JP2003214822A - Apparatus and method for measuring depth and cutting device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To enable measuring of a difference of positions of a depth direction between two points by an efficient and accurate method. <P>SOLUTION: The method for measuring the depth comprises the steps of emitting a laser beam 24 to a surface W1 of an object W to be measured via a convex lens to image the surface W1 while approaching an optical system 11 to or separating the system 11 from the object W, detecting a focus in which the illuminating area becomes minimum, obtaining the position of the system 11 at that time as a first focus detecting position, then illuminating the beam 24 to the bottom G1 of the groove formed on the surface W1 of the object W via the convex lens to image the bottom G1 while approaching the system 11 to or separating the system 11 from the object W, detecting the focus in which the illuminating area becomes minimum, obtaining the position of the system 11 at that time as a second focus detecting position, calculating the difference between the first focus detecting position and the second focus detecting position, and obtaining the depth of the groove. <P>COPYRIGHT: (C)2003,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a depth measuring device and a depth measuring method capable of obtaining a position difference in the depth direction between two points, and a cutting device equipped with the depth measuring device.

[0002]

2. Description of the Related Art As shown in FIG. 7, a semiconductor wafer W1 in which a plurality of circuits such as ICs and LSIs are divided by streets S and formed to dice by vertically and horizontally cutting the streets S, each circuit is individually cut. Of semiconductor chips C.

Further, as shown in FIG. 8, a cutting groove G having a depth corresponding to the thickness of the final semiconductor chip is formed in advance on the street of the semiconductor wafer W2, and the back surface is ground to form the cutting groove G. Similarly, the semiconductor chip C can be formed for each circuit by a technique called “pre-dicing” in which the semiconductor chip C is exposed and divided into individual semiconductor chips C.

As described above, the cutting groove G is divided into individual semiconductor chips C by the so-called pre-dicing technique.
Must be formed to a depth corresponding to the final thickness of the semiconductor chip C. Therefore, when forming the cutting groove G, the cutting direction of the cutting blade used for cutting (vertical direction in the illustrated example) The cutting groove G having a desired depth is formed by accurately controlling the position.

[0005]

However, since an error may occur in the control of the position of the cutting blade in the cutting direction, the depth of the cutting groove assumed based on the position of the cutting blade and the actual formation of the cutting groove. The actual depth of the cut groove may not match, and in this case, the accuracy of the thickness of the final semiconductor chip is adversely affected.

Also in normal dicing, V-shaped grooves (V grooves) are formed in advance on the streets so that the corners on the surface side of the final semiconductor chip are chamfered, and then completely cut. In this case, it is necessary to detect the position of the V-groove forming blade in the cutting direction when forming the V-groove so that the final shape of the semiconductor chip does not vary.

However, in this case as well, an error may occur in the control of the position of the blade for forming the V-groove, so that the depth of the V-groove actually formed cannot be accurately recognized. There's a problem.

On the other hand, although it is possible to measure the depth of the cutting groove by using a stylus sensor or the like, there is a problem that the semiconductor wafer may be damaged and it is inefficient. Such a problem is not limited to the processing of semiconductor wafers, and is not limited to the groove depth, and is common when measuring the position difference in the depth direction between two points having different heights. It is also a problem to have.

Therefore, when measuring the difference in position in the depth direction between two points, there is a problem in that the depth value can be accurately measured by an efficient method.

[0010]

As a concrete means for solving the above problems, the present invention provides a convex lens for enlarging the first surface and the second surface of an object to be measured and the first surface and the second surface. An optical system including at least an imaging unit for imaging a surface, a laser light irradiation unit for irradiating a measurement object with a laser beam via a convex lens, and a drive for moving the optical system toward and away from the measurement object Provided is a depth measuring device including a driving unit for driving the image processing unit and an image processing unit for processing an image acquired by an optical system.

In this depth measuring apparatus, the image processing means binarizes the acquired image and the number of pixels for calculating the number of pixels at the laser light irradiation position from the binarized image. A calculation unit, a storage unit that stores the calculated number of pixels, a focus detection unit that detects each focus position based on each number of pixels, and a position difference of the optical system when each focus position is detected are calculated. And a depth calculation means for calculating the depth from the first surface to the second surface, wherein the optical system moves in a direction of approaching or separating from the object to be measured. Images are sequentially acquired for each of the two surfaces, the storage unit stores a plurality of images acquired for the first surface and the second surface in association with the position of the optical system, and the focus detection unit sets Of multiple images of the It is an additional requirement to select the image with the smallest number of pixels from each and calculate the difference in the position of the optical system corresponding to each selected image as the depth from the first surface to the second surface. .

The present invention also provides an optical system including at least a convex lens for enlarging the first surface and the second surface of an object to be measured, and an image pickup means for picking up an image of the first surface and the second surface, and a convex lens. A laser light irradiation unit that irradiates a measurement object with a laser beam via a drive unit, a driving unit that drives the optical system in a direction to move toward and away from the measurement object, and image processing that processes an image acquired by the optical system. A depth measuring method for measuring a depth from a first surface to a second surface by using a depth measuring device configured by Alternatively, the surface of the object to be measured is irradiated with laser light through a convex lens while being separated, and the surface is imaged, and the image processing means detects the focus at which the irradiation area is the minimum, and the position of the optical system at that time is set to the first position. First as the focus detection position Focus detection process and imaging the groove bottom by irradiating the groove bottom of the groove formed on the surface of the measurement object with a laser beam through a convex lens while moving the optical system toward or away from the measurement object by the driving means. Then, the second focus detection step of detecting the focus at which the irradiation area is minimized by the image processing means and obtaining the position of the optical system at that time as the second focus detection position, the first focus detection position, and the second focus detection position. And a depth calculation step of calculating the difference between the focus detection position and the depth of the cutting groove.

Further, in this depth measuring method, the focus which minimizes the irradiation area of the laser beam is detected by selecting the image having the smallest number of pixels at the irradiation position from the plurality of picked-up images. Requirements.

Further, the present invention provides a cutting device including at least a chuck table for holding a workpiece, a cutting means for cutting the workpiece held on the chuck table, and the depth measuring device. .

In this cutting device, the workpiece is a semiconductor wafer, and the depth measuring device additionally measures the depth of the cutting groove formed on the surface of the semiconductor wafer by the action of the cutting means. As a requirement.

According to the depth measuring device and the depth measuring method configured as described above, the laser light is irradiated to two points having different depths of the object to be measured, and the irradiation position is imaged for each of them to focus the light. By detecting, the difference in position in the depth direction between any two points can be measured efficiently and accurately.

Further, in the cutting device configured as described above, the depth measuring device has an alignment function for detecting a street where a cutting groove is to be formed and a depth measuring function for measuring the depth of the cutting groove. You can have both.

[0018]

BEST MODE FOR CARRYING OUT THE INVENTION As an example of an embodiment of the present invention, a depth measuring device 10 shown in FIG. 1 will be described. The measuring device 10 includes an optical system 11 for enlarging and imaging the surface of an object to be measured, and a driving unit 12 for driving the optical system 11.
An image processing unit 13 for processing the captured image and a laser light irradiation unit 14 for irradiating the measurement object with laser light.

In the optical system 11, the laser light irradiation section 1
The laser light emitted from the laser beam 4 is reflected by the half mirror 15, is condensed by the convex lens 16, and is irradiated onto the measurement object. The reflected light is picked up by the image pickup means 17.

The image pickup means 17 is, for example, (256 × 25).
It is composed of a CCD camera or the like having a pixel number of about 6), converts the formed image into an image composed of digital information, and transfers the image to the image processing means 13.

The driving means 12 is composed of a pulse motor 18 and a ball screw 19 connected to the rotary shaft of the pulse motor 18, and the ball screw 19 has an optical system 11 therein.
A nut (not shown) provided on the support portion 20 for supporting the ball screw 19 is screwed, and the ball screw 19 is driven by the pulse motor 18 that rotates upon receiving the supply of the pulse signal to rotate the ball screw 19 so that the support portion 20 moves up and down. The optical system 11 moves up and down accordingly.

The measuring object 22 is the holding unit 2 in the illustrated example.
The laser beam 24 emitted from the laser beam irradiator 14 is radiated to the measurement object 22 via the convex lens 16.

The image processing means 13 is provided with a CPU, a memory, etc., and is capable of performing calculations on the image information transferred from the image pickup means 17 and storing the information. Functionally, a binarization unit 25 that binarizes the acquired image, a pixel number calculation unit 26 that calculates the pixel number of the binarized image, and a storage unit 27 that stores the calculated pixel number.
And a focus detection unit 28 that detects a focus based on the number of pixels stored in the storage unit 27, and a depth calculation unit 29 that calculates a depth based on the position of the detected focus.
Further, the depth value obtained by the depth calculation means 29 can be displayed on a display device or the like, or can be transferred to a device that performs control using the depth value.

As an example of the measuring method using the measuring device 10 of FIG. 1, a method of measuring the depth of the cutting groove G formed in the semiconductor wafer W as shown in FIG. 2 will be described.

In FIG. 2, the semiconductor wafer W is suction-held by the holder 30, and on the surface thereof, streets are formed in a lattice pattern at predetermined intervals, and a large number of chip regions (later described) are divided by the streets. A circuit such as an IC or an LSI is formed in C which is divided into semiconductor chips. Further, a cutting groove G which does not penetrate to the back surface is formed along the street.

Next, the procedure for measuring the depth of the cutting groove G will be described with reference to the flowchart shown in FIG.
As shown in FIG. 4, first, laser light 24 is passed from the optical system 11 to the surface W1 of the semiconductor wafer W, that is, a region (first surface) where the cutting groove G is not formed, via the convex lens 16.
To illuminate and image. Then, the image acquired by imaging is transferred to the image processing means 13 (step S11).

In the image processing means 13, the images picked up by the binarizing means 25 are successively binarized (step S12). By binarizing, it is possible to distinguish between a portion irradiated with laser light and a portion not irradiated with laser light by a pixel, and therefore the number of pixels of the portion irradiated with laser light is calculated by the pixel number calculation means 26 (step). S1
3). Then, the calculated number of pixels is stored in the storage unit 27 (step S14). Steps S11, S12, S
13 and S14 are sequentially executed while moving the optical system 11 in the vertical direction by the drive by the driving unit 12, that is, moving in the direction of moving toward or away from the semiconductor wafer W, and an image is individually displayed for each position of the optical system. The number of pixels of each image obtained is calculated and stored in the storage means 27.

For example, each time the pulse signal is supplied once to the pulse motor 18 constituting the driving means 12 and the optical system 11 moves in the vertical direction, steps S11, S12 and S1 are performed.
3, S14 is executed, the number of pixels of the portion irradiated with the laser beam is sequentially calculated, and the obtained number of pixels is stored in the storage unit 13 included in the image processing unit 13.

Since the vertical position of the optical system 11 is determined by the number of pulses supplied to the pulse motor 18, as shown in FIG. 5, the vertical position of the optical system 11 and the position of the optical system 11 obtained based on the number of pulses are determined. The calculated number of pixels is stored in the storage means 27 in association with each other. In the pulse motor 18, the amount of rotation per pulse is set, and the ball screw 19 rotates accordingly, so that the image processing means 1
In 3, the amount of movement of the optical system 11 can be obtained according to the number of pulses. The number of pulses can be taken out from the pulse motor 18 or a control unit (not shown) that controls the pulse motor 18 and recognized.

The vertical position of the optical system 11 is determined by predetermining a certain reference position and using the displacement amount with the reference position as the reference. Therefore, it is necessary to set the reference position before measurement. The reference position can be set arbitrarily, and for example, the optical system 11 when focusing on the surface of the holding unit 30.
The position of is the reference position Z0.

After the reference position is determined, while the optical system 11 is moved toward or away from the semiconductor wafer W from the reference position, the surface is repeatedly imaged until, for example, N pulses are reached (step S15). The number of pixels is stored in the storage means 27.

After all the pixel numbers for N times have been stored in the storage means 27, the one having the smallest pixel number is selected from among them (step S16). When the number of pixels becomes the minimum,
Since the irradiation area of the laser beam is the minimum, the focus detection means 28 considers that the surface W1 of the image captured at this time is the most expanded and in-focus, and the optical system 11 at that time is in focus. The position is set to the first focus detection position Z1 (step S17). Steps S11 to S17 are the first focus detection step P1.

Then, the optical system 11 is placed directly above the cutting groove G.
Is moved to image the groove bottom G1, the focus of the groove bottom G1 is detected by the same method as described above, and the position of the optical system 11 at that time is obtained. In this case, the reference position uses the previously determined position.

Explaining with reference to the flow chart of FIG. 3, while moving the optical system 11 in the direction of moving toward or away from the semiconductor wafer W, the groove bottom G1 of the cutting groove G (second surface ) Is imaged (step S2
1) After binarizing the image (step S22), the number of pixels in the portion irradiated with the laser light is sequentially calculated (step S22).
23), the vertical position of the optical system and the obtained number of pixels are associated and stored in the storage means 27 provided in the image processing means 13 (step S24).

The steps S21 to S24 are repeated until the number of pulses reaches N times, for example, and the image of the groove bottom G1 is taken (step S25), and the number of pixels for each of them is stored in the storage means 27.

After all the pixel numbers for N times have been stored in the storage means 27, the one having the minimum pixel number is selected (step S26). Then, in the focus detection means 28, the image captured at this time is regarded as the one that is most expanded and focused at the groove bottom G1, and the optical system 1 at that time is considered.
The position of 1 is set as the second focus detection position Z2 (step S
27). Steps S21 to S27 are the second focus detection step P2.

Since both the surface W1 of the semiconductor wafer W and the groove bottom G1 of the cutting groove G are imaged using the optical system 11, their focal lengths are equal. Therefore, when the first focus detection position Z1 and the second focus detection position Z2 are obtained, (Z1
By calculating -Z2), the depth in relation to the surface W1 of the groove bottom G1 can be obtained (depth calculation step P3).

Next, the depth measuring device 10 is mounted on the cutting device 40 shown in FIG. 6, the cutting groove is formed on the surface of the semiconductor wafer W, and the depth of the cutting groove is measured.

The cutting device 40 shown in FIG. 6 has a cassette mounting area 42 in which a wafer cassette 41 in which the semiconductor wafer W is accommodated is mounted, and the semiconductor wafer W is unloaded from the wafer cassette 41 and transferred to the wafer cassette 41. The carrying-in / out means 43 for carrying in the semiconductor wafer W, the carrying means 44 for carrying the semiconductor wafer W, the chuck table 45 for holding the semiconductor wafer W, and the street on which the cutting groove is to be formed are detected and formed. The depth measuring device 10 for measuring the depth of the cutting groove and the cutting means 46 for cutting the street are provided.

When a cutting groove is to be formed on the surface of the semiconductor wafer W by using the cutting device 40, a plurality of semiconductor wafers W are stored in the wafer cassette 41, and the semiconductor wafer W is unloaded from the wafer cassette 41 by the loading / unloading means 43. It is temporarily placed in the temporary placement area 47. Then, after that, the particles are adsorbed by the conveying means 44 and are conveyed to the chuck table 45 by the turning movement of the conveying means 44, placed thereon, and held by suction.

The chuck table 45 holding the semiconductor wafer W is positioned immediately below the depth measuring device 10 by moving in the + X direction. Then, the depth measuring device 10 detects a street to be cut.

Here, the depth measuring device 10 is constructed as shown in FIG. 1 and has a function of measuring the depth of the cutting groove and a function of detecting the street.

A key pattern image indicating the position of the street is stored in advance in the storage means 27 provided in the image input means 13 constituting the depth measuring device 10. On the other hand, the image actually captured by the optical system 11 while moving in the Y-axis direction is binarized by the binarizing means 25 to clarify the actual shape. Then, by performing pattern matching with the previously stored key pattern image previously stored in the storage means 27 in the image processing means 13, the street to be cut can be detected.

In the cutting means 46, a cutting blade 48 is attached to the tip of a spindle arranged in the Y-axis direction, and the cutting blade 48 rotates as the spindle rotates.
Is configured to rotate.

The depth measuring device 10 and the cutting means 46 are integrally formed, and the cutting means 46 is the measuring device 10.
Since the movement of the cutting blade 48 is interlocked with the movement of the cutting blade 48 in the Y-axis direction, the street to be cut is detected by the pattern matching, and at the same time, the street and the cutting blade 48 are automatically aligned (aligned). ).

Therefore, the chuck table 45 is further + X.
When the cutting means 46 descends while moving in the direction and the cutting blade 48 rotates at high speed, a cutting groove is formed on the detected street. At this time, a cutting groove having a desired depth is formed by precisely controlling the cutting depth of the cutting blade 48.

When the chuck table 45 is reciprocally moved in the X-axis direction and the cutting means 46 is moved in the Y-axis direction at every street interval to form cutting grooves on each street, the cutting is performed on all the streets in the same direction. Grooves are formed.

Further, when the chuck table 45 is rotated by 90 degrees and the same cutting as described above is performed, cutting grooves G are formed vertically and horizontally on all the streets like the semiconductor wafer W2 shown in FIG.

The depth of the cutting groove G formed in this way is not always uniform due to an error in the cutting depth of the cutting blade 48 and the like, so it is necessary to measure the depth in the cutting groove G later. May occur. In that case, the chuck table 15 holding the semiconductor wafer W in which the first one cutting groove G is formed is moved in the −X direction and positioned directly below the depth measuring device 10, and the procedure of the flowchart of FIG. 3 is performed. Therefore, the depth of the cutting groove G can be measured. Then, by displaying the depth value on the monitor 49 or the like, it is possible to confirm whether or not the depth is the desired depth.

Therefore, if the depth is not accurate, the depth can be precisely controlled by adjusting the depth of the cutting groove on the street from the second and performing cutting. In particular, when the depth measuring device 10 is mounted on the cutting device 40, the depth can be measured immediately after the formation of the cutting groove, and if the depth is not the desired depth, the adjustment can be performed immediately before cutting. , Very efficient.

When a semiconductor chip is manufactured by a so-called dicing method, the depth of the cutting groove corresponds to the final thickness of the semiconductor chip. Therefore, the depth of the cutting groove is accurately controlled as described above. What to do is especially important.

In the present embodiment, the case where the depth measuring device is mounted on the cutting device has been described as an example.
The depth measuring device can be used alone or mounted on another device.

Further, the depth measuring apparatus and the depth measuring method of the present invention are not limited to the case of measuring the depth of the groove,
It can be widely used when measuring the step difference between any two points.

[0054]

As described above, according to the depth measuring device and the depth measuring method of the present invention, laser light is irradiated to two points having different depths on the object to be measured, and the irradiation positions are respectively set. By imaging and detecting the focus, any 2
Since the position difference in the depth direction between the points can be efficiently and accurately measured, the depth of the groove or the like can be easily managed.

Further, when the depth measuring device according to the present invention is mounted on a cutting device, it has an alignment function for detecting a street where a cutting groove is to be formed and a depth measuring function for measuring the depth of the cutting groove. Since both can be provided, the cutting device can be constructed at low cost without increasing the size of the device.

Further, since the depth of the cutting groove formed by cutting can be measured immediately, the cutting groove having an accurate depth can be efficiently formed.

[Brief description of drawings]

FIG. 1 is an explanatory diagram showing a configuration of a depth measuring device according to the present invention.

FIG. 2 is a perspective view showing how the depth of a cutting groove formed in a semiconductor wafer is measured using the same depth measuring device.

FIG. 3 is a flowchart showing a depth measuring method according to the present invention.

FIG. 4 is a front view showing a state in which the depth of a cutting groove formed in a semiconductor wafer is measured using the same depth measuring method.

FIG. 5 is an explanatory diagram showing information stored in a storage unit of an image processing unit that constitutes the depth measuring device according to the present invention.

FIG. 6 is a perspective view showing a cutting device according to the present invention.

FIG. 7 is a perspective view showing a semiconductor wafer.

FIG. 8 is a perspective view showing a semiconductor wafer having a cutting groove formed on its surface.

[Explanation of symbols]

10 ... Measuring device 11 ... Optical system 12 ... Driving means 13 ... Image processing means 14 ... Laser light irradiation unit 15 ... Half mirror 16 ... Convex lens 17 ... Imaging means 18 ... Pulse motor 19 ... Ball screw 20 ... Support section 22 ... Object to be measured 23 ... Holding part 24 ... Laser light 25 ... Binarizing means 26 ... Pixel number calculating means 27 ... Storage means 28 ... Focus detection means 29 ... Depth calculation means 30 ... Holding unit 40 ... Cutting device 41 ... Wafer cassette 42 ... Cassette area 43 ... Loading / unloading means 44 ... Conveying means 45 ... Chuck table 46 ... Cutting means 47 ... Temporary storage area 48 ... Cutting blade 49 ... Monitor W ... Semiconductor wafer G ... Cutting groove G1 ... Groove bottom Z0 ... reference position Z1 ... first focus detection position Z2 ... Second focus detection position

Claims (7)

[Claims] 1. An optical system including at least a convex lens for enlarging a first surface and a second surface of an object to be measured and an image pickup means for picking up an image of the first surface and the second surface, and the convex lens. A laser light irradiating unit that irradiates the measurement target with laser light, a driving unit that drives the optical system in a direction to move toward and away from the measurement target, and an image acquired by the optical system. Depth measuring device including image processing means for 2. The image processing means, the binarizing means for binarizing the acquired image, the pixel number calculating means for calculating the number of pixels at the irradiation position of the laser beam from the binarized image, and the calculation. The storage means for storing the number of the selected pixels, the focus detection means for detecting the respective focus positions based on the respective number of the pixels, and the difference between the positions of the optical system when the respective focus positions are detected by the first The depth measuring device according to claim 1, further comprising: a depth calculating unit that calculates a depth from the surface to the second surface. 3. The optical system sequentially acquires images for each of the first surface and the second surface while moving in the direction of approaching or separating from the measurement object, and the storage means stores the first surface and the second surface. The plurality of images acquired for the second surface are stored in association with the position of the optical system, and the focus detection unit has the largest number of pixels among the plurality of images for the first surface and the second surface. 3. The depth according to claim 2, wherein each of the images having a small number of pixels is selected, and the difference in the position of the optical system corresponding to each of the selected images is calculated as the depth from the first surface to the second surface. Measuring device. 4. An optical system, which comprises at least a convex lens for enlarging the first surface and the second surface of an object to be measured, and an imaging means for imaging the first surface and the second surface, and the convex lens. A laser light irradiating unit that irradiates the measurement target with laser light, a driving unit that drives the optical system in a direction to move toward and away from the measurement target, and an image acquired by the optical system. A depth measuring method for measuring a depth from the first surface to the second surface by using a depth measuring device including an image processing means for controlling the optical system by the driving means. While approaching or moving away from the measurement target, the surface of the measurement target is irradiated with laser light through the convex lens to capture an image of the surface, and the image processing means detects a focal point that minimizes the irradiation area. The position of the optical system at that time as the first focus detection position And a first focus detection step which is performed by the driving means while moving the optical system toward or away from the measurement target through the convex lens at the bottom of a groove formed on the surface of the measurement target. Second focus detection in which the bottom of the irradiation area is detected by the image processing means by irradiating laser light to image the groove bottom, and the position of the optical system at that time is obtained as the second focus detection position. A depth measuring method comprising: a step; and a depth calculating step of calculating a difference between the first focus detection position and the second focus detection position to obtain a depth of the groove. 5. The depth measuring method according to claim 4, wherein the focus that minimizes the irradiation area of the laser light is detected by selecting the image with the smallest number of pixels at the irradiation position from the plurality of captured images. . 6. A cutting device comprising at least a chuck table for holding a workpiece, a cutting means for cutting the workpiece held on the chuck table, and the depth measuring device according to claim 1. 7. The workpiece is a semiconductor wafer, and the depth measuring device measures the depth of the cutting groove formed on the surface of the semiconductor wafer by the action of the cutting means.
The cutting device described in.