embodiment. The upper first region 75 and the second region 77 have a phase gradient that increases away from each other, and the lower first region 76 and the second region 78 increase in phase so as to approach each other. Has a gradient.

  FIG. 16 shows an output image of the image sensor 105 for each defocus amount when the mask 103e is illuminated with vertically incident coherent light. FIG. 17 is a graph showing a result of plotting the defocus amount with respect to this output image, using the difference obtained by subtracting the lower two inter-peak distances from the upper two inter-peak distances as an evaluation value.

  If the peak is a feature point, the position of the feature point is moved according to the defocus amount. If measurement is performed for a known defocus amount in advance, and the information shown in FIG. 17 is acquired and stored in a storage unit (not shown), the corresponding measurement is performed by acquiring the evaluation value by the measurement method of this embodiment. The position defocus amount can be identified. For example, the defocus amount at the position calculated as the evaluation value of −25 μm of the test surface is identified as −200 μm.

  The evaluation value need not be limited to the difference in distance, and may be an index that quantitatively reflects the relative position of the peak of the image intensity. The advantages over the conventional method, simultaneous measurement at a plurality of positions, imaging of a plurality of sheets, illumination conditions, and application to a projection exposure apparatus are the same as in the first embodiment. Furthermore, the phase gradient of the mask 103e may be a stepwise phase distribution that approximates this, and the same effect can be obtained in this case.

  Example 6 uses the same conditions as Example 1 except that the wavelength used for measurement and the mask 103f are used instead of the mask 103a. The illumination light has a single wavelength of 550 nm, and the mask 103f has a transmittance distribution shown in FIG. 18A and a phase distribution shown in FIG. As shown in FIG. 18A, the mask 103f has two openings 81 and 82 and a light shielding part 80 other than that. The left opening 81 functions as a first opening (first light passage), and the right opening 83 functions as a second opening (second light passage). Each opening is a square with a side of 1.0 μm, and the interval is 0.5 μm. A first region 85 having the same shape is provided at the same position below the first opening, and has a first phase of -0.785 radians. A second region 87 having the same shape is provided at the same position below the second opening, and has a first phase of 0.785 radians. That is, the phase difference is 1.57 radians. The means for realizing the phase difference is the same as in the first embodiment.

  FIG. 19 shows a one-dimensional cross section of the output image of the image sensor 105 with respect to defocus amounts of ± 200 μm and 0 μm when the mask 103f is illuminated with normal-incidence monochromatic coherent light. A dotted line corresponds to a defocus amount of −200 μm, a solid line corresponds to 0 μm, and a one-dot chain line corresponds to +200 μm. It can be seen that the magnitude relationship between the two peaks is reversed depending on the defocus direction.

  In FIG. 20, the left peak position image intensity (second value) is subtracted from the right peak position image intensity (first value) of this output image, and divided by the maximum value of the total intensity, the unit is expressed as%. It is a graph which shows the result of having plotted the evaluated value with respect to the defocus amount.

  If measurement is performed for a known defocus amount in advance, and the information shown in FIG. 20 is acquired and stored in a storage unit (not shown), an evaluation value corresponding to the evaluation value is acquired by acquiring the evaluation value by the measurement method of this embodiment. The defocus amount at the measurement position corresponding to the value can be acquired. For example, the defocus amount at the position calculated as an evaluation value of -2.7% of the test surface is identified as -100 μm.

  Advantages over the conventional method, simultaneous measurement at a plurality of positions, imaging of a plurality of sheets, illumination conditions, and application to a projection exposure apparatus are the same as in the first embodiment. For example, it is preferable that the phase difference between the first phase and the second phase in vacuum with respect to the wavelength of the illumination light is 0.61 radians or more and 3.24 radians or less.

  As mentioned above, although the preferable Example of this invention was described, this invention is not limited to these Examples, A various deformation | transformation and change are possible within the range of the summary. For example, in the above-described embodiments, the imaging surface of the imaging device is the test surface, but the present invention is not limited to this, and may be a part of a sample or mask, the surface of a stage, or the like. In each of the embodiments, the mask provided with the opening as the light passing portion through which the light is transmitted has been described. However, the present invention is not limited thereto, and a light reflecting pattern may be provided as the light passing portion. .

  The present invention can be applied to, for example, a virtual slide and an exposure apparatus.

DESCRIPTION OF SYMBOLS 103 ... Mask for measurement, 104 ... Imaging optical system (imaging optical system), 35, 36 ... 1st area | region, 37, 38 ... 2nd area | region

Claims (20)

A measurement method for measuring a defocus amount, which is a deviation amount in a direction of an optical axis from a focal position of an imaging optical system at a measurement position of a test surface,
Obtaining an evaluation value based on an image of a mask formed by the imaging optical system;
Using the relationship between the defocus amount and the evaluation value acquired in advance, obtaining the defocus amount corresponding to the evaluation value;
Have
The mask includes a first region having a first phase and a second region having a second phase different from the first phase;
The evaluation value is based on a first value representing an imaging state of an image corresponding to the first region and a second value representing an imaging state of an image corresponding to the second region. A measurement method characterized by being acquired.   The measurement method according to claim 1, wherein the evaluation value is a value that reflects a shape of an image of the mask that changes in accordance with the defocus amount. The mask includes a plurality of each of the first region and the second region,
The first value is a width of the image of the plurality of first regions in the arrangement direction of the plurality of first regions, and the second value is in the arrangement direction of the plurality of second regions. The measurement method according to claim 2, wherein the width is an image width of the plurality of second regions.   The measurement method according to claim 3, wherein the plurality of first regions and the plurality of second regions are arranged so as to intersect with each other, and the intersecting portion functions as a light shielding unit.   5. The device according to claim 1, wherein the first phase is constant in the first region, and the second phase is constant in the second region. 6. Measurement method.   The measurement method according to claim 1, wherein the mask has a phase distribution represented by a Bessel function.   The measurement method according to claim 1, wherein the evaluation value is a value that reflects an image intensity of the mask that changes in accordance with the defocus amount.   The first value is generated from the coordinates of the point at which the brightness of the image corresponding to the first area is maximum, and the second value is the brightness of the image corresponding to the second area. The measurement method according to claim 7, wherein the measurement method is generated from the coordinates of a point that becomes maximum. The mask includes a first inclined surface and a second inclined surface having an inclination angle different from that of the first inclined surface;
A first slit that functions as the first region is formed on the first inclined surface, and a second slit that functions as the second region is formed on the second inclined surface. The measurement method according to claim 7 or 8, wherein the measurement method is performed.   The measurement method according to claim 9, wherein at least one of the first inclined surface and the second inclined surface is approximated by a step shape.   The measurement method according to claim 1, wherein the first phase has a linear distribution in the first region.   The measurement method according to claim 11, wherein the first phase has a distribution approximated to a step shape.   The measurement method according to claim 1, wherein the second phase has a linear distribution in the second region.   The measurement method according to claim 13, wherein the second phase has a distribution approximated to a step shape. The first phase has a linear distribution in the first region and the second phase has a linear distribution in the second region;
The measurement method according to claim 8, wherein the mask includes a plurality of first regions arranged in a diagonal direction and a plurality of second regions arranged in a diagonal direction.   The first value is generated from the intensity of the point where the brightness of the image corresponding to the first area is maximum, and the second value is the brightness of the image corresponding to the second area. The measurement method according to claim 7, wherein the measurement method is generated from an intensity of a point that becomes maximum.   The measurement method according to claim 16, wherein the first phase is constant in the first region, and the second phase is constant in the second region.   When one pattern including the first region and the second region is formed, the mask has a plurality of patterns, and the defocus amounts at a plurality of measurement positions on the test surface are set once. The measurement method according to any one of claims 1 to 17, wherein the measurement method is acquired by measurement.   The phase difference between the first phase and the second phase in vacuum with respect to the wavelength of illumination light that illuminates the mask is 0.61 radians or more and 3.24 radians or less. 17. The measuring method according to 17. A measurement step of measuring the defocus amount using the measurement method according to any one of claims 1 to 19,
Adjusting the position of the test surface in the optical axis direction of the imaging optical system so that the defocus amount measured by the measurement step is reduced;
A position adjusting method characterized by comprising: