JPH0240541A - Surface checking apparatus - Google Patents

Abstract

PURPOSE:To make it possible to detect the height of a defect in addition to the position and the size of the defect by providing the constant state of a distance between an optical system and a surface to be checked under the focused state, scanning the surface to be checked with light, detecting the reflected light, and comparing the amount of the light with a reference value. CONSTITUTION:A surface to be checked 6a of test specimen 6 is scanned with converging light which is projected from an optical system 1. The distribution of the reflected light rays from the surface to be checked 6a in a focus control circuit 12 is detected. A distance between the optical system 1 and the test specimen 6 is kept constant. Then, a defect is detected by the comparison between the reflected light rays and a specified standard value in a comparator 16, and a signal is generated. The position and the size of the defect are detected with a defect processing circuit 17. The position of the optical system 1 is fixed with a focus locking circuit 29. Thereafter, the detected defect is scanned again with a crossfeed motor controller 27. The maximum value of the fluctuation in distribution of the reflected light rays from the surface to be checked 6a in the second scanning is detected with a peak holding circuit 31. Then, the height of the defect is computed by a personal computer 28 from the maximum value of the fluctuation.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of Industrial Application The present invention is used for defect inspection on the surface of flat substrates such as glass and wafers, and for defect inspection of master stampers with group pits such as optical disks. The present invention relates to a surface inspection device that inspects minute defects on a surface by scanning converged light and detecting an increase or decrease in the amount of reflected light.

BACKGROUND ART Conventionally, as an apparatus or method for inspecting minute defects on a flat substrate surface, there is a method of scanning light and detecting the scattered light, as shown in Japanese Patent Laid-Open No. 61-133843, for example. , JP-A-62-35249, JP-A-57
There is a method of irradiating a laser beam in a spiral manner and detecting a defect based on an increase or decrease in the amount of reflected light, as shown in Japanese Patent No. 94636.

Problems to be Solved by the Invention However, according to these surface inspection methods, although the position and size of a defect can be detected, the height (depth) of the defect and the type of unevenness cannot be determined.

In this regard, as shown in Japanese Patent Application Laid-Open No. 62-188948, there is a method in which a laser beam is incident perpendicularly and the type of defect is determined by comparing the reflected light and scattered light from the substrate. According to the method disclosed in Japanese Patent No. 58-177373, in which a laser beam is scanned in a spiral manner and the scattered light from the substrate is detected through an optical system that focuses the light into an elliptical shape, the type of unevenness can be determined. The height of the defect cannot be detected.

Here, in the resist coating process for semiconductor wafers, the resist coating process for optical disks, etc., and the recording agent coating process, even minute defects (for example, 5μ
m or less), if the height of the defect is above a certain level, the coating agent will become uneven due to the defect (this is due to the formation of an area where the coating agent does not spread on the outer periphery from the defect location during coating), which can be fatal. It becomes a defect. Therefore, it is very important to detect the height in defect inspection.

However, according to conventional methods, no matter which method is used, it is not possible to detect the height of the defect as described above, so if a defect is found, even if it is a minute defect, the board is disposed of as a defective product. Alternatively, a person must measure the height of the defect using a microscope to determine whether the defect is acceptable.

The former method results in a low yield, while the latter method requires a great deal of effort and time.

Means for Solving the Problem An optical system for irradiating converged light and an object to be inspected having a surface to be inspected that is scanned by the light by this optical system are provided, and the distribution of reflected light from the surface to be inspected is determined. means for detecting and maintaining a constant distance between the optical system and the object to be inspected, detecting reflected light from the surface to be inspected and comparing it with a predetermined reference value to detect the presence or absence of a defect and generate a defect signal. and means for detecting the position and size of the detected defect, and
means for rescanning the detected defect; means for fixing the position of the optical system immediately before rescanning the detected defect; and means for controlling the reflected light from the surface to be inspected when rescanning the detected defect. Means for detecting the maximum value of the variation in the distribution and means for calculating the height of the defect based on the detected maximum value of the variation are provided.

Operation First, while keeping the distance between the optical system that irradiates convergent light and the surface to be inspected of the object to be inspected in a constant state of focus, the surface to be inspected is scanned with light, and the light from the surface to be inspected is By detecting the reflected light and comparing the amount of detected light with a predetermined reference value, the presence or absence of a defect larger than a predetermined size is detected. Once the presence of a defect is detected, its location is also detected. This allows the location and size of the defect to be determined. Furthermore, when such a defect is detected, the defect is rescanned by, for example, positioning the optical system at the same scanning position using a rescanning means. After fixing the focused optical system in position and locking the focus, the defect is rescanned. When the maximum value of the variation in the distribution of reflected light from the surface to be inspected is detected during this rescanning, the focus is locked, so the distance in the thickness direction of the object to be inspected (in this case, the height of the defect) is larger, the more the focus is shifted. This corresponds to a large value, and shows the maximum value at the defective part depending on its height. Therefore, the height of the defect is calculated by calculating based on the maximum value of this detected variation, and the height of the defect is also detected.

Example An embodiment of the present invention will be described based on the drawings.

First, the configuration and operation of the first part of the optical pickup, which is the optical system in a conventionally known surface defect inspection device, will be explained in the second section.
This will be explained with reference to the drawings and FIG. This optical pickup 1 converts a laser beam emitted from a laser light source (not shown) into a parallel beam using a collimator lens 2, transmits it through a beam splitter 3, and then passes it through a quarter-wave plate 4 through an objective lens 5. As a result, the convergent light is condensed and scanned onto the surface 6a to be inspected of the object 6 to be inspected. The light reflected by the surface to be inspected 6a passes through the objective lens 5 again, and then passes through the 1/4 wavelength plate 4 again, becoming light with a phase delay of 90° relative to the incident light, and is sent to the beam splitter 3. Re-enter. This beam splitter 3 reflects light having a 90° phase delay with respect to the incident light, and a cylindrical lens 7 with different horizontal and vertical magnifications is provided on the optical path of the reflected light. A four-part photodiode 8 divided into two light receiving areas (see FIG. 3) is provided.

Such an optical pickup 1 has a focus servo function to always position the focal position of the objective lens 5 on the surface to be inspected 6a, and a defect signal generation function (inspection) to generate a defect signal in accordance with a defect occurring on the surface to be inspected 6a. means). That is, in the focus servo, the objective lens 5
When the focal point position no longer matches the surface to be inspected 6a, the shape of the projection light spot generated on the four-division photodiode 8 changes, and the detection signal generated from the four light-receiving areas of the four-division photodiode 8 changes. As shown in FIG.
When the calculation unit 11 calculates the difference between these addition results, a focus error signal is output (so-called astigmatism method), and this is sent to the focus control circuit 12.
(See FIG. 1) to maintain the in-focus state by feeding back to the actuator 13 for the objective lens 5 and performing servo control so that the focus error signal becomes O. Such a focus control circuit 12 detects the distribution of reflected light from the surface to be inspected 6a and adjusts the light pickup 1.
This constitutes a means for keeping the distance between the objective lens 5 and the object to be inspected 6 constant.

Further, the signals from the arithmetic units 9 and 10 are added by the arithmetic unit 14 and used for defect detection. That is, this computing unit 1
4 to a reference voltage generator 15 as shown in FIG.
The presence or absence of a defect is detected by comparing it with a reference voltage from the comparator 16. In this way, the surface to be inspected 6
An inspection means is configured to detect the reflected light from a, compare it with a predetermined reference value, detect the presence or absence of a defect, and generate a defect signal. That is, when the objective lens 5 is focused on a defect-free area on the inspection surface 6a of the inspection object 6 and a laser spot is generated in that area, most of the optical signal is specularly reflected and added by the calculator 14. Since the level is greater than the reference voltage,
The output of comparator 16 is held at L level. However, when the focal position of the objective lens 5 reaches the defect area, the laser beam is scattered in the defect area, so the beam component that is specularly reflected decreases, and the level of the optical signal added by the calculator 14 decreases. It becomes smaller than the reference voltage. Therefore, the output of the comparator 16 changes to H level and becomes a defect signal. In other words, this is similar to detecting the presence or absence of pits on an optical disc.

In this embodiment, in order to also detect the height of the detected defect, in addition to the function of detecting the defect, the focus is locked just before re-scanning the detected defect and the current is applied to the actuator 16. A function to keep the focus constant, a function to detect the difference signal when the optical pickup 1 rescans over the defect in such a focus locked state, that is, the maximum value of the fluctuation of the focus error signal from the arithmetic unit 11, It also has the function of calculating the height of a defect from the maximum value.

Here, objective lens NA=0.45, incident spot diameter=
When using the optical pickup 1 designed to have a diameter of 2 μm and an incident beam wavelength of 830 nm, the relationship between the peak value of the difference signal obtained from the arithmetic unit 11 and the height of the defect is as shown in FIG. The size of the detected defect is larger than the spot diameter, that is, larger than 2 μm. However, in order to further inspect the height of micro defects, it is sufficient to increase the objective lens NA or shorten the wavelength.

The configuration of the surface defect inspection apparatus of this embodiment including such functions is shown in FIG. First, the flat surface to be inspected 6
An object 1 to be inspected, such as a glass master disk or a semiconductor wafer, is mounted and supported on a turntable 2o. This turntable 20 is rotationally driven by a turntable motor 21. This turntable motor 21
An encoder 22 is provided on the axis of the turntable, and the rotational speed of the turntable motor 21 is controlled (for example, PLL control) based on the rotational position signal from the encoder 22 to prevent phenomena such as rotational jitter. A controller 23 is provided.

Further, the above-mentioned optical pickup 1 is provided opposite the object to be inspected 6 on the turntable 20. The optical pickup 1 is mechanically connected to a feed device 26 that includes a transverse feed motor 24 and a position detector 25 for linearly feeding the object 6 to be inspected in a radial direction at a constant speed. The lateral feed motor 24 of this feed device 26 is connected to the position detector 2
It is controlled by a lateral feed motor controller 27 based on the position detection signal etc. from 5. In this way, the object to be inspected 6
is rotated together with the turntable 20, and the optical pickup 1 is transversely fed in the radial direction of the turntable 20 at a constant speed, so that the surface 6a to be inspected of the object 6 to be inspected is spirally shaped by the laser beam from the optical pickup 1. It will be scanned. The entire surface of the object to be inspected 6 can be scanned by such spiral scanning.

These controllers 23, 27 are controlled by a personal computer 28.

Further, the defect signal obtained from the comparator 16 is used to calculate the size and position (radius and angular position) by the defect processing circuit 17.
is performed, and the results are sent to the personal computer 28. In other words,
This defect processing circuit 17 serves as means for detecting the position and size of the detected defect.

On the other hand, the defect signal from the comparator 16 is also sent to the lateral feed motor controller 27, and is configured to stop the lateral movement of the optical pickup 1 by the lateral feed motor 24 at the same time as the defect is detected. The lateral feed motor controller 27, which stops the lateral feeding (radial direction feeding) of the optical pickup 1 at the same time as defect detection, serves as defect rescanning means. In addition, under normal conditions, focus servo is performed using a difference signal from the arithmetic unit 11 based on the detection of the optical pickup 1, but when a defect is detected, the optical pickup 1, whose traverse feed has been stopped, moves over the same defect again. However, a focus lock circuit 29 is provided to stop the focus servo control and fix the position of the objective lens 5 just before passing over the defect. More specifically, the defect signal from the comparator 16 is delayed through a delay circuit 30 and then input to the focus lock circuit 29. That is, when the defect signal is input with a delay of a time slightly shorter than the time required for one rotation of the turntable 20, the focus lock circuit 29 cuts off the input of the difference signal (focus error signal) to the focus control circuit 12. , the objective lens 5 can be fixed at a timing immediately before the optical pickup 1 passes over the defect again.

On the other hand, the defect signal delayed by the delay circuit 30 is also input to the peak hold circuit 31, and a computing unit 11 calculates the maximum value of the variation in the distribution of reflected light from the surface to be inspected 6a when the detected defect is rescanned. The detection operation is started based on the difference signal output from the.

Specifically, the peak hold circuit 31 is reset by the delayed defect signal to hold the peak value. At this time, the signal from the arithmetic unit 11 has a wavelength of about 1 to 10, for example.
By inputting the signal to the peak hold circuit 31 through a bandpass filter 32 that passes 00 μm, minute fluctuations in the difference signal and offsets of the difference signal due to surface runout are removed. The peak value held by the peak hold circuit 31 is measured by a multimeter 33, and the value is sent to the personal computer 28.

If the peak value of the difference signal in the focus lock state is detected in this way, the height of the defect can be determined by calculating on the personal computer 28 based on the peak value, as shown in FIG. .

After the peak value of the fluctuation of the difference signal is detected by the peak hold circuit 31, the focus lock is released, the focus control circuit 12 returns to the focus servo control, and the spiral-shaped optical pickup 1 is moved horizontally. Restart scanning and continue defect detection. By repeating such operations and measuring the entire surface 6a of the object 6 to be inspected, the size and position (radius, angle) and height of the defect existing on the surface 6a to be inspected can be determined.

Note that the sum signal from the arithmetic unit 14 based on the optical pickup 1 is A, and the comparator 16 receives this sum signal as one input.
B is the defect signal from C1, C1 is the difference signal from the arithmetic unit 11, and D is the signal after the difference signal C has passed through the bandpass filter 32.
, when the output from the peak hold circuit 31 is E, a timing chart of these signals accompanied by a delay operation by the delay circuit 30 is shown in FIG.

Effects of the Invention As described above, the present invention provides an inspection means for generating a defect signal, a means for detecting the position and size of the detected defect, a means for rescanning the detected defect, and a means for rescanning the detected defect immediately before rescanning the detected defect. means for fixing the position of the optical system, means for detecting the maximum value of variation in the distribution of reflected light from the surface to be inspected of the object to be inspected when rescanning the detected defect; Since we have provided a means to calculate the height of the defect based on the value, it is possible to measure and detect not only the position and size of the defect, but also the height of the defect, and it is possible to determine whether the defect is acceptable or not. It can be submitted for certification, etc.

is a characteristic diagram showing the relationship between the peak value of the signal and the height difference of the defect,
FIG. 5 is a timing chart.

■... Optical system, 6... Object to be inspected, 6a... Surface to be inspected, 12... Distance constant holding means, 16... Inspection means, 17... Position and size detection means, 27...Rescanning means, 28...Calculating means, 29...Position fixing means, 31...Maximum value detection means Applicant Rico Co., Ltd. -

[Brief explanation of the drawing]

Claims (1)

[Claims] An optical system that irradiates converged light, an object to be inspected that has a surface to be inspected that is scanned by the light by this optical system, and an object that detects the distribution of reflected light from the surface to be inspected. means for maintaining a constant distance from the surface to be inspected; inspection means for detecting the reflected light from the surface to be inspected and comparing it with a predetermined reference value to detect the presence or absence of a defect and generating a defect signal; and the position of the detected defect. and a means for detecting the size of the detected defect, a means for rescanning the detected defect, a means for fixing the position of the optical system immediately before rescanning the detected defect, and a means for rescanning the detected defect. A surface inspection device comprising means for detecting the maximum value of variation in the distribution of reflected light from the surface to be inspected, and means for calculating the height of the defect based on the detected maximum value of variation. .