JPS6322255B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a defect inspection device for detecting defects occurring in an object having regularly arranged basic patterns.

Conventionally, inspection of regularly arranged basic patterns has mainly relied on visual inspection using a microscope. This visual inspection requires constant adjustment of the microscope, which takes time and makes high-speed inspection difficult. Furthermore, since the operator must look through a microscope to look for defects in the basic pattern, there is a risk of severe eye fatigue and reduced inspection accuracy.

In recent years, instead of visual inspection, an inspection apparatus using coherent light and a spatial filter has been developed that separately displays periodic information of a substrate pattern and aperiodic information due to defects.

Conventionally, this type of device utilizes the property that Fourier transformed images of regularly arranged basic patterns (hereinafter referred to as regular patterns) are regularly distributed in a lattice pattern to create light-opaque areas distributed in a lattice pattern. There is a method that uses a spatial filter to block regular pattern periodic information light and detect non-periodic defect information from the see-through portion of the spatial filter.

However, with this type of filter, the Fourier transform image of the subject must be precisely aligned with the opaque area of the spatial filter, and the synchronization information light must be completely blocked, making positioning of the spatial filter difficult and making inspection difficult. This method has the drawback that the basic pattern to be targeted must be held extremely accurately perpendicular to the optical axis, making the operation extremely complicated.

Furthermore, since the regular pattern of periodic information light must be accurately aligned with the opaque region of the spatial filter, a highly accurate lens with little distortion on the Fourier plane must be used, resulting in an expensive lens.

Next, as a device that separates and displays periodic information of regular patterns and aperiodic information due to defects, we focused on meshes among regular patterns, and used coherent light and a ring-shaped spatial filter to display meshes. Devices have been devised to detect defects quickly and with high precision.

This device has the characteristics that the optical system is relatively simple and does not require high precision, but when the repeating period of the basic pattern is extremely small (for example, inspection of pregroups of optical disks), it is necessary to use the diffraction of the object. The light spreads out too much. Therefore, a large diameter and bright lens is required, which increases the price of the lens. Furthermore, bright lenses have drawbacks such as having a shallow depth of focus, requiring a highly accurate focusing mechanism, and requiring time for focusing, making it impossible to speed up inspections.

This invention was made to eliminate the above-mentioned drawbacks, and there is at least a gap between the n-th and n+1-th diffracted light obtained by irradiating substantially coherent light onto an object having regularly arranged basic patterns. A single lens is provided, and the light re-imaged by the lens is processed by a signal processing unit via at least one photoelectric converter, and the type of defect is detected based on the photoelectric output at different far-field coordinates of the lens. By determining the
It is an object of the present invention to provide a defect inspection device that can speed up inspection because the depth of focus becomes deep and can easily perform focusing, and can also discriminate the type of defect.

Hereinafter, one embodiment of the present invention will be described with reference to the drawings. In FIG. 1, reference numeral 1 denotes an object having a regularly arranged basic pattern, and in the illustrated example, it is a video disc having a relatively simple basic pattern, in which a large number of grooves are regularly arranged. Reference numeral 2 denotes a rotational drive device for rotating this video disc 1, and this device 2 is composed of a rotating motor 2a and a holding portion 2b that is provided on the rotating shaft of the motor 2a and holds the video disc 1 in a removable manner. . A substantially coherent laser light source 3 made of a HeNe laser device, for example, is disposed on the side of the rotary drive device 2. Laser light R emitted from this light source 3
is reflected by a half mirror 4 and irradiated onto a regular pattern on the video disc 1.

At this time, the diffracted light from the regular pattern obtained by irradiating the regular pattern of the video disc 1 with the laser beam R strengthens each other in a direction shifted by one wavelength.
Concentrates on integer order diffracted light. In addition, the diffracted light from defective parts that deviate from the regular pattern does not strengthen each other and disappears naturally, and the diffracted light generated from the defective parts that stand up, such as defects in unevenness, dust, etc., spreads over the entire surface.

From this, it can be seen that a video disk is formed between the n-th order diffraction light and the n+1-th order diffraction light, for example, the 0th-order diffraction light 5a and the first-order diffraction light 5b, in the diffraction pattern obtained by irradiating the laser beam R in a regular pattern. 1
An objective lens 7 is provided to receive diffracted light 6 generated by the defective portion. Diffracted light 6 received by objective lens 7 is reimaged on photoelectric converter 9 via pinhole 8 . This diffracted light 6 is converted by a photoelectric converter 9 into an electrical signal according to the amount of light. This electrical signal is sent to the analog-to-digital converter 1
After being converted into a digital signal by 0, the signal is sent to the signal processing computer 11. This digital signal is subjected to threshold value processing in a signal processing computer 11, with a signal exceeding a predetermined value considered as a defect, and is displayed on a display section 12. Here, reference numeral 13 denotes a pulse motor that moves the video disk 1 in parallel for each round of inspection while rotating the rotary motor 2a. By repeating the signal to the pulse motor 13 and sequentially moving the feed rotation drive device 2 in parallel, the entire surface of the regular pattern on the video disc 1 can be inspected.

The test object used here is a video disk 1 in which a large number of grooves 1b are regularly arranged on a substrate 1a as shown in FIG. 2, and the width dimension of the grooves is a, the depth dimension is h, and the pitch dimension is Assume that p is a defect in some of the grooves (for example, pitch unevenness deviated from the regular pattern by τ).

Now, when (2N+1) grooves are irradiated with laser light R at the same time, a pattern of diffracted light is projected at a certain distance (hereinafter referred to as a far-field pattern).
F _T (ξ) is given by the following equation.

F _T (ξ) = F (ξ) + G (ξ) ... (1) Here, ξ is the coordinate of the position where the far-field pattern is projected (hereinafter referred to as far-field coordinate), and F (ξ) is the regular pattern G(ξ) is the far-field pattern from the defect. Further, F(ξ) is given by the following equation.

F(ξ)={(p-a) sin c[(p-a)ξ] +a sin c(aξ)exp[-jπ(4h/λ+pξ)]
}×sin {2π(N+1/2)pξ}/sin(πpξ)……(
2) However, sin c(x)=sin πx/πx, where λ is the wavelength of the irradiated light and j is the imaginary unit. From this, |F(ξ)| ² is a function that has a maximum value at ξ=n/p, and diffracted light of orders other than integers (n=0, 1, 2, 3...) has a small value, ｜G(ξ)｜ ²
will be approximately the same. That is, the amount of light output from the objective lens 7 is based on a regular pattern,
The level of damage is almost the same as that caused by defects.

At this time, the diffracted light due to the defect is transmitted to the objective lens 7.
The image is refocused to the point corresponding to the defect. That is, as shown in FIG. 2, if there is only one defective part, the image will be focused on one point. On the other hand, the integer-order diffracted light due to the regular pattern spreads over the entire imaging plane at the imaging position of the objective lens 7 corresponding to the regular pattern. That is, N=
When the laser beam R is irradiated onto a regular pattern of 100 points, the diffracted light from the regular pattern spreads to 201 points.

From this, the amount of light passing through the objective lens 7 is almost the same for the diffracted light 6 due to the defect and the diffracted light from the regular pattern, so the amount of light from the defect that is imaged at one point via the objective lens 7 will be much stronger than the regular pattern spread over 201 points. Thereby, the amount of light due to the regular pattern on the imaging plane of the objective lens 7 can be ignored.

Therefore, the bright spot of light observed on this imaging plane is considered to be due to a defect, and the amount of light is determined by setting the far-field coordinate of the objective lens 7 as ξ ₀ and assuming that the aperture of the objective lens 7 is sufficiently small. It is given by the following equation.

I=｜G(ξ ₀ )｜ ² = 8/π ² ξ ₀ ² sin ² (πτξ ₀ )si
n ² (πaξ ₀ ) {1−cos (4h/λπ)}...(3) From this equation (3), the amount of light at the position of pinhole 8 is proportional to the square of the sin function of the pitch unevenness r. I understand. Therefore, this amount of light is photoelectrically converted by the photoelectric converter 9, and the value obtained by A/D conversion by the analog/digital converter 10 is applied to the signal processing computer 11, and the pitch unevenness can be determined by equation (3). In addition, in order to increase the speed, if you calculate the formula (3) in advance and record it as a numerical table in the signal processing computer 11, the processing time can be reduced by drawing the numerical table instead of calculating the formula (3). It can be significantly shortened.

In addition, even if there is a vertical defect due to dust, scratches, etc., the bright spot observed on the imaging plane is considered to be due to the defect, as described above, and the light intensity I is determined by ξ ₀ with the far-field coordinates of the objective lens 7, and the defect Let the size of be d,
If the aperture of the objective lens 7 is sufficiently small, it is given by the following equation.

The size of the defect can be determined from this equation (4). However, in actual inspection, it is not necessary to know the exact size of a defect, and more importance is placed on determining the number of defects larger than a certain size. Therefore, it is sufficient to obtain the threshold value in advance using equation (4) and compare it.

Next, if we change the far field coordinate ξ ₀ of the objective lens 7, we get
As shown in FIG. 3, the maximum value point of the light amount I changes.
FIG. 3 shows the amount of light I for the pitch unevenness calculated using equation (3) when the far-field coordinates ξ ₀ =0.2 and ξ ₀ =0.4. From this figure, the size of pitch unevenness τ
is less than 12 μm (the point showing the maximum value of the light intensity I at far-field coordinates ξ ₀ =0.4), the sensitivity is much better when ξ ₀ =0.4 than when far-field coordinates ξ ₀ =0.2. I understand. On the other hand, if the far-field coordinate ξ ₀ is changed from ξ ₀ = 0.4 to ξ ₀ = 0.2, the maximum value point of the light amount I changes from the pitch unevenness size τ = 1.2 μm to τ =
Since it moves to 2.5μm, the measurement range can be widened. From this, the sensitivity and measurement range can be changed by moving the far field coordinate ξ ₀ of the objective lens 7.

Next, the effects of this invention will be listed.

A part of the diffracted light occurring between the n-th order diffracted light and the n+1-th order diffracted light can be sufficiently detected using a dark lens with a small aperture diameter. The price can be significantly reduced.

Since the depth of focus of the objective lens 7 is deep, once the focus is set, there is no need to refocus during the inspection, and the focus can be easily achieved, which is advantageous in speeding up the inspection.

Since there is no need for a highly accurate focusing mechanism as in the prior art, the structure can be simplified and the cost can be reduced.

Since the objective lens 7 may be disposed between the n-th order diffracted light and the n+1-th order diffracted light, the objective lens 7 can be easily aligned.

Highly accurate measurement becomes possible by moving the far-field coordinate ξ ₀ and appropriately selecting the sensitivity and measurement range depending on the object of measurement.

Furthermore, when comparing Equation (3) and Equation (4) above, it can be seen that there is a difference in the distribution of the amount of light I depending on the far-field coordinate ξ ₀ . Here, Fig. 4a is a curve diagram showing the relationship between the far-field coordinates of equation (3) and the light amount I, and Fig. 4b is a curve diagram showing the relationship between the far-field coordinates of equation (4) and the light amount I. It is a diagram. From this figure, in equation (3), ξ ₀ = 0 and I
= 0, and in equation (4), I takes the maximum value when ξ ₀ =0.

In view of this, for example, a pair of objective lenses is disposed between the nth-order diffracted light and the n+1-th order diffracted light of the diffraction pattern obtained by irradiating coherent light onto a regular pattern, as in the above embodiment. , a pinhole, a photoelectric converter, etc. are arranged in correspondence with this pair of objective lenses. At this time, by setting the first objective lens or the second objective lens to different far-field coordinates ξ ₀₁ or ξ ₀₂ , the respective light quantities I ₀ can be detected from different far-field coordinates ξ ₀ . From this, if the light intensity when the far-field coordinate of the first objective lens is ξ ₀₁ is I ₀₁ , and the light intensity when the far-field coordinate of the second objective lens is ξ ₀₂ is I ₀₂ , then from FIG.
If <ξ ₀₁ <ξ ₀₂ and I ₀₁ >I ₀₂ , it is a raised defect such as dust or scratches, and if 0 <ξ ₀₁ <ξ ₀₂ and I ₀₁ <I ₀₂ , it is a pitch uneven defect.

Therefore, by using a plurality of photoelectric conversion outputs at different far-field coordinates of the lens in this way, it is possible to determine whether the defect is a pitch unevenness or an individual defect such as dust or scratches.

Next, another embodiment of the invention will be described.

FIG. 5 shows an example in which the half mirror in the above embodiment is replaced with a mirror 21, and the other configurations are the same as in FIG. 1. In FIG. 5, the same parts as those in FIG. 1 are designated by the same reference numerals for convenience of explanation, and the explanation thereof will be omitted. In this case, coherent light is used to prevent the n-th order diffraction light and the n+1-th order diffraction light in the diffraction pattern obtained by irradiating the regular pattern of the video disc 1 with coherent light (laser light R) from being reflected by the mirror 21. It is necessary to adjust the illumination angle of the light.

Therefore, according to such a configuration, in addition to the effects mentioned in the above embodiment, the objective lens 7 can be disposed between the video disk 1 and the mirror 21, so that the objective lens is not attached to the subject. 7 can be brought close together, and a lens with a short working distance can be used.

In FIG. 6, an objective lens 7 is disposed between n-th order diffracted light 32a and n+1-th order diffracted light 32b obtained by transmitting coherent light through a regular pattern of an object 31 made of a transparent member. The other configurations are the same as in the case of FIG.

Therefore, according to such a configuration, by irradiating the subject 31 with coherent light and detecting the transmitted and diffracted light 33, the same effects as in the above embodiment can be achieved.

Note that the present invention is not limited to the above-mentioned embodiments, and can be implemented with various modifications without changing the gist.

For example, in the above embodiment, the sensitivity and measurement range were changed by moving the far-field coordinate ξ of the diffracted light generated from the defect using one lens, but the present invention is not limited to this. For example, by arranging multiple lenses and changing the far-field coordinates of each lens, and by arranging a photoelectric converter corresponding to each lens, and by switching the photoelectric converter, the sensitivity and measurement range can be adjusted according to the measurement target. can be selected arbitrarily.

Further, in the above embodiments, a video disk in which grooves are regularly arranged is used as the subject, but it goes without saying that the basic pattern formed on the subject is not necessarily limited to grooves. Further, the light source is not limited to a laser light source, and any light source that emits coherent light or a white light source can also be used.

As described above, according to the present invention, a lens is disposed between the n-th and n+1-th order diffracted lights obtained by irradiating substantially coherent light onto an object having a regularly arranged basic pattern, thereby detecting defects. By detecting the diffracted light of the area, a small-diameter lens can be used, significantly reducing the cost, and the depth of focus becomes deeper, making it easier to focus, which speeds up inspection. It is possible to provide a defect inspection device that can perform

[Brief explanation of the drawing]

Fig. 1 is a schematic configuration diagram showing an embodiment of the present invention, Fig. 2 is an enlarged vertical cross-sectional view of a part of the subject used in the embodiment, and Fig. 3 is an explanation of the embodiment. FIG _. 4 is a waveform diagram showing the relationship between the light amount I and the pitch unevenness size τ for explaining other embodiments, and FIG. FIG. 6 is a schematic configuration diagram showing other embodiments. 1... Video disk, 1a... Board, 1b...
Groove, 2... Rotation drive device, 2a... Rotation motor, 2b
...Holding part, 3...Laser light source, 4...Half mirror,
5a... 0th order diffraction light, 5b... 1st order diffraction light, 6...
Diffracted light, 7... Objective lens, 8... Pinhole, 9...
Photoelectric converter, 10...analog-digital converter,
DESCRIPTION OF SYMBOLS 11... Signal processing computer, 12... Display unit, 13... Pulse motor, 21... Mirror, 31... Subject, 32
a... nth order diffracted light, 32b... n+1st order diffracted light,
33...Transmitted and diffracted light.

Claims (1)

[Claims] 1. A subject having basic patterns arranged regularly;
A light source that irradiates the basic pattern of the object with substantially coherent light, and a light source that is placed between the nth-order diffracted light and the n+1st-order diffracted light in the diffraction pattern obtained by irradiating the basic pattern with substantially coherent light. at least one lens that reimages the diffracted light generated from the defective portion of the object; at least one photoelectric converter that converts the light reimaged by the lens into an electrical signal; and the photoelectric converter a signal processing unit that determines the type of defect in the defective portion from the output of the lens, and the defect inspection is characterized in that the type of defect is determined using a plurality of photoelectric conversion outputs at mutually different far-field coordinates of the lens. Device. 2. The defect inspection device according to claim 1, wherein a plurality of lenses are arranged at mutually different far-field coordinates, and a plurality of photoelectric converters are arranged in correspondence with each lens. 3. The defect inspection apparatus according to claim 1 or 2, wherein the diffracted light generated from the defective part of the object to be inspected is guided to the photoelectric converter via a pinhole.