JPS60245240A - Fault inspection and apparatus thereof - Google Patents

Abstract

PURPOSE:To curtail the inspection time by detecting location of fault through scanning of the scanning line at a high speed and executing the close examination through low speed scanning of scanning lines for the area near the detected fault location. CONSTITUTION:A fault detection circuit 20 detects a fault by obtaining a difference signal through comparison between the current signal sent from a modulator 17 and a delayed signal before a predetermined period. However, this fault detection signal indicates only the location of fault on the wafer 1. When the coarse inspection by high speed scanning as described above completes, the X-Y table 2 is changed to the predetermined low speed operation, and the fault detection part 20 obtains the final data such as normal location or size, etc. of the fault 27 obtained in the coarse inspection based on the picture signal. As explained above, since a low speed scanning period can be curtailed as a whole, the inspection period can also be curtailed as the total inspection period.

Description

[Detailed Description of the Invention] [Technical Field] The present invention relates to a defect inspection technique, and in particular to a technique of scanning a scanning line (for example, light, etc.) on the surface of an object to be inspected to obtain a signal and detecting defects based on the signal. , for example, relates to a technique effective for use in detecting defects in the appearance of wafers in the manufacture of semiconductor devices.

[Background Art] In the manufacture of semiconductor devices, as a technology for inspecting defects on the appearance of wafers, there is a technology in which a wafer and a one-dimensional photosensor are scanned relative to each other to obtain an image signal, and defects are detected based on this image signal. is possible.

However, in such defect inspection technology, as the minimum defect detection accuracy becomes finer due to the miniaturization of integrated circuits in semiconductor devices, the inspection speed becomes slower, resulting in a significantly longer inspection time. The inventor has clarified that there is a point.

A document that describes optical inspection equipment is the 1983 special edition of Electronic Materials (published by Kogyo Kenkyukai, January 1983).
Published January 18th, P2O4-P2O9).

[Object of the Invention] An object of the present invention is to provide a defect inspection technique that can shorten inspection time.

The above and other objects and novel features of the present invention will become apparent from the description of this specification and the accompanying drawings.

[Summary of the Invention] A brief overview of typical inventions disclosed in this application is as follows.

That is, first, in order to roughly inspect the entire surface of the object to be inspected for defects, a scanning line is scanned at high speed to detect each defect position on the object. In order to accurately inspect the defects, the area to be inspected is limited to a small area by performing a detailed inspection that scans the scanning line at a low speed to detect the size of defects, etc., thereby reducing the overall inspection time. This is what I did.

[Embodiment] FIG. 1 is a block diagram showing a defect inspection apparatus which is an embodiment of the present invention, and FIGS. 2, 3, and 4 are explanatory diagrams for explaining the operation.

In this embodiment, this defect inspection apparatus is equipped with an xy table for holding a wafer 1 as an object to be inspected, and the xy table 2 is operated by a pair of servo motors 3 and 4.
It is designed to be driven in the X and Y directions via a feed mechanism or the like. Servo motor 3.4 is controller 7
The controller 7 performs feed track control based on the measurement results of the laser range finders 5 and 6. The controller 7 has the first
A changeover switch 8 is connected, and this switch 8 switches between a high-speed power source 9 and a low-speed power source 10.

A light source 11 such as a mercury lamp is provided above the XY table 2, and the light source 11 emits light onto the wafer 1 on the table 2 via a nof mirror 12 and a lens 3. A charge-coupled device (COD>14) as a one-dimensional photo sensor is placed directly above the table 2 on a half-mirror.
12 and is disposed to directly face the wafer 1 with a lens 13 in between, and the CCD 14 can scan in the X and Y directions relative to the wafer 1 by moving the XY table 2.

An analog signal processor 15 is connected to the CCD 14, and an A/D converter 16 is connected to this processor 15. A defect detection section 20 is connected to the A/D converter 16 via a second changeover switch 19, and the second changeover switch 19 switches between a path including a modulator 17 and a path 18 without it. It has become.

The defect detection section 20 is connected to a design pattern generation section 22 via a third switching switch 21, and the third switching switch 21 is connected to a path having a second modulator 23 and a path 24 not having it. It is designed to switch between. A timing signal generator 25 is connected to the defect detection section 20, and this generator 25 is also connected to the controller 7 and the design pattern generation section 22 (paths not shown).

The controller 7, defect detection section 20, design pattern generation section 22, timing signal generator 25, and each changeover switch 8, 19.21 receives instructions from a command section 26 comprised of a computer, etc., and performs operations as described below. The command unit 26 is programmed with control sequences, function instructions, etc., and also includes a memory portion (not shown) for storing test results.

Next, the operation of the defect inspection apparatus will be explained, and a defect inspection method that is an embodiment of the present invention will be explained.

Here, the defect detection unit 20 detects the position and size of the defect by taking the difference between the current image signal from the CCD 14 and the delayed image signal from a predetermined time ago, and also detects the position and size of the defect by taking the difference between the current image signal from the CCD 14 and the delayed image signal from a predetermined time ago. It is assumed that the arrangement is such that the position of the defect relative to the layer of the wafer is detected by superimposing the signals.

In this embodiment, first, according to a command from the command unit 26,
The first switch 8 is switched to the high-speed power supply 9 side, the second switch 19 is switched to the modulation section 17 side, and the third switch 21 is switched to the modulator 23 side. The servo motor 3.4 of the XY table holding the wafer 1 is supplied with power from the high speed power supply 9 via the controller 7, and
is driven in the XY direction. As a result, the CCD 14 scans the wafer 1 at a relatively high speed.

Now, the shift speed S of the CCD 14 at the normal scanning speed of the CCD 14 with respect to the wafer 1 (hereinafter referred to as low speed scanning), and the scanning speed X in the direction perpendicular to the CCD 1'4 (hereinafter referred to as the X direction). It is assumed that the relationship is as shown in FIG. That is, CCD14
It is assumed that the scanning speed X is set so that the CCD 14 advances by the width W of the picture element 14a at the time when one shift is completed. As a result, the CCD 14 collects images so as to cover the entire surface of the wafer 1. Then, the size of the picture element 14a of the CCD 14 relatively projected onto the wafer 1 is set to, for example, 0.2 μrn x 0
．． 2 μm, and the minimum size of the defect 27 to be detected is, for example, 0.6 μm in diameter.
The size of 7 is detected by the CCD 14 as indicated by the diagonal lines in FIG.

However, the scanning speed of the CCD 14 with respect to the wafer 1 is
For example, when the low-speed scanning is doubled (X is extended in the scanning X direction, and detection of the smallest defect 27 becomes impossible.

Therefore, in this embodiment, the inability to detect defects due to high-speed scanning is overcome by inserting the modulator 17 between the COD 14 and the defect detection section 20 by switching the second changeover switch 19. There is. That is, the modulator 17 modulates the image signal from the CCD 14 that is input via the analog signal processing section 15 and the A/D converter 16 so as to compensate for the state in which it is extended in the scanning Send to. Modulation modes for compensation include, for example, averaging the image signal components of the adjacent picture elements 14a, 14a, sampling the image signal of each picture element every other picture element, and averaging the image signal components of each picture element. Possible methods include doubling the amount.

The defect detection unit 20 detects defects by comparing the current signal from the modulator 17 and the delayed signal from a predetermined time ago and obtaining a difference signal. However, since this defect detection signal is based on a modulated signal, it does not accurately display the size of the defect 27, but only the position of the defect on the wafer 1.

At the same time, the defect detection section 20 includes a design pattern generation section 22.
A pattern signal from the wafer 1 is modulated by the second converter 23 in a manner corresponding to the modulation of the first modulator 17 and is transmitted, and the position of the defect 27 in the layer of the wafer 1 is detected by this signal.

Each position of the defect detected by the defect detection section 20 is transmitted to the command section 26 and stored in a part of its memory as shown virtually in FIG. 4.

When the rough inspection by high-speed scanning is completed, the command unit 26
reads the position of the defect 27 stored in a part of the memory.

Subsequently, the command unit 26 relatively moves the CCD 14 at high speed, for example, near the position of the defect 27 closest to the beginning of the scan. That is, the command unit 26 sends a command to the controller 7 to move the XY table 2 in a desired direction at high speed.

Rangefinder 5.6 indicates that CCD 14 has reached the vicinity of defect 27.
When confirmed, the command unit 26 switches the first to third changeover switches 8, 9, and 21 to the low speed side. By this switching, the XY table 2 is operated at the expected low speed, and the CCD 44 relatively performs regular image acquisition scanning on the wafer 1 as shown in FIG. The image signal sampled during this regular scanning is transferred to the second changeover switch 1.
9 has been switched, the signal is transmitted to the defect detection section 20 via a path 18 that bypasses the modulator 17. The defect detection unit 20 detects a defect 27 detected in the rough inspection based on the image signal.
Contain final data such as official position and size. At this time, since the third changeover switch 21 is switched to the defect detection section 20, a regular pattern signal is sent from the design pattern generation section 22 via the path 24.
This allows the layer position of the defective wafer to be determined.

When the detailed inspection by low-speed scanning at that position is completed, the command unit 26 relatively moves the CCD 14 near the position of the next defect 27 at high speed.

When the CCD 14 reaches the vicinity of the second defect 27, the command unit 2
6, the CCD 14 is used to perform relatively low-speed scanning, and the detailed inspection as described above is carried out.

Thereafter, a detailed inspection is performed for each defect position determined by the rough inspection, and the position, size, etc. of defects on the entire surface of the wafer are precisely determined.

[Effects] (1) After locating the defect through a rough inspection, by performing a detailed inspection using low-speed scanning only for that defect position, the overall low-speed scanning time can be shortened, which reduces the overall inspection time. As a result, inspection time can be shortened.

(2) By shortening the overall inspection time,
Since it is also possible to reduce the scanning speed during detailed inspection to an extremely low speed, inspection accuracy can be further improved.

Although the invention made by the present inventor has been specifically explained above based on Examples, it goes without saying that the present invention is not limited to the Examples and can be modified in various ways without departing from the gist thereof. Nor.

For example, sensors that collect signals by relatively scanning an object to be inspected are not limited to -dimensional photosensors, but also image pickup tubes,
It may be an X-ray sensor, an electron beam sensor, an infrared sensor, an ultrasonic sensor, or the like.

The defect detection unit is not limited to one that detects defects based on the difference between the current input signal and the past input signal, but also one that detects defects based on the difference between the sampled signal and the design pattern signal, and one that detects defects based on the difference between the two sampled signals. It may also be possible to detect defects using the following method.

[Field of Application] In the above explanation, the invention made by the present inventor was mainly applied to the field of application, which is the field of application, which is the background of the invention, which is the visual defect inspection technology of wafers. It can also be applied to external defect inspection technology for masks and reticles, internal flaw detection technology for structural materials, etc.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an embodiment of the present invention, FIG.
FIGS. 3 and 4 are explanatory diagrams for explaining the action. 1... Wafer (object to be inspected), 2... XY table,
3.4...Servo motor, 5.6...Distance finder, 7.
... Controller, 8, 19. 21 ... Selector switch, 9 ... High speed power supply, lO ... Low speed power supply, 11.
...Light source, 12... Half mirror 113... Lens, 14... CCD (-dimensional photo sensor), 15.
...Analog signal processor, 16...-A/D converter,
17°23...Modulator, 18.24...Path, 20
... Defect detection section, 22 ... Design pattern generation section, 2
5... Timing signal generator, 26... Command unit, 2
7...Defect.

Claims (1)

[Claims] 1. - A method for inspecting defects on an object to be inspected, which includes a first inspection in which the entire surface of the object to be inspected is scanned with a scanning line to obtain a signal, and rough defect positions are detected. After that, ``a second inspection is performed in which a scanning line is scanned near the rough defect at a scanning interval narrower than the scanning interval of the scanning line in the first inspection to obtain a signal, and a precise defect position is detected. A defect inspection method characterized by performing precise defect inspection on an object to be inspected. 2. Claim 1 characterized in that the signal obtained by the first inspection is modulated to compensate for distortion of the signal in accordance with the ratio to the scanning speed during the detailed inspection. Defect inspection method described in section. 3. A sensor that scans the surface of the inspected object with a scanning line to detect information from the inspected object and converts it into an electrical signal, and a defect detection section that detects defects in the inspected object based on the electrical signal from this sensor. A defect inspection device comprising: a speed switching section that switches the scanning speed of the scanning line; a storage section that stores a defect position on the object detected by the first scan; and the speed switching section. and a command section that switches the speed of the second inspection to a low speed and causes a second inspection to be performed in the vicinity of the defect position stored in the storage section. 4. The defect inspection device according to claim 3, wherein the sensor is a two-dimensional photo sensor.