JP7807271B2 - Photomask inspection equipment - Google Patents

Description

This disclosure relates to a photomask inspection device.

Inspection devices for inspecting photomasks (reticles) have been proposed in the past (Patent Documents 1 and 2).

In Patent Document 1, a light source is provided that outputs light including g-, h-, and i-rays. Light from the light source is irradiated onto a photomask through a wavelength-selective filter, and light that passes through the photomask is incident on the imaging surface of an imaging means. The wavelength-selective filter includes a first filter that transmits only g-rays, a second filter that transmits only h-rays, and a third filter that transmits only i-rays. Light selectively passes through these first, second, and third filters. When light passes through the first filter, only g-rays enter the imaging means through the photomask, causing the imaging means to generate an image using the g-ray. When light passes through the second filter, the imaging means generates an image using the h-ray, and when light passes through the third filter, it generates an image using the i-ray. In Patent Document 1, these three images are combined using wavelength calculations to generate a composite image using light including the g-, h-, and i-rays, and the photomask is inspected based on the composite image.

In Patent Document 2, multiple illumination sources are provided, each of which includes multiple laser diode arrays. The beams output from the multiple illumination sources are combined by beam combining optics, causing the beams to be incident on the sample with a desired illumination profile. The beams reflected from the sample are incident on a detector and detected by the detector. The inspection device determines sample parameters, such as the critical dimensions of the sample, based on the signals output from the detector.

Japanese Patent Application Laid-Open No. 2008-256671 Japanese Patent Application Laid-Open No. 2020-064063

However, in Patent Document 1, g-line images, h-line images, and i-line images are captured at different times. As a result, positional variations occur in the captured images due to mechanical vibrations of the inspection device. As a result, the composite image contains errors caused by this positional variation, resulting in a problem of reduced inspection accuracy.

In Patent Document 2, a laser diode array is used as the light source. Each laser diode array is provided with multiple light-emitting elements, which results in variations in at least one of the wavelength and phase of the light emitted from each illumination source among the multiple elements. This reduces the coherence of the light emitted from each illumination source.

Therefore, even if the illumination profile is made closer to the illumination profile of the light source in the exposure tool, the diffraction phenomenon that occurs on the sample in the inspection tool will be significantly different from the diffraction phenomenon in the exposure tool. This causes the image detected by the detector to differ from the image in the exposure tool, resulting in a problem of reduced inspection accuracy. Furthermore, if an abnormality occurs in any one of the multiple elements, the entire system must be replaced, which reduces reliability.

The present disclosure therefore aims to provide a highly reliable photomask inspection device that can detect photomasks with higher detection accuracy.

a first light source including a single first semiconductor light emitting element that emits first light having a first peak wavelength; a second light source including a single second semiconductor light emitting element that emits second light having a second peak wavelength different from the first peak wavelength; a mixer that mixes the first light from the first light source and the second light from the second light source; an illumination optical system that guides the light obtained by mixing the first light and the second light by the mixer to the photomask; a movement driver that changes the relative position of the photomask and the illumination optical system in an optical axis direction; and an objective lens that forms an image on which the light from the photomask is incident. The optical system includes an image sensor that receives the light incident through the imaging optical system and generates an image, and an arithmetic processing unit that inspects the photomask based on the image, wherein the arithmetic processing unit determines a focus evaluation value for each of the multiple imaged images generated by the image sensor while changing the distance between the photomask and the illumination optical system, calculates mask characteristics based on each of the multiple imaged images, and judges the quality of the photomask based on the mask characteristics that correspond to a range of a predetermined focus deviation amount in an exposure device in which the photomask is used from the relative position at which the focus evaluation value is maximum among the multiple imaged images.
The second aspect is a photomask inspection device according to the first aspect, which includes a light guide through which the light mixed with the first light and the second light by the mixing section enters at an incident end, travels inside, and exits at an exit end, and an illumination optical system that guides the light emitted from the light guide to the photomask.

In this photomask inspection device, the mixing unit mixes first and second light beams having different peak wavelengths. The mixed light then passes through the photomask and is incident on the light-receiving surface of the image sensor via the imaging optical system. Therefore, the captured image includes a projected image of the photomask formed by light beams having the first and second peak wavelengths. In other words, a single image capture can obtain a captured image including projected images formed by light beams having multiple peak wavelengths.

As a result, there is no need to combine multiple captured images obtained by sequentially irradiating light with different peak wavelengths, as in Patent Document 1. Therefore, unlike Patent Document 1, positional variations for each peak wavelength do not occur in principle in the captured images. Therefore, the calculation processing unit can inspect the photomask based on the captured images with higher inspection accuracy.

Furthermore, in this photomask inspection device, the first light source and the second light source each include a single semiconductor light-emitting element. Therefore, wavelength or phase variations in a diode array, as in Patent Document 2, do not occur in principle, and the first light source can emit a more coherent first light, and the second light source can emit a more coherent second light. This makes it possible to produce a diffraction phenomenon in the light passing through the photomask that is equivalent to the diffraction phenomenon in an exposure device. This allows the processing unit to inspect the photomask with higher inspection accuracy. Furthermore, the reliability of the first light source and the second light source is also high.

FIG. 1 is a perspective view schematically illustrating an example of the configuration of a photomask inspection apparatus. FIG. 1 is a diagram schematically illustrating an example of an optical configuration of a photomask inspection apparatus. FIG. 1 is a diagram schematically illustrating an example of the optical configuration of an exposure apparatus. 10 is a graph showing the transmittance and reflectance of a first dichroic mirror. 10 is a graph showing the transmittance and reflectance of a second dichroic mirror. FIG. 2 is a diagram schematically illustrating an example of a captured image. 10 is a flowchart showing an example of the operation of the photomask inspection apparatus. 10A and 10B are diagrams illustrating the luminance distribution of a projected image on a line of each captured image. 10 is a graph showing the relationship between the Z-axis position and the focus evaluation value, and the relationship between the Z-axis position and the width of the projected image. FIG. 10 is a diagram schematically illustrating an example of a captured image including a defect. 10 is a graph showing the relationship between the Z-axis position and the focus evaluation value, and the relationship between the Z-axis position and the defect brightness.

The following describes the embodiments in detail with reference to the drawings. Note that in the drawings, the dimensions and number of parts are exaggerated or simplified as necessary for ease of understanding. Parts with similar configurations and functions are given the same reference numerals, and duplicate explanations will be omitted below. The drawings also show XYZ Cartesian coordinates as appropriate to indicate the positional relationships of each component. For example, the Z axis is aligned vertically, while the X and Y axes are aligned horizontally. In the following description, one side of the Z axis direction will be referred to as the +Z side, and the other side as the -Z side. The same applies to the X and Y axes.

Furthermore, in the following description, similar components are illustrated with the same symbols, and their names and functions are also the same. Therefore, detailed descriptions of them may be omitted to avoid duplication.

Furthermore, even if ordinal numbers such as "first" or "second" are used in the following description, these terms are used for convenience to facilitate understanding of the contents of the embodiments, and are not intended to be limited to the ordering that may result from these ordinal numbers.

When expressions indicating relative or absolute positional relationships (e.g., "in one direction," "along one direction," "parallel," "orthogonal," "center," "concentric," "coaxial," etc.) are used, unless otherwise specified, these expressions not only express that exact positional relationship, but also express a state in which there is a relative displacement in terms of angle or distance within a range that provides tolerance or equivalent functionality. When expressions indicating an equal state (e.g., "identical," "equal," "homogeneous," etc.) are used, these expressions not only express a state in which there is strict quantitative equality, but also express a state in which there is a difference that provides tolerance or equivalent functionality, unless otherwise specified. When expressions indicating shape (e.g., "rectangular shape" or "cylindrical shape," etc.) are used, these expressions not only express the exact geometric shape, but also express shapes with irregularities, chamfers, etc. within a range that provides equivalent effects, unless otherwise specified. When expressions such as "comprise," "include," "have," "includes," "includes," or "have" are used to describe one component, these expressions are not exclusive expressions that exclude the presence of other components. When the expression "at least one of A, B, and C" is used, the expression includes A only, B only, C only, any two of A, B, and C, and all of A, B, and C.

<Photomask inspection equipment>
Fig. 1 is a perspective view that schematically shows an example of the configuration of a photomask inspection apparatus 1, and Fig. 2 is a diagram that schematically shows an example of the optical configuration of the photomask inspection apparatus 1. The photomask inspection apparatus 1 is an apparatus that inspects a photomask 80.

<Photomask>
First, the photomask 80 to be inspected will be described. The photomask 80 is a photomask used in the exposure apparatus 1000. FIG. 3 is a diagram schematically showing an example of the optical configuration of the exposure apparatus 1000. The exposure apparatus 1000 performs an exposure process on the substrate W using the photomask 80, thereby transferring the pattern of the photomask 80 onto the substrate W. The substrate W is, for example, a substrate for a flat panel display. It is also possible to use various substrates for the substrate W, such as a semiconductor substrate and a substrate for a solar cell.

The photomask 80 has a plate-like shape, e.g., a rectangular shape in a plan view. The length of one side of the photomask 80 is set to, for example, several meters. As a more specific example, the lengths of the sides of the photomask 80 are 1.8 m and 2.0 m, respectively, and the thickness of the photomask 80 is, for example, 2.1 mm. A light-shielding film (not shown) is formed in a predetermined pattern on one main surface of the photomask 80. That is, the photomask 80 has transmissive portions that transmit light and blocking portions that block light. Note that the photomask 80 may also be a phase-shift mask provided with a phase-shift film that transmits light with a lower transmittance than the transmissive portions.

The photomask 80 is held by a mask holder (not shown) in the exposure apparatus 1000. The mask holder holds the photomask 80 in a position where its thickness direction is aligned with the Z-axis direction. The mask holder supports, for example, only the peripheral edge of the photomask 80.

As shown in FIG. 3, the exposure apparatus 1000 includes an illumination unit 1100 and an imaging unit 1200. The illumination unit 1100 and the imaging unit 1200 are located on opposite sides of the photomask 80. In the example of FIG. 3, the illumination unit 1100 is located on the +Z side of the photomask 80, and the imaging unit 1200 is located on the -Z side of the photomask 80.

The illumination unit 1100 includes a light source 1110 and an illumination optical system 1120. The light source 1110 emits exposure light toward the illumination optical system 1120. The exposure light is, for example, ultraviolet light, and the light source 1110 is, for example, an ultraviolet irradiator such as a mercury lamp (e.g., a high-pressure mercury lamp). The light from the light source 1110 is irradiated onto the photomask 80 via the illumination optical system 1120.

In the example of Figure 3, the illumination optical system 1120 includes a condenser lens 1130, a field stop 1140, a condenser lens 1150, an aperture stop 1160, and a condenser lens 1170. The condenser lens 1130, the field stop 1140, the condenser lens 1150, the aperture stop 1160, and the condenser lens 1170 are arranged in this order, increasing in distance from the light source 1110 in the Z-axis direction. In other words, the condenser lens 1130 is arranged closest to the light source 1110. Light from the light source 1110 is condensed by the condenser lens 1130 and passes through the field stop 1140. The aperture diameter of the field stop 1140 is variable, allowing the illumination range to be adjusted. Light that passes through the field stop 1140 is condensed by the condenser lens 1150 and passes through the aperture stop 1160. Light that passes through the aperture stop 1160 is condensed onto the photomask 80 by the condenser lens 1170. The aperture diameter of the aperture stop 1160 is variable, allowing the numerical aperture of the illumination optical system 1120 to be adjusted.

In addition, an optical filter that transmits only light within a predetermined wavelength range from the light source 1110 may be provided in the illumination optical system 1120.

Exposure light from the illumination unit 1100 passes through the transparent portions of the photomask 80. The patterned light that passes through the transparent portions of the photomask 80 enters the imaging unit 1200 (imaging optical system). The imaging unit 1200 includes an objective lens 1210, an aperture stop 1220, and an imaging lens 1230. The objective lens 1210, aperture stop 1220, and imaging lens 1230 are arranged in this order with increasing distance from the photomask 80 in the Z-axis direction. The light that passes through the transparent portions of the photomask 80 is enlarged via the objective lens 1210 and the imaging lens 1230. The aperture diameter of the aperture stop 1220 is variable, allowing the numerical aperture of the imaging unit 1200 to be adjusted. When the substrate W is a substrate for a flat panel display, the numerical aperture of the imaging unit 1200 in the exposure apparatus 1000 is set small, for example, to approximately 0.1.

The substrate W is provided on the opposite side of the photomask 80 from the imaging unit 1200. In the example of Figure 3, the substrate W is provided on the -Z side of the imaging unit 1200. The substrate W is held in a horizontal position by a substrate holder (not shown). Here, a horizontal position means that the thickness direction of the substrate W is aligned with the Z-axis direction.

The patterned light from the imaging unit 1200 is irradiated onto the +Z side main surface of the substrate W. This exposes the resist formed on the main surface of the substrate W in a patterned manner. In other words, the pattern of the transparent portion of the photomask 80 is transferred onto the main surface of the substrate W.

If the photomask 80 is defective, the pattern transferred to the substrate W will deviate from the designed pattern, which is undesirable. Therefore, the photomask inspection device 1 inspects the photomask 80.

<Photomask inspection system overview>
2 , the photomask inspection apparatus 1 includes an image sensor (optical sensor) 25 that receives light transmitted through the photomask 80. As will be described in detail later, the photomask inspection apparatus 1 simulates, on the light-receiving surface (imaging surface) of the image sensor 25, the pattern of light that is irradiated onto the substrate W when the photomask 80 to be inspected is used in the exposure apparatus 1000. The image sensor 25 detects the light on the light-receiving surface, and the photomask inspection apparatus 1 inspects the photomask 80 based on the detection results. In other words, the photomask inspection apparatus 1 can simulate the inspection of the photomask 80 based on the light on the substrate W in the exposure apparatus 1000.

The photomask inspection system 1 comprises an illumination unit 10, a detection unit 20, a movement mechanism 40, a control unit 50, a lifting mechanism 60, and a holding unit 90. Below, we will first provide an overview of the configuration of the photomask inspection system 1, and then provide a detailed description of an example of each component.

The holding unit 90 is a member that holds the photomask 80. This holding unit 90 holds the photomask 80 so that the thickness direction of the photomask 80 is along the Z-axis direction. In the example shown in Figure 1, the holding unit 90 supports only the peripheral edge of the photomask 80. However, the holding unit 90 may also support the entire underside of the photomask 80 using a translucent member.

The illumination unit 10 and the detection unit 20 are located on opposite sides of the photomask 80 in the Z-axis direction. In the example of Figures 1 and 2, the illumination unit 10 is located on the -Z side of the photomask 80, and the detection unit 20 is located on the +Z side of the photomask 80.

The illumination unit 10 emits light L2 toward the photomask 80. In the example of FIG. 2, the illumination unit 10 includes a simulated light irradiation unit 11 and an illumination optical system 17. The simulated light irradiation unit 11 emits light L2 having a spectrum similar to the spectrum of the light used for exposure within the wavelength range of light emitted by the light source 1110 of the exposure apparatus 1000. In other words, the light L2 is light that simulates the light used to expose the substrate W in the exposure apparatus 1000. This light L2 is incident on the photomask 80 from the -Z side via the illumination optical system 17. In other words, the illumination optical system 17 guides the light L2 from the illumination unit 10 to the photomask 80. The patterned light L2 that passes through the transparent portions of the photomask 80 is incident on the detection unit 20.

The detection unit 20 includes an imaging optical system 21 and an image sensor 25. The imaging optical system 21 images light L2 transmitted through the photomask 80 onto the light-receiving surface of the image sensor 25. Sigma (= N1/N2), which is the ratio of the numerical aperture N1 of the illumination optical system 17 to the numerical aperture N2 of the imaging optical system 21, is set to the same value as Sigma (= N10/N20), which is the ratio of the numerical aperture N10 of the illumination optical system 1120 to the numerical aperture N20 of the imaging unit 1200 of the exposure apparatus 1000. As a result, the luminance distribution of light L2 incident on the light-receiving surface of the image sensor 25 mimics the luminance distribution of the patterned light irradiated onto the substrate W. In other words, the luminance distribution of light irradiated onto the substrate W when the photomask 80 is used in the exposure apparatus 1000 is reproduced in a pseudo manner on the light-receiving surface of the image sensor 25.

<Lighting Department>
As illustrated in FIG. 2, the simulated light irradiation unit 11 includes a plurality of light sources 12 and a mixer 14 .

The multiple light sources 12 emit light L1 having different peak wavelengths. In the example of Figure 2, three light sources 12a to 12c are provided as the multiple light sources 12. Each light source 12 includes a single semiconductor light-emitting element 121. The semiconductor light-emitting element 121 includes, for example, a light-emitting diode element or a laser element. Such a single semiconductor light-emitting element 121 includes, for example, a single semiconductor stack structure having a single p-type semiconductor layer and a single n-type semiconductor layer.

Light source 12a emits light L1a having a first peak wavelength. The first peak wavelength is, for example, 365 nm (the so-called i-line wavelength). Light source 12b emits light L1b having a second peak wavelength different from the first peak wavelength. The second peak wavelength is, for example, 405 nm (the so-called h-line wavelength). Light source 12c emits light L1c having a third peak wavelength different from both the first and second peak wavelengths. The third peak wavelength is, for example, 436 nm (the so-called g-line wavelength). The optical spectral distribution (spectrum) of light L1 emitted by each light source 12 may have only one peak wavelength. In other words, light L1 may be light of a single wavelength. Specifically, the spectrum of light emitted by each light source 12 may have a steep mountain-like shape with the intensity at the peak wavelength being the maximum intensity. If the semiconductor light-emitting element 121 is a light-emitting diode element, it emits light L1 having a spectrum with a somewhat broad peak; if the semiconductor light-emitting element 121 is a laser element, it emits light L1 having a spectrum with a relatively narrow peak.

Each semiconductor light-emitting element 121 is individually supplied with power from a power supply unit (not shown). Each semiconductor light-emitting element 121 emits light L1 at a brightness corresponding to the power from the power supply unit. The power supply unit is, for example, a current source and includes, for example, a switching power supply circuit. In this case, the power supply unit controls the brightness of light L1 by controlling the current supplied to the semiconductor light-emitting element 121. Since each power supply unit is individually controlled by the control unit 50, the brightness of light L1 from each light source 12 is individually controlled by the control unit 50.

Each light source 12 may be provided with a sensor (not shown) that measures the intensity (or illuminance, light amount) of light L1 from the semiconductor light-emitting element 121. The sensor outputs a signal indicating the detected value to the control unit 50. The control unit 50 controls the power supply unit based on the detected value, and controls the current supplied from the power supply unit to the semiconductor light-emitting element 121. This allows the control unit 50 to control the intensity of light L1 emitted from the light source 12 with greater precision.

Light L1a emitted from light source 12a enters mixing unit 14 through lens 13a, light L1b emitted from light source 12b enters mixing unit 14 through lens 13b, and light L1c emitted from light source 12c enters mixing unit 14 through lens 13c.

The mixer 14 mixes the multiple light beams L1 incident from the multiple light sources 12. Mixing light here means substantially aligning the optical paths of the multiple light beams L1. In the example of Figure 2, the mixer 14 includes a first dichroic mirror 141 and a second dichroic mirror 142.

Light L1a and light L1b are incident on the first dichroic mirror 141. In the example of Figure 2, light source 12a is located on the -X side of the first dichroic mirror 141 and emits light L1a toward the +X side along the X-axis. Light source 12b is located on the +Z side of the first dichroic mirror 141 and emits light L1b toward the -Z side along the Z-axis. The first dichroic mirror 141 has a plate-like shape and is positioned so that its thickness direction extends from the +X and +Z sides toward the -X and -Z sides. The first dichroic mirror 141 reflects light L1a and transmits light L1b in the same direction as the reflection of light L1a. Figure 4 is a graph showing the transmittance and reflectance of the first dichroic mirror 141. The first dichroic mirror 141 has low transmittance and high reflectance for light L1a of the first peak wavelength (365 nm in this case). On the other hand, the first dichroic mirror 141 has high transmittance and low reflectance for light L1b with the second peak wavelength (here, 405 nm) and light L1c with the third peak wavelength (here, 436 nm).

By this first dichroic mirror 141, light L1a is reflected toward the -Z side along the Z-axis direction, while light L1b passes through the first dichroic mirror 141 and travels toward the -Z side. This ideally causes the optical paths of light L1a and light L1b to coincide. In other words, light L1a and light L1b are mixed.

This mixed light and light L1c are incident on the second dichroic mirror 142. In the example of Figure 2, the second dichroic mirror 142 is located on the -Z side of the first dichroic mirror 141. The light source 12c is located on the -X side of the second dichroic mirror 142 and emits light L1c to the +X side along the X-axis direction. The second dichroic mirror 142 has a plate-like shape and is positioned such that its thickness direction runs from the +X and +Z sides toward the -X and -Z sides. The second dichroic mirror 142 reflects the mixed light and transmits light L1c in the same direction as the reflection of the mixed light. Figure 5 is a graph showing the transmittance and reflectance of the second dichroic mirror 142. The second dichroic mirror 142 has low transmittance and high reflectance for light L1a and light L1b. On the other hand, the second dichroic mirror 142 has high transmittance and low reflectance for light L1c.

By this second dichroic mirror 142, the mixed light containing light L1a and light L1b is reflected toward the +X side along the X-axis direction, while light L1c passes through the second dichroic mirror 142 and travels toward the +X side. As a result, the optical paths of light L1a, light L1b, and light L1c ideally coincide. In other words, light L1a, light L1b, and light L1c are mixed. The mixed light containing light L1a, light L1b, and light L1c corresponds to light L2.

In the example of Figure 2, light L2 is focused by lens 15 onto the input end of light guide 16. Light L2 travels inside light guide 16 and exits from the output end of light guide 16. Light guide 16 may be, for example, a liquid light guide including a tube filled with a light-transmitting liquid, or a fiber light guide in which multiple optical fibers are bundled together. In the example of Figure 2, the output end of light guide 16 corresponds to the output end of pseudo-light irradiation unit 11.

The multiple light sources 12 are controlled so that the spectrum of light L2 emitted by this simulated light irradiation unit 11 resembles the spectrum of the light used for exposure in the exposure apparatus 1000. As a specific example, a case will be described in which the light emitted by the illumination unit 1100 in the exposure apparatus 1000 includes i-rays, h-rays, and g-rays. In this case, light sources 12a, 12b, and 12c are controlled so that the intensity peaks of the i-rays, h-rays, and g-rays contained in the light match the intensity peaks of light L1a, light L1b, and light L1c, respectively. This allows the simulated light irradiation unit 11 to emit light L2 that mimics the light in the exposure apparatus 1000.

Note that target values for the peak intensity of light L1 from each light source 12 may be set in advance and recorded in a non-transitory storage unit (e.g., a memory or a hard disk) (not shown). Alternatively, a user may input the peak intensity of each light L1 using an input device (e.g., a keyboard or a mouse) (not shown). The control unit 50 controls the light source 12 based on the detection value of a sensor included in the light source 12 and the target value. This allows the spectrum of light L2, which includes light L1a, light L1b, and light L1c , to approach the spectrum of light in the exposure apparatus 1000 with greater accuracy.

In the example of Figure 2, the output end of the light guide 16 is located on the -Z side of the illumination optical system 17, and emits light L2 toward the illumination optical system 17. In the example of Figure 2, the illumination optical system 17 includes a condenser lens 171, a field stop 172, a condenser lens 173, an aperture stop 174, and a condenser lens 175. The condenser lens 171, field stop 172, condenser lens 173, aperture stop 174, and condenser lens 175 are arranged in this order in the Z-axis direction, increasing in distance from the output end of the light guide 16.

Light L2 from the output end of the light guide 16 is collected by a collecting lens 171 and passes through a field stop 172. The aperture diameter of the field stop 172 is variable by an aperture mechanism, allowing the illumination range to be adjusted. Light L2 that passes through the field stop 172 is collected by a collecting lens 173 and passes through an aperture stop 174. Light L2 that passes through the aperture stop 174 is collected onto the photomask 80 by a condenser lens 175. The aperture diameter of the aperture stop 174 is variable by an aperture mechanism, allowing the numerical aperture N1 of the illumination optical system 17 to be adjusted. The aperture mechanisms of the field stop 172 and aperture stop 174 may be controlled by the control unit 50.

<Detection unit>
The patterned light L2 transmitted through the transmitting portion of the photomask 80 is incident on the detecting unit 20. The detecting unit 20 includes an imaging optical system 21 and an image sensor 25.

The imaging optical system 21 forms an image of the patterned light L2 that has passed through the transparent portions of the photomask 80 on the light-receiving surface of the image sensor 25. In the example of FIG. 2, the imaging optical system 21 includes an objective lens 22, an aperture stop 23, and an imaging lens 24. The objective lens 22, aperture stop 23, and imaging lens 24 are arranged in this order, increasing in distance from the photomask 80 in the Z-axis direction. The numerical aperture of the objective lens 22 is equal to or greater than the numerical aperture of the objective lens 1210 in the exposure apparatus 1000. The light that has passed through the photomask 80 is enlarged via the objective lens 22 and the imaging lens 24. The aperture diameter of the aperture stop 23 is variable via an aperture mechanism, allowing the numerical aperture of the imaging optical system 21 to be adjusted. The aperture mechanism of the aperture stop 23 may be controlled by the control unit 50.

Light L2 transmitted through the imaging lens 24 is incident on the light receiving surface of the image sensor 25. The image sensor 25 is, for example, a CCD (Charge Coupled Device) image sensor or a CMOS (Complementary Metal Oxide Semiconductor) image sensor. The image sensor 25 generates a captured image IM1 based on the light incident on its light receiving surface and outputs the captured image IM1 to the control unit 50. The control unit 50 inspects the photomask 80 based on the captured image IM1.

<Movement mechanism>
The moving mechanism 40 moves the holding unit 90 within the XY plane. As a result, the photomask 80 held by the holding unit 90 also moves within the XY plane. The moving mechanism 40 has, for example, a ball screw mechanism and is controlled by the control unit 50. As the photomask 80 moves within the XY plane, the illumination unit 10 and the detection unit 20 can be scanned relative to the photomask 80. As a result, the illumination unit 10 irradiates each measurement area on the photomask 80 with light L2, and the image sensor 25 can generate an image IM1 of each measurement area on the photomask 80. Therefore, the control unit 50 can inspect multiple measurement areas on the photomask 80. Note that the moving mechanism 40 only needs to have the function and structure to move the photomask 80 relative to the illumination unit 10 and the detection unit 20. For example, the illumination unit 10 and the detection unit 20 may be moved integrally.

<Lifting mechanism>
The lifting mechanism 60 raises and lowers the holding unit 90 in the Z-axis direction. As a result, the photomask 80 held by the holding unit 90 also raises and lowers. The lifting mechanism 60 has, for example, a ball screw mechanism, and is controlled by the control unit 50. By raising and lowering the photomask 80 with the lifting mechanism 60, the photomask 80 can be moved to the focal point of the objective lens 22. Note that the lifting mechanism 60 only needs to have the function and structure to raise and lower the photomask 80 relative to the detection unit 20, and may raise and lower the detection unit 20, for example.

<Control unit>
The control unit 50 can control the entire photomask inspection apparatus 1. For example, the control unit 50 controls the illumination unit 10, the moving mechanism 40, and the lifting mechanism 60 as described above. The control unit 50 also functions as an arithmetic processing unit that determines the mask characteristics of the photomask 80 based on the captured image IM1 generated by the image sensor 25. The mask characteristics include, for example, the line width of the pattern formed on the photomask 80, the spacing between patterns, or various defects. How to determine the mask characteristics will be described in detail later.

The control unit 50 is an electronic circuit device and may have, for example, an arithmetic processing unit and a memory unit. The arithmetic processing unit may be, for example, a central processing unit (CPU). The memory unit may have a non-transitory memory unit (for example, a read-only memory (ROM) or a hard disk) and a temporary memory unit (for example, a random access memory (RAM)). The non-transitory memory unit may store, for example, a program that defines the processing to be performed by the control unit 50. The processing unit executes this program, allowing the control unit 50 to perform the processing defined in the program. Of course, some or all of the processing performed by the control unit 50 may be performed by hardware.

<Mask characteristic calculation method>
Next, an example of a method for calculating the mask characteristics of the photomask 80 based on the captured image IM1 captured by the image sensor 25 will be described.

Figure 6 is a diagram showing an example of a captured image IM1. The captured image IM1 contains a pattern of light L2. The brightness distribution of this light L2 corresponds to the projected image of the transparent portion of the photomask 80. In the example of Figure 6, the projected image contains a vertical portion extending vertically and a horizontal portion extending to the right from midway along the vertical portion.

The example in Figure 6 also shows a schematic example of the luminance distribution along line A, which extends in the width direction of the vertical portion. The control unit 50 determines the position of the rising edge, where the luminance value rises from zero, and the position of the falling edge, where the luminance value falls to zero, in the luminance distribution, and calculates the width of the vertical portion (projected image width) based on these positions.

However, if the position of the objective lens 22 relative to the photomask 80 shifts in the Z-axis direction from the focal position, the projected image in the captured image IM1 may vary. In other words, if the lifting mechanism 60 stops the holder 90 at a position shifted from the focal position, the projected image may differ from the image projected at the focal position.

When the substrate W is a substrate for a flat panel display, the numerical aperture of the imaging unit 1200 in the exposure apparatus 1000 is small, at approximately 0.1, and therefore the depth of focus of the imaging unit 1200 is large. However, as will be described in detail later, even a slight deviation of approximately 10 μm from the focal position will cause the width of the projected image in the captured image IM1 to fluctuate.

Here, the Z-axis position of the holder 90 relative to the objective lens 22 (hereinafter referred to as the Z-axis position) is sequentially changed, and each time, the image sensor 25 generates a captured image IM1, and the control unit 50 performs inspection based on these multiple captured images IM1. An example of an inspection method will be described in detail below.

Figure 7 is a flowchart showing an example of the operation of the photomask inspection apparatus 1. Here, the moving mechanism 40 moves the holder 90 to a predetermined measurement position. The control unit 50 causes the multiple light sources 12 to emit light L1 (step S1). As a result, light L2 simulating the light in the exposure apparatus 1000 passes through the transparent portions in the predetermined measurement area of the photomask 80, and the pattern-shaped light L2 passes through the imaging optical system 21 and is incident on the light-receiving surface of the image sensor 25.

Next, the control unit 50 controls the lifting mechanism 60 and the image sensor 25 to generate captured images IM1 at each Z-axis position (step S2). For example, the lifting mechanism 60 moves the holding unit 90 within a predetermined lifting range (focus range) that includes the focal position, while the image sensor 25 sequentially generates captured images IM1. Note that the lifting mechanism 60 may also stop the holding unit 90 at the image capture timing. This makes it possible to obtain multiple captured images IM1 corresponding to multiple Z-axis positions.

Next, the control unit 50 calculates mask characteristics based on the multiple captured images IM1 (step S3). Here, the projected image width is calculated as the mask characteristic. Figure 8 shows the luminance distribution of the projected image on line A of each captured image IM1. More specifically, Figure 8 shows the luminance distribution LD1 at the focal position and the luminance distribution at Z-axis positions 10 μm away from the focal position on the +Z and -Z sides. As can be seen from Figure 8, when defocus occurs, the luminance distribution fluctuates depending on the amount of defocus. As a result, the projected image widths calculated based on each luminance distribution differ from each other.

The control unit 50 then calculates a focus evaluation value for each of the multiple captured images IM1. The focus evaluation value is an index that indicates the degree of focus, with the higher the focus evaluation value, the better the focus. The focus evaluation value may be, for example, an index indicated by the high-frequency components among the frequency components obtained by Fourier transforming the luminance distribution of the captured image IM1. The higher the high-frequency components, the more abruptly the luminance distribution changes, and therefore the clearer the contours of the projected image and the better the focus. Alternatively, various indices such as the contrast and sharpness of the captured image IM1 may be used as the focus evaluation value.

Since multiple captured images IM1 correspond to multiple Z-axis positions, the focus evaluation value calculated based on the luminance distribution of each captured image IM1 also corresponds to the Z-axis position. Similarly, the projected image width calculated based on the luminance distribution on line A of each captured image IM1 also corresponds to the Z-axis position. Figure 9 is a graph showing the relationship between the Z-axis position and the focus evaluation value, and the relationship between the Z-axis position and the projected image width. The example of Figure 9 also shows approximate curves G1 and G2. Approximate curve G1 is calculated based on plot points showing the relationship between the Z-axis position and the focus evaluation value, and approximate curve G2 is calculated based on plot points showing the relationship between the Z-axis position and the projected image width.

The focus evaluation value reaches its maximum value when the Z-axis position is the focus position P0. Therefore, the control unit 50 first determines the Z-axis position (i.e., the focus position P0) at which the focus evaluation value reaches its maximum value. For example, the control unit 50 calculates an approximate curve G1 using an approximation method such as the least squares method based on multiple focus evaluation values corresponding to multiple Z-axis positions. The control unit 50 then calculates the Z-axis position on the approximate curve G1 at which the focus evaluation value reaches its maximum value Fmax as the focus position P0.

Next, the control unit 50 calculates the projected image width LW at the focal position P0. First, the control unit 50 calculates the projected image width for each captured image IM1. This makes it possible to obtain multiple projected image widths corresponding to multiple Z-axis positions. Next, the control unit 50 calculates an approximation curve G2 based on the multiple Z-axis positions and multiple projected image widths using an approximation method such as the least squares method. Then, the control unit 50 calculates the projected image width LW at the focal position P0 based on the approximation curve G2 and the focal position P0.

Next, the control unit 50 determines whether the photomask 80 is good or bad based on the mask characteristics (here, the projected image width LW) (step S4). For example, the control unit 50 determines whether the difference between the projected image width LW and the target width is equal to or less than a predetermined width tolerance, and if the difference is greater than the width tolerance, determines that the photomask 80 is defective. The width tolerance is, for example, set in advance and stored in the memory unit.

<Effects of the embodiment>
As described above, according to the photomask inspection apparatus 1, the illumination unit 10 emits light L2 that simulates the light of the exposure apparatus 1000, and the sigma of the photomask inspection apparatus 1 is set to be the same as the sigma of the exposure apparatus 1000. This allows the light (projected image) on the substrate W in the exposure apparatus 1000 to be reproduced in a pseudo manner on the light receiving surface of the image sensor 25. Therefore, the photomask inspection apparatus 1 can determine the quality of the photomask 80 based on the projected image on the substrate W.

Furthermore, according to the photomask inspection device 1, the mixing unit 14 mixes multiple light beams L1 having different peak wavelengths. The illumination unit 10 then irradiates the photomask 80 with light beam L2, which is the mixed light, and the light beam L2 that passes through the photomask 80 is incident on the light receiving surface of the image sensor 25. Therefore, the captured image IM1 includes a projected image of the photomask 80 formed by the light beams L2 having multiple peak wavelengths. In other words, a single image capture can obtain the captured image IM1 that includes a projected image formed by the light beams L2 having multiple peak wavelengths.

As a result, there is no need to combine multiple captured images obtained by sequentially irradiating light with different peak wavelengths, as in Patent Document 1. Therefore, unlike Patent Document 1, positional variation for each peak wavelength does not occur in principle in the captured image IM1. Therefore, the control unit 50 can calculate the projected image width with high accuracy, and the photomask 80 can be inspected with even higher inspection accuracy.

Furthermore, according to the photomask inspection apparatus 1, each light source 12 includes a single semiconductor light-emitting element 121. Therefore, wavelength or phase variations, such as those in a diode array as in Patent Document 2, do not occur in principle, and each light source 12 can emit more coherent light L1. As a result, a diffraction phenomenon equivalent to that in the exposure apparatus 1000 can be caused in the light L2 passing through the photomask 80. This allows the control unit 50 to inspect the photomask 80 with higher inspection accuracy.

Furthermore, since the light source 12, which includes a single semiconductor light-emitting element 121, does not, in principle, have any variation in the emission direction, it is possible to suppress variation in the light incident on the light guide 16. This also allows the control unit 50 to inspect the photomask 80 with higher inspection accuracy. Furthermore, the light source 12 is highly reliable.

<Mask characteristics: defect brightness>
In the above example, the inspection method was described using the projected image width as the mask characteristic. However, the defect brightness may also be used as the mask characteristic. FIG. 10 is a diagram schematically illustrating an example of a captured image IM1 including a defect D1. In FIG. 10, the defect D1 is a defect in the light-shielding film formed on the photomask 80. Specifically, the defect D1 is a defect in which the light-shielding film is not properly formed and light L2 passes through, and is also called a white defect.

Here, the brightness value of defect D1 is calculated. Hereinafter, the brightness value of defect D1 will also be referred to as defect brightness. This defect brightness also varies depending on the position of the holder 90 in the Z-axis direction, so the control unit 50 may calculate the defect brightness based on multiple captured images IM1, similar to the projected image width.

Specifically, in step S3, the control unit 50 first calculates the defect brightness and focus evaluation value for each captured image IM1. This makes it possible to obtain multiple defect brightness values corresponding to multiple Z-axis positions, and multiple focus evaluation values corresponding to multiple Z-axis positions. Figure 11 is a graph showing the relationship between the Z-axis position and the focus evaluation value, and the relationship between the Z-axis position and the defect brightness. The example in Figure 11 also shows approximate curves G1 and G3. Approximate curve G3 is calculated based on the plot points of the Z-axis position and the defect brightness.

As described above, the control unit 50 determines the focal position P0 where the focus evaluation value reaches the maximum value Fmax based on the approximate curve G1. Next, the control unit 50 calculates the approximate curve G3 based on multiple Z-axis positions and multiple defect intensities using an approximation method such as the least squares method. The control unit 50 then calculates the defect intensity DL at the focal position P0 based on the approximate curve G3 and the focal position P0.

Next, in step S4, the control unit 50 determines whether the photomask 80 is good or bad based on the mask characteristics (here, the defect brightness DL). For example, the control unit 50 determines whether the defect brightness DL is equal to or less than a predetermined defect tolerance, and determines that the photomask 80 is defective if the defect brightness DL is greater than the defect tolerance.

As described above, the photomask inspection device 1 can determine the quality of the photomask 80 based on the brightness value of the defect D1. Note that in the above example, a white defect was used as the defect D1, but various other defects, such as a black defect, can also be used as the defect D1.

<Variations in Projected Images in Exposure Apparatus 1000>
The photomask inspection apparatus 1 artificially reproduces the luminance distribution (projected image) of the patterned light on the substrate W in the exposure apparatus 1000 on the light receiving surface of the image sensor 25. In other words, the captured image IM1 when defocus occurs in the photomask inspection apparatus 1 corresponds to the projected image on the substrate W when a similar defocus occurs in the exposure apparatus 1000. Therefore, the photomask inspection apparatus 1 can also evaluate the variation in the projected image on the substrate W caused by defocus.

For example, the relationship between the Z-axis position and the projected image width in Figure 9 can be considered to be the relationship between the Z-axis position and the projected image width in the exposure apparatus 1000. In other words, in the exposure apparatus 1000, the relationship between the position (Z-axis position) of the photomask 80 relative to the objective lens 1210 and the projected image width on the substrate W corresponds to the relationship in Figure 9.

The defocus amount ΔZ (maximum value) that can occur due to mechanical errors of the exposure apparatus 1000 can be determined in advance, so the variation in the projected image width that occurs in the exposure apparatus 1000 can be determined based on the relationship in Figure 9. The defocus amount ΔZ may be set in advance and stored in a memory unit, for example, or the user may input the defocus amount ΔZ using an input device (e.g., a keyboard or mouse) not shown.

The control unit 50 calculates the maximum value LWmax and minimum value LWmin of the projected image width in the range between a position shifted by the defocus amount ΔZ from the focal position P0 to the -Z side and a position shifted by the defocus amount ΔZ from the focal position P0 to the +Z side, based on the approximation curve G2. The range from the minimum value LWmin to the maximum value LWmax corresponds to the variation in the projected image width that occurs in the exposure apparatus 1000.

The control unit 50 may determine whether the photomask 80 is defective based on the minimum value LWmin and maximum value LWmax of the projected image width. Specifically, the control unit 50 determines whether the minimum value LWmin is equal to or greater than the allowable lower limit of the variation in the projected image width on the substrate W, and whether the maximum value LWmax is equal to or less than the allowable upper limit. The control unit 50 determines that the photomask 80 is defective when the minimum value LWmin is less than the allowable lower limit and/or when the maximum value LWmax is greater than the allowable upper limit.

Similarly, the control unit 50 may calculate the variation in defect brightness caused by the defocus amount ΔZ. Specifically, the control unit 50 calculates the maximum value DLmax and minimum value DLmin of the defect brightness in the range between a position shifted by the defocus amount ΔZ from the focal position P0 toward the -Z side and a position shifted by the defocus amount ΔZ from the focal position P0 toward the +Z side, based on the approximation curve G3 (see also Figure 11). The range from the minimum value DLmin to the maximum value DLmax corresponds to the variation in defect brightness that occurs in the exposure apparatus 1000.

The control unit 50 may determine whether the photomask 80 is good or bad based on the maximum defect brightness value DLmax. Specifically, the control unit 50 determines whether the maximum value DLmax is equal to or less than the defect tolerance. The control unit 50 may determine that the photomask 80 is defective when the maximum value DLmax is greater than the defect tolerance.

Here, the terms in the "Means for Solving the Problem" section correspond to the terms in the "Mode for Carrying Out the Invention" section. The first light source corresponds to any one of light sources 12a to 12c, and the second light source corresponds to one of light sources 12a to 12c that is different from the first light source. The first semiconductor light-emitting element corresponds to semiconductor light-emitting element 121 belonging to the first light source, and the second semiconductor light-emitting element corresponds to semiconductor light-emitting element 121 belonging to the second light source. The first light corresponds to light L1 emitted by the first semiconductor light-emitting element, and the second light corresponds to light L1 emitted by the second semiconductor light-emitting element.

As mentioned above, the photomask inspection apparatus 1 has been described in detail, but the above description is merely an example in all respects, and this disclosure is not limited thereto. Furthermore, the various modifications described above can be applied in combination as long as they are not mutually inconsistent. It is understood that numerous modifications not illustrated can be envisioned without departing from the scope of this disclosure.

1 Photomask inspection device 10 Illumination unit 12, 12a to 12c First light source, second light source (light source)
121 First semiconductor light emitting element, second semiconductor light emitting element 14 Mixing unit 17 Illumination optical system 21 Imaging optical system 22 Objective lens 25 Image sensor 50 Arithmetic processing unit (control unit)
80 Photomask 90 Holding part

Claims (2)

a holder for holding a photomask;
a first light source including a single first semiconductor light emitting element that emits first light having a first peak wavelength;
a second light source including a single second semiconductor light emitting element that emits second light having a second peak wavelength different from the first peak wavelength;
a mixer that mixes the first light from the first light source and the second light from the second light source;
an illumination optical system that guides the light obtained by mixing the first light and the second light by the mixing unit to the photomask;
a movement drive unit that changes the relative position between the photomask and the illumination optical system in the optical axis direction;
an imaging optical system including an objective lens onto which the light from the photomask is incident;
an image sensor that receives the light incident through the imaging optical system and generates a captured image;
a processing unit that inspects the photomask based on the captured image;
Equipped with
the arithmetic processing unit determines a focus evaluation value for each of the plurality of captured images generated by the image sensor while changing the distance between the photomask and the illumination optical system, calculates mask characteristics based on each of the plurality of captured images, and determines the quality of the photomask based on the mask characteristics corresponding to a range shifted by a predetermined focus shift amount in an exposure apparatus in which the photomask is used from the relative position at which the focus evaluation value is maximum among the plurality of captured images. 2. The photomask inspection apparatus according to claim 1 ,
a light guide into which the light obtained by mixing the first light and the second light by the mixing section is incident from an incident end, travels inside, and is emitted from an exit end;
an illumination optical system that guides the light emitted from the light guide to the photomask.