JPS6159258A - Apparatus for detecting surface flaw of thin piece - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an apparatus for detecting surface defects in a thin piece of silicon wafer or the like by ultrasonic flaw detection.

[Conventional technology]

Conventionally, when detecting surface defects on thin pieces such as silicon wafers, there is a method of fixing the thin piece immersed in water and scanning it by moving the probe in a zigzag pattern or in a rectangular shape in the X-Y axis direction. It had been taken.

[Problem that the invention seeks to solve]

However, the conventional flaw detection method described above is inefficient, as it takes about 300 seconds to detect flaws on the surface of a thin piece with a diameter of 6 inches.

In addition, in the flaw detection method using underwater immersion, it is important that there are few air bubbles and that there are no foreign substances such as floating objects mixed in.

However, what about underwater Ti? At present, in the ultrasonic flaw detection method using R, no specific measures have been taken to prevent the above-mentioned generation of bubbles and contamination of foreign matter.

[Means for solving problems]

The present invention has been made to improve and eliminate the above-mentioned drawbacks of the conventional art, and includes a conveying device for loading a thin slice to be tested, and a conveying device installed at the terminal end of the conveying device with a thin slice to be tested placed thereon. A device for taking out a thin piece to be tested is equipped with a rotating disk and an elevating mechanism for the disk, and a thin piece to be tested placed on the disk is chucked from a direction perpendicular to the conveying direction to center the thin piece to be tested. A positioning device, a water tank, a probe feeding device that supports a probe disposed in the water tank and moves the probe back and forth in a straight line, and a suction part for the test thin piece at the lower tip. a sample holder with a double tube structure that can be rotated and raised and lowered; a handler device that rotates the sample holder between the disk and the water tank; By configuring a thin section surface defect detection device with a conveying device, flaw detection is performed by moving the probe in a straight line without rotating the thin section at high speed, reducing the time required for crucible flaws and achieving high accuracy. The aim is to realize flaw detection.

The configuration of the device of the present invention will be described in more detail below based on the embodiment shown in the drawings.

〔Example〕

Figure 1 is a plan view showing the entire structure of the surface defect detection device;
The figure is a front view, and FIG. 3 is a right side view of the same. The surface defect detection device A includes a transport device B for loading a thin piece 1 to be tested such as a silicon wafer, and a transport device WtC for unloading the thin piece 1 to be tested. , a positioning device WD, a water tIIE, a feeding device ytF for the probe 2 disposed in the water tank E, a sample holder 〇, a handler device H for rotating the sample holder G, and a sample holder H for rotating the sample holder G; It is composed of a transport device J for departure.

The charging conveyance device B has a set part of a carrier 3 formed by laminating a large number of thin pieces 1 to be tested, and a conveyance path 5 having a belt 4 having a circular cross section arranged at both ends is continuous with the set part. are doing. 6 is a table installed between the belts 4, 4.

Retrieval of test thin section 1? The iIC is placed on the terminal end side of the transport device 5iB, and is configured as shown in FIGS. 4 and 5. That is, the take-out device C is
It has an air cylinder 8 fixedly installed on the behead 7. An underframe 9 is attached to the tip of the piston rod of the cylinder 8, and a motor 10 for rotating the specimen thin section is fixed to the underframe 9.

Reference numeral 11 denotes a disk for the test thin piece g1 attached to the tip of the rotating shaft of the motor 10. This disk 11 is located in a circular hole 12 formed on the terminal end side of the table 6, and therefore passes through the circular hole 12 in accordance with the operation of the air cylinder 8 and moves to the position indicated by the chain line in FIG. Go up and down between the solid line positions. In the state indicated by the solid line, it is located below the thin piece 1 to be tested, which is placed on the conveyor belt 4.4 and conveyed. In the state shown by the chain line, the disk 11 pushes up the thin piece 1 to be tested on the conveyor belt 4.4 and transfers the thin piece 1 to be tested from the U-transport belt 4.4.

In FIG. 4, reference numeral 43 denotes a detection element such as a photoelectric sensor for checking the placement state of the test thin piece 1 placed on the disk 11. In this embodiment, the position of the notch 1a of the test thin piece 1 is detected by detecting the level of reflected light irradiated onto the test thin piece 1.

The positioning device is as shown in Fig. 4 and Fig. ff15.
A chuck 13.14 is provided at the end of the conveyor belt 4.4 and is disposed in a direction perpendicular to the one-leg feeding direction.

The chuck 13 has a bifurcated shape and is fixed to the tip of a rod 16 which is slidably supported at the upper end of the support post 15. Reference numeral 17 indicates a rod 16 for sliding the rod 16. It is an air cylinder.

Further, the chuck 14 has a flat plate shape, and has a support boss H5.
A rod 1B is slidably supported at the upper end of the rod 18 and fixed to the tip of the rod 18. These chucks 13 and 14, which may be air cylinders for sliding the opening 7 and 18, are designed to slide by a certain amount using the same method.

FIG. 10 shows another embodiment of the positioning device. This embodiment consists of an air cylinder 50 and a rod 5.
1 are arranged in parallel in the horizontal direction, and a chuck fixing member 52 is connected to the tip of the piston rod 50a and the tip of the loft 51. Then, an arc-shaped chuck 54 is attached to the chuck fixing member 52 with bolts 53 or the like in plan view. The curvature of the inner circumferential surface of the chuck 54 is the same as the curvature of the outer circumference of the thin piece 1 to be tested. Further, two chucks 54 are installed facing each other. With the chucks 54 and 54 having such a configuration, it is possible to hold the thin piece 1 to be tested in close contact with the outer peripheral end face of the thin piece 1, and there is no need to worry about relative positional deviation between the two. It is possible to improve alignment accuracy.

As is clear from Figures 2 and 3, tank E is
The upper end portion is formed in a corrugated shape, so that fresh water (the supply part is not shown) supplied from the lower surface of the tank can overflow uniformly from the entire circumference of the tank. Moreover, the upper part of the water tank E is designed to create a horizontal water flow. This is achieved by circulating the water inside the tank using a pump installed outside the tank.

The feeding mechanism F of the probe 2 is installed at a position close to the water tank E, as shown in FIGS. 1 to 3 and 6,
It has a slider 21 placed on a guide bed 20. A screw shaft 23 attached to a feed motor 22 is threaded through one end of the slider 21 . Further, a vertical support arm 24 of the probe 2 is screwed and vertically provided on the other end side of the slider 21, and a horizontal support arm 25 is supported on the lower end side of the vertical support arm 24. Three probes 2 are attached to this horizontal support arm 25.

The angle of each probe 2 is determined so that the ultrasonic beams converge at the same point on the test thin section l.

The flaw detection method will be described later.

Next, we will discuss the sample holder 〇, which rotates the sample holder 〇 by immersing it in water tME while holding the specimen 1 by suction, and the handler device H, which rotates the cough sample holder 〇 between the extraction device Wc and the water tank E. explain. As shown in FIGS. 1 to 3 and 6 to 9, the handler device tl has a rotating shaft 27 of a swing arm 26 rotatably attached to a turntable 28.
A swing motor 29 is connected to the lower end side of the swing motor 29. In this embodiment, the pivot angle of the pivot arm 26 is 9G''.

Sump 7L/yh) Lt Dar G is outside ff130.
Inner vI31 (7) It has a double tube structure, and the outer tube 3o
is attached to the tip of the swing arm 26 through the
It can be raised and lowered freely.

The inner cylinder 31 is rotatably supported inside the outer cylinder 30. 32 is a housing for attaching the sample holder G to the rotating arm 26. A motor 33 for raising and lowering the sample holder 0 is fixedly attached to the side surface of the casing 32, and a pinion 34 is attached to the tip of the rotating shaft. The pinion 34 meshes with a rack 35 carved in the outer cylinder 30. A gear box 36 is attached to the upper end side of the outer cylinder 3°. 37 is a motor attached to the upper end side of the gear box 36, and a gear 38 is fixed to the tip of its rotating shaft. 39 is inner v
131, which is a gear fixed to the upper part of the motor 37.
meshes with the gear 38 of.

Therefore, the sample holder 〇 can be freely raised and lowered by the rack and binion mechanism, and the inner tube 31 can be freely rotated within the outer tube 30.

The upper end side of the inner 'i'fj 31 passes through the gear box 36 and communicates with a negative pressure source (not shown) via a rotary joint 40.

In addition, on the lower end side of the inner v131, test thin section 1 is placed 'ill@
A jig 41 for holding is attached. Reference numeral 42 denotes a cushioning material, such as rubber, installed in the recess of the jig 41. It is a matter of course that these jigs 41 and 111H material 42 communicate with the negative pressure source via the inner w131.

As shown in FIG. 1, the conveying device J for carrying out the thin section to be examined is installed within the swing range of the swing arm 26. The basic configuration is the same as that of the charging transport device rfiB, so the explanation here will be omitted.

Next, a description will be given of how to install the surface defect detection apparatus for the test thin piece 1 constructed as described above.

By the way, at the time of starting operation, the swing arm 26 has turned to the side of the take-out device C, and the sample holder G is waiting above the disk 11 of the take-out device C.

First, when the specimen 1 is conveyed by the charging carrier wave ffB, the air cylinder 8 of the take-out device C rises at the terminal end thereof, pushes up the disk 11, and lifts the conveyor belt 4.
．． 4 is transferred onto the disk 11. Then, in this state, the rotation motor 10 starts rotating. At the same time, the detection element 43 detects the shape of the thin piece 1 to be tested.

In detecting the shape, by detecting the level of reflected light irradiated onto the test thin piece 1 with the detection element 43, it is possible to know which rotation angle position the notch portion 1a of the test thin piece 1 is located.

After detecting the position of the notch 1a in this manner, the rotation of the disk 11 is stopped at a position where the cough notch 1a faces the chuck 14 of the positioning device ffD. The reason for detecting the position of the notch portion 1a is to know where a defect is located on the thin piece 1 to be inspected when detecting a surface defect, which will be described later. After f8, the air cylinders 17, 19 of the positioning device are operated to clamp the thin piece 1 to be tested between the chucks 13, 14, thereby centering the thin piece 1 to be tested.

After the centering of the thin piece 1 to be tested is completed, the thin piece 1 to be tested is attracted and held through the inner cylinder 31 which communicates with a negative pressure source (not shown). At this time, the chucks 13 and 4 are respectively moved back and returned.

Next, the swing motor 29 of the handler device H is operated to move the swing arm 26 clockwise by 90° in FIG.
The specimen 1 is rotated and held by suction, and the specimen 1 is moved to the upper part of the water tank E. Then, the motor 33 is operated at this position, and the pinion 34. The sample holder 〇 is lowered through the rack 35 of the outer cylinder 30, and the test specimen l is immersed into the water tank E as shown in Figs. 2, 3, 6, etc.

Thereafter, the rotation motor 37 is operated to rotate the inner cylinder 31, thereby rotating the thin piece 1 to be inspected which is attracted to the lower end suction jig 41 of the inner cylinder 31. Then, the probe 2 is excited to irradiate the surface of the test thin piece 1 (the lower surface of the sixth W!J) with an ultrasonic beam. In this embodiment, each probe 2 is positioned so that the ultrasonic beams emitted from the three probes 2 converge at one point on the surface of the test thin piece 1.

In flaw detection, two probes 2 are alternately excited at the same time, and after the excitation, reflected echoes are received by all the probes 2. This is to improve flaw detection accuracy.

The reflected echoes received by the probe 2 are diffusely reflected by surface defects. Therefore, by receiving these reflected echoes, it is possible to detect the presence or absence of defects, and by detecting the reception level of the reflected echoes, defects can be detected. It is possible to detect the size of

However, when the probe 2 is fixed, only a part of the thin section 1 to be inspected can be inspected.Therefore, with this device A, the inspection can be performed in a straight line from the outer edge of the rotating thin section toward the center. By moving the probe 2, front flaw detection on the surface of the test thin piece 1 can be performed in a short time. When carried out using this method, the flaw detection time is about 10 seconds for a test thin piece l having a diameter of 6 inches, for example. The above linear movement of the probe 2
The rotation of the motor 22, which is performed by operating the motor 22 shown in FIGS. By the way, the position of the notch 1a of the thin piece 1 to be tested is determined in advance, and since the thin piece 1 to be tested is rotated from that state for flaw detection, the angle of rotation of the thin piece 1 to be tested and the probe 2 are It is possible to know the position of the defect from the feed amount. If necessary, the position signal of this defect can be extracted and displayed.

Therefore, during this flaw detection, in this embodiment, fresh water is supplied from the bottom side of the water tank E and uniformly overflows from the upper edge, thereby removing floating objects and eliminating the generation of bubbles. Further, when the thin piece 1 to be tested is immersed, air bubbles adhering to the surface of the thin piece 1 to be tested are removed by a horizontal flow of fresh water formed at the top of the water tank. This results in
It is possible to improve flaw detection accuracy.

After detecting defects on the surface of the test thin piece 1 in this way, the rotating motor 37 is stopped, the lifting motor 33 is reversed, the sample holder G is raised, and the test thin piece 1 is placed in the water tank E. move it above. At this time, the motor 22 of the feed mechanism F rotates in reverse, causing the probe 2 to move back and return in a straight line.

Next, the rotation motor 29 is reversed, and the rotation arm 26
1 to the position shown by the chain line in FIG. That is, the test thin piece l after flaw detection is moved to the transport device J for carrying out. After that, by cutting off the communication to the negative pressure source, the test thin section l
Release the suction hold and transfer the thin piece 1 to be tested onto the conveyor belt 4.4. The conveyor belt 4.4 conveys the test specimen 1 to the empty carrier 3 and stores it therein. After the rotating arm 26 has transferred the thin piece 1 to be tested, it retreats back to above the disk 11 of the removal bag rtC, and waits at this position for the next cycle.

This completes one cycle. Thereafter, by repeating the above-mentioned operations, all of the test thin pieces 1 in the carrier 3 on the loading side are inspected for flaws, and the carrier 3 is replaced with a new carrier 3. This also applies to the unloading side.

〔Effect of the invention〕

As explained above, according to the present invention, all defects on the surface of a thin piece such as a silicon wafer can be automatically detected. In addition, defects are detected by the probe by rotating the thin section to be tested and moving the probe in a straight line from the outer edge of the thin section to the center, which significantly shortens the time required for flaw detection. Is possible. In addition, it is possible to detect where the defect is present in the thin section to be inspected from the rotation angle signal of the thin section to be inspected and the position signal of the probe.

In addition, fresh water is supplied from the bottom side of the water tank and overflows from the upper periphery, making it possible to remove air bubbles and floating objects that may impede flaw detection. Furthermore, air bubbles that adhere to the surface (lower surface side) of the test thin section when the test thin section is immersed in the water tank can be removed by the horizontal flow at the top of the water tank.
Highly accurate ultrasonic flaw detection is possible.

[Brief explanation of drawings]

Fig. 1 is a plan view showing the entire defect detection device, Fig. 2 is a front view thereof, Fig. 3 is a right side view thereof, Fig. 4 is a plan view showing the positioning device, and Fig. 5 is a plan view showing the extracting device. Right side view, Figure 6 is a front view showing the Hendler device, sample holder, and feeding device, Figure 7 is an enlarged partial cross-sectional view of the sample holder, Figure 8 is a vertical cross-sectional view of the sample holder, Figure 9 is a front view showing the Hendler device, sample holder, and feeding device. FIG. 8 is a sectional view taken along the line ■-■ in FIG. 8, and FIG. 10 is a plan view showing another embodiment of the positioning device. 1...Test thin section B...Charging conveyance device 11
...Circle ic...Takeout device D...Positioning device E...Water tank 2...Probe
F...Feeding device G...Sample holder H...Handler device J...Transportation device for carrying out Patent applicant Sumitomo Kinmei Industries Co., Ltd. Agent Patent attorney Toshihiko Uchida Figure 1. Figure 2 3 ■ ``゛1

Claims (1)

[Claims] 1. A conveying device for loading a thin piece to be tested, a thin piece taking-out device that is installed at the terminal end of the conveying device and is equipped with a disk that rotates with a thin piece to be tested placed thereon, and a mechanism for lifting and lowering the disk; a positioning device that chucks the thin piece placed on the disk from a direction perpendicular to the conveying direction and centers the thin piece to be tested;
It has a water tank, a probe feeding device that supports a probe disposed in the water tank and reciprocates the probe in a straight line, and has a suction part for the test thin piece at the lower tip, and rotates. and a sample holder with a double tube structure that can be raised and lowered freely, a handler device that rotates the sample holder between the disk and the water tank, and a transport device for carrying out the test thin section installed within the rotation range of the sample holder. A thin section surface defect detection device characterized by comprising: