JPH11338124A - Defect shape detection method and device, and defect repair method and device using the same - Google Patents

Abstract

(57) [Problem] To provide a high quality phase shift reticle by correcting an irregular shape defect generated on a phase shift reticle with high accuracy. A method and an apparatus for measuring the shape of an optically transparent defect on an optically transparent substrate include irradiating the substrate with illumination light having different wavelengths, and comparing the substrate with the illumination light to be irradiated. The reflected light of the illumination light reflected on the substrate or the transmitted light transmitted through the substrate while moving the substrate, and detects the optically transparent defect on the substrate based on the heterodyne interference interference of the detected reflected light or transmitted light. The shape was measured. Further, an optically transparent defect on the optically transparent substrate is detected at a first resolution, and the shape of the detected defect on the substrate is measured at a second resolution higher than the first resolution. A defect repair method and apparatus for repairing a defect based on information on the position and shape of the defect on the substrate measured at the second resolution.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a technique for correcting a defect generated on a reticle, a photomask or the like (hereinafter referred to as a reticle or the like), and more particularly, to detecting a defect shape of a transparent phase defect generated by a phase shift reticle. The present invention relates to a method and an apparatus for performing the method, and a defect correcting method and an apparatus using the same.

[0002]

2. Description of the Related Art In a manufacturing process of a photomask such as a reticle used for manufacturing an LSI or a printed circuit board, a defect is strictly inspected, and after the defect is corrected, it is shipped. This is because even when a micro defect of, for example, a micron order exists on a photomask such as a reticle, the circuit pattern of the photomask such as the reticle is not normally transferred to a wafer due to the defect. This is because there is a problem. This problem has become more apparent with the recent increase in the degree of integration of LSIs, and the existence of finer sub-micron-order defects or smaller sub-micron-order defects has become unacceptable. In recent years, in order to improve the transfer resolution of a circuit pattern on a reticle formed of a metal thin film of chrome or the like, a transparent or translucent thin film called a phase shift film or a phase shifter between circuit patterns on the reticle (generally an exposure light source). (Having a film thickness having an optical path length which is an odd multiple of 1/2 of the wavelength) is used (a phase shift reticle).

[0003] In the process of forming a circuit pattern such as a photomask such as a reticle, it is common that several defects are generated. Therefore, detection and correction of defects are always performed.

[0004] In the case of a light-shielding film made of a metal thin film, a technique has been established for correcting a lack defect by depositing a light-shielding material by FIB (converged ion beam) by CVD or the like.

Further, in the case of a surplus defect, an energy beam such as a laser or FIB is more simply irradiated, and a large energy is used by utilizing the difference in material (metal) between quartz as a base material and the defect. Techniques for irradiating and evaporating and removing metal (heat-sensitive) defect portions have been established for some time.

[0006] Gas assist is also used for a defect of a phase shifter on a phase shift reticle, for a defect that is completely lacking, or for a defect that has a surplus shape similar to a circuit pattern. A modified technique using the FIB has been devised (for example, Japanese Patent Application Laid-Open No. 4-288542).

Japanese Patent Application Laid-Open Nos. Hei 6-174550 and Hei 6-13065 disclose techniques for inspecting the degree of formation of a phase shifter film in a somewhat large area (about several microns).
3, JP-A-7-128842, JP-A-6-3313
No. 21 discloses a technique for measuring the thickness of a phase shifter using an optical interference technique.

[0008]

As a feature of the defect generated in the shifter pattern on the phase shift reticle, there is a feature that the material is the same as that of the base material. For this reason, it is not possible to perform correction using an energy beam utilizing a difference in material.

Further, even when the FIB gas-assisted correction is performed on a surplus defect, it is required to remove only the surplus defect without cutting the substrate portion. For this purpose, it is necessary to control the correction amount in accordance with the shape (thickness) of the surplus defect. Generally, when forming a phase shifter, a part of the surplus defect remains in a portion where the phase shifter is not originally formed due to a process defect, so that the shape (thickness) is not appropriate. It has a fixed shape, and its material is often the same as the phase shifter and the base material.

The FIB can be used not only as a processing means but also as a high-resolution observation means by observing secondary electrons generated from a sample. Therefore, in order to correct a defect composed of a substance having a different generation of secondary electrons, the correction can be performed with high accuracy by performing the correction while observing the FIB itself. However, the defect of the phase shifter cannot be observed by FIB because it is basically the same material as the base material. For this reason, the shape of the defect is measured by other means at a high resolution (specifically, submicron or less in the plane direction and several tens of nanometers or less in the height direction), and the correction amount based on the measurement result is It is necessary to perform control.

An object of the present invention is to solve the above-mentioned problems of the prior art and to provide a technique for correcting a defect generated on a substrate having a circuit pattern such as a phase shift reticle with high accuracy.

[0012]

In order to achieve the above object, the present invention provides a method for measuring the shape of an optically transparent defect on an optically transparent substrate. Irradiating the light and moving the substrate and the illuminating light relative to each other, detecting reflected light of the illuminating light reflected from the substrate or transmitted light transmitted through the substrate, and detecting heterodyne interference of the detected reflected light or transmitted light; The shape of the optically transparent defect on the substrate is measured based on the interference.

This method is characterized in that illumination light having different wavelengths is light of two different wavelengths, and the size of the irradiation area of the light of two different wavelengths on the substrate is different.

According to the present invention, there is provided a method for measuring the shape of an optically transparent defect on an optically transparent substrate, comprising:
The substrate is irradiated with illumination light of a predetermined wavelength, and while relatively moving the substrate and the illumination light to be irradiated, reflected light of the illumination light reflected by the substrate or transmitted light transmitted through the substrate is detected, and the detected reflection is detected. The shape of the optically transparent defect on the substrate is measured based on a detection signal corresponding to the level of light or transmitted light.

Further, according to the present invention, there is provided an apparatus for measuring the shape of an optically transparent defect on an optically transparent substrate, comprising: mounting means capable of mounting the substrate and moving in an XY plane; Irradiating means for irradiating the sample placed on the mounting means with illumination light having different wavelengths; detecting means for detecting heterodyne interference of reflected light of the illumination light irradiated by the irradiation means and reflected by the substrate; The position of the optically transparent defect on the substrate based on the detection signal of the reflected light of the illumination light reflected from the substrate detected by the detection means while moving the substrate with respect to the illumination light irradiated on the substrate by the placement means And a defect measuring means for measuring the temperature.

According to the present invention, there is provided an apparatus for measuring the shape of an optically transparent defect on an optically transparent substrate, comprising: mounting means capable of mounting the substrate and moving in an XY plane; Irradiating means for irradiating the sample placed on the mounting means with illumination light having different wavelengths; detecting means for detecting heterodyne interference of transmitted light of the illuminating light irradiated by the irradiating means and transmitted through the substrate; The shape of an optically transparent defect on the substrate is measured based on a detection signal of transmitted light transmitted through the substrate detected by the detection device while moving the substrate with respect to the illumination light applied to the substrate by the mounting device. And a defect measuring means.

An apparatus for measuring the shape of the defect is characterized in that the irradiating means irradiates the sample with light of two wavelengths having different sizes of the irradiation area on the sample.

An apparatus for measuring the shape of the defect is as follows.
The detection means has an optical path length correction unit for correcting the optical path length of the transmitted light according to the thickness of the substrate.

According to the present invention, there is provided an apparatus for measuring the shape of an optically transparent defect on an optically transparent substrate, comprising: mounting means capable of mounting the substrate and moving in an XY plane; Irradiation means for irradiating the sample placed on the mounting means with illumination light,
Detecting means for detecting reflected light of illumination light irradiated by the irradiation means and reflected by the substrate; and a substrate detected by the detection means while moving the substrate with respect to the illumination light irradiated to the substrate by the mounting means. And a defect measuring means for measuring a shape of an optically transparent defect on the substrate based on a detection signal obtained by detecting a light-dark level of reflected light of the substrate.

According to the present invention, there is provided an apparatus for measuring the shape of an optically transparent defect on an optically transparent substrate, comprising: mounting means capable of mounting the substrate and moving in an XY plane; Irradiation means for irradiating the sample placed on the mounting means with illumination light,
Detecting means for detecting the transmitted light of the illuminating light emitted by the irradiating means and transmitted through the substrate; and a substrate detected by the detecting means while moving the substrate with respect to the illuminating light applied to the substrate by the mounting means A defect measuring means for measuring a shape of an optically transparent defect on the substrate based on a detection signal detecting a light-dark level of the transmitted light.

Further, according to the present invention, an optically transparent defect on an optically transparent substrate is detected at a first resolution, and the shape of the detected defect on the substrate is detected at a second resolution higher than the first resolution.
And a defect is corrected based on the information on the position and shape of the defect on the substrate measured at the second resolution.

Further, according to the present invention, an optically transparent defect on an optically transparent substrate is detected at a first resolution, and the shape of the detected defect on the substrate is determined by a second resolution higher than the first resolution.
The resolution of the defect is measured, and the material of the detected defect is analyzed.
The defect correction method is characterized in that the defect is corrected on the basis of the information on the position and shape of the defect on the substrate measured at a resolution of and the information on the material of the analyzed defect.

The defect correction method is characterized in that the shape of the defect is measured based on heterodyne interference of light reflected on the substrate by illuminating the substrate with illumination light.

Further, the defect correcting method is characterized in that the shape of the defect is measured based on heterodyne interference of transmitted light transmitted through the substrate by illuminating the substrate with illumination light.

Further, the defect repair method is characterized in that the defect is repaired by using a focused ion beam.

Further, the defect repair method is characterized in that the material of the defect is analyzed by mass analysis of the defect.

According to the present invention, there is provided a defect detecting means for detecting a defect on a sample, a defect shape measuring means for measuring the shape of the defect on the sample detected by the defect detecting means at a higher resolution than the defect detecting means, And a defect correcting unit for correcting a defect based on information from the defect shape measuring unit.

Further, according to the present invention, a defect detecting means for detecting a defect on a sample, a defect shape measuring means for measuring the shape of the defect on the sample detected by the defect detecting means at a higher resolution than the defect detecting means, A defect comprising analyzing means for analyzing the material of the defect based on information from the defect detecting means, and defect correcting means for correcting the defect based on information from the defect shape measuring means and the analyzing means; It was a correction device.

In this defect correcting apparatus, the sample is an optically transparent substrate, and the defect is the same optically transparent material as the substrate. The defect shape measuring means includes an illumination unit for illuminating the sample. A reflected light detector for detecting light reflected from the substrate and light reflected from the defect by the illumination of the illumination unit; and detecting an optically transparent defect on the substrate based on the reflected light detected by the reflected light detector. It is characterized by measuring the shape.

In this defect correcting apparatus, the sample is an optically transparent substrate, and the defect is the same optically transparent material as the substrate, and the defect shape detecting means is an optically transparent substrate using heterodyne interference. It is characterized in that the shape of a transparent defect is measured.

In this defect correction apparatus, the sample is an optically transparent substrate, and the defect is an optically transparent material the same as the substrate. The defect shape detecting means irradiates the substrate with illumination light. And detecting the shape of an optically transparent defect by detecting light transmitted through the substrate and light reflected by the substrate.

Further, the defect repair apparatus is characterized in that the defect repair means is a repair means using a focused ion beam.

Further, the defect correcting apparatus is characterized in that the analyzing means includes a mass spectrometer.

[0034]

FIG. 1 shows an embodiment of the present invention.
The single-wavelength light output from the linearly polarized laser oscillator 13 is converted by the frequency shifter 12 into two light beams having polarization planes orthogonal to each other and having slightly different light frequencies. The delicate frequency difference is often selected from several tens of kHz to several tens of MHz, but the content of the present invention does not change depending on the frequency. One light is projected on the sample as a reference light and the other light is projected as a measurement light.
In this case, which light emitted from the frequency shifter is used as the reference light is arbitrary.

With respect to the oscillation wavelength of the laser oscillator 13,
It is carefully selected because it greatly affects the spatial resolution when measuring the amount of phase difference of a defect according to the present invention. In general, in the heterodyne interference system, a wavelength of 633 nm of a He-Ne laser is selected from the ease of handling and the spread of peripheral optical elements. However, in the defect phase difference amount measuring apparatus (or defect shape measuring apparatus) according to the present invention, the spatial phase difference distribution of the defect (here, the spatial distribution means a phase difference distribution on a substrate plane in a photomask such as a reticle). Is selected from the viewpoint of ensuring the spatial resolution determined by the wavelength of the light source and the NA of the detection objective lens at the resolution required for the purpose.

Specifically, for a process-level LSI represented by elements such as 256M to 1GDRAM,
It is necessary to correct at a resolution of 0.25 μm or less. Light source wavelength λ, object side NA of detection objective lens optical system and resolution D
, A relation of D = 0.5λ / NA is approximately established. When applied to this, λ when the optical system with NA = 0.5 is 250 nm or less, and λ when the NA = 0.75 optical system is 375 nm or less. As a laser light source that performs practical continuous oscillation at a wavelength in this range,
There is an Ar ion laser. 351 with Ar ion laser
nm and those having an oscillation wavelength in the range of 364 to 351 nm are commercially available. Further, in the case of a laser in which an Ar ion laser and a double harmonic crystal are combined, 229 nm to 264 nm is used.
Those having several oscillation wavelengths down to nm are commercially available. Further, a wavelength of 266 nm is obtained with the fourth harmonic of the Nd: YAG laser.

When a light source having a wavelength of 250 nm is used to obtain a resolution of 0.15 μm, an NA of about 0.8 is required.

The light emitted from the frequency shifter is split by the half mirror 5 into reference light 15 and probe light 16. The reference light passes through the analyzer 17 as it is, and
At 8, interference occurs on the reference light detector 14, and the interference beat waveform is converted into an electric signal waveform. On the other hand, the probe light is focused on the sample by the objective lens 3.

At this time, on the sample, as shown in FIG.
The reference spot light 202 and the measurement spot light 203 based on the reference light divided by the frequency shifter 12 are collected with different spot diameters. It is desirable that the measurement spot light 203 be focused to the diffraction limit of the optical system in order to increase the spatial resolution. In addition, since the reference spot light 202 determines a reference height, a spot diameter having a sufficiently large size is required. In this case, the reference height is determined as an average of the heights of the planes included in the spot. For this reason, the light-shielding portion and the transmitting portion may be mixed in the spot.

Further, when the measurement spot light 203 is arranged at the center of the reference spot light 202, the measurement has no directivity, and the measurement can be performed regardless of the direction of the pattern. This is an advantageous feature for a type of measurement in which the entire surface of the sample is automatically and continuously scanned.

The role of the birefringent lens 4 is to collect the same light by the same optical system and to form spots of different sizes. The birefringent lens 4 has different focal lengths for light of different orthogonal polarization planes, taking advantage of the characteristic that the refractive index is different between one polarization plane, which is a characteristic of a birefringent crystal, and the polarization plane orthogonal thereto. Operates as a lens. In the drawings in the present invention, for clarity of explanation, one is shown as a lens having a finite focal length, and the other is shown as having an infinite focal length.

Since the birefringent lens has a special function as described above, the degree of freedom is not as high as that of a general optical element, and the design of the apparatus may be restricted in some cases. FIG. 11B shows a device in which the function of the birefringent lens 4 shown in FIG. 11A is substituted by a general optical element. The incident light is
The light is split by the polarization beam splitter 1101 according to the polarization plane, one of which is passed through a lens 1103, and the other is passed through a lens having a different focal length (not shown in the drawing).
The beam is again combined into one beam by the polarizing beam splitter 1104 from the mirror 1102.

Next, the reference light and the measurement light reflected on the sample are:
The analyzer 6 allows only the interfering polarized light components to pass therethrough, is condensed on the pinhole of the pinhole portion 8 by the imaging lens 7, and the interference output is converted into an electric signal waveform by the probe light detector 9. Is converted to For the above purpose, the analyzer 6 moves the polarization planes of the reference light and the measurement light orthogonal to each other,
It is installed in such a direction that light of a polarization plane having a 45-degree inclination passes through. This is the same for the analyzer 17.

Now, the output 302 of the reference light detector 14 and the output 301 of the probe light detector 9 are both beat signals of two lights shifted by the frequency shifter 12, and become sine waves of the same frequency. However, the signals have different phases as shown in FIG.

This phase difference ΔφB is given by ΔφB = 4πnD / λ
Becomes Here, n is the refractive index, λ is the wavelength of the measuring light, and D is the defect height from the reference plane. In the case of the reflection type measurement shown in FIG. 1, n is the refractive index (n 雰 囲 気 1) of the atmosphere in which the sample is placed, generally, air. In addition, in this measurement method, a probe light detector is arranged on a transparent substrate such as a photomask such as a reticle as shown in FIG. 8 on the opposite side of the sample from the illumination light source side (in the case of FIG. 8). The light is illuminated from the front side of the sample substrate, and the probe light detector 9 is disposed on the back side.) In this case, n is the refractive index of the defect material.

In the transmitted light type measurement, since the probe light is shielded at the light-shielded portion, the phase of the reference light is determined only by the light-transmitted portion, so that the measurement is stabilized without being affected by the light-shielded portions having different heights. There are advantages. Since the phase defect is originally a defect existing only in the light transmitting portion, there is no problem in eliminating the data in the light shielding portion.

The result so far is the height of only one minute area of the defect, but the XY stage or the XY stage
By scanning the YZ stage in the XY directions, the shape of the entire defect (which may be expressed as a height or a phase difference with respect to a reference plane) can be measured. Of course, this scanning fixes the sample and the probe light 16
Alternatively, a method of scanning the entire optical system including the probe light 16 or a combination thereof may be used. Although measurement of one point of measurement spot light data has been described so far, in the present invention, measurement light is condensed on a sample as slit-like measurement slit light 401 as shown in FIG. Is also good. In this case, if the output of the entire slit is detected, the average height of the slit will be measured. However, as shown in FIG. By forming an image on the sensor 501, height information in the slit can be decomposed and measured for each pixel of the linear sensor. Therefore, the linear height distribution can be measured at a time, and there is an advantage that the scanning direction can be limited to, for example, only the Y direction when measuring the height distribution of the entire defect. In this case, the birefringent lens has a cylindrical property.

Further, the measuring light is increased as shown in FIG.
For example, a wide-field measurement spot 701 covering the entire defect is provided.
If the detector is a two-dimensional area sensor and the image 702 of the two-dimensional area sensor projected on the sample is as shown in FIG.
The height of the entire defect can be measured without scanning.

In FIG. 5, it is assumed that a single sensor having a single output such as a CCD is used as a linear sensor. However, in the detection using heterodyne interference as in the present invention, it is necessary to know the phase difference between signals. The detection signal must be sampled as a waveform, which requires that the waveform be measured multiple times. Therefore, it is necessary to perform high-speed sampling. However, depending on the speed, a parallel output linear sensor 601 which can take out data of each pixel in parallel as shown in FIG. (For example, a photodiode array) may be used.

As described above, measurement using heterodyne interference has a feature of high accuracy, but it is necessary to perform detection using a waveform. Required height resolution is λ / 500-
The heterodyne interference method is preferable when the value exceeds λ / 1000. However, when the resolution is λ / 100 to λ / 500 or less, it is possible to cope without using heterodyne interference. (Defect correction accuracy of phase shift reticle is ± 3 degrees ~
Accuracy of 5 degrees is required, which is the accuracy unless the heterodyne is used. FIG. 32 shows a configuration that does not use heterodyne interference. The configuration is simpler than that of FIG. 1 except for a part related to frequency shifter and reference light detection. The detection signal in this case is detected as a light / dark level, and a light output is obtained at the same height as the reference plane, and a dark output is obtained at a height having a phase difference of 180 degrees from the reference plane. In the case of more than that, the operation is repeated, and the relation between the output and the height does not become one-to-one. It is considered that the phase difference defect generated in the phase shift reticle has the maximum thickness of the phase shifter. The reason is that the etching residue that is not properly etched when forming a normal phase shifter occupies most of the phase difference defect.

The thickness of a normal phase shifter is λ /
Since the thickness gives a phase difference of 2, it can be considered that the height of the phase difference defect does not exceed λ / 2 when measured by the transmitted light method. The thickness d is given by d = λ / 2 (n−1), where n is the refractive index of the phase shifter, 1 is the refractive index of air, and λ is the exposure wavelength. When the height of the phase shifter is measured by a reflection method, the optical path difference is 2d, and 2d = λ / (n−1). Since n is a value near 1.5, if it is set to 1.5, the optical path difference will be 2d = 2λ. That is, in the reflection type, measurement can be performed only up to 1/4 the thickness of the transmission type without special measures.

Therefore, the transmission type measurement is more advantageous in view of the measurement range. However, considering that the size of the defect is minute, it is difficult to imagine that the thickness is as large as a normal phase shifter, and it is considered that the reflection type can be used. Conversely, it can be said that when measurement is performed by the reflection method, measurement can be performed with four times the sensitivity (resolution) as compared with the transmission method. This point should be determined from the balance between the resolution and the measurement range. If both of them are desired, the two measurement methods may be combined.

The difficulty in the configuration of the transmitted light type is that the objective lens (for example, the lens 80 in FIG.
3). This lens must focus on the substrate surface from the rear surface of the substrate, so that the focal position becomes farther by the thickness of the substrate, and a large working distance is required. Therefore, it is necessary to design a lens in consideration of the thickness of the substrate, and it is difficult to obtain a high-resolution lens.

As shown in FIG. 15, a high N
This problem can be avoided by arranging a lens 1501 of high resolution A and thus a high resolution lens 1501 and arranging a lens 1502 of low NA on the back side, which is easy to design or has little influence on light passing through the substrate. it can. Specifically, in the configuration in which illumination is performed from the front side as shown in FIG. 8, the objective lens 3 on the front side can have a high resolution, and if a minute measurement spot can be formed on the sample, the detection side can emit light from the spot. It is sufficient that the lens has a resolution sufficient to collect the light, and the resolution does not necessarily need to be high. However, if the detector is a linear sensor or area sensor, it is necessary to accurately form an image on the sample for each pixel of the sensor. In this case, a high-resolution objective lens is provided on the sensor side. Required. On the other hand, since the spot of the measurement light is large, the lens on the illumination side has a low resolution and is good.

Therefore, it is desirable to employ a configuration in which illumination is performed from the back side as shown in FIG. FIG. 14 assumes a CCD as a linear sensor. Therefore, a waveform memory 1401 is added to detect the detection signal as a waveform on the time axis.

In any case, when the present invention is applied to a photomask such as a reticle or the like, a plurality of layers having a thickness of 2.3 mm, 6.3 mm, 9.0 m are used at the same time.
m), which makes lens design difficult. This is because it is not possible to design a focal point to be formed with good resolution over a substrate having a plurality of thicknesses. Therefore, a lens is designed according to the thickest substrate (for example, 9.0 mm), and when measuring a substrate having another thickness, the optical path difference correction plate 121 is inserted between the objective lens 1403 and the substrate. I do.

The material and thickness of the optical path length correction plate 121 shown in FIG. 12 are selected so that the optical path lengths of the substrate and the optical path length correction plate are equal for each substrate thickness. For example, the substrate is the same as the sample substrate (for example, synthetic quartz) or the substrate having the same refractive index, and has the largest thickness (for example, 9.0 m).
m) and the difference between the other substrates (for example, 2.3 mm and 6.3 mm) (accordingly, 6.7 mm and 2.7 mm) are defined as the thickness. At this time,
Excluding the optical path length correction plate when measuring a 9.0 mm thick substrate, a 2.7 mm correction plate is inserted when measuring a 6.3 mm thick substrate, and 6.7 when measuring a 2.3 mm thick substrate.
Insert an optical path length correction plate of mm.

However, as shown in FIG. 13, the objective lenses 1301 and 13 designed so as to correspond to the substrates of different thicknesses, respectively.
02 and 1303 may be exchanged according to the thickness of the substrate.

FIG. 16 shows an example in which an axicon lens 1601 is used as a lens which is not affected at all by the thickness of the substrate. This is the 1602 part of the lens and 16
This lens forms a beam narrowed to the diffraction limit over a long distance by forming light fringes in the space from light emitted from the 03 portion.

The problem caused by the difference in substrate thickness is not limited to the problem of measuring the height of a defect, but is common to, for example, the inspection of a circuit pattern shape of a photomask such as a reticle. Therefore, these inventions are common to all techniques for forming an image with light transmitted through an image on a transparent substrate or condensing the light.

Next, variations in the relationship between the position and the shape of the reference light and the measurement light on the sample will be described.

Although the relationship shown in FIGS. 2, 4 and 7 enables detection without directionality, a special optical system such as a birefringent lens is required to realize different spot shapes. If a certain degree of human intervention is allowed in the defect shape measurement work, or if the reference / measurement spots are arranged by recognizing the directionality of the circuit pattern for each measurement defect using image processing technology, a directional spot The arrangement is also possible.

FIG. 17 shows a reference light spot 22 and a measurement light spot 23 both having the same spot diameter. The two spots are separated from each other, the reference spot is projected on a reference plane or a normal portion without defects, and the measurement spot is projected on a defective portion, and each has height information of the projected position. In this case, since the reference spot is a minute point,
Although the stability is deteriorated and the directionality of measurement comes out, a special optical system such as a birefringent lens is not required if the same spot is used as described above. In order to form such a minute spot, a birefringent prism 180 as shown in FIG.
Generally, 1 is used.

FIG. 19 shows an illumination shape when a linear sensor is used as a detector. Reference slit light 1901
Both the measurement slit light 1902 and the measurement slit light 1902 have the same slit shape. In addition to the advantage of speeding up the measurement, the average value of a wider range than the case where the reference slit light is the spot light is used as the reference plane, so that the measurement is stabilized.

From the viewpoint of stabilization, reference area light 2001 as shown in FIG. 20 is better than slit light. Of course, the stability is the same even when the measurement side is the measurement spot light 2101 as shown in FIG. However, in any case, since the reference light and the measurement light have different beam diameters, a device such as a birefringent lens is required.

FIG. 22 shows the reference area light 2001 and the measurement area light 220 in consideration of stability and easy formation of a spot.
One combination is shown. In this case, it is desirable to use a linear sensor or an area sensor for the detector.

In FIG. 23, the reference area light 2301 and the measurement area light 2302 are set larger for more stable detection.
It is not preferable for accurate measurement. In contrast, FIG.
As shown in (2), the beam shape may be prevented from reaching the light-shielding portion by the field stop, or may be detected by transmitted illumination from the back side.

Next, a procedure for correcting a defect using the above-described defect shape measurement technique will be described. As shown in FIG. 9, first, a defect is detected by a defect inspection apparatus or a foreign substance inspection apparatus 901. The defect shape is measured by a defect shape measuring device 902 based on the detection result (for example, coordinate data, matching with stage coordinates) 903. Here, the reason why the defect detection and the shape measurement are separated into different devices is mainly due to the difference in the characteristics of each device. There is a demand for a defect inspection apparatus or a foreign substance inspection apparatus to perform inspection at a resolution as low as possible to detect the presence of a defect and at a higher speed. On the other hand, the defect shape measuring device is required to measure the shape data required for the defect correcting operation at a higher resolution. Of course, as long as both specifications can be satisfied, there is no problem in providing one device with both functions.

The obtained defect shape data 904 is sent to a defect correction device 905, where the defect is corrected. Since the shape data at this point is data measured with a very high resolution, the shape cannot be reproduced only by matching the coordinates on the stage. Therefore, alignment is required. However, in the FIB device assumed as the repair device 905, the difference in the material is visualized, so that if the normal portion and the defective portion are made of the same material, the defect cannot be visualized.
Therefore, an alignment mark is provided for positioning the two devices.

In FIG. 9, data 903 such as defect detection coordinates from a defect inspection apparatus or a foreign substance inspection apparatus 901 are shown.
And defect shape data 9 from the defect shape measuring device 902
04 is directly transferred to the next device, that is, the defect shape measuring device 902 and the defect correcting device 905, respectively. The data transferred to the defect correcting device 905 or processed by the server may be sent to the defect shape measuring device 902 and the defect correcting device 905, respectively.

FIG. 10 shows an example of a specific defect positioning mark (alignment mark) 1001 in a photomask such as a phase shift reticle. In FIG. 10, an extremely fine (for example, 0.1 μm) which has no influence on the exposure transfer onto the light-shielding portion.
Were made in two or three places to expose a different material (synthetic quartz) to the light-shielding portion material (metal thin film). Since this hole has a size smaller than the resolution by light, it is not transferred at the time of exposure, but the FIB can be observed because the resolution can be higher than that of light. The holes are made by evaporating the metal thin film with a laser focused on an extremely fine spot. Further, it may be performed by a charged particle beam apparatus such as FIB.

Alternatively, a material different from the metal thin film may be deposited on the light shielding portion by laser CVD or FIB. Larger alignment marks (eg, 0.5 μm or larger) since there is no transfer potential
Can be made.

FIG. 31 illustrates a method of processing an alignment mark by laser processing. In the figure, 1 is a sample to be corrected, 3140 is a gas supply system, and 3141 is a gas supply source. 3142 and 3144 are valves. 3143 is a gas flow control unit. 3145 is a nozzle for supplying gas to the sample surface. Reference numeral 3146 denotes a supplied gas. 3102 is a vacuum chamber. Reference numeral 3103 denotes a vacuum exhaust port for evacuating the vacuum chamber 3102. 3101 is a laser source. 310
Reference numeral 4 denotes a lens for condensing a laser beam. Reference numeral 3105 denotes a stage on which a sample can be placed and movable in the X and Y directions. 3106
Is a laser beam. 3107 is a mirror for changing the direction of the laser beam. Reference numeral 3108 denotes a window for introducing a laser beam into the chamber.

There are several types of laser processing. First
Is a removal process using a laser. Processing is performed by removing the material at the position irradiated with the laser beam. In this case, the gas supply system 3140 is unnecessary. The second is laser CVD processing. In this case, the supplied gas is Mo
(CO) 6, W (CO) 6, or Pt (CO) 6, Au
Using (CO) 6, TEOS, or the like as a material gas, CVD is performed at a position irradiated with a laser beam, and a material is deposited to form a convex alignment mark.

The position for forming the alignment mark can be controlled by controlling the position of the stage. The size of the processed alignment mark can be controlled by the processing time or the gas flow rate. Also, by moving the stage during processing, it is possible to control the planar shape of the alignment mark.

Next, the function of the correction device will be described. FIG.
8 illustrates a processing method using a charged particle beam such as a focused ion beam (FIB) or an electron beam (EB). In the figure, 1 is a substrate sample having a surface defect. Reference numeral 2830 denotes a gas supply system, which is used when a defect is dented as compared with a normal portion and is deposited to make the height equal to the normal portion. 2831 is a gas supply source. 2832 and 2834 are valves. 2833
Is gas flow control means. 2835 is a nozzle for supplying gas to the surface of the sample. 2836 is a supplied gas. 281
1 is a vacuum chamber. Reference numeral 2812 denotes a vacuum exhaust port for evacuating the vacuum chamber 2801. 2813 is a charged particle beam source. An electrostatic lens 2814 focuses the charged particle beam. Reference numeral 2815 denotes a stage on which a sample can be placed and movable in the X and Y directions. Reference numeral 2816 denotes a charged particle beam.

The alignment mark detection unit 2
801, a defect coordinate positioning unit 2802, and a charged particle beam control unit 2803.

There are several types of processing using a charged particle beam. The first is a sputtering process using a charged particle beam. In this case, the processing is performed by removing the surface material by sputtering, and a simulated defect having a concave shape or a simulated defect having a convex shape is formed by removing a part of the simulated defect. In this case, the gas supply system 2830 is unnecessary.

The second is gas assisted etching using a charged particle beam. In this case, chlorine gas (Cl2), xenon difluoride (XeF2) and iodine gas (I
2) Using bromine gas (Br2) or water vapor (H2O) as an assist gas, gas-assisted etching is performed to remove the material on the surface, and processing is performed to leave a concave-shaped simulated defect or a part. By removing the periphery, a simulated defect having a convex shape is formed.

The third is CVD using a charged particle beam. In this case, the supplied gas is pyrene gas, W (CO)
6, or TEOS + 02 mixed gas, or TEO
Using a mixed gas of S + O3 or the like as a material gas, CVD is performed at a position irradiated with a charged particle beam, and a material is deposited to form a simulated defect having a convex shape. The position at which the simulated defect is formed becomes possible by controlling the position of the stage. The size of the processed simulated defect depends on the processing time,
Alternatively, it can be controlled by the gas flow rate. Also,
By moving the stage during processing, it is also possible to control the planar shape of the simulated defect.

In FIG. 25, an apparatus for analyzing the material constituting the defect is added to the process shown in FIG. 9, and after the defect is detected, the detected defect is analyzed. In the removal and repair of the surplus defect by the FIB, the repair amount is d (micron), and the charge amount of the charged particle beam (the product of the beam current and the processing time) is C
(Coulomb), the processing area S (square micron), and the processing speed coefficient obtained from the material constituting the defect is α (cubic micron / coulomb), which is roughly represented by d = α · C / S. For example, α in SiO 2 is 0.25. Since this value varies depending on the material of the defect, it is necessary to analyze the defect material in order to deal with defects of various materials.

The defects are analyzed, and the processing conditions are determined from the prepared material and the table of the processing speed count. This analysis result is not only for correction but also for preventing defects from occurring by feeding back to the process according to the material of the generated defect. The feedback includes, for example, cleaning of the resist coating machine if the resist remains, and replacement of the cleaning chemical if the cleaning agent remains.

This analysis is performed as shown in FIG.
The measurement may be performed after measuring the shape of the defect.

For analysis, a mass spectrometer or the like is used. If SIMS (secondary electron mass spectrometer) technology is used, the probe beam can be FIB, and FIG.
As shown in (1), it is possible to configure a device that serves as both an analyzer and a correction device. SIMS used at this time includes:
A time-of-flight mass analyzer, a magnetic field mass analyzer, a quadrupole mass analyzer, or the like is used.

[0085]

According to the present invention, a defect generated in a phase shift reticle or the like is measured and analyzed for a three-dimensional shape distribution by a high-resolution optical measuring device, and the defect is corrected with high accuracy according to the material and shape. In the prior art, it is possible to accurately correct a defect having an irregular shape, which has been difficult to correct.

[Brief description of the drawings]

FIG. 1 is a schematic front sectional view showing an overall schematic configuration of an embodiment of a defect shape measuring apparatus according to the present invention.

FIG. 2 is a plan view of a substrate showing a state of illumination light for detection according to the present invention.

FIG. 3 is a diagram showing a waveform of a detection signal according to the present invention.

FIG. 4 is a plan view of a substrate showing a state of illumination light for detection according to the present invention.

FIG. 5 is a schematic front sectional view showing a detector portion according to an embodiment of the present invention.

FIG. 6 is a schematic front sectional view showing a detector portion according to an embodiment of the present invention.

FIG. 7 is a plan view of a substrate showing a state of illumination light for detection according to the present invention.

FIG. 8 is a schematic front sectional view showing an overall schematic configuration of another embodiment of the defect shape measuring apparatus according to the present invention.

FIG. 9 is a block diagram showing a situation of an embodiment of a correction process according to the present invention.

FIG. 10 is a plan view of a substrate showing a state of a positioning mark according to the present invention.

FIG. 11 is a schematic side sectional view showing one element part of the optical system according to the present invention.

FIG. 12 is a schematic front sectional view showing one element part of the optical system according to the present invention.

FIG. 13 is a schematic front sectional view showing one element part of the optical system according to the present invention.

FIG. 14 is a front sectional view showing the overall schematic configuration of another embodiment of the defect shape measuring apparatus according to the present invention.

FIG. 15 is a schematic front sectional view showing one element part of the optical system according to the present invention.

FIG. 16 is a schematic front sectional view showing one element part of the optical system according to the present invention.

FIG. 17 is a plan view of a substrate showing a state of illumination light for detection according to the present invention.

FIG. 18 is a schematic front sectional view showing one element part of an optical system according to the present invention.

FIG. 19 is a plan view of a substrate showing a state of illumination light for detection according to the present invention.

FIG. 20 is a plan view of a substrate showing a state of illumination light for detection according to the present invention.

FIG. 21 is a plan view of a substrate showing a state of illumination light for detection according to the present invention.

FIG. 22 is a plan view of a substrate showing a state of illumination light for detection according to the present invention.

FIG. 23 is a plan view of a substrate showing a state of illumination light for detection according to the present invention.

FIG. 24 is a plan view of a substrate showing a state of illumination light for detection according to the present invention.

FIG. 25 is a block diagram showing a situation of an embodiment of a correction process according to the present invention.

FIG. 26 is a block diagram showing a situation of an embodiment of a correction step according to the present invention.

FIG. 27 is a block diagram showing a situation of an embodiment of a correction process according to the present invention.

FIG. 28 is a front sectional view showing a schematic configuration of an embodiment of a correction device according to the present invention.

FIG. 29 is a front sectional view showing a schematic configuration of an embodiment of a correction device according to the present invention.

FIG. 30 is a block diagram showing a situation of an embodiment of a correction process according to the present invention.

FIG. 31 is a front sectional view showing a schematic configuration of an embodiment of a correction device according to the present invention.

FIG. 32 is a front sectional view showing a schematic configuration of an embodiment of a defect shape measuring apparatus according to the present invention.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Phase shift reticle, 2 ... Phase defect, 3 ... Objective lens, 4 ... Birefringent lens optical system, 5 ... Half mirror, 6 ...
Analyzer 7 Image forming lens 8 Pinhole 9 Detector for measuring light 10 Phase difference detecting unit 11 Beam expander 12 Frequency shifter 13 Laser oscillator 14 Reference light Detector, 15: Reference light, 16: Probe light, 17: Analyzer, 18: Condensing lens, 19: Output of phase difference amount, 201: Circuit pattern, 202: Reference spot light, 203: Measurement spot light, 301 ... Detector output for measurement light, 302: Detector output for reference light, 401: Measurement slit light, 501: Linear sensor, 601: Parallel output type linear sensor, 701: Wide-field measurement spot, 702 ...
The image of the two-dimensional area sensor projected on the sample, 803 ...
Substrate back side objective lens, 901: Defect inspection device or foreign matter inspection device, 902: Defect shape measurement device, 903: Data such as defect detection coordinates, 904: Defect shape data, 90
5 Defect repair device, 1001 Defect positioning mark, 1
101: polarizing beam splitter, 1102: mirror, 1
103: lens, 1104: polarizing beam splitter, 1
301 ... 4.6 mm thick substrate lens, 1302 ... 6.3 mm thick substrate lens, 1303 ... 9.0 mm thick substrate lens, 1
401: waveform memory, 1501: high NA illumination lens, 1
Reference numeral 502: low NA detection lens, 1503: minute illumination spot, 1601: axicon lens, 1801: birefringent prism, 1901: reference slit light, 1902: measurement slit light, 2001: reference area light, 2101: measurement spot light, 2201 ... Measurement area light, 2301 ... Reference area light, 2302 ... Measurement area light, 2401 ... Reference area light, 2402 ... Reference area light, 2501 ... Defect material analysis device, 2701 ... Defect repair device with defect material analysis function, 2801 ... Alignment Mark detection unit, 28
02: Defect coordinate positioning unit, 2803: Ion beam control unit, 2901: SIMS, 2902: Ion beam control parameter calculation unit, 3001: Defect shape measuring device with defect detection function, 3101: Laser beam control unit

──────────────────────────────────────────────────の Continued on the front page (51) Int.Cl. ⁶ Identification code FI G01N 21/88 G01N 21/88 DE H01L 21/66 H01L 21/66 J (72) Inventor Junzo Higashi Totsuka-ku, Yokohama-shi, Kanagawa 292 Yoshidacho Inside Hitachi, Ltd. Production Technology Laboratory

Claims (22)

[Claims] 1. A method for measuring the shape of an optically transparent defect on an optically transparent substrate, the method comprising: irradiating the substrate with illumination light having different wavelengths; And detecting the reflected light of the illumination light reflected by the substrate or the transmitted light transmitted through the substrate while relatively moving the substrate, based on the detected reflected light or the heterodyne interference of the transmitted light on the substrate. A defect shape detection method, comprising measuring the shape of an optically transparent defect. 2. An illumination light having two different wavelengths, wherein said two different wavelengths of light are illuminated by different wavelengths on the substrate. 2. The defect shape detection method according to 1. 3. A method for measuring the shape of an optically transparent defect on an optically transparent substrate, the method comprising: irradiating the substrate with illumination light of a predetermined wavelength; And detecting the reflected light of the illumination light reflected by the substrate or the transmitted light transmitted through the substrate while relatively moving the
A defect shape detection method, comprising: measuring a shape of an optically transparent defect on the substrate based on a detection signal corresponding to the detected light or dark level of the reflected light or transmitted light. 4. An apparatus for measuring the shape of an optically transparent defect on an optically transparent substrate, comprising: mounting means capable of mounting the substrate and moving in an XY plane; Irradiating means for irradiating illumination light having a different wavelength to the sample mounted on the mounting means, and detecting means for detecting heterodyne interference of reflected light of the illumination light illuminated by the irradiating means and reflected by the substrate Moving the substrate with respect to the illumination light illuminated on the substrate by the placing means; and detecting the reflected light of the illumination light reflected from the substrate detected by the detection means on the substrate. And a defect measuring means for measuring the shape of the optically transparent defect. 5. An apparatus for measuring the shape of an optically transparent defect on an optically transparent substrate, comprising: mounting means for mounting said substrate and moving in an XY plane; Irradiating means for irradiating illumination light having a different wavelength to the sample mounted on the mounting means, and detecting means for detecting heterodyne interference of transmitted light of the illumination light emitted by the irradiating means and transmitted through the substrate; Optically moving the substrate based on a detection signal of transmitted light transmitted through the substrate detected by the detection means while moving the substrate with respect to the illumination light irradiated on the substrate by the mounting means. A defect shape detecting device comprising: a defect measuring means for measuring a shape of a transparent defect. 6. The defect shape detecting apparatus according to claim 4, wherein said irradiating means irradiates the sample with light of two wavelengths having different sizes of an irradiation area on the sample. . 7. The shape defect detecting apparatus according to claim 5, wherein said detecting means has an optical path length correcting section for correcting an optical path length of said transmitted light according to a thickness of said substrate. 8. An apparatus for measuring the shape of an optically transparent defect on an optically transparent substrate, comprising: mounting means for mounting said substrate and moving in an XY plane; Irradiating means for irradiating the sample placed on the placing means with illumination light, detecting means for detecting reflected light of the illumination light irradiated by the irradiating means and reflected on the substrate, and An optically transparent defect on the substrate based on a detection signal obtained by detecting a light / dark level of reflected light from the substrate detected by the detection means while moving the substrate with respect to the illumination light applied to the substrate. And a defect measuring means for measuring the shape of the defect. 9. An apparatus for measuring the shape of an optically transparent defect on an optically transparent substrate, comprising: mounting means capable of mounting the substrate and moving in an XY plane; Irradiating means for irradiating the sample placed on the placing means with illumination light, detecting means for detecting transmitted light of the illuminating light transmitted by the irradiating means and transmitted through the substrate; and While the substrate is moved with respect to the illumination light irradiated on the substrate, an optically transparent substrate on the substrate is detected based on a detection signal that detects a light / dark level of transmitted light transmitted through the substrate detected by the detection unit. A defect shape detecting device, comprising: a defect measuring means for measuring a shape of a defect. 10. An optically transparent defect on an optically transparent substrate is detected at a first resolution, and the shape of the detected defect on the substrate is detected at a second resolution higher than the first resolution. And correcting the defect based on information on the position and shape of the defect on the substrate measured at the second resolution. 11. An optically transparent defect on an optically transparent substrate is detected at a first resolution, and the shape of the detected defect on the substrate is detected at a second resolution higher than the first resolution. In the analysis, the material of the detected defect is analyzed, and the defect is determined based on the information on the position and shape of the defect on the substrate measured at the second resolution and the information on the material of the analyzed defect. A defect repair method characterized by correcting. 12. The defect repair method according to claim 11, wherein the shape of the defect is measured based on heterodyne interference of light reflected on the substrate after illuminating the substrate with illumination light. 13. The defect repair method according to claim 11, wherein the shape of the defect is measured based on heterodyne interference of transmitted light transmitted through the substrate by illuminating the substrate with illumination light. 14. The defect repair method according to claim 10, wherein the repair of the defect is performed using a focused ion beam. 15. The defect repair method according to claim 11, wherein the material of the defect is analyzed by mass analysis of the defect. 16. A defect detecting means for detecting a defect on a sample, a defect shape measuring means for measuring the shape of the defect on the sample detected by the defect detecting means at a higher resolution than the defect detecting means, and A defect correcting unit for correcting the defect based on information from a shape measuring unit. 17. A defect detecting means for detecting a defect on a sample, a defect shape measuring means for measuring the shape of the defect on the sample detected by the defect detecting means at a higher resolution than the defect detecting means, Analyzing means for analyzing the material of the defect based on information from the detecting means; and defect correcting means for correcting the defect based on information from the defect shape measuring means and the analyzing means. And a defect repair device. 18. The method according to claim 18, wherein the sample is an optically transparent substrate, the defect is the same optically transparent material as the substrate, and the defect shape measuring means comprises: A reflected light detection unit that detects light reflected from the substrate and light reflected from the defect due to illumination of the illumination unit, and an optically transparent light on the substrate based on the reflected light detected by the reflected light detection unit. 18. The defect repair apparatus according to claim 16, wherein a shape of the defect is measured. 19. The apparatus according to claim 19, wherein the sample is an optically transparent substrate, the defect is the same optically transparent material as the substrate, and the defect shape detecting means uses heterodyne interference to detect the optically transparent material. 18. The defect repairing apparatus according to claim 16, wherein a shape of the transparent defect is measured. 20. The sample according to claim 1, wherein the sample is an optically transparent substrate, the defect is the same optically transparent material as the substrate, and the defect shape detecting means irradiates the substrate with illumination light. 18. The defect repairing apparatus according to claim 16, wherein a shape of the optically transparent defect is measured by detecting light transmitted through the substrate and light reflected by the substrate. 21. The defect repair apparatus according to claim 16, wherein said defect repair means is a repair means using a focused ion beam. 22. An apparatus according to claim 17, wherein said analyzing means comprises a mass spectrometer.