KR20200141510A - Photomask inspection device and photomask inspection method - Google Patents

Abstract

It provides a photomask inspection apparatus capable of detecting a diffraction pattern more suitable for measurement. The photomask inspection apparatus measures the pattern characteristics of a phase shift part of a phase shift mask. The photomask inspection apparatus includes a holding unit, an irradiation unit, a slit mask, a Fourier transform lens, and a first optical sensor. The holding unit holds the phase shift mask. The irradiation unit irradiates light to a region including the light transmitting unit and the phase shift unit. The slit mask has a slit, and is disposed at a position where light that has passed through the slit partly in the width direction of the light-transmitting portion and the entire phase shifting portion in the width direction passes through the slit. The light passing through the slit is incident on the Fourier transform lens. The first optical sensor detects a diffraction pattern of light from the Fourier transform lens at a plurality of timings.

Description

Photomask inspection device and photomask inspection method

The present invention relates to a photomask inspection apparatus and a photomask inspection method.

In recent years, in order to transfer a pattern with high resolution to a substrate such as a semiconductor substrate or a substrate for a display display, a phase shift mask has been used. In this phase shift mask, a phase shift film is formed to delay the phase of light by half a wavelength.

In Patent Document 1, a photomask inspection apparatus for measuring a phase delay (phase difference) due to a phase shift film is described. In this photomask inspection apparatus, light is irradiated to a photomask through a variable aperture stop, and light that has passed through the photomask is imaged on a photoelectric converter (sensor) through a Fourier transform lens. Accordingly, the photoelectric converter detects a Fourier transform image (diffraction pattern).

The photomask inspection apparatus first irradiates light to a region in which a phase difference does not occur in the photomask (a region of only a transparent portion or a region of only a phase member (phase shift film)), and a Fourier transform in the case where a phase difference does not occur. Remember the image as a reference image. Then, the photomask inspection apparatus irradiates light to both the transparent portion of the photomask and the phase member, and calculates the phase difference due to the phase member based on the comparison of the Fourier transform image obtained by the irradiation and the reference image. .

Japanese Laid-Open Patent Publication No. Hei 4-229863

However, in the technique of Patent Document 1, it cannot be said that a Fourier transform image (diffraction pattern) suitable for calculating a phase difference can be detected. This is because it cannot be said that it is possible to position them so that the relative positions of the variable aperture stop and the photomask are optimal. Since the required precision of this positioning increases as the width of the pattern becomes narrower, it is difficult to obtain an optimum Fourier transform image (diffraction pattern), especially for a photomask having a fine pattern.

Therefore, an object of the present invention is to provide a photomask inspection apparatus and a photomask inspection method capable of detecting a diffraction pattern more suitable for measurement.

A first aspect of the photomask inspection apparatus is a light-transmitting portion that transmits light, a light-shielding portion that blocks light, and is formed between the light-transmitting portion and the light-shielding portion, and transmits the light while transmitting light. A photomask inspection apparatus for measuring pattern characteristics of the phase shift portion of a phase shift mask having a phase shift portion that shifts a phase with respect to light, comprising: a holding portion holding the phase shift mask; the light transmitting portion; An irradiation unit that irradiates light to a region including the phase shift unit, and a slit, and a position in which the light transmitted through a part in the width direction of the light transmitting unit and the entire phase shift unit in the width direction passes through the slit A slit mask disposed in the slit, a Fourier transform lens into which light passing through the slit is incident, and a first optical sensor for detecting a diffraction pattern of light from the Fourier transform lens at a plurality of timings.

A second aspect of the photomask inspection apparatus is the photomask inspection apparatus according to the first aspect, further comprising a movement mechanism for relatively moving the slit mask and the phase shift mask when viewed from a plan view, and the first The optical sensor detects the diffraction pattern at a plurality of timings in the midst of the movement mechanism relatively moving the slit mask and the phase shift mask.

A third aspect of the photomask inspection apparatus is a photomask inspection apparatus according to the second aspect, wherein the movement mechanism moves relative to the slit mask and the phase shift mask along a direction inclined with respect to the width direction. .

A fourth aspect of the photomask inspection apparatus is a photomask inspection apparatus according to any one of the first to third aspects, wherein the intensity of light at a central position among a plurality of diffraction patterns detected by the first optical sensor is The smallest diffraction pattern is selected as the selected diffraction pattern, and based on the selected diffraction pattern, an operation processing unit for calculating at least one of the width of the phase shift unit and the phase difference due to the phase shift unit is further provided as the pattern characteristic do.

A fifth aspect of the photomask inspection apparatus is a photomask inspection apparatus according to the fourth aspect, wherein the calculation processing unit calculates a width of the phase shift unit based on a pitch of intensity of light intensity in the selected diffraction pattern. do.

A sixth aspect of the photomask inspection apparatus is a photomask inspection apparatus according to the fourth or fifth aspect, wherein the operation processing unit includes both of a plurality of peak values or a plurality of bottom values of the intensity of light in the selected diffraction pattern. Based on the difference of, the phase difference due to the phase shift unit is calculated.

A seventh aspect of the photomask inspection apparatus is a photomask inspection apparatus according to the fourth aspect, wherein the operation processing unit includes an intensity distribution of light transmitted through the light transmitting unit and the phase shift unit, a width of the phase shift unit, and the phase A first step of setting a phase difference by a shift unit, a second step of calculating a calculation diffraction pattern using a fast Fourier transform based on the intensity distribution, the width, and the phase difference, and the calculation diffraction pattern selected When it is determined that the calculation diffraction pattern is not similar to the selected diffraction pattern in the third step of determining whether it is similar to the diffraction pattern, and in the third step, the second step by changing the width and the phase difference And a fourth step of executing the third step.

An eighth aspect of the photomask inspection apparatus is a photomask inspection apparatus according to the seventh aspect, wherein in the first step, the operation processing unit has a constant intensity of light passing through each of the phase shift unit and the light transmitting unit. Set the intensity distribution so that

A ninth aspect of the photomask inspection apparatus is the photomask inspection apparatus according to the seventh aspect, wherein the operation processing unit comprises, in the first process, a light intensity at a boundary between the phase shift unit and the light transmitting unit, wherein the phase The intensity distribution is set so as to gradually increase from the shift portion toward the light transmitting portion.

A tenth aspect of the photomask inspection apparatus is a photomask inspection apparatus according to the seventh aspect, which is formed between a second optical sensor and the slit mask and the phase shift mask, and partially receives light from the phase shift mask. An optical element for guiding to the second optical sensor is further provided, and the calculation processing unit sets the intensity distribution based on an image captured by the second optical sensor in the first step.

An eleventh aspect of the photomask inspection method is a light-transmitting portion that transmits light, a light-shielding portion that blocks light, and is formed between the light-transmitting portion and the light-shielding portion, and transmits the light while transmitting the light-transmitting portion. A photomask inspection method for measuring pattern characteristics of the phase shift unit of a phase shift mask having a phase shift unit that shifts a phase with respect to light, wherein the irradiation unit includes light in a region including the light transmitting unit and the phase shift unit. A step of irradiating, and the first optical sensor, through a slit formed in a slit mask, and a Fourier transform lens, of the light transmitted through a part in the width direction of the light transmitting part and the whole in the width direction of the phase shift part. And a step of detecting a diffraction pattern at a plurality of timings.

According to the first aspect of the photomask inspection apparatus and the eleventh aspect of the photomask inspection method, since the relative position between the slit mask and the phase shift mask actually fluctuates slightly, the first optical sensor detects the diffraction pattern at a plurality of timings. By doing so, it is possible to detect a plurality of diffraction patterns corresponding to a plurality of relative positions. Therefore, compared to the case where the diffraction pattern is detected only once, it is easier to detect a diffraction pattern suitable for calculating the pattern characteristics of the phase shift unit.

According to the second aspect of the photomask inspection apparatus, since the relative position of the slit mask and the phase shift mask can be controlled by the movement mechanism, the relative position optimal for the movement range can be included. Therefore, the first optical sensor is easy to detect a diffraction pattern more suitable for specific calculation of the phase shift unit.

According to the third aspect of the photomask inspection apparatus, the relative velocity component in the width direction can be set low. Therefore, the first optical sensor is easy to detect the diffraction pattern at a relative position close to the optimum relative position.

According to the fourth aspect of the photomask inspection apparatus, at least one of the width of the phase shift unit and the phase difference due to the phase shift unit can be calculated with high accuracy.

According to the fifth aspect of the photomask inspection apparatus, it is possible to calculate the width of the phase shift portion by simple calculation.

According to the sixth aspect of the photomask inspection apparatus, the phase difference due to the phase shift unit can be calculated by simple calculation.

According to the seventh aspect of the photomask inspection apparatus, the width of the phase shift unit and the phase difference due to the phase shift unit can be calculated with higher precision.

According to the eighth aspect of the photomask inspection apparatus, the intensity distribution can be easily set.

According to the ninth aspect of the photomask inspection apparatus, the width of the phase shift unit and the phase difference due to the phase shift unit can be calculated with higher precision.

According to the tenth aspect of the photomask inspection apparatus, the width of the phase shift unit and the phase difference due to the phase shift unit can be calculated with higher precision.

1 is a perspective view schematically showing an example of a configuration of a photomask inspection apparatus.
2 is a diagram schematically showing an example of a configuration of a photomask inspection apparatus.
3 is a plan view schematically showing an example of a configuration of a slit mask.
4 is a graph schematically showing an example of a plurality of diffraction patterns.
5 is a graph schematically showing an example of a plurality of diffraction patterns.
6 is a flowchart showing an example of the operation of the photomask inspection device.
7 is a flowchart showing an example of a method of calculating pattern characteristics.
8 is a diagram for explaining an example of a relative movement direction of a slit mask and a phase shift mask.
9 is a diagram schematically showing an example of a simulation model.
10 is a graph schematically showing an example of a computational diffraction pattern.
11 is a flowchart showing an example of a method of calculating pattern characteristics.
12 is a diagram schematically showing an example of a simulation model.
13 is a flowchart showing an example of the operation of the photomask inspection device.
14 is a flowchart showing an example of a method of calculating pattern characteristics.

Hereinafter, embodiments will be described in detail with reference to the drawings. In addition, in the drawings, for the purpose of easy understanding, the dimensions and numbers of each part are exaggerated or simplified as necessary. In addition, the same reference numerals are assigned to portions having the same configuration and function, and redundant descriptions are omitted in the following description. In addition, in the figure, in order to show the positional relationship of each structure, XYZ rectangular coordinates are shown suitably. For example, the Z axis is arranged along the vertical direction, and the X axis and Y axis are arranged along the horizontal direction. In the following description, one side in the Z-axis direction is also referred to as a +Z side, and the other side is also referred to as a -Z side. The same is true for the X and Y axes.

1 is a perspective view schematically showing an example of a configuration of a photomask inspection device 1, and FIG. 2 is a diagram schematically showing an example of a configuration of a photomask inspection device 1. This photomask inspection device 1 is a device that inspects the phase shift mask 80. Here, first, an example of the phase shift mask 80 used as an inspection object will be described.

<Phase Shift Mask>

The phase shift mask 80 is a photomask used in an exposure apparatus not shown. The exposure apparatus can transfer the pattern to the predetermined substrate by performing exposure treatment on the predetermined substrate using the phase shift mask 80. The predetermined substrate is, for example, a semiconductor substrate or a substrate for a flat panel display.

As illustrated in FIG. 2, the phase shift mask 80 includes a base material 81, a phase shift film 82, and a light shielding film 83. The substrate 81 has translucency to exposure light (for example, ultraviolet rays such as i-rays), and is formed of, for example, quartz glass or the like. The base material 81 has a plate-like shape, and has a rectangular shape, for example, when viewed in a plan view (that is, viewed along the thickness direction). The length of one side of the phase shift mask 80 is set to about several [m], for example.

The phase shift film 82 is formed on one main surface of the substrate 81 in a predetermined pattern. Although the phase shift film 82 has translucency with respect to light for exposure, its transmittance is smaller than that of the substrate 81. The transmittance of the phase shift film 82 is, for example, about several [%] (more specifically, 5 [%]). The phase shift film 82 shifts the phase of the light transmitted through itself by approximately 180 degrees with respect to the phase of the light transmitted through the light transmitting portion 8a. Such a phase shift film 82 is formed of, for example, tantalum oxide.

The light shielding film 83 is formed on the phase shift film 82 in a predetermined pattern, for example. The light shielding film 83 is formed in a region inside the outline of the phase shift film 82 when viewed in plan. The light-shielding film 83 has light-shielding properties for exposure light, and is formed of, for example, chromium or chromium oxide.

Hereinafter, a region of the phase shift mask 80 in which the phase shift film 82 is not formed in a plan view is referred to as a light-transmitting portion 8a, and a region in which the light-shielding film 83 is formed in a plan view is referred to as a light-shielding part. It is called (8c), and the area between the light-transmitting part 8a and the light-shielding part 8c is called a phase shift part 8b. The light-transmitting portion 8a, the phase shifting portion 8b, and the light-shielding portion 8c are each formed in a predetermined pattern when viewed in a plan view. The width of the light-transmitting portion 8a (the width along the X-axis direction in FIG. 2) is set to, for example, about 2 to 4 [µm], and the width of the phase shift portion 8b (in FIG. 2, the width along the X-axis direction) ) Is set to, for example, about 0.3 to 0.5 [µm].

When exposure is performed in the exposure apparatus using this phase shift mask 80, on the substrate, light transmitted through the light transmitting part 8a and the light transmitted through the phase shift part 8b interfere at the boundary, (Cancer (暗)) occurs. As a result, the contrast of the projected image of the light transmitting part 8a can be increased. Therefore, by using the phase shift mask 80 in the exposure apparatus, the pattern can be transferred to a predetermined substrate with higher resolution compared to the case where the phase shift mask 80 is not used.

In this phase shift mask 80, the pattern shape of the phase shift part 8b is directly related to the transfer capability. For example, when the thickness of the phase shift film 82 in the phase shift unit 8b deviates from the design value, the phase difference due to the phase shift unit 8b deviates from 180 degrees. This reduces the effect of interference, lowers the resolution, and, consequently, makes the resolution of the pattern on the transferred substrate unstable, and ultimately causes a number of obstacles, such as lowering the yield of manufacture or impairing the quality of the product. . Moreover, even if the width of the phase shift part 8b deviates from a design value, the same trouble arises. Therefore, in order to determine the good or bad quality of the phase shift mask 80, the pattern characteristics of the phase shift section 8b formed in the phase shift mask 80 (specifically, the width of the phase shift section 8b and It is preferable to measure the phase difference) by the phase shift unit 8b and to properly manage the mask manufacturing process. Further, the phase shift film 82 changes over time due to oxidation, and due to this change, the phase difference due to the phase shift film 82 can also change over time. Therefore, it is desirable that the phase shift mask 80 be inspected regularly.

<Photomask inspection device>

The photomask inspection apparatus 1 measures the pattern characteristic of the phase shift part 8b formed in this phase shift mask 80. 1 and 2, the photomask inspection device 1 includes an irradiation unit 10, a detection unit 20, a moving mechanism 40, a control unit 50, an elevating mechanism 60, and a display unit 70. And a holding part (90).

The holding part 90 is a member that holds the phase shift mask 80. This holding unit 90 holds the phase shift mask 80 so that the thickness direction of the phase shift mask 80 follows the Z-axis direction. In the example of FIG. 1, the holding portion 90 holds only the peripheral portion of the phase shift mask 80. In addition, the holding part 90 may support the lower surface of the phase shift mask 80 as a whole by a light-transmitting member.

The irradiation unit 10 and the detection unit 20 are formed on opposite sides of the phase shift mask 80 in the Z-axis direction. In the example of FIGS. 1 and 2, the irradiation portion 10 is formed on the -Z side with respect to the phase shift mask 80, and the detection portion 20 is formed on the +Z side with respect to the phase shift mask 80.

The irradiation unit 10 irradiates light along the Z-axis direction, and makes the light incident on a part of the phase shift mask 80. As the light, for example, light having a wavelength equal to that of light for exposure (for example, i-line) is employed. The irradiation unit 10 includes, for example, a light source 11, a condensing lens 12, a band pass filter 13, a relay lens 14, a pinhole plate 15, a reflecting plate 16, and a condenser lens 17. Are doing.

The light source 11 irradiates light. The light source 11 is an ultraviolet irradiator, for example. As this ultraviolet irradiator, a mercury lamp can be adopted, for example. The light irradiation/stopping of the light source 11 is controlled by the control unit 50.

The condensing lens 12, the band pass filter 13, the relay lens 14, the pinhole plate 15, the reflecting plate 16, and the condenser lens 17 are between the light source 11 and the phase shift mask 80. In the above, they are arranged in this order.

The condensing lens 12 is a convex lens and is arranged so that its focal point is positioned on the light source 11. The light irradiated from the light source 11 becomes collimated light or light with a small diffusion angle by the condensing lens 12, and this light enters the band pass filter 13. The band pass filter 13 transmits only light having a predetermined wavelength band (transmission band) among the light. As this wavelength band, a wavelength band of exposure light (for example, a wavelength band including i-line) can be adopted. The wavelength band of the band pass filter 13 is set to be narrow, and light having a substantially short wavelength (so-called monochromatic light) passes through the band pass filter 13. The light transmitted through the band pass filter 13 enters the relay lens 14.

The relay lens 14 is a convex lens and condenses incident light into the pinhole 151 of the pinhole plate 15. The pinhole 151 penetrates the pinhole plate 15 in the thickness direction. The pinhole plate 15 is disposed at a position where the pinhole 151 becomes the focus of the relay lens 14. The light that has passed through the pinhole 151 becomes substantially irradiated from a point light source, and is incident on the reflective surface of the reflective plate 16. The reflecting plate 16 is formed to change the traveling direction of light, and causes the light to enter the condenser lens 17. The condenser lens 17 is a convex lens and is disposed at a position where the focus becomes substantially the pinhole 151. The condenser lens 17 converts the incident light into collimated light or light with a small diffusion angle. The NA (number of apertures) of light from the condenser lens 17 is set to an appropriate value by the condenser lens 17 and the pinhole 151. The irradiation unit 10 irradiates this light onto a part of the phase shift mask 80 along the Z-axis direction.

The detection unit 20 detects the light transmitted through the phase shift mask 80 and detects a diffraction pattern by the light. The detection unit 20 includes, for example, an objective lens 21, an imaging lens 22, a prism 23, a slit mask 24, a Fourier transform lens 25, a relay lens 26, and an image sensor (optical sensor). (27, 28) is provided.

The objective lens 21, the imaging lens 22, the prism 23, the slit mask 24, the Fourier transform lens 25 and the image sensor 27 move away from the phase shift mask 80 in the Z-axis direction. Therefore, they are arranged in this order.

The light that has passed through this part of the phase shift mask 80 is expanded through the objective lens 21 and the imaging lens 22. Some of the light from the imaging lens 22 is reflected by the prism 23 to the image sensor 28 side. In short, the prism 23 is an optical element that guides a part of the light from the phase shift mask 80 to the image sensor 28. This optical element is not limited to the prism 23, and may be a mirror or a half mirror.

The slit mask 24 is disposed at the focal point of the imaging lens 22. The light incident on the slit mask 24 from the imaging lens 22 passes through the slit 24a formed in the slit mask 24. This slit mask 24 blocks light in regions other than the slit 24a and allows light to pass only through the slit 24a, thereby exerting the function of a visual field stop to narrow the field of view. This slit 24a has an area such that only light from a region including only the phase shift portion 8b and its vicinity is transmitted.

As illustrated in FIG. 2, the slit mask 24 has a substrate 241 and a light shielding film 242. The base material 241 has translucency to light for exposure, and is formed of, for example, quartz glass or the like. The base material 241 has a plate-like shape, and has, for example, a rectangular shape when viewed in plan. The substrate 241 is formed in a posture along the Z-axis direction in its thickness direction.

The light shielding film 242 is formed on one main surface of the substrate 241. The light-shielding film 242 has a light-shielding property for exposure light, and is formed of, for example, chromium or chromium oxide. This light-shielding film 242 is formed avoiding a partial region of the substrate 241 when viewed in a plan view. In this part of the region, a slit 24a through which light is passed is formed. The slit 24a has an elongate shape when viewed in plan.

3 is a plan view schematically showing an example of the configuration of the slit mask 24. In FIG. 3, in order to show an example of the optical positional relationship of the phase shift mask 80 with respect to the slit mask 24, the light-transmitting part 8a and the phase shifting part 8b are also shown virtually by a double-dashed line. . In short, this two-dot chain line represents a projection image in which the light transmitting portion 8a and the phase shifting portion 8b are projected onto the slit mask 24 via the objective lens 21 and the imaging lens 22. In the following, the projected image obtained by projecting the light transmitting part 8a onto the slit mask 24 is referred to as the light transmitting part image 80a, and the projected image projecting the phase shift part 8b onto the slit mask 24 is a phase shift part. Call it image (80b).

In the example of FIG. 3, the longitudinal direction of the slit 24a is along the extension direction of the light-transmitting portion 8a, and the slit 24a faces the phase shifting portion 8b. More specifically, inside the slit 24a, the whole of one phase shift part image 80b in the width direction (herein the X-axis direction) and adjacent to the one phase shift part image 80b A part of the light-transmitting part image 80a in the width direction is included. In other words, the light transmitted through the entire phase shift section 8b in the width direction and a part of the transmissive section 8a adjacent to the phase shift section 8b in the width direction is transmitted through the slit 24a. ).

Referring again to FIG. 2, the light passing through the slit 24a is formed on the imaging surface of the image sensor 27 through the Fourier transform lens 25. The image sensor 27 is disposed so that the imaging surface is positioned at the focal point of the Fourier transform lens 25.

The image sensor 27 is, for example, a CCD image sensor or the like, and generates a captured image IM1 based on the light formed on its own imaging surface, and outputs the captured image IM1 to the control unit 50. Since light is formed on the image sensor 27 through the Fourier transform lens 25, in this captured image IM1, the light that has passed through the light-transmitting section 8a and the light that has passed through the phase shift section 8b. A diffraction pattern is taken. Further, the image sensor 27 is not limited to an image sensor having pixels arranged in two dimensions, but may be a line sensor having pixels arranged in one dimension. In short, the image sensor 27 may be an optical sensor capable of converting the luminance distribution of the light intensity pattern (diffraction pattern) formed in the X-axis direction into digital data.

The control unit 50 calculates the pattern characteristics (width and phase difference) of the phase shift unit 8b based on this diffraction pattern. Specific examples of the diffraction pattern and specific examples of the calculation method will be described later in detail.

The light passing through the relay lens 26 from the prism 23 forms an image on the imaging surface of the image sensor 28. The image sensor 28 is arranged so that its imaging surface is positioned at the focal point of the relay lens 26. The image sensor 28 is, for example, a CCD image sensor or the like, and generates a captured image IM2 based on the light formed on its own imaging surface, and outputs the captured image IM2 to the control unit 50. In the picked-up image IM2, the measurement target area of the phase shift mask 80 is taken. The control unit 50 may display the captured image IM2 on the display unit 70. Accordingly, it is possible to visually recognize which area of the phase shift mask 80 the worker is measuring.

The movement mechanism 40 moves the holding part 90 in the XY plane. Accordingly, the phase shift mask 80 held by the holding unit 90 also moves in the XY plane. The moving mechanism 40 has, for example, a ball screw mechanism, and is controlled by the control unit 50. When the phase shift mask 80 moves in the XY plane, the irradiation unit 10 and the detection unit 20 can be scanned with respect to the phase shift mask 80. Accordingly, the pattern characteristics of the phase shift unit 8b can be measured in a plurality of measurement regions of the phase shift mask 80. In addition, the movement mechanism 40 should just have a function and a structure of relatively moving the phase shift mask 80 with respect to the irradiation unit 10 and the detection unit 20, and for example, the irradiation unit 10 and the detection unit 20 ) May be moved integrally.

The lifting mechanism 60 raises and lowers the holding part 90 in the Z-axis direction. Accordingly, the phase shift mask 80 held by the holding unit 90 is also raised and lowered. The lifting mechanism 60 has a ball screw mechanism, for example, and is controlled by the control unit 50. The phase shift mask 80 can be moved to the focal point of the objective lens 21 by raising and lowering the phase shift mask 80 by the lifting mechanism 60. In addition, the lifting mechanism 60 should just have a function and a structure of relatively lifting the phase shift mask 80 with respect to the detection part 20, and for example, the detection part 20 may be raised and lowered.

The display unit 70 is, for example, a display device such as a liquid crystal display or an organic EL display, and the contents of the display are controlled by the control unit 50. For example, the control unit 50 outputs an image signal including the measurement result to the display unit 70. The display unit 70 displays the measurement result based on the image signal. Further, as described above, the display unit 70 may display the captured image IM2 under the control of the control unit 50.

The prism 23, the slit mask 24, the Fourier transform lens 25, the relay lens 26, and the image sensors 27 and 28 are incorporated in the optical head 30 of FIG. 1.

By the way, since the pattern of the light-transmitting part 8a formed in the phase shift mask 80 is suitably set by the design of a board|substrate, the extension direction of the light-transmitting part 8a differs depending on the measurement target position of a pattern. Therefore, in each position of the XY plane, in order to make the longitudinal direction of the slit 24a follow the extending direction of the light-transmitting part 8a, the slit mask 24 may be formed rotatably.

For example, the optical head 30 has an upper member 31 and a lower member 32 that are rotatably connected to each other, and the above optical elements incorporated in the optical head 30 are attached to the upper member 31. It may be built-in. The lower member 32 may be fixed non-rotatably with respect to the casing of the photomask inspection apparatus 1, and the upper member 31 may be rotatably connected with respect to the lower member 32. According to this, by rotating the upper member 31 in the XY plane, the longitudinal direction of the slit 24a of the slit mask 24 incorporated in the upper member 31 can be adjusted. Further, a rotation drive mechanism (for example, a motor) for rotating the upper member 31 with respect to the lower member 32 may be provided. This rotation drive mechanism is controlled by the control unit 50.

The control unit 50 can control the photomask inspection apparatus 1 as a whole. For example, as described above, the control unit 50 controls the irradiation by the irradiation unit 10, the movement by the movement mechanism 40, the lifting by the lifting mechanism 60, and the rotation of the optical head 30. . Further, the control unit 50 also functions as an arithmetic processing unit that calculates the pattern characteristics of the phase shift unit 8b based on the captured image IM1 generated by the image sensor 27.

The control unit 50 is an electronic circuit device and may have, for example, an arithmetic processing device and a storage medium. The arithmetic processing device may be an arithmetic processing device such as a CPU (Central Processor Unit), for example. The storage unit may have a non-transitory storage medium (for example, a ROM (Read Only Memory) or a hard disk) and a temporary storage medium (for example, a RAM (Random Access Memory)). In the non-transitory storage medium, for example, a program defining a process to be executed by the control unit 50 may be stored. When the processing device executes this program, the control unit 50 can execute the processing specified in the program. Of course, some or all of the processing executed by the control unit 50 may be executed by hardware.

<Measurement method based on diffraction pattern>

4 is a graph schematically showing an example of a plurality of diffraction patterns DP1 to DP5. In Fig. 4, since the vertical axis represents the intensity of light and the horizontal axis represents the position in the X-axis direction, the diffraction pattern can also be referred to as a luminance profile.

The diffraction patterns DP1 to DP5 are diffraction patterns when the relative positions of the slit 24a and the phase shift mask 80 when viewed from a plane are changed. Here, as a parameter indicating the relative position, the distance d (see Fig. 3) is introduced. In the example of FIG. 3, the light transmitting part image 80a and the phase shift part image 80b are located inside the slit 24a when viewed from the top. Specifically, inside the slit 24a, the light-transmitting part image 80a is located on the -X side, and the phase shift part image 80b is located on the +X side. It is the distance from the distance d, the short side of the -X side of the slit 24a to the boundary between the light-transmitting part image 80a and the phase shift part image 80b. The larger this distance d is, the larger the ratio occupied by the light-transmitting portion image 80a in the slit 24a. In other words, most of the light passing through the slit 24a is occupied by the light from the light transmitting portion 8a.

The diffraction patterns DP1 to DP5 are diffraction patterns obtained when the distance d is changed. The distance (d) corresponding to the diffraction patterns DP1 to DP5 is shorter as the number at the end of the sign of the diffraction pattern is smaller. In short, the diffraction pattern DP1 is a diffraction pattern corresponding to the shortest distance (d), and the diffraction pattern DP5 is a diffraction pattern corresponding to the longest distance (d).

In addition, the range of change of the distance d is set as follows. That is, as the whole of the phase shift part image 80b in the width direction is contained in the inside of the slit 24a, the change range is set. In short, the range of change of the distance d is set so that a part of the phase shift unit image 80b does not protrude from the slit 24a in the width direction. In other words, the width (width along the X-axis direction) of the slit 24a is, when the distance d is changed within the range of change, the whole of the phase shift unit image 80b in the width direction is the slit 24a Is set to be included inside of.

As illustrated in FIG. 4, the diffraction patterns DP1 and DP2 have a shape that is convex upward (that is, one mountain shape). This is because, when the distance d is short, the light from the light-transmitting portion 8a is shielded by the slit 24a, so that the light passing through the slit 24a becomes only light from the phase shift portion 8b. Therefore, the diffraction pattern becomes a diffraction pattern with a simple rectangular opening, and thus has such a distribution shape. Further, in the diffraction patterns DP1 and DP2, the peak value of the intensity of light is relatively small. This is because the aperture is small and the transmittance of the phase shift unit 8b is low.

In the example of Fig. 4, the diffraction patterns DP3 to DP5 have two mountain shapes having two peaks. This is because the distance d becomes longer, so that not only the light from the phase shift unit 8b but also the light from the light transmitting unit 8a sufficiently passes through the slit 24a, and the phases thereof are shifted by almost 180 degrees. This is because an interference pattern is generated. In addition, in the example of Fig. 4, although two mountain shapes are shown, when the region of the horizontal axis is taken wider, new peaks appear on both sides thereof (see Fig. 5).

Hereinafter, the position at which the light intensity takes the bottom value between the highest peak value and the next highest peak value is referred to as a center position (x0).

Each peak value of the diffraction patterns DP3 to DP5 and the bottom value at the center position x0 increase as the distance d increases. This is because, in the light passing through the slit 24a, light increases from the light transmitting portion 8a having a high transmittance.

In the diffraction pattern DP3, the intensity (bottom value) of light at the center position x0 is zero. This means that the complex amplitude of the light flux passing through the light transmitting part 8a and the slit 24a in this order and the complex amplitude of the light flux passing through the phase shift part 8b and the slit 24a in this order are the same. Are doing. In short, when both complex amplitudes are equal to each other, the light intensity (bottom value) becomes zero since the light is weakened to each other in the same amount at the center position (x0).

When the complex amplitudes coincide with each other, the following equation is established.

ws' = w'·√t… (One)

Here, with reference to FIG. 3 also, ws' denotes the width of a region located inside the slit 24a of the light-transmitting portion image 80a (that is, the distance d), and w'denotes the phase shift portion image ( 80b) represents the width, and t represents the transmittance of the phase shift portion 8b.

In consideration of the magnification factor α by the objective lens 21 and the imaging lens 22, the actual width w (= w'/α) of the phase shift unit 8b and corresponding to the slit 24a By introducing the width ws (= ws'/α) of the light transmitting portion 8a, equations (1) to (2) are derived.

ws = w·√t… (2)

For example, when the width w of the phase shift unit 8b is 0.4 [µm] and the transmittance t of the phase shift unit 8b is 0.05 (= 5 [%]), it corresponds to the slit 24a The width ws of the light-transmitting portion 8a is 0.089 [µm]. In other words, if the slit mask 24 can be positioned with respect to the phase shift mask 80 so that the width ws becomes 0.089 [µm], the image sensor 27 is the picked-up image IM1 including the diffraction pattern DP3. ) Can be created. In short, the diffraction pattern DP3 can be detected.

By the way, the pitch of strength and weakness in the diffraction pattern DP3 (for example, the distance between the peak positions of the intensity of light) (Δdx) is the light-transmitting part image 80a and the phase shift part in the inside of the slit 24a It depends on the distance (pitch) Δx' between the centers of the image 80b (see FIG. 3 ). Specifically, the pitch (Δdx) is theoretically proportional to the center-to-center distance (Δx'). The proportionality coefficient β1 can be obtained in advance by simulation or experiment. Therefore, when the pitch Δdx is obtained from the diffraction pattern DP3, the center-to-center distance Δx' can be obtained based on the pitch Δdx.

In addition, this center-to-center distance Δx' geometrically satisfies the following relational expression (refer to FIG. 3 ).

w'+ ws' = 2·Δx' ... (3)

Equation (3) is a parameter for each light-transmitting part image 80a and phase shift part image 80b in the slit mask 24, so these are the parameters for the light-transmitting part 8a and the phase shift part 8b. Convert. Specifically, by substituting w = w'/α, ws = ws'/α, and Δx = Δx'/α into the equation (3), the following equation can be derived.

w + ws = 2·Δx… (4)

The following equation can be derived from equation (2) and equation (4).

w = 2·Δx / (1 + √t)… (5)

Assuming that the transmittance (t) of the phase shift unit 8b is approximately equal to the design value of the film or the transmittance (t^) of a test pattern in the vicinity, the equation (Δx) (= Δx'/α) can be obtained. Based on (5), the width w of the phase shift unit 8b can be calculated.

Further, the phase difference (θ) due to the phase shift unit 8b can also be obtained based on the diffraction pattern. 5 is a graph schematically showing an example of a plurality of diffraction patterns DP3 and DP31 to DP34. The diffraction patterns DP3 and DP31 to DP34 are diffraction patterns obtained when the phase difference θ is changed in the state in which the equation (2) is satisfied. Since Equation (2) holds, the bottom value at the center position (x0) is zero in any of the diffraction patterns DP3, DP31 to DP34. The diffraction pattern DP3 shows a diffraction pattern when the phase difference θ is 180 degrees, and the diffraction patterns DP31 to DP34 each have a phase difference θ of 144 (= 360 × 0.4) degrees, 162 (= 360 × 0.45). ), 198 (= 360 × 0.55) and 216 (= 360 × 0.6) diffraction patterns are shown.

As illustrated in FIG. 5, the waveforms of the diffraction patterns DP3 and DP31 to DP34 differ depending on the phase difference θ. In other words, the phase difference (θ) can be obtained based on the detected diffraction pattern waveform. For example, the center position (x0) fluctuates according to the phase difference (θ). Specifically, the center position (x0) moves toward the +X side as the phase difference (θ) increases. Therefore, the center position (x0) when the phase difference (θ) is 180 degrees is previously set as the reference position, and the relationship between the difference between each center position (x0) and the reference position, and the phase difference (θ), for example, It is set in advance by simulation or experiment. And, when the difference between the center position (x0) of the detected diffraction pattern and the reference position is obtained, the phase difference (θ) can be obtained based on the obtained difference and the relationship.

In addition, since each peak value and each bottom value in the diffraction pattern take a value according to the phase difference (θ), instead of the center position (x0), the phase difference (θ) may be obtained based on each peak value or each bottom value. . For example, in the region on the +X side from the center position (x0), each peak value decreases as the phase difference θ increases, and the bottom value also decreases as the phase difference θ increases. In contrast, in a region on the -X side from the center position (x0), each peak value increases as the phase difference θ increases, and the bottom value also increases as the phase difference θ increases.

In the following, the peak value closest to the center position (x0) in the +X side region is referred to as a first-order peak value, and the peak value closest to the center position (x0) in the region on the -X side is referred to as -1 order peak. It is called the value.

Here, as an example of the parameter for calculating the phase difference θ, the peak difference Δp obtained by subtracting the first order peak value from the -1 order peak value is adopted. In Fig. 5, as an example, the peak difference Δp with respect to the diffraction pattern DP34 is shown. This peak difference Δp increases as the phase difference θ increases. For example, the peak difference (Δp) in the diffraction pattern (DP3) is zero, the peak difference (Δp) in the diffraction patterns (DP31, DP32) has a negative value, and the diffraction pattern (DP31) The peak difference (Δp) in is smaller than the peak difference (Δp) in the diffraction pattern DP32. In addition, the peak difference Δp in the diffraction patterns DP33 and DP34 has a positive value, and the peak difference Δp in the diffraction pattern DP34 is the peak difference in the diffraction pattern DP33 Greater than (Δp). The relationship between the peak difference Δp and the phase difference θ can be set in advance by, for example, simulation or experiment. Therefore, when the peak difference Δp of the detected diffraction pattern is obtained, the phase difference θ can be calculated based on the peak difference Δp.

In addition, it is not always necessary to adopt the peak difference Δp between the first peak value and the -1 order peak value, and a difference between any of a plurality of peak values may be adopted. However, since the amount of variation with respect to the phase difference θ of the peak difference Δp of the first and negative first peak values is larger than that of the other two, the phase difference θ can be obtained with high accuracy.

Moreover, you may adopt a bottom value instead of a peak value. Specifically, a difference between both of a plurality of bottom values may be employed. However, since the amount of variation with respect to the phase difference θ of the difference between the bottom values is smaller than the peak difference Δp, it is preferable to employ the peak difference Δp from the viewpoint of improving accuracy.

As described above, based on the waveform of the diffraction pattern (e.g., diffraction patterns DP3, DP31 to DP34, etc.) when the bottom value at the center position (x0) is zero, the width of the phase shift unit 8b ( The phase difference θ by w) and the phase shift unit 8b can be calculated.

However, in order to detect this diffraction pattern, the position at which the width ws is the optimum value so that the interference pattern by the two beams of light passing through the light transmitting part 8a and the phase shifting part 8b appears remarkably, or It is necessary to position the slit mask 24 relative to the phase shift mask 80 in the vicinity thereof. The required precision for positioning is higher as the width ws is narrower. For example, when the width ws is 0.089 [µm], an accuracy of about several to several tens [nm] is required. The width ws also depends on the width w of the phase shift unit 8b (Equation (2)), and it can be said that the higher the accuracy required for positioning, the more the phase shift mask 80 has a finer pattern width. have.

By the way, under the above-described condition (transmittance (t) = 0.05), as can be understood from equation (5) for those skilled in the art, the calculated width w of the phase shift unit 8b includes the slit mask 24 and A calculation error (measurement error) of about 0.8 times the error (number to several tens [nm]) of positioning between the phase shift mask 80 occurs. In short, a maximum calculation error of about 16 [nm] may occur. By the way, in a pattern where the required accuracy of the line width is particularly high, since the entire line width including the phase shift unit 8b is directly connected to the resist process and is strictly managed, a measurement accuracy of about 5 to 10 [nm] is required. However, with respect to the individual width w of the phase shift unit 8b, the required accuracy is not as high as the entire line width, and the positioning accuracy (from several to several tens [nm]) is sufficient for practical use.

Therefore, in the photomask inspection apparatus 1, the image sensor 27 changes the diffraction pattern at different timings while changing the relative position between the slit mask 24 and the phase shift mask 80 over time. Repeat detection. Accordingly, a diffraction pattern corresponding to the optimal relative position or its vicinity is detected. Hereinafter, an example of the operation of the specific photomask inspection device 1 will be described.

<Example of operation of photomask inspection device>

6 is a flowchart showing an example of the operation of the photomask inspection device 1. Here, it is assumed that the control unit 50 initially irradiates the irradiation unit 10 with light.

First, in step S1, the control unit 50 controls the movement mechanism 40, and performs coarse positioning with respect to the phase shift mask 80 in the XY plane. Specifically, the movement mechanism 40 moves the phase shift mask 80 in the XY plane so that the irradiation unit 10 and the detection unit 20 face the measurement target region of the phase shift mask 80 in the Z-axis direction. Move. Both of the light-transmitting portion 8a and the phase shift portion 8b are included in the measurement target region. In addition, this alignment is not alignment with an accuracy that can always detect a diffraction pattern in which the bottom value at the center position (x0) becomes zero, but is more coarse alignment. In addition, the control unit 50 controls the rotational drive mechanism of the optical head 30 to rotate the optical head 30 so that the longitudinal direction of the slit 24a follows the extending direction of the light transmitting portion 8a in the measurement target area. Let it.

Next, in step S2, the control unit 50 controls the lifting mechanism 60 to perform an autofocus process. Specifically, the lifting mechanism 60 adjusts the position of the phase shift mask 80 in the Z-axis direction so that the distance between the objective lens 21 and the phase shift mask 80 becomes a focal length.

Next, in step S3, in a state in which the slit mask 24 and the phase shift mask 80 are relatively fine, the image sensor 27 captures a diffraction pattern at a plurality of timings, and the captured image IM1 is taken. It outputs to the control part 50. In other words, the image sensor 27 detects the diffraction pattern at a plurality of timings while changing the relative positions of the slit mask 24 and the phase shift mask 80 when viewed from the plane as time passes. More specifically, the control unit 50 controls the movement mechanism 40 so that the phase shift mask 80 is relative to the slit mask 24 along the width direction of the slit 24a (here, the X-axis direction). Move. In FIG. 3, the moving direction D1 of the phase shift mask 80 is schematically shown by a block arrow.

In addition, this moving range includes an optimal relative position at which the width ws becomes an optimal value (for example, 0.089 [µm]) with a high probability or the vicinity thereof. Therefore, in step S3, there is a timing at which the relative position of the slit mask 24 and the phase shift mask 80 becomes optimal or becomes its vicinity. Therefore, in any of the plurality of picked-up images IM1 generated by the image sensor 27, a diffraction pattern corresponding to a relative position close to the optimum relative position is included.

Further, the relative speed (relative speed in step S3) between the phase shift mask 80 and the slit mask 24 during imaging is preferably low, and is set lower than the relative speed in step S1, for example. According to this, it is easy to detect a diffraction pattern corresponding to a relative position close to the optimum relative position.

Next, in step S4, the control unit 50 selects a diffraction pattern used for measurement from a plurality of diffraction patterns each included in the plurality of captured images IM1. Hereinafter, the selected diffraction pattern is also referred to as a selective diffraction pattern (SP1). More specifically, the control unit 50 selects, as the selective diffraction pattern SP1, the diffraction pattern having the smallest light intensity (bottom value) among the plurality of diffraction patterns at the center position (x0).

Next, in step S5, the control unit 50 calculates the pattern characteristics (width w and phase difference θ) of the phase shift unit 8b based on the selected diffraction pattern SP1. 7 is a flowchart showing a specific example of a method of calculating the pattern characteristics of the phase shift unit 8b.

In step S51, the control unit 50 obtains the pitch (Δdx) of the intensity of light in the selective diffraction pattern SP1. For example, in the selected diffraction pattern SP1, the control unit 50 determines the difference between the position when the intensity of light takes the first peak value and the position when the intensity of light takes the -1st peak value by pitch ( It is calculated as Δdx).

Next, in step S52, the control unit 50 calculates the width w of the phase shift unit 8b based on the pitch Δdx calculated in step S51. More specifically, the control unit 50 obtains the center-to-center distance (Δx) based on the pitch (Δdx) calculated in step S51, and the phase shift is based on the center-to-center distance (Δx) and equation (4). The width w of the part 8b is calculated. Further, the relationship between the pitch Δdx and the center-to-center distance Δx is preset, for example by simulation or experiment, and is stored, for example, in a storage medium of the control unit 50 or the like.

Next, in step S53, the control unit 50 calculates the peak difference Δp of the selected diffraction pattern SP1. For example, in the selected diffraction pattern SP1, the control unit 50 calculates a value obtained by subtracting the first order peak value from the -1 order peak value as the peak difference Δp.

Next, in step S54, the control unit 50 calculates the phase difference θ by the phase shift unit 8b based on the peak difference Δp calculated in step S53. Further, the relationship between the peak difference Δp and the phase difference θ is set in advance by, for example, simulation or experiment, and is stored, for example, in a storage medium of the control unit 50 or the like. The control unit 50 calculates the phase difference θ by the phase shift unit 8b based on the peak difference Δp calculated in step S53 and the relationship.

In addition, these series of calculations may be performed based on one selected diffraction pattern SP1, or may be performed based on a plurality of diffraction patterns in the vicinity of a position where the relative position is optimal. As a specific example, among the detected M diffraction patterns (M is 3 or more), the upper N (N is 2 or more and less than M) diffraction patterns having a small bottom value at the center position (x0) are selected from the selected diffraction pattern ( It can be selected as SP1). Alternatively, N diffraction patterns in which the bottom value at the center position becomes equal to or less than a predetermined reference value may be selected as the selective diffraction pattern SP1. According to this, it is possible to use N diffraction patterns in which the effect of interference is relatively strong. Then, pattern characteristics may be statistically obtained from the results of the selected N selected diffraction patterns SP1. For example, as statistics, an average or regression analysis can be employed. As a specific example, one diffraction pattern may be calculated by averaging N selected diffraction patterns SP1, and the pattern characteristics of the phase shift unit 8b may be obtained based on the diffraction pattern as described above.

Next, in step S6, the control unit 50 displays the calculated pattern characteristics (width (w) and phase difference (θ)) of the phase shift unit 8b on the display unit 70. Thereby, the worker can judge the quality of the phase shift part 8b of the phase shift mask 80. Further, the control unit 50 may determine whether or not the calculated pattern characteristic of the phase shift unit 8b is within a preset good range, and display the determination result on the display unit 70. According to this, the worker can quickly know the good or bad of the phase shift unit 8b.

Further, the processing of steps S1 to S6 may be repeatedly performed while sequentially changing the measurement target region. Thereby, the entire surface of the phase shift mask 80 can be inspected.

As described above, according to the photomask inspection apparatus 1, the diffraction pattern is detected at a plurality of timings while changing the relative positions of the slit mask 24 and the phase shift mask 80 when viewed in a plane. Accordingly, the detected plurality of diffraction patterns includes a diffraction pattern corresponding to a relative position close to the optimum relative position. Therefore, the pattern characteristic of the phase shift part 8b can be calculated based on a more suitable diffraction pattern.

By the way, in Patent Document 1, a reference image (reference diffraction pattern) detected when light is irradiated with respect to a region without a phase difference is used. In the case of using this reference diffraction pattern, it is necessary to measure the diffraction pattern of the region to be measured at a time different from the measurement time of the reference diffraction pattern. Since each measurement time is different, a difference (for example, optical axis shift, etc.) may occur in the state of the optical system at each measurement time due to heat generated in the device during that period. In short, the optical conditions at both measurement times may differ from each other. When the optical conditions are different in this way, a measurement error occurs with respect to the pattern characteristics of the phase shift unit 8b. On the other hand, in the photomask inspection apparatus 1, it is not necessary to use such a reference diffraction pattern. Accordingly, the occurrence of the measurement error can be avoided, and the pattern characteristics of the phase shift unit 8b can be calculated with high precision.

In addition, in the above-described example, the control unit 50 selects, as the selective diffraction pattern SP1, the diffraction pattern having the smallest light intensity at the center position x0 among the plurality of diffraction patterns. According to this, it is possible to select a diffraction pattern corresponding to the relative position closest to the optimum position. Therefore, compared with the case of using other diffraction patterns, the pattern characteristics of the phase shift unit 8b can be calculated with high precision.

In addition, in the above-described example, the width w of the phase shift unit 8b is calculated based on the pitch of the strength and weakness in the selective diffraction pattern SP1. According to this, the width w of the phase shift unit 8b can be calculated by simple processing.

In addition, in the above-described example, the phase difference θ by the phase shift unit 8b is calculated based on the difference in peak values in the selective diffraction pattern SP1. According to this, the phase difference θ by the phase shift unit 8b can be calculated by a simple process.

In addition, in the example described above, the diffraction pattern of light transmitted through one slit 24a is analyzed to calculate the width w and the retardation θ. Therefore, compared to the case of calculating the pattern characteristics of the phase shift unit based on the interference of light transmitted through the slit using two slits, the distance between the light transmitting units 8a (distance between patterns) is narrow It is easy to apply to the shift mask 80 as well. In short, the photomask inspection apparatus 1 can also be applied to the phase shift mask 80 for a line-and-space pattern or a hole pattern array with narrow intervals.

<Presence or absence of control of moving mechanism>

In the above-described example, the image sensor 27 captures the captured image IM1 at a plurality of timings in the midst of the movement mechanism 40 relatively moving the phase shift mask 80 relative to the slit mask 24. It was created (step S3). However, movement by the movement mechanism 40 is not necessarily required. The shaking of the amount of fine movement required in the present embodiment is constantly occurring due to bending of the device structural material (for example, the moving mechanism 40 or the like). Alternatively, since residual vibration occurs in the period until the device is corrected after the relative movement of the phase shift mask 80 with respect to the slit mask 24, the diffraction pattern may be detected during this period. In short, you may detect a plurality of diffraction patterns each corresponding to a plurality of relative positions by using this residual vibration.

In short, regardless of whether the relative position of the slit mask 24 and the phase shift mask 80 fluctuates under controllable, a plurality of image sensors 27 are provided in a state in which the relative position fluctuates with the passage of time. The diffraction pattern may be sequentially detected at the timing of In short, in step S3, in a state in which the moving mechanism 40 is not performing the moving operation, the image sensor 27 sequentially generates the captured image IM1 at a plurality of timings, and detects a plurality of diffraction patterns. You can do it.

Also by this, it is possible to detect a plurality of diffraction patterns each corresponding to a plurality of relative positions. In short, compared to the case where the diffraction pattern is detected only once, the image sensor 27 is more likely to detect a diffraction pattern more suitable for calculation of the pattern characteristics of the phase shift unit 8b. And by selecting a diffraction pattern more suitable for measurement from among a plurality of diffraction patterns, the pattern characteristics of the phase shift unit 8b can be calculated with higher precision.

On the other hand, in the absence of control by the movement mechanism 40, since the variation range of the relative position depends on the surrounding environment or the like, it is not known whether the optimum relative position is included in the variation range. On the other hand, when the movement mechanism 40 moves the phase shift mask 80 relative to the slit mask 24, the phase shift mask 80 is slit so that the optimal relative position is contained within the movement range. It can be moved relative to the mask 24. Accordingly, the plurality of detected diffraction patterns can include a diffraction pattern closer to the optimum relative position, and further, the pattern characteristics of the phase shift unit 8b can be calculated with higher precision.

<Direction of movement>

As described above, the precision of positioning of the slit mask 24 and the phase shift mask 80 may require an accuracy of several tens [nm] or less. Therefore, in order to properly detect the diffraction pattern at a timing at which the relative position of the slit mask 24 and the phase shift mask 80 falls within the corresponding precision, it is preferable that the relative speed of the phase shift mask 80 and the slit mask 24 is low. Do.

So, the moving mechanism 40 moves the phase shift mask 80 with respect to the slit mask 24 along a direction inclined with respect to the width direction of the slit 24a (in other words, the width direction of the phase shift unit 8b). You may move it. 8 is a diagram for explaining the movement direction D1 of the phase shift mask 80 with respect to the slit mask 24. In Fig. 8, the moving direction D1 is schematically shown by a block arrow. This movement direction D1 intersects with respect to the width direction of the slit 24a within a range of, for example, about 30 to 60 degrees. When the phase shift mask 80 is moved with respect to the slit mask 24 along this movement direction D1, the relative velocity component along the width direction of the slit 24a can be reduced. According to this, it is easy to detect a diffraction pattern corresponding to a more optimal relative position.

<Another example of the calculation method of the pattern characteristic of the phase shift part 8b>

Next, another example of a method of calculating the pattern characteristics of the phase shift unit 8b based on the selective diffraction pattern SP1 will be described. Here, first, the outline will be described. The control unit 50 sets each initial value as the value of the unknown width w and the phase difference θ, and calculates a diffraction pattern (hereinafter referred to as a computational diffraction pattern) using the initial values. Then, the control unit 50 determines whether the calculated diffraction pattern is similar to the selected diffraction pattern SP1. In other words, the control unit 50 determines whether or not the difference between the computational diffraction pattern and the selected diffraction pattern SP1 is large. When determining that they are not similar, that is, that the difference is large, the control unit 50 changes the value of the width w and the value of the phase difference θ and calculates the calculation diffraction pattern again. The control unit 50 repeatedly executes the above operation until the computational diffraction pattern is similar to the selected diffraction pattern, that is, until the difference becomes smaller than the reference value. When the arithmetic circuit pattern is similar to the selected diffraction pattern, the value of the width w and the value of the phase difference θ represent the measured values.

<Simulation model>

9 is a diagram schematically showing an example of a simulation model M1 for calculating a computational diffraction pattern. This simulation model M1 shows the intensity distribution of light in a region of the phase shift mask 80 corresponding to the slit 24a. The intensity of light in the light-transmitting portion 8a, the phase shift portion 8b, and the light-shielding portion 8c is set in advance based on each transmittance (eg, pattern design value). In the simulation model M1 of FIG. 9, the intensity of light in each of the light-transmitting portion 8a, the phase shift portion 8b, and the light-shielding portion 8c is set constant. Therefore, the intensity of light at the boundary between the light-transmitting portion 8a and the phase shifting portion 8b is rapidly rising, and similarly, at the boundary between the phase shifting portion 8b and the light-shielding portion 8c The intensity is rising rapidly. In this simulation model M1, the widths w and ws satisfy the equation (2), and the width w becomes unknown. In addition, in this simulation model M1, the phase difference θ in the phase shift unit 8b is also unknown.

<Calculated diffraction pattern>

The control unit 50 calculates a diffraction pattern corresponding to the simulation model M1 using a known simulator. It is obvious that this calculation can be easily performed by fast Fourier transform. 10 is a graph schematically showing computational diffraction patterns AP1 to AP4. The computational diffraction patterns AP1 to AP4 are computational diffraction patterns obtained when the phase difference θ is changed. Specifically, the computational diffraction patterns AP1 to AP4 have a phase difference of 180 degrees, 208.8 degrees (= 360 × 0.58), 216 degrees (= 360 × 0.6), and 223.2 degrees (= 360 × 0.62). It's a pattern. In the example of FIG. 10, an example of the selective diffraction pattern SP1 is also shown for reference. In the example of Fig. 10, the selective diffraction pattern SP1 is similar to the computational diffraction pattern AP3.

<Operation of control unit>

11 is a flowchart showing an example of the operation of the control unit 50. This flow corresponds to a specific example of step S5 in FIG. 6. First, in step S501, the control unit 50 sets the value of the width w of the phase shift unit 8b and the value of the phase difference θ by the phase shift unit 8b as respective initial values. The initial value may be set in advance, for example.

Next, in step S502, the control unit 50 calculates a calculation diffraction pattern based on the values of the width w and the phase difference θ. Specifically, the control unit 50 applies a simulator using a fast Fourier transform to the simulation model M1 to calculate a computational diffraction pattern.

Next, in step S503, the control unit 50 determines whether or not the calculated diffraction pattern calculated in step S502 is similar to the selected diffraction pattern SP1. For example, the control unit 50 generates difference information indicating a difference between the calculated diffraction pattern and the selected diffraction pattern SP1, and determines whether the difference is smaller than a reference value. The difference information is not particularly limited, but for example, the sum of the absolute values of the difference in intensity of light at each position of the computational diffraction pattern and the selective diffraction pattern SP1 can be employed. The smaller the sum is, the smaller the difference. Alternatively, as difference information, for example, the first order of the pitch (Δdx) in the computational diffraction pattern and the pitch (Δdx) in the selected diffraction pattern (SP1), and the peak difference in the computational diffraction pattern ( A second difference between Δp) and the peak difference Δp in the selective diffraction pattern SP1 may be adopted. The smaller the difference between them, the smaller the difference between the computational diffraction pattern and the selective diffraction pattern SP1.

When it is determined that the calculation diffraction pattern is not similar to the selected diffraction pattern SP1, in step S504, the control unit 50 changes at least one of the values of the width w and the phase difference θ, and the simulation model M1 ) Is updated. Next, the control unit 50 executes step S503. In short, when the computational diffraction pattern is not similar to the selected diffraction pattern SP1, at least one of the values of the width (w) and the phase difference (θ) is considered to be still apart from the measured value, so the value is changed, The calculation diffraction pattern is calculated again (step S503), and it is determined whether or not the calculated calculation diffraction pattern is similar to the selected diffraction pattern SP1 (step S504). By repeating steps S502 to S504, a certain computational diffraction pattern is similar to the selective diffraction pattern SP1.

When it is determined in step S503 that the computational diffraction pattern is similar to the selected diffraction pattern SP1, in step S6, the control unit 50 uses the latest width w and phase difference θ as the respective measured values. 70).

As described above, the computational diffraction pattern similar to the selective diffraction pattern SP1 is calculated using the fast Fourier transform for the simulation model M1. According to this, the width w of the phase shift unit 8b and the phase difference θ due to the phase shift unit 8b can be obtained with higher precision.

In addition, in the simulation model M1, the intensity of each light of the light transmitting portion 8a and the phase shifting portion 8b is set constant. Accordingly, the setting of the intensity distribution is simple, and the calculation process can be simplified.

<The width of the phase shift section and the method of determining the phase difference by the phase shift section>

In order to efficiently calculate the computational diffraction pattern similar to the selected diffraction pattern SP1, the control unit 50 may determine the values of the width w and the phase difference θ in step S504 based on the difference information. That is, the control unit 50 may determine the values of the width w and the phase difference θ so that the difference between the computational diffraction pattern and the selected diffraction pattern SP1 is small. For example, as difference information, a case of employing a first order and a second order is considered. In this case, the control unit 50 changes the value of the width w so that the first difference becomes small, and changes the value of the phase difference θ so that the second difference decreases.

More specifically, when the pitch (Δdx) in the computational diffraction pattern is larger than the pitch (Δdx) in the selected diffraction pattern, in order to reduce the pitch (Δdx) in the computational diffraction pattern calculated next, The control unit 50 changes the width w to a smaller value. In addition, when the peak difference (Δp) in the computational diffraction pattern is larger than the peak difference (Δp) in the selected diffraction pattern, in order to reduce the peak difference (Δp) in the diffraction computational pattern calculated next, the control unit (50) changes the phase difference θ to a smaller value.

According to this, the calculation diffraction pattern calculated next can be brought close to the selective diffraction pattern SP1. Therefore, a computational diffraction pattern similar to the selected diffraction pattern SP1 can be calculated earlier.

<Another example of a simulation model>

In the simulation model M1, the intensity of light in each of the light-transmitting portion 8a and the phase shift portion 8b was set constant. However, in reality, at the boundary between the light transmitting part 8a and the phase shifting part 8b, the intensity of light is thought to have an inclination and gradually increase as it goes from the phase shift part 8b to the light-transmitting part 8a. . The boundary between the light shielding portion 8c and the phase shifting portion 8b is also the same. Therefore, such a simulation model may be utilized.

12 is a diagram schematically showing an example of the simulation model M2. In the simulation model M2, the intensity of light increases as it goes from the light-shielding portion 8c to the phase shifting portion 8b at the boundary portion between the light-shielding portion 8c and the phase shifting portion 8b. The slope becomes steeper toward the phase shift unit 8b side. Similarly, the intensity of light increases as it goes from the phase shift part 8b to the light-transmitting part 8a at the boundary part between the phase shift part 8b and the light-transmitting part 8a, and the inclination of the light-transmitting part 8a ), the more steep it is. Such intensity distribution of light may be set in advance, for example. In addition, as the width of the phase shift unit 8b in this case, a width from a position at which the intensity of light becomes a first preset value to a position to be a second preset value can be adopted.

If the simulation model M2 is adopted, the control unit 50 can more calculate the arithmetic diffraction pattern based on the actual condition, and the width w of the phase shift unit 8b and the phase difference due to the phase shift unit 8b ( θ) can be calculated with higher precision.

This simulation model M2 may be set in advance. Alternatively, the control unit 50 is based on the pattern design value of the phase shift mask 80 (transmittance of the light transmitting part 8a, the transmittance of the phase shifting part 8b, the width w of the phase shifting part 8b, etc.) Thus, a simulation model M2 (light intensity distribution) may be generated using a predetermined image simulator.

<Another example of a simulation model>

With reference to FIG. 2, the photomask inspection apparatus 1 is provided with an image sensor 28, and this image sensor 28 captures the measurement target area of the phase shift mask 80, and the picked-up image IM2 Is creating. Thus, the control unit 50 may set the light intensity distribution of the simulation model based on the captured image IM2. As a specific example, the pixel value of each pixel in the measurement target region included in the captured image IM2 may be employed as the intensity distribution of light in the simulation model. According to this, it is possible to set a simulation model based on actual conditions more.

In addition, since the intensity distribution of light passing through the light transmitting part 8a and the phase shifting part 8b is almost the same in the vicinity of the measurement target area and the extended image, the light transmitting part 8a and the phase in the vicinity of the measurement target area The pixel value of each pixel of the shift unit 8b may be employed. For example, when it is difficult to specify the measurement target area in the captured image IM2 due to deviations in the optical system, or the measurement target area is not included in the captured image IM2, the area in the vicinity thereof is included. When there is a pixel value in a region adjacent thereto, the pixel value may be employed. As a more specific example, the pixel values of each of the pixels of the light-transmitting portion 8a and the phase shift portion 8b positioned on the extension of the measurement target region may be employed as the intensity distribution of light in the simulation model.

<Operation of photomask inspection device>

13 is a flowchart showing an example of the operation of the photomask inspection device 1. First, in step S11, the control unit 50 controls the movement mechanism 40 to perform step movement. This step movement represents the movement of the image sensor 28 to a position suitable for imaging the measurement target area (or the area in the vicinity thereof). Next, in step S12, similarly to step S2, the control unit 50 controls the lifting mechanism 60 to perform autofocus processing.

Next, in step S13, the image sensor 28 generates a captured image IM2, and outputs the captured image IM2 to the control unit 50.

Next, in step S14, the control unit 50 stores an image (or an image in its vicinity) corresponding to the measurement target region among the captured image IM2 in the storage medium. For example, a predetermined area of the captured image IM2 is extracted as a measurement target area (or a nearby area), and the image is stored in a storage medium.

Next, in step S15, similarly to step S1, the control unit 50 controls the movement mechanism 40 to place the phase shift mask 80 on the slit mask 24 so that the slit 24a faces the region to be measured. It is moved relative to, and positioning in the XY plane is performed.

Next, in step S16, similarly to step S3, in a state where the slit mask 24 and the phase shift mask 80 are relatively fine, the image sensor 27 generates the captured image IM1 at a plurality of timings, The captured image IM1 is output to the control unit 50.

Next, in step S17, similarly to step S4, the control unit 50 selects a diffraction pattern (selective diffraction pattern SP1) from a plurality of diffraction patterns.

Next, in step S18, the control unit 50 calculates the pattern characteristics of the phase shift unit 8b based on the selected diffraction pattern SP1. 14 is a flowchart showing a specific example of this calculation method. First, in step S511, the control unit 50 sets the light intensity distribution of the simulation model based on the captured image IM2. As a more specific example, each pixel value of the image stored in step S14 is employed as the light intensity distribution of the simulation model. The phase difference θ may be set constant at each position of the phase shift unit 8b, or may be set at the same slope as the intensity distribution of light at each boundary unit.

Next, the control unit 50 executes Steps S512 to S515. Steps S512 to S515 are the same as steps S501 to S504, respectively, so repeated explanation is avoided.

When it is determined in step S514 that the computational diffraction pattern is similar to the selected diffraction pattern SP1, in step S19, as in step S6, the control unit 50 uses the latest width w and phase difference θ as measured values. It is displayed on the display part 70.

According to this, since the intensity distribution of the light in the simulation model is set based on the captured image IM2, the calculation diffraction pattern can be calculated more based on the actual condition, and the width w and the phase difference θ are determined with higher precision. Can be calculated.

As described above, the photomask inspection apparatus and the photomask inspection method have been described in detail, but the above description is an illustration in all aspects, and this disclosure is not limited thereto. In addition, the various modified examples described above can be applied in combination as long as they do not contradict each other. And it is construed that many modifications which are not illustrated can be assumed without departing from the scope of this disclosure.

1: photomask inspection device
8a: light transmitting part
8b: phase shift unit
8c: light shielding part
10: investigation department
24: slit mask
24a: slit
25: Fourier transform lens
27: first optical sensor (image sensor)
28: second optical sensor (image sensor)
40: moving mechanism
50: operation processing unit (control unit)
80: phase shift mask

Claims (11)

A light-transmitting portion that transmits light, a light-shielding portion that blocks light, and a phase shift portion formed between the light-transmitting portion and the light-shielding portion to transmit light and shift a phase with respect to the light transmitted through the light-transmitting portion. A photomask inspection apparatus for measuring pattern characteristics of the phase shift portion of a phase shift mask formed in a predetermined pattern,
A holding unit for holding the phase shift mask,
An irradiation unit for irradiating light to a region including the light transmitting unit and the phase shift unit,
A slit mask having a slit and disposed at a position in which light that has passed through the slit part of the transmissive portion in the width direction and the entire phase shift portion in the width direction passes through the slit;
A Fourier transform lens into which light passing through the slit is incident,
A first optical sensor for detecting a diffraction pattern of light from the Fourier transform lens at a plurality of timings
With a photomask inspection device. The method of claim 1,
Further provided with a moving mechanism for relatively moving the slit mask and the phase shift mask when viewed in a plan view,
The first optical sensor detects a diffraction pattern at a plurality of timings while the moving mechanism is relatively moving the slit mask and the phase shift mask. The method of claim 2,
The photomask inspection apparatus, wherein the movement mechanism moves relative to the slit mask and the phase shift mask along a direction inclined with respect to the width direction. The method according to any one of claims 1 to 3,
Among the plurality of diffraction patterns detected by the first optical sensor, a diffraction pattern having the smallest light intensity at a central position is selected as a selected diffraction pattern, and the width and the phase of the phase shift part are based on the selected diffraction pattern. A photomask inspection apparatus further comprising an arithmetic processing unit that obtains at least one of the phase differences due to the shift unit as the pattern characteristic. The method of claim 4,
The calculation processing unit calculates a width of the phase shift unit based on a pitch of strength and weakness of the intensity of light in the selected diffraction pattern. The method according to claim 4 or 5,
The operation processing unit calculates a phase difference by the phase shift unit based on a difference between a plurality of peak values or a plurality of bottom values of the intensity of light in the selected diffraction pattern. The method of claim 4,
The calculation processing unit,
A first step of setting an intensity distribution of light passing through the light transmitting unit and the phase shift unit, a width of the phase shift unit, and a phase difference due to the phase shift unit,
A second step of calculating a calculation diffraction pattern using a fast Fourier transform based on the intensity distribution, the width, and the phase difference, and
A third step of determining whether the computational diffraction pattern is similar to the selected diffraction pattern, and
In the third step, when it is determined that the calculation diffraction pattern is not similar to the selected diffraction pattern, a fourth step of performing the second step and the third step by changing the width and the phase difference
A photomask inspection device that runs. The method of claim 7,
The calculation processing unit, in the first step, sets the intensity distribution so that the intensity of light transmitted through each of the phase shift unit and the light transmitting unit is constant. The method of claim 7,
The calculation processing unit, in the first step, sets the intensity distribution so that the intensity of light gradually increases at the boundary between the phase shift unit and the light-transmitting unit from the phase shift unit to the light-transmitting unit. Mask inspection device. The method of claim 7,
A second optical sensor,
An optical element formed between the slit mask and the phase shift mask and guiding a part of light from the phase shift mask to the second optical sensor
And additionally,
The calculation processing unit, in the first step, sets the intensity distribution based on the image captured by the second optical sensor. A light-transmitting portion that transmits light, a light-shielding portion that blocks light, and a phase shift portion formed between the light-transmitting portion and the light-shielding portion to transmit light and shift a phase with respect to the light transmitted through the light-transmitting portion. A photomask inspection method for measuring a pattern characteristic of the phase shift portion of a phase shift mask formed in a predetermined pattern,
A step of irradiating light to a region including the light transmitting unit and the phase shift unit by an irradiation unit,
The first optical sensor, through a slit formed in a slit mask and a Fourier transform lens, transmits a part in the width direction of the light transmitting part and the whole in the width direction of the phase shift part at a plurality of timings. Detection process
With a photomask inspection method.

Images