JP2003015270A - Device and method for inspecting phase shift mask - Google Patents

Abstract

PROBLEM TO BE SOLVED: To securely detect the local phase defect of a phase shift mask by an inexpensive inspecting device. SOLUTION: The whole inspection area of a phase shift mask being the object of inspection is irradiated by an illumination system 20 with almost uniform light quantity. The image of the phase shift mask is image-picked up by an image pickup optical system 8 having the resolution limit of Rayleigh where a pattern being the object of inspection on the phase shift mask is not resolved and by a CCD camera 4, and a picture processor 3 analyzes the light intensity distribution of the image. Thus, the local phase defect of the phase shift mask is inspected.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a local phase defect of a phase shift mask, which is a type of photomask often used in an exposure step of photolithography processing when manufacturing a semiconductor device such as a semiconductor integrated circuit (IC). The present invention relates to a phase shift mask inspecting apparatus and a phase shift mask inspecting method.

[0002]

2. Description of the Related Art A semiconductor integrated circuit is manufactured by using a photolithography technique. However, with the advancement of its performance and the increase in capacity, there are many problems to be solved.
Photolithography is a technique in which an original plate called a photomask is first created by photography, and then a large amount of it is transferred and copied using photography. As a next-generation memory that supports multimedia and IT (Information Technology) technology, a large-capacity memory of GB (gigabyte) or more is desired. The rule (line width) required in this large capacity memory is 0.13 μm or less. However, it is difficult to fabricate such an extremely fine semiconductor only with the current photolithography technology.
Even if a transfer optical system having a high numerical aperture (NA) of 0.6 is used with an ArF laser stepper having a high resolving power, the transfer image obtained by using a photomask having a normal structure has the same contrast. Does not reach 0.6. Therefore, a phase shift mask (PSM / Phas) that increases the image resolving power by controlling not only the intensity of the “wave” but also its phase by utilizing the nature of the “wave” of the light passing through the photomask
e Shift Mask) has been developed in recent years.

Here, regarding the phase shift mask,
This will be described with reference to FIGS. 17 to 22. FIG. 17 is a diagram showing calculated values of the light intensity distribution of an image when transfer is performed using a normal photomask, and FIG. 18 is an image when the same pattern as in FIG. 17 is transferred using a phase shift mask. FIG. 19 is a sectional view schematically showing a configuration example of a normal photomask, FIG. 20 is a sectional view schematically showing a configuration example of a phase shift mask, and FIG. 21 is a phase diagram. FIG. 22 is a cross-sectional view schematically showing another configuration example of the shift mask, and FIG. 22 is a diagram showing an example of one inspection range when the phase shift mask is inspected by the conventional method.

FIG. 17 shows the light of a transferred image when an L & S (line and space) pattern having a line width of 0.13 μm is transferred using an ordinary photomask by using an ArF laser stepper with NA of 0.6. The calculated value of the intensity distribution is shown. Here, the polygonal line 201 indicates the width of the pattern,
Reference numeral 02 indicates the light intensity distribution of the transferred image. Also, Imi
n and Imax represent the minimum light amount and the maximum light amount of each peak, respectively. A contrast of 0.7 or more is desirable for stable fabrication of a semiconductor circuit, but as shown in FIG. 17, a normal photomask can obtain a contrast of less than 0.6. On the other hand,
8 shows the calculated value of the intensity distribution of the transferred image when the transfer is performed using the phase shift mask with the same apparatus configuration as above. Here, the polygonal line 203 shows the width of the pattern, and the curve 204 shows the light intensity distribution of the transferred image. This Figure 1
As is clear from 8, if a phase shift mask is used,
With the same laser stepper as described above, it is possible to obtain a contrast of 0.9 or higher.

As shown in FIG. 19, a normal photomask has a metal (chrome) 212 arranged on a glass substrate 211 for blocking light, and the metal portion blocks light.
The portion without the metal 212 transmits light. On the other hand, in the phase shift mask, as shown in FIG. 20, a transparent material is appropriately arranged (alternately in the case of the Levenson type) as a phase shifter 213 in a portion that transmits light, and the phase of transmitted light is changed to change the resolution. Is to increase. Normally, the Levenson-type phase shift mask is configured so that the phases of the adjacent transmissive portions are shifted by 180 degrees. As a result, when the transmitted light that passes through the adjacent transmission parts overlaps in the non-transmission parts, they cancel each other due to interference,
As a result, the resolution is improved. This is the principle of the phase shift mask. Also, as shown in FIG. 21, the phase shifter 2
Instead of providing 13, the glass substrate may be eroded to provide the phase shift unit 214 to obtain the phase difference.

In order for the phase shift mask to exert its performance, the phase difference of the transmitted light passing through the adjacent transmitting portions must be 180 degrees. Therefore, the phase shifter 21
If the phase difference deviates from 180 degrees due to a change in the thickness of No. 3 or a change in the corrosion depth of the phase shift portion 214, a defect will occur in the transferred image at that portion. This is hereinafter referred to as a local phase defect. Such a local phase defect is fatal to the manufacturing process of the semiconductor integrated circuit,
It is difficult to quickly find this local phase defect in the mask inspection process. This is one of the reasons for hesitating to adopt the phase shift mask, and is also the reason for hindering the increase of the memory capacity.

The local phase defect has a size smaller than the control limit in the manufacturing process of the phase shift mask, and it is very difficult to find it in the manufacturing process. Therefore, it is inspected in the final inspection process. The conventional method of phase defect inspection is to measure the thickness of the phase shifter on the phase shift mask and the corrosion depth of the glass substrate using a film thickness meter or a depth gauge to inspect whether they are equal to the design values. is there. Since the so-called dimension measurement in the thickness and depth directions is an extremely minute portion, the conventional dimension inspection device requires a huge amount of inspection time and is not suitable for practical use. Here, as an example, consider the case of inspecting the entire 100 mm ^square phase shift mask. If the circuit size of the semiconductor is 0.2 μm, the corresponding phase shift mask is usually drawn four times as large, so the corresponding phase shift mask has a circuit pattern of 0.8 μm. Is drawn. This is subjected to reduction exposure on a photosensitive material on a wafer by a transfer exposure device such as a stepper, and developed to form a semiconductor circuit pattern.

Individual phase shifters 21 of such size
In order to measure the thickness of No. 3, one inspection area is
As shown by the arrow X in FIG. 2, the magnification must be set so as to be about 5 μm ^□ and the magnification must be increased. Then,
Number of tests required to test the entire surface of the phase shift mask will be 400 million times (one side 100 mm / 5 [mu] m = 20000, therefore the number of inspections is 20000 ² times). Even if the time required for one inspection is 0.001 seconds, the inspection time is 400,000 seconds, that is, about 5 days or less. However, the inspection of the phase shift mask is not practical unless it is completed within several tens of minutes to several hours. That is, it is difficult to inspect the phase defect by the conventional method or apparatus. Furthermore, there is a problem in that the cost of the inspection device is increased because expensive equipment must be used in order to create a dimension inspection device with high resolution.

[0009]

As described above, the local phase defect is a defect of the phase shift mask which is difficult to inspect. As the semiconductor circuit pattern is further miniaturized, the size of the local phase defect itself becomes smaller, and this problem remains as a major unsolved problem. Furthermore, as described above, it is difficult to solve this problem by the conventional method of "dimension measurement". The present invention solves such a problem, provides a phase shift mask inspection apparatus and inspection method capable of detecting even minute defects such as local phase defects in a short time, and achieves a significant cost reduction of the apparatus. The purpose is to

[0010]

In order to achieve the above object, the phase shift mask inspection apparatus of the present invention comprises an illuminating means for illuminating a phase shift mask to be inspected with a substantially uniform light amount over the entire inspection area. An objective lens for observing the phase shift mask, an image pickup unit for reading an image of the phase shift mask by the objective lens, and an image analysis unit for performing image analysis of the image read by the unit, the illumination Means, the objective lens, and the imaging means, the Rayleigh resolution limit of the optical system is a value that does not resolve the pattern to be inspected on the phase shift mask, and the image analysis means uses the imaging means. Means for inspecting the local phase defect of the phase shift mask by analyzing the light intensity distribution of the image read by the means. To.

In such a phase shift mask inspection apparatus, the image analysis means performs mask processing on a portion of the image read by the image pickup means other than the inspection object, and masking means for extracting only the inspection object portion. It is preferable to have a means for inspecting the local phase defect only in the inspection target portion. Further, when the image analysis means finds a local phase defect on the phase shift mask, means for calculating the size of the local phase defect by comparing the light intensity of the defective portion with the light intensity of the surrounding area. It is good to have.

Alternatively, in the case of a phase shift mask inspection device for inspecting a phase shift mask having a plurality of identical patterns formed thereon, in order to move the phase shift mask at least on a plane perpendicular to the observation direction of the objective lens. Of the phase shift mask by analyzing the difference in the light intensity distribution in the images of the corresponding portions of the two patterns on the phase shift mask read by the image pickup means. It is preferable to use it as a means for inspecting local phase defects. Furthermore, the light of the image read by the image pickup means from the design value of the pattern formed on the phase shift mask and the characteristic value of the optical system configured by the illumination means, the objective lens, and the image pickup means. Providing a light intensity distribution calculating means for calculating the intensity distribution, and analyzing the difference between the light intensity distribution of the image read by the image capturing means and the light intensity distribution calculated by the light intensity distribution calculating means by the image analyzing means. May be a means for inspecting the local phase defect of the phase shift mask.

In these phase shift mask inspection apparatuses, when the image analysis means finds a local phase defect on the phase shift mask, it compares the light intensity difference at the defect portion with the light intensity difference at the periphery thereof. Therefore, a means for calculating the size of the local phase defect may be provided. Further, it is preferable that the illuminating means is an illuminating means by ring-shaped epi-illumination. Further, when the image analysis unit finds a local phase defect on the phase shift mask, it is preferable to provide a unit for outputting position information of the local phase defect to an external device.

In order to achieve the above-mentioned object, the phase shift mask inspection method according to the present invention illuminates the phase shift mask to be inspected with a substantially uniform light amount over the entire inspection region, and the image of the phase shift mask is obtained. Is imaged by an optical system having a Rayleigh resolution limit that does not resolve the pattern to be inspected on the phase shift mask, and the local phase defect of the phase shift mask is analyzed by analyzing the light intensity distribution of the image. It is characterized by inspecting. In such a phase shift mask inspection method, a mask process is performed on a portion other than the inspection target of the image captured by the optical system to extract only the inspection target portion, and the local phase defect is inspected only on the inspection target portion. You should do it.

Furthermore, when a local phase defect is found on the phase shift mask, the size of the local phase defect is calculated by comparing the light intensity of the defective portion with the light intensity of the surrounding area. Good. The phase shift mask inspecting method according to the present invention also provides a method of inspecting a phase shift mask having a plurality of identical patterns,
The entire area of the inspection area is illuminated with a substantially uniform amount of light, and the image of the phase shift mask is formed by an optical system having a Rayleigh resolution limit that does not resolve the pattern to be inspected on the phase shift mask. The local phase defect of the phase shift mask is inspected by imaging two patterns of the same pattern at corresponding positions and analyzing the difference in the light intensity distribution of the two images.

Alternatively, the phase shift mask to be inspected is illuminated over the entire surface of the inspection area with a substantially uniform light amount so that the image of the phase shift mask does not resolve the pattern to be inspected on the phase shift mask. An image is taken by an optical system having a Rayleigh resolution limit, and the light intensity of the image is calculated from the light intensity distribution of the image, the design value of the pattern formed on the phase shift mask, and the characteristic value of the optical system. The local phase defect of the phase shift mask may be inspected by analyzing the difference from the distribution. In these phase shift mask inspection methods, when a local phase defect is found on the phase shift mask, the local phase defect is compared by comparing the difference in light intensity at the defective portion with the difference in light intensity at the surroundings. It is advisable to calculate the dimension of. Further, it is preferable that the illumination is performed by a ring-shaped epi-illumination.

[0017]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention will be described below with reference to the drawings. [Inspection Principle: FIG. 7 to FIG. 12] The phase shift mask inspection apparatus and the inspection method according to the present invention do not measure the thickness of the phase shifter on the phase shift mask and the corrosion depth of the glass substrate as in the related art, but The light distribution created by the phase shift mask is set so that the relationship between the imaging system having a low numerical aperture and a low magnification and the wavelength of the illumination light is smaller than the resolution corresponding to the Rayleigh resolution limit, and is uniform. By imaging in combination with an illumination optical system that can illuminate, and measuring and calculating the light amount distribution after passing through the phase shift mask,
The inspection is for a local phase defect on the phase shift mask.

Therefore, first, the principle of this inspection will be described with reference to FIGS. FIG. 7 is a diagram showing an example of a local phase defect to be found by the phase shift mask inspection apparatus and inspection method of the present invention, and FIG. 8 is a defect-free L
FIG. 9 is a diagram showing a calculated value of a light intensity distribution when observing an & S pattern, FIG. 9 is a diagram showing a calculated value of a light intensity distribution when observing an L & S pattern having a local phase defect, and FIG. 10 is a defect-free L & S pattern. FIG. 11 is a diagram for explaining the light amount distribution of the light passing through the light source, FIG. 11 is a diagram for explaining the light amount distribution of the light passing through the L & S pattern having the local phase defect, and FIG. It is a figure for demonstrating the light amount distribution at the time of imaging a pattern.

A typical example of the local phase defect of the phase shift mask for semiconductor memory fabrication is shown in FIG. In this example, the depth of a part of the phase shift part 214 provided by the corrosion of the glass substrate is changed to a local phase defect part 215 due to a defect or an error in the manufacturing process or "fluctuation". is there. here,
The dimensional error of the local phase defect portion 215 is set to D nm, and the line width of the pattern itself is L μm at the transmission portion and S μm at the non-transmission portion.
Then, as an example of a local phase defect that transfers a spot on the wafer, for example, L & S = 1 μ
m, D = 60 nm. Defects of this size are
It cannot be seen with an ordinary optical microscope due to lack of resolution. In addition, it is extremely difficult to find a defect of this size with a conventional defect inspection apparatus. This is because the conventional defect inspection apparatus measures the dimensions, and as described in the section of the conventional technique, the dimension measurement accuracy is insufficient, or even if there is accuracy, the measurement time is enormous. Because. Hereinafter, description will be given mainly using the above-mentioned typical local phase defect example.

According to the present invention, if the local phase defect on the phase shift mask is transferred, the light distribution has fluctuations therein. Therefore, the light distribution produced by the phase shift mask is analyzed in detail. By doing so, an attempt is made to find a local phase defect that can be transferred.
Therefore, an optical system having a low numerical aperture and a low magnification is used, and the illumination wavelength is set so that the resolution of the optical system becomes smaller than the Rayleigh resolution limit. Then, L without phase defects
The image of the phase shift mask on which the & S pattern is formed by the objective lens has a constant and uniform light amount distribution as shown by a straight line 122 in FIG. Here, the polygonal line 121 indicates the width of the L & S pattern.

This constant value changes depending on the phase difference and the degree of coherence. In this example, the phase difference is 180 degrees, and the constant amount of light on the image plane in this case is about 0 to about 0.15 depending on the size of the L & S pattern and the coherence ratio. However, “1” is the image plane light amount (maximum transmitted light amount) of the mask having no pattern. On the other hand, when the phase shift mask has a local phase defect as shown in FIG. 7, the light amount distribution changes from a constant value. When the light amount distribution of the image of such a phase shift mask is calculated, it becomes as shown by the curve 124 in FIG. However, L, S = 1 μm, defect size D = 60 nm, NA = 0.09, illumination wavelength λ =
The magnification of the objective lens was 578 nm and the magnification was 5 times. A polygonal line 123 shows the width of the L & S pattern.

As can be seen from FIG. 9, by using an objective lens having a low numerical aperture, an image is formed by an optical system having a Rayleigh resolution limit that does not resolve the pattern to be observed. Can be expanded. In this case, the vertical defect of 60 nm on the phase shift mask is enlarged to 20 μm in the horizontal direction on the image plane as indicated by the arrow Y. Since the 5 × objective lens is used, the width of the L & S pattern is 5 μm.
Usually, the local phase defect itself on the mask is difficult to see even with a high-power microscope. However, local phase defects may be found by visual observation with a magnifying glass with low magnification. This is a case where the optical system including the eyes satisfies the conditions of low magnification and low numerical aperture, and a spot is formed on the retina. It is for this reason that experts in the manufacture of phase shift masks try to find local phase defects with a low-magnification loupe or stereomicroscope.

Furthermore, with a low-power lens, the inspection area for one inspection can be enlarged. When a 5 × objective lens and a 2/3 inch CCD camera are used, the inspection area for one inspection is 1.2 mm ^□ . Then, 100m
When inspecting the entire surface of the m ^square phase shift mask, the number of inspections is about 7,000 (one side is 100 mm / 1.2 mm = 8).
3, therefore the number of inspections is 83 ^twice). Therefore, if the time required for one inspection is 0.1 seconds, the inspection time is about 1
It will be less than 2 minutes. Assuming 0.001 seconds per time as assumed in the description of the prior art, the inspection time is 10 seconds or less.
As described above, according to the present invention, it is possible to configure a high-speed apparatus that completes an inspection that requires several days to several tens of days in the case of a high-resolution apparatus in a short time of several minutes to several tens of minutes. it can.

According to the present invention, the size of the local phase defect can be measured by analyzing the light intensity distribution. Next, this point will be described. When an optical system having a low numerical aperture and a low magnification is used and the illumination wavelength is such that the resolution of the optical system becomes smaller than the Rayleigh resolution limit, the phase of the local phase defect and the defect size D
Can be calculated by comparing the light intensity generated by the defective portion with the light intensity generated by the peripheral portion. A concrete description of the phase difference χ of the defective portion is as shown in Expression 1, which does not depend on the illumination method.

[0025]

[Equation 1]

Here, φ is the phase difference when there is no defect, and C is the contrast between the defective portion and the peripheral portion. The contrast is a measurable parameter defined by the mathematical expression 2 from the light intensity I _χ of the defect spot and the peripheral light intensity I _φ .

[0027]

[Equation 2]

In the case of transmitted illumination, the size D of the defect can be obtained from the light intensity I _{χ of the} spot, the light intensity I _{φ of the} periphery, the illumination wavelength λ and the refractive index n of the glass substrate by the formula 3, In the case of (reflection) illumination, the refractive index of air is n, and the incident angle of the epi-illumination with respect to the substrate is θ.

[0029]

[Equation 3]

[0030]

[Equation 4]

This point will be described in more detail. First, the case of using transmitted illumination will be described. In the phase shift mask, the phases of the lights transmitted through the adjacent transmission parts are different by about 180 degrees, so that they cancel each other when they are superposed on the image plane. When an optical system with a low numerical aperture and a low magnification is used and the illumination wavelength is set so that the resolution of the optical system is smaller than the Rayleigh resolution limit, the objective lens of the phase shift mask on which the L & S pattern having no local phase defect is formed. The image has a constant and uniform light amount distribution. This constant value changes depending on the size of L & S, the phase difference and the coherence degree, but is about 0 to about 0.15 when the phase difference is 180 degrees. However, "1" is standardized to the image plane light amount of a mask without a pattern. If there is a local phase defect, the light distribution of the defect portion changes from this constant value, but the magnitude of the change depends on the size D of the defect.

It is assumed that adjacent light beams A and A'having a phase difference as shown in FIG. 10 are superposed on the image plane.
Since the amplitudes a of the light waves A and A ′ are the same and the phases thereof are different by φ, the light waves A and A ′ are each expressed by a mathematical expression 5 using a cosine function.
You can write it like this. Here, ω is the angular velocity of the light wave, t is time, and α is the initial phase.

[0033]

[Equation 5]

For the sake of simplicity, if ωt + α = Ω, the light intensity A + A 'on the image plane, that is, on the sensor plane is as shown in equation 6.

[0035]

[Equation 6]

What is actually measured is the time average I of the light intensity.
_φ , but cos (Ω + φ
/ 2) only, and the root mean square of this is 1/2, and therefore, Equation 7 is obtained. Here, <> represents a time average.

[0037]

[Equation 7]

Therefore, when an optical system having a low numerical aperture and a low magnification is used and the illumination wavelength is set so that the resolution of the optical system becomes smaller than the Rayleigh resolution limit, the objective lens of the phase shift mask having the L & S pattern is used. The image has a constant and uniform light amount distribution. Next, consider the light intensity of the image formed on the sensor surface by the light waves B and B'transmitted through the local phase defect portion 215 as shown in FIG. If the phase difference between the light waves B and B ′ is _χ , the light intensity I _χ is expressed by Equation 7
Since the calculation is performed in the same manner as in, the spot having a brightness different from that of the portion having no phase defect is formed.

[0039]

[Equation 8]

The contrast C of the spot is expressed by equation (9).

[0041]

[Equation 9]

When this is solved for cos ² (χ / 2), the following equation 10 is obtained.

[0043]

[Equation 10]

Therefore, since φ is a designed phase difference (when there is no defect), the spot contrast C
By measuring, the phase difference χ between the light waves B and B ′ can be obtained. When written concretely, it becomes as shown in Expression 11.

[0045]

[Equation 11]

The defect size D is 2π / λ for the optical path length nD.
Multiplying is the phase difference, and since Equation 12 holds,
It can be calculated by Equation 13. Here, n is the refractive index of the glass substrate, and λ is the illumination wavelength.

[0047]

[Equation 12]

[0048]

[Equation 13]

Alternatively, if the amplitude of the incident light is standardized to "1" in the equation (5), a = 1, and instead of the equation (13), from the spot light intensity _I.chi. And the peripheral light intensity _I.phi. , 14 can also be obtained.

[0050]

[Equation 14]

Next, the case of using epi-illumination will be described. A local phase defect portion 215 as shown in FIG.
Consider the light intensity of the image formed on the sensor surface by the light wave B and the light wave B ′ reflected by. Letting θ be the angle formed by the incident light with respect to the normal to the glass substrate, the phase difference Ψ between a portion without a phase defect and a portion with a phase defect is Ψ = 4nDπ / (λcos θ ) And the phase difference Ψ is Ψ = χ−φ, the defect size D can be obtained by Equation 15. When the incident amplitude is standardized to “1”, it can also be obtained by Expression 16.

[0052]

[Equation 15]

[0053]

[Equation 16]

In the above example, the phase shift mask of the type in which the glass substrate is ground to have the phase difference is explained, but the same formula is established for the type in which the phase shifter is provided on the substrate. . The basic equation is Equation 14 for transmitted illumination, and the refractive index n may be that of a medium through which light passes. In epi-illumination, this is multiplied by the factor cos θ / 2. Therefore, in the case of epi-illumination, cosθ / 2 (<
1) It is possible to detect defects of about double the size.
This is because the optical path difference becomes oblique along with the length of the round trip, and the phase difference is expanded to 2 / cos θ, which is an increase in the optical path difference. Therefore, in implementing the present invention, it is preferable to use epi-illumination. In addition, when the incident angle θ is increased, cos θ / 2 is decreased and even a defect having a smaller size can be detected, and the reflectance on the surface of the glass substrate is increased, so that the observable light amount is increased and the finer size is obtained. Since it becomes possible to observe a large difference in light intensity, it is effective to increase the incident angle θ in the case of epi-illumination.

A numerical example will now be given. For example,
The light intensity is converted to an 8-bit digital signal, the maximum intensity is 255, the minimum intensity is 0, and the illumination wavelength λ = 0.5.
48 μm, the refractive index of the glass substrate is 1.68,
If the measured I _χ and I _φ are 50 and 100, respectively, then I _χ = 50/255 = 0.196 I _φ = 100/255 = 0.392, so D = 22.6 nm Become. The phase difference at this time is about 30 degrees. As described above, according to the present invention, it is possible to detect the local phase defect at high speed and calculate the size of the defect from the light intensity distribution.

[Embodiment: FIGS. 1 to 3] Next, an embodiment of the phase shift mask inspection apparatus of the present invention will be described with reference to FIGS. 1 to 3. 1 is a front view showing a schematic configuration of a phase shift mask inspection apparatus according to an embodiment of the present invention, FIG. 2 is a diagram showing a configuration of an optical system thereof, and FIG. 3 is a diagram showing another configuration example of the optical system. is there.

As shown in FIG. 1, the phase shift mask inspection apparatus of this embodiment comprises an inspection machine body 1, a loader 2 and an image processing apparatus 3. The inspection machine body 1 is a CCD camera 4,
It is composed of an XYZ table 5, a mask holder 6, an imaging optical system 8, a pedestal 13, and an illumination system 20. The CCD camera 4 is an image pickup means, and is a device that converts an image formed by the imaging optical system 8 into image data. The CCD camera 4 is connected to the image processing device 3, and the acquired image data is transferred to the image processing device 3 for analysis.

The XYZ table 5 is a moving means, holds the mask holder 6 for fixing the phase shift mask 7 to be inspected, and adjusts the position of the imaging optical system 8 with the objective lens. . This XYZ table 5
Is composed of an electric XY table for adjusting the position in the horizontal direction to adjust to the inspection area and an electric Z table for adjusting the distance between the objective lens and the phase shift mask 7 to adjust the focus.

The mask holder 6 is a holder for fixing the phase shift mask 7 to be inspected, and is fixed to the electric XY table constituting the XYZ table 5. The setting of the phase shift mask 7 on the mask holder 6 may be performed manually or by an automatic mounting device called a loader. The removal from the mask holder 6 may be performed manually or by an automatic removal device called an unloader. In this embodiment, the loader 2 having both the functions of the loader and the unloader described above is used.

The image forming optical system 8 is an optical system for forming an image of the phase-shifting mask 7 that is transmitted and illuminated on the light receiving surface of the CCD camera 4, and the illumination system 20 is for performing the transmission illumination. The illumination means, which is an optical system, will be described later in detail. The gantry 13 is for firmly holding the entire inspection machine body 1 and has a structure for preventing shaking and vibration.

The loader 2 is an apparatus for mounting the phase shift mask 7 on and off the mask holder 6 mounted on the pedestal 14, but it is not essential to the present invention. The image processing device 3 is an image analysis means placed on the pedestal 15,
A computer main body 10 including a CPU, ROM, RAM and the like, and a display device (monitor) 11 include a normal personal computer or the like. The image data acquired by the CCD camera 4 is transferred to the image processing device 3 and subjected to analysis processing, which will be described later, such as mask processing and filter processing, to determine the presence or absence of a local phase defect in the phase shift mask 7. Further, the image processing device 3 is also a control means for integrally controlling the entire phase shift mask inspection device.

When a local phase defect is inspected by such a phase shift mask inspection apparatus, if the phase shift mask 7 has a defect, the image of the CCD camera 4
As shown in FIG. 9, a light amount fluctuation that is brighter or darker than the surroundings occurs. Therefore, when the image processing device 3 discovers the light amount fluctuation by analysis processing, there is a local phase defect at that position. Remember its position as a thing. When outputting the details of the defect, the image data of the recorded defect portion is processed by the method described above, and necessary data such as the defect size is calculated and displayed.

As described above, the phase shift mask inspection apparatus of the present invention detects and measures the presence and size of the light by changing the light intensity distribution produced by the local phase defect. Then, as is clear from the numerical example explained in the section of the inspection principle, in order to detect the local phase defect of several nm, it is necessary to detect the light amount difference of several%. Therefore, the fluctuation of the image plane illuminance by the illumination system 20 and the imaging optical system 8 must be made sufficiently smaller than the fluctuation of the light intensity distribution created by the local phase defect. Otherwise, the fluctuation of the light intensity distribution created by the local phase defect will be covered by the fluctuation of the optical system.

In the phase shift mask inspection apparatus of this embodiment, the fluctuation of the light intensity distribution in the optical system is reduced,
The optical system shown in FIG. 2 is used for uniform illumination. Next, this optical system will be described. In this optical system, the light emitted from the lamp 21 is radially diffused by the Kyle prism 22. This is because the central light amount of the lamp 21 is stronger than the peripheral light amount, so that the central light amount is dispersed to the peripheral portion to average the light intensity.
The light that has passed through the Kyle prism 22 enters the first kaleidoscope 23. The kaleidoscope is a kaleidoscope cell, which is an optical element for making a plurality of virtual images of a lamp and multiplexing lights to make the illumination light amount uniform.

The light that has passed through the first kaleidoscope 23 passes through the optical fiber 24 in order to prevent the fluctuation due to heat by separating the lamp house from the optical system, and further passes through the second kaleidoscope 25. After that, the phase shift mask 7 is illuminated through an interference filter 26 for converting the illumination light into a single wavelength to reduce the objective lens to a Rayleigh resolution or less and a condenser 27 for performing Koehler illumination with parallel rays. Here, the XYZ table 5 and the mask holder 6 are not shown in FIG. 2, and the illumination system 20 shown in FIG. 1 has a configuration from an optical fiber 24 to a condenser 27.

The light transmitted through the phase shift mask 7 is condensed by the objective lens 28 and is focused on the light receiving surface of the CCD camera 4 via the relay lens 29. The relay lens 29 is an optical element for suppressing the occurrence of moire by slightly changing the magnification. If the image pattern of the phase shift mask and the pixel pitch of the CCD camera have an integer ratio relationship, a moire pattern may occur on the image plane. To prevent this, for example, 0.8 times to 1.2 times. The relay lens 29 of a certain degree is used to change the magnification of the image to prevent the occurrence of moire.

By the way, as described above, it is effective to apply the epi-illumination to the phase shift mask inspection apparatus of the present invention. Therefore, an example of this epi-illumination will be described next. The illumination system of this epi-illumination is shown in FIG. 3, but in FIG. 3, the configuration of the light source unit is shown in simplified form.
Although not shown, the lamp 21 to the interference filter 26 shown in FIG. 2 are provided. The relay lens is not shown.

The light emitted from the lamp 21, which is the light source shown in FIG. 2, and converted into a uniform single-wavelength light by the structure up to the interference filter 26 becomes a parallel light beam by the condenser 27 and becomes a ring shape by the ring slit 31. Be shaped.
A fly-eye lens or the like may be arranged in the slit portion of the ring slit 31 to make the illumination light amount uniform. The light emitted from the slit is a ring-shaped perforated mirror 32.
And is condensed by the ring-shaped condenser 33 to illuminate the phase shift mask 7. The light reflected from the surface of the phase shift mask 7 is reflected by the ring-shaped condenser 33.
Is collected by the objective lens 34 arranged inside the
An image is formed on the light receiving surface of the CCD camera 4 via a relay lens (not shown). Such an illumination system is called a ring-shaped epi-illumination.

As described above, the larger the incident angle θ of the epi-illumination is, the higher the detection sensitivity of the local phase defect is. Therefore, it is preferable that θ is large, but in order to do so, the ring slit 31 and the perforated mirror 32 are required. Since the diameter of the ring-shaped capacitor 33 has to be increased, there are problems in arrangement space and workability. Therefore, it is preferable to select an appropriate θ in consideration of these factors.

[Operation Example 1: FIG. 4 to FIG. 6, FIG. 13 and FIG.
4] Next, as an operation example of the phase shift mask inspection apparatus of this embodiment, an actual inspection method will be described with reference to FIGS.
Also, description will be made with reference to FIGS. FIG. 4 is a diagram for explaining the inspection region of the phase shift mask, FIGS. 5 and 6 are diagrams for explaining the data processing in the phase shift mask inspection device of this embodiment, and FIG. 13 is the phase shift mask inspection device. 14 is a flowchart showing the processing of the phase shift mask inspection in FIG. 14, and FIG. 14 is a diagram showing a display example of the inspection result in the phase shift mask inspection apparatus.

The operation example described below applies to both the case of using the transillumination and the case of using the epi-illumination, but the case of using the transillumination will be described here. Further, here, first, an example of a case of inspecting a phase shift mask for producing a memory circuit (hereinafter, abbreviated as “memory mask”) will be described. As described above, an objective optical system having a low magnification and a low numerical aperture is adopted. In this embodiment, a 5 × objective lens is used and a CCD camera having a 2/3 inch is used. In this case, the size of the image of the inspection area taken in at one time is about 1.2 m.
m ^□ .

From the characteristics of the captured image of the phase shift mask (memory mask), as shown in FIG.
There are a rough pattern area 111 corresponding to a portion where a wiring pattern and the like are formed, and a fine pattern area 112 corresponding to a portion where a memory pattern and the like are formed. Among these, the local pattern defect does not occur because the phase shift is not performed in the rough pattern region 111. Therefore, the inspection target is
It is the fine pattern region 112 in which a local phase defect may occur. Therefore, the density distribution (transmitted light intensity distribution) of the image is inspected, and an appropriate density portion corresponding to the fine pattern is cut out. In other words, a mask that shields the rough pattern area 111 is created, and the mask processing is performed on the entire image. The standard density of the fine pattern part is
Since it is obtained from the size and phase difference of the pattern and the degree of coherence as described above, it is only necessary to mask a long continuous region where the portion greatly deviated from the density. This mask can also be created based on the design pattern of the phase shift mask to be inspected. It should be noted that the “density” here means the white level of the image. Therefore,
In the case of "high density", the image is white and the mask has a high transmittance.

As described above, the phase shift mask inspecting apparatus of this embodiment uses the objective lens having a low numerical aperture so that the size of the fine pattern on the phase shift mask becomes less than the Rayleigh resolution limit. Therefore, the fine pattern region 112 is not resolved, and its density becomes constant as described with reference to FIG. If there is a local phase defect, the enlarged area of that part will have different density values as described with reference to FIG. The phase shift mask inspection apparatus of this embodiment detects a local phase defect by detecting a portion having a different density value. For that purpose, image processing is performed using an absolute value differential filter of some kind, and the density value change To amplify. However, the filter used here does not have an average level of 0 unlike a Sobel filter, and it is desirable that the average level is not changed.

By performing this filtering process, the changed portion can be emphasized and the density of the defective portion can have the maximum value. In addition, even if there is an error due to illuminance non-uniformity of the illumination system on the image plane, it is possible to emphasize the change in the density of the defect portion which is generally steeper than the change due to the error. Therefore, when there is a local phase defect, the location where the density value after filtering is maximized can be determined to identify the location. If there is no defect, the density is theoretically a constant value, but the density actually measured has a width due to the non-uniformity of illuminance of the illumination system and the dark current noise of the CCD camera. Is Δ.

When the inspection image is binarized at a level lower than the maximum density value by Δ, if there is no defect, the curve 10 in FIG.
As shown in FIG. 5, almost all parts of the inspection area are “1”. Here, "1" represents a white level having a density higher than the threshold value. On the other hand, when there is a local phase defect, as shown by the curve 106 in FIG. 6, only the defective part is stably “1” even if binarized at a level lower than the maximum value by Δ. Here, the curves 105 and 106 are straight lines 12 in FIG.
The image data indicated by the curve 124 in FIG. 2 and FIG. 9 has been subjected to the above-mentioned image processing, and the polygonal lines 101 and 103 are also shown in FIGS. 5 and 6 to show the width of the pattern. If a stable "1" level location is created in the "0" level by performing such processing and binarization, a local phase defect exists at that location and the level becomes "1" level overall. If so, it is determined that there is no defect.

Next, an actual inspection procedure using the phase shift mask inspection apparatus of this embodiment will be described. The user first sets the inspection target phase shift mask 7 in the mask holder 6 of the phase shift mask inspection apparatus shown in FIG. 1 manually or automatically using the loader 2, and the image processing apparatus 3 is not shown. Use the keyboard and mouse to properly set the total area to be inspected, the amount of illumination light, and the focus. Then, when the start of the inspection is instructed, this apparatus starts the processing shown in the flowchart of FIG.

First, in step S1, the phase shift mask 7 to be inspected is moved to the inspection start point by the XYZ table 5, and in step S2, the image in the inspection area is CC-scanned.
It is captured as image data by the D camera 4 and transferred to the image processing apparatus 3. Then, in step S3, the above-described masking process is performed on the image data to extract only the data of the fine pattern area to be inspected. Then, in step S4, a filtering process for emphasizing the change point is performed on the inspection target area. A kind of absolute value differential filter is used for this filter processing. Therefore, the change point of the density (brightness) is emphasized. The local phase defect is converted to be brighter than the surroundings. If there is no defect, this filtering process makes the inspection areas have almost the same density.

Then, in step S5, the maximum value of the density of the filtered image data is obtained, and the image data is binarized by using the density that is smaller than the maximum value by Δ as a threshold value to create binarized data. Here, Δ means that the density has a width due to the non-uniformity of illuminance of the illumination system and the dark current noise of the CCD camera. Therefore, the size of these noises remaining after the filtering process is Δ, and Δ is only the maximum density. Binarization processing is performed using the lower level as a threshold.

As described above, if almost all the screen is "1" at this time, there is no local phase defect. If there is a local phase defect, only that point is "1" and the remaining area is "0". Therefore, in step S6, it is determined whether or not the binarized data is almost "1". If No, the process proceeds to step S7 and it is determined that there is a defect. here,
"1" is the maximum brightness, "0" is the minimum brightness (density)
That is. Further, the “most” criterion is to appropriately set the number of dots whose data is “1”, which is very unlikely to have that many defects in the inspection area, for example, “50”. .

Then, the process proceeds from step S7 to step S8, and the position where the data is "1" is recorded as the defect position. If the area of data “1” is wide to some extent, the center position of the area may be recorded. Further, in step S9, the size of the defect is calculated from the magnitude of the density change of the image before the filter processing by the above-described method, and the defect size is recorded together with the position. In step S10,
By slightly increasing Δ, the binarized data obtained by binarizing the image data after the filtering process is binarized again with the density smaller than the maximum value by Δ as a threshold value. Then, the process returns to step S6 to repeat the process.

In the first determination in step S6, even if there are a plurality of local phase defects, only the one that has the maximum density change in the image can be detected. However, when the local phase defect is found, the threshold value is set in this way. By lowering and performing the determination again, it becomes possible to detect a local phase defect having a smaller density change in the image. If steps S8 and S9 are performed twice or more at a certain inspection position, only the newly discovered local phase defect data may be recorded. Here, if a local phase defect is detected, step S1
When Δ is increased with 0, the number of local phase defects detected gradually (that is, the number of dots with data “1”)
However, if the number becomes a number that cannot be actually present in the inspection area, it is considered that the newly detected local phase defect is noise.

Therefore, "50" shown as an example of the judgment criterion in step S6 is such a small number of dots that it cannot be said that it is "almost" the full screen, but such noise was detected. In this case, it is a suitable criterion for judging that there are no more local phase defects. If there is no local phase defect or if the threshold value is decreased as a result of increasing Δ by less than the density of the region without local phase defect, the data immediately becomes “1” over the entire area,
In this case, the number of dots whose data is "1" is of course 50.
Above, it is judged that the data is almost "1".

If Yes in step S6, it is determined in step S11 whether the determination in step S6 is the first time at the current inspection position. If it is the first time, the process proceeds to step S12, it is determined that there is no local phase defect at the current inspection position, and the process proceeds to step S14. If it is not the first time, the process proceeds to step S13 and it is determined that there is no local phase defect more than the one detected so far at the current inspection position, and step S13 is performed.
Proceed to 14. In step S14, it is determined whether or not the current position is the final position of the inspection, and if it is the final position, the process ends. If it is not the final position, the process proceeds to step S15 and XY
The phase shift mask 7 is moved to a position for inspecting the next inspection area by the Z table, and the process proceeds to step S2 to repeat the process.

By such processing, the position and size of the local phase defect on the phase shift mask can be detected. In this process, steps S3 to S13
The image processing and the other data import processing are performed in sequence. However, if the storage means for storing the image data has sufficient capacity, the image data import and the image processing are performed in parallel. Alternatively, the image data may be stored first. Defective part detected by such processing (or inspection result that there is no defect)
FIG. 14 shows a display example of. This display is displayed on the display screen of the display device 11 of the image processing device 3 shown in FIG.

Phase shift mask inspection apparatus of this embodiment
Then, a 5x objective lens and a 2/3 inch CCD camera
Since it is used, the inspection area that can be inspected at one time is 1.2
mm ^□Is. The image display area 131 on the left side of FIG.
1.2 mm^□The image of the inspection area of
Three local phase defects were detected in the area surrounded by the corners.
Is displayed. However, the defective part is emphasized in the table.
It is shown. Moreover, the part shown by cross hatching is
This is a non-inspection area. Further, in this embodiment, 1.
2 mm^□The inspection area has the origin at the bottom left edge in the X and Y directions
Both are divided into 512 grid points.

In the Phase Error Information frame on the right side,
The details of this local phase defect are displayed. In this example, the position of the first defect is X = 93, Y = 75 in the coordinate system.
And the light amount difference between the defective portion and its surroundings is 11.861.
%, The depth variation of this defect calculated from this value is 68.
It is displayed that it is 2 nm. Inspection Condi
The tions frame shows the conditions for local phase defect inspection. Mask Type is the type of phase shift mask, Sensitivi
ty indicates the inspection sensitivity setting. Critical Dimensi
Line and Space of on indicate dimensions of the non-transmissive portion and the transmissive portion of the L & S pattern, respectively. Since the other displays are detailed, I will omit the explanations individually, but the user
By appropriately changing the settings displayed in the Inspection Condition frame, more appropriate inspection can be performed.

Although the objective lens having a magnification of 5 is used in this embodiment, it goes without saying that the objective lens is not limited to this.
However, if the magnification is too low, the light quantity fluctuation due to the local phase defect cannot be enlarged enough to be analyzed by the CCD camera.
If the magnification is too high, the inspection area that can be inspected at one time becomes small. Therefore, it is preferable to set the magnification so that the light quantity fluctuation spot due to the local phase defect can be enlarged to the size of about two pixels of the CCD camera. Further, there may be provided means for transferring the position information of the found defect to an external device such as an electron microscope capable of performing a more detailed analysis even though the inspection range is narrow. By doing so, the details of the defect can be known by effectively utilizing the high-resolution inspection apparatus.

[Operation Example 2: FIGS. 15 and 16] Next, referring to FIG.
5 and FIG. 16, an operation example in the case of inspecting a phase shift mask (hereinafter also referred to as “logic mask”) for forming a general logic circuit will be described. Figure 15
Is a diagram for explaining the data processing in the inspection mode of this operation example, and FIG. 16 is a flowchart showing the processing of the phase shift mask inspection in the inspection mode of this operation example. In the case of the memory mask, a constant pattern repeats, so the light intensity fluctuation of the local phase defect part was compared with the light intensity in the surrounding area.However, the same pattern was hardly repeated in the logic mask, and the light intensity in the surrounding area also took various values accordingly. ing. Therefore, in this case, the same detection as the memory mask cannot be performed.

By the way, usually, in a photomask including a phase shift mask, a plurality of the same circuit patterns are formed in different places. Further, since the local phase defect is caused by the fluctuation in the manufacturing process of the phase shift mask, the probability that the defect is generated at the same place of the same pattern formed at another place is extremely small and may be regarded as zero. Therefore,
In the case of the logic mask, the local phase defect can be detected by comparing the light distributions in the two patterns on the phase shift mask. If both patterns are normal, the difference in light distribution is 0, and if one has a defect, the difference in light distribution is detected.

An example of this detection is shown in FIG. When the light intensity distributions of the normal pattern and the pattern having the local phase defect are measured and the difference is obtained, the variation of the light distribution as shown in FIG. 15 is obtained. The vertical axis represents the difference in light intensity between the two patterns, and the horizontal axis represents the position. In the case of this example, the difference in light intensity is large because there is a defect near the center, but the difference is 0 in the peripheral part where there is no defect. Moreover, when there is no local phase defect, the difference in light intensity is 0 as a whole. However, in order to perform such an analysis, it is necessary to accurately match the positions of the two light intensity distributions to obtain the difference. If the alignment accuracy is insufficient, a difference that depends on the light intensity distribution itself of the pattern that does not depend on the defect is visible. Therefore, the alignment must be performed so that the difference is sufficiently smaller than the fluctuation of the light amount due to the local phase defect.

Next, the procedure for inspecting the logic mask in the phase shift mask inspecting apparatus of this embodiment will be described with reference to FIG.
This will be described using 6. This inspection procedure has many points similar to the inspection procedure of the memory mask described in the operation example 1. Therefore, in the flowchart of FIG. 16, parts corresponding to the processing of the flowchart of FIG. The explanation is omitted or simplified. Further, in this example, it is assumed that two same patterns are formed on one logic mask.

The user first sets the phase shift mask 7 to be inspected in the mask holder 6 of the phase shift mask inspection apparatus shown in FIG. 1 manually or automatically using the loader 2, and the image processing apparatus 3 is illustrated. Using the keyboard and mouse without the above, set all areas to be inspected, illumination light amount, focus, etc. appropriately. Then, when the start of the inspection is instructed, this apparatus starts the processing shown in the flowchart of FIG. First, in step S1, the phase shift mask 7 to be inspected is moved to the inspection start point, and in step S21 the image in the inspection area of the first pattern is taken by the CCD camera 4.
It is captured as image data by and transferred to the image processing apparatus 3.

Then, in step S22, the phase shift mask 7 is moved to the corresponding inspection position in the second pattern, and the image in the inspection area of the second pattern is similarly captured by the CCD camera 4 as image data.
Transfer to the image processing apparatus 3. Then, in step S23,
The absolute value of the difference between the two captured image data is obtained and used as the image data to be inspected. Then, the process from step S4 is performed on this image data in substantially the same manner as the case of the memory mask inspection described with reference to FIG. However,
This is different from the inspection of the memory mask in that the density of the image data is almost 0 when there is no defect.

By such processing, the position and size of the defect on the phase shift mask can be detected. However, since the absolute value of the difference between the image data of the two patterns is processed, it is not possible to determine which of the two patterns is actually defective. Note that the flow chart of FIG. 9 shows an example in which two patterns of image data are taken in for each inspection position area, but if there is enough capacity in the storage means for storing the image data, the first The image processing may be performed by capturing one line or all of the image data of the pattern, and then capturing the image data of the corresponding area of the second pattern.

By doing so, the distance for moving the phase shift mask is reduced, and the inspection time can be shortened. Further, the image data acquisition and the image processing may be performed in parallel as in the case of the memory mask inspection. When three or more identical patterns are formed on one phase shift mask, two patterns are selected and the above inspection procedure is repeated, or all the image data for the first pattern is stored. Well,
Other patterns may be sequentially compared to the pattern.

When inspecting a phase shift mask that does not form a plurality of identical circuit patterns, the light distribution that the design pattern will produce through the inspection objective lens and the phase shift mask are actually imaged. You may make it analyze the difference of the light distribution of the image. By designating a position on the phase shift mask, the designed pattern shape and size at that position can be obtained. Since the numerical aperture, aberration, illumination wavelength and coherence ratio of an optical system are known, the H.264 standard is known. H. The light intensity distribution of the image due to the pattern can be calculated using Hopkins' partially coherent imaging theory. As the light intensity distribution calculation means for performing this calculation,
The image processing device 3 may be used, or data calculated using an external computer or the like may be input to the image processing device 3. This calculated value and the actual C
Even if the difference in the light amount distribution measured by the CD camera is obtained, the local phase defect can be detected as in the above case.

According to such a phase shift mask inspection apparatus, the local phase defect on the phase shift mask is detected at high speed for any phase shift mask, and the light amount distribution is measured and calculated to calculate the size of the defect. It becomes possible. Of course, the inspection target is not limited to the phase shift mask for semiconductor circuit formation. Further, in the above-described embodiment and operation example, the example used for the inspection of the local phase defect has been described. However, even in the case of a defect due to dust, scratches, or the like, the variation of the light amount distribution naturally occurs. It goes without saying that you can do it.

[0098]

As is apparent from the above description, according to the phase shift mask inspection apparatus and inspection method of the present invention,
Local phase defects, which are extremely difficult to detect with conventional inspection equipment, can be detected with high speed and high accuracy, and the inspection efficiency of phase shift masks can be dramatically improved. It makes a great contribution. Furthermore, from the fluctuation of light quantity due to local phase defects,
If partial line coherent imaging theory calculation is used to obtain and display line width variations in defects, the lack of precision in conventional dimension measurement is compensated for, and fine dimension variations of several tens to several nm are inspected and displayed inline. You can also In addition, by transferring the data of the location and size of the defect to an electron microscope or other inspection device, it is possible to efficiently observe the local phase defect, clarify the process of the local phase defect occurrence, and correct the defect. You can also give the possibility to do.

[Brief description of drawings]

FIG. 1 is a front view showing a schematic configuration of a phase shift mask inspection apparatus according to an embodiment of the present invention.

FIG. 2 is a diagram showing a configuration of the optical system.

FIG. 3 is a diagram showing another configuration example of the optical system.

FIG. 4 is a diagram for explaining an inspection region of a phase shift mask.

FIG. 5 is a diagram for explaining data processing in the phase shift mask inspection apparatus according to the embodiment of the present invention.

FIG. 6 is another diagram for explaining the data processing in the phase shift mask inspection apparatus according to the embodiment of the present invention.

FIG. 7 is a diagram showing an example of a local phase defect to be found by the phase shift mask inspection apparatus and inspection method of the present invention.

FIG. 8 is a diagram showing calculated values of light intensity distribution when a defect-free L & S pattern is observed.

FIG. 9 is a diagram showing a calculated value of a light intensity distribution when an L & S pattern having a local phase defect is observed.

FIG. 10 is a diagram for explaining a light amount distribution of light transmitted through a defect-free L & S pattern.

FIG. 11 is a diagram for explaining a light amount distribution of light transmitted through an L & S pattern having a local phase defect.

FIG. 12: L with local phase defect due to epi-illumination
It is a figure for demonstrating the light amount distribution at the time of imaging & S pattern.

FIG. 13 is a flowchart showing the processing of the phase shift mask inspection in the phase shift mask inspection apparatus according to the embodiment of the present invention.

FIG. 14 is a diagram showing a display example of inspection results in the phase shift mask inspection apparatus.

FIG. 15 is a diagram for explaining data processing in another inspection mode.

FIG. 16 is a flowchart showing a process of a phase shift mask inspection in the inspection mode.

FIG. 17 is a diagram showing calculated values of the light intensity distribution of an image when transfer is performed using a normal photomask.

FIG. 18 is a diagram showing a calculated value of a light intensity distribution of an image when the same pattern as in FIG. 17 is transferred using a phase shift mask.

FIG. 19 is a cross-sectional view schematically showing a configuration example of a normal photomask.

FIG. 20 is a sectional view schematically showing a configuration example of a phase shift mask.

FIG. 21 is a sectional view schematically showing another configuration example of the phase shift mask.

FIG. 22 is a diagram showing an example of one inspection range when a phase shift mask is inspected by a conventional method.

[Explanation of symbols] 1: Inspection machine body 2: Loader 3: Image processing device 4: CCD camera 5: XYZ table 6: Mask holder 7: Phase shift mask 8: Imaging optical system 10: Computer main body 11: Display device 13,14,15; pedestal 20: lighting system 21: Lamp 22: Kyle prism 23: First Kaleidoscope 24: optical fiber 25: second kaleidoscope 26: Interference filter 27: Capacitor 28, 34: Objective lens 29: Relay lens 31: Ring slit 32: Perforated mirror 33: Ring-shaped capacitor

   ─────────────────────────────────────────────────── ─── Continued front page    (72) Inventor Mitsuo Eguchi             94 Kanda 504, Karatsu City, Saga Prefecture (72) Inventor Yuto Kobayashi             14 at 150, Tsurugi Tsuji, Saitama City, Saitama Prefecture (72) Inventor Norio Ikeda             1-22 Uetake, Saitama City, Saitama Prefecture (72) Inventor Masamitsu Kiyohara             25, 174 Kitahonjuku, Kitamoto City, Saitama Prefecture (72) Inventor Yuji Aida             230-20 Shimoyasumatsu, Tokorozawa, Saitama Prefecture (72) Inventor Kozo Takahashi             3-17-5 Harayama, Saitama City, Saitama Prefecture             Mour Urawa No. 301 (72) Inventor Morihisa Homoto             2-2-1 Fukuoka, Kamifukuoka City, Saitama Prefecture Dainichi             Inside the Semiconductor Products Division of this printing company (72) Inventor Hideki Yamamoto             2-2-1 Fukuoka, Kamifukuoka City, Saitama Prefecture Dainichi             Inside the Semiconductor Products Division of this printing company (72) Inventor Fumasa Murai             2-2-1 Fukuoka, Kamifukuoka City, Saitama Prefecture Dainichi             Inside the Semiconductor Products Division of this printing company F term (reference) 2G051 AA56 AB02 BA01 BB17 CA03                       CA04 CB02 DA07 EA08 EA11                       EA14 EC03 EC05 ED01 FA10                 2H095 BB03 BD04 BD15 BD18 BD27

Claims (15)

[Claims] 1. An illuminating means for illuminating a phase shift mask to be inspected with a substantially uniform light amount over the entire inspection region, an objective lens for observing the phase shift mask, and the phase shift mask by the objective lens. Image pickup means for reading the image of the image and an image analysis means for analyzing the image read by the image pickup means, and the Rayleigh resolution limit of the optical system constituted by the illumination means, the objective lens and the image pickup means. Is a value that does not resolve the pattern to be inspected on the phase shift mask, and the image analysis unit analyzes the light intensity distribution of the image read by the image pickup unit to analyze the local phase defect of the phase shift mask. An apparatus for inspecting a phase shift mask, which is a means for inspecting. 2. The phase shift mask inspection apparatus according to claim 1, wherein the image analysis unit performs a mask process on a portion of the image read by the image pickup unit other than the inspection target, and extracts only the inspection target portion. A phase shift mask inspection apparatus comprising mask processing means and means for inspecting a local phase defect only in the portion to be inspected. 3. The phase shift mask inspection apparatus according to claim 1, wherein, when the image analysis unit finds a local phase defect on the phase shift mask, the light intensity of the defective portion and the surrounding light intensity of the defective portion are detected. An apparatus for inspecting a phase shift mask, comprising means for calculating the size of the local phase defect by comparing light intensities. 4. A phase shift mask inspecting apparatus for inspecting a phase shift mask having a plurality of identical patterns formed thereon, wherein the illuminating means illuminates the phase shift mask to be inspected with a substantially uniform light amount over the entire inspection region. An objective lens for observing the phase shift mask, an imaging means for reading an image of the phase shift mask by the objective lens, and moving the phase shift mask at least on a plane perpendicular to the observation direction of the objective lens. And a moving means for moving the image, and an image analyzing means for analyzing the image read by the image pickup means. The Rayleigh resolution limit of an optical system configured by the illumination means, the objective lens and the image pickup means. Is a value that does not resolve the pattern to be inspected on the phase shift mask, and the image analysis means is the imaging means. A phase shift mask for inspecting a local phase defect of the phase shift mask by analyzing a difference in light intensity distribution in images of corresponding portions of the two patterns on the phase shift mask read by Mask inspection device. 5. An illuminating means for illuminating a phase shift mask to be inspected over the entire inspection region with a substantially uniform light amount, an objective lens for observing the phase shift mask, and the phase shift mask by the objective lens. Image pickup means for reading the image of the image, an image analysis means for analyzing the image read by the means, a design value of a pattern formed on the phase shift mask, the illumination means, the objective lens, and the image pickup means. And a light intensity distribution calculating means for calculating a light intensity distribution of an image read by the image pickup means from a characteristic value of an optical system composed of the phase shift mask and the Rayleigh resolution limit of the optical system. The image analysis means is a value that does not resolve the pattern to be inspected above, and the image analysis means is a light intensity distribution of the image read by the imaging means and the light intensity distribution. A phase shift mask inspection apparatus, which is means for inspecting a local phase defect of the phase shift mask by analyzing a difference from the light intensity distribution calculated by the cloth calculation means. 6. The phase shift mask inspection apparatus according to claim 4 or 5, wherein when the image analysis unit finds a local phase defect on the phase shift mask, the light intensity difference in the defect portion and its surroundings. 2. A phase shift mask inspection apparatus comprising means for calculating the size of the local phase defect by comparing the light intensity differences in. 7. The phase shift mask inspection apparatus according to claim 1, wherein the illumination unit is an illumination unit using a ring-shaped epi-illumination. 8. The phase shift mask inspection apparatus according to claim 1, wherein when the image analysis unit finds a local phase defect on the phase shift mask, the position of the local phase defect is detected. An apparatus for inspecting a phase shift mask, which is provided with means for outputting information to an external device. 9. A phase shift mask to be inspected is illuminated with a substantially uniform light amount over the entire inspection region so that the image of the phase shift mask does not resolve the pattern to be inspected on the phase shift mask. A phase shift mask inspecting method, which comprises inspecting a local phase defect of the phase shift mask by imaging with an optical system having a Rayleigh resolution limit and analyzing a light intensity distribution of the image. 10. The phase shift mask inspection method according to claim 9, wherein a mask processing is performed on a portion of the image captured by the optical system other than the inspection target to extract only the inspection target portion, and only the inspection target portion. The method for inspecting a phase shift mask, comprising: 11. The method for inspecting a phase shift mask according to claim 9, wherein when a local phase defect is found on the phase shift mask, the light intensity of the defective portion is compared with the light intensity around it. A method for inspecting a phase shift mask, characterized in that the dimension of the local phase defect is calculated thereby. 12. A phase shift mask having a plurality of identical patterns to be inspected is illuminated with a substantially uniform light amount over the entire inspection area, and an image of the phase shift mask is inspected on the phase shift mask. An optical system having a Rayleigh resolution limit that does not resolve a pattern captures images of two patterns of the same pattern at corresponding positions, and analyzes the difference in the light intensity distribution of the two images to obtain the phase. A method of inspecting a phase shift mask, which comprises inspecting a local phase defect of a shift mask. 13. A phase shift mask to be inspected is illuminated with a substantially uniform light amount over the entire surface of the inspection region so that an image of the phase shift mask does not resolve a pattern to be inspected on the phase shift mask. An image is taken by an optical system having a Rayleigh resolution limit, and the light intensity of the image is calculated from the light intensity distribution of the image, the design value of the pattern formed on the phase shift mask, and the characteristic value of the optical system. A method for inspecting a phase shift mask, comprising inspecting a local phase defect of the phase shift mask by analyzing a difference from a distribution. 14. The method for inspecting a phase shift mask according to claim 12, wherein when a local phase defect is found on the phase shift mask, a difference in light intensity at the defect portion and a difference in light intensity around the defect portion. A method for inspecting a phase shift mask, wherein the size of the local phase defect is calculated by comparing 15. The phase shift mask inspection method according to claim 9, wherein the illumination is performed by ring-shaped epi-illumination.