TWI761609B - Wafer processing method - Google Patents

Abstract

The problem of the present invention is that it can be solved even if the operator does not select the processing conditions during wafer processing. The solution is to have: a process of selecting one processing condition from the processing condition list (K); comparing the memory macroscopic image memorized in the selected processing condition with the first image captured by the low-magnification imaging means (201) , the first judgment process of judging whether they are consistent; When the first judgment process judges that the memory macroscopic image is inconsistent with the first image, it returns to the first repeated process of the selection process; In the first judgment process, it is judged that the memory macroscopic image and the first image are If they match, compare the memory microscopic image memorized for the processing conditions used in the first judgment process with the second image captured by the high-magnification imaging means (202), and judge whether they are consistent in the second judgment process; In the second judgment process When it is judged that the memory micro-image is inconsistent with the second image, it returns to the second repetition process of the selection process; and when the second judgment process determines that the memory micro-image is consistent with the second image, the processing conditions used in the second judgment process are used. Processing engineering for processing wafers.

Description

Wafer processing method

The present invention relates to a processing method for processing wafers.

A cutting device that cuts a plate-shaped wafer with a cutting blade along a planned dividing line transfers the wafer to a holding table, and performs alignment in order to identify the planned dividing line where the cutting edge is to be cut (for example, refer to Patent Document 1). .

When starting the cutting process, the operator (worker) inputs the cutting feed rate, the cutting depth, the index feed amount, and the machining conditions (device data) for setting the target pattern used for the alignment, etc., which are necessary for the cutting process. Alternatively, the operator can select from a machining condition list (device data list) previously input to the cutting device and tabulated.

On the other hand, when starting the cutting process, instead of asking the operator to input or select the processing conditions, a two-dimensional code is arranged on the cassette for accommodating the wafer or the ring frame for supporting the wafer, so that the A method in which a reading unit included in a cutting device reads the two-dimensional code and selects processing conditions (for example, refer to Patent Document 2). [Prior Art Literature] [Patent Literature]

[Patent Document 1] Japanese Patent Application Laid-Open No. 07-106405 [Patent Document 2] Japanese Patent Application Laid-Open No. 09-306873

(The problem to be solved by the invention)

When the operator selects the processing conditions from the processing condition list, the selection of the processing conditions may be wrong, and the processing conditions corresponding to the wafers held by the holding table are not selected. For example, the index feed amount is different from the appropriate value. An alignment error occurs and the fully automatic operation of the cutting device may be interrupted. In this case, it is necessary to start the fully automatic operation of the cutting device after the operator selects the correct machining conditions again. A problem occurs when the two-dimensional code is not placed in the box or ring frame when the cutting device is used to read the two-dimensional code.

Therefore, when processing wafers (for example, cutting processing or laser processing), it is possible to solve the problem even if the operator does not select the processing conditions, and even if the two-dimensional code is not arranged in a box or a ring frame, it can be solved. The subject of selection of processing conditions. (means to solve the problem)

The present invention for solving the above-mentioned problems is based on holding a wafer on which devices are formed in a region defined by a planned dividing line by a holding platform, and a wafer using a processing device processed along the planned dividing line of the wafer. A processing method, characterized in that: the processing apparatus is provided with a low-magnification imaging means and a high-magnification imaging means, and includes: a holding process for holding a wafer by the holding platform; A selection process for selecting one processing condition; a first judgment for judging whether the two images are consistent by comparing the memory macroscopic image stored in the processing condition selected in the selection process with the first image captured by the low-magnification imaging means process; when the first judgment process determines that the memory macroscopic image is inconsistent with the first image In the second judgment process of comparing the memory microscopic image memorized for the processing conditions used in the first judgment process with the second image captured by the high-magnification imaging means, and judging whether the two images are consistent; When the judgment process determines that the memory microscopic portrait is inconsistent with the second portrait, return to the second repeat process of the selection process; and when the second judgment process determines that the memory microscopic portrait is consistent with the second portrait, use memory as useful The processing process of processing the wafer under the processing conditions of the memory microscopic image in the second judgment process. [Effect of invention]

The wafer processing method of the present invention is implemented using a processing apparatus provided with a low-magnification imaging means and a high-magnification imaging means, and includes: a holding process for holding the wafer by the holding platform; A selection process for selecting a processing condition from a list of processing conditions; comparing the memory macroscopic image stored in the processing condition selected in the selection process with the first image captured by the low-magnification imaging method, and determining whether the two images are consistent. 1. Judgment process; When the 1st judgment process judges that the memory macroscopic image is inconsistent with the first image, it returns to the 1st repetition process of the selection process; When the 1st judgment process judges that the memory macroscopic image is consistent with the first image, the comparison is memorized It is used in the second judgment process of judging whether the two images are consistent with the memory microscopic image and the second image captured by the high-magnification imaging means for the processing conditions of the first judgment process; When the 2 images are inconsistent, return to the second repeat process of the selection process; and when the second judgment process judges that the memory microscopic image is consistent with the second image, process the crystal with the processing conditions of the memory microscopic image that is useful in the second judgment process. For the machining process of circles, it is not necessary to select the machining conditions of the operator, and the selection of the machining conditions can be performed even if there is no QR code in the ring frame, etc.

The processing apparatus 1 shown in FIG. 1 used in order to implement the wafer processing method of the present invention cuts and is held on the holding table 30 by a first processing means 61 and a second processing means 62 having a rotating cutting blade 613 . The device of the wafer W. In addition, the processing apparatus used in order to implement the wafer processing method of the present invention is not limited to a cutting apparatus such as the processing apparatus 1, and may irradiate the wafer with a laser of a predetermined wavelength to form a modified layer or Laser processing equipment for cutting wafers.

On the base 10 of the processing apparatus 1, the processing feed means 13 which moves the 1st processing means 61 and the holding table 30 relatively in the X-axis direction of a processing feed direction is arrange|positioned. The processing and feeding means 13 is composed of a ball screw 130 having an axis in the X-axis direction, a pair of guide rails 131 arranged in parallel with the ball screw 130, a motor 132 for rotating the ball screw 130, and a nut inside. The movable plate 133 is formed by the ball screw 130 and the bottom of which is slidably connected to the guide rail 131 . Then, when the motor 132 rotates the ball screw 130 , the movable plate 133 is guided by the guide rails 131 to move in the X-axis direction, and the holding platform 30 arranged on the movable plate 133 follows the movement of the movable plate 133 . And move in the X-axis direction.

The holding table 30 for holding the wafer W is, for example, a circular plate shape having an outer shape, and includes a suction portion 300 formed of a porous member for sucking the wafer W, and a frame 301 for supporting the suction portion 300 . The exposed surface, that is, the horizontal holding surface 300a that is in line with the surface of the frame body 301, is sucked and held by the wafer W. The holding table 30 is rotatable around the axis in the Z-axis direction by the rotating means 32 arranged on the bottom surface side of the holding table 30, and the rotation center of the holding table 30 and the center of the holding surface 300a are aligned. Around the holding table 30 , four fixing clamps 34 for holding and fixing the annular frame F that supports the wafer W are arranged equally in the circumferential direction.

On the rear side on the base 10 , the gate column 14 is erected so as to straddle the movement path of the holding platform 30 . On the front face of the gate post 14 are disposed: first index feeding means 15 for reciprocating the first machining means 61 in the Y-axis direction orthogonal to the X-axis direction and the Z-axis direction; and second machining means 62 The second index feed means 16 that reciprocates in the Y-axis direction.

The first index feeding means 15 includes, for example, a ball screw 150 having an axis in the Y-axis direction, a pair of guide rails 151 arranged in parallel with the ball screw 150 , and one end connected to the +Y direction side of the ball screw 150 . The motor (not shown) and the internal nut are screwed to the ball screw 150 and the side part is slidably connected to the movable plate 153 of the guide rail 151 . Then, when the ball screw 150 is rotated by a motor (not shown), the movable plate 153 is guided by the guide rails 151 to move in the Y-axis direction, and is disposed on the movable plate 153 via the first incision feeding means 17 The first machining means 61 of , is indexed and fed in the Y-axis direction.

The first incision feeding means 17 is capable of moving the first machining means 61 in the Z-axis direction orthogonal to the holding surface 300a of the holding table 30, and includes a ball screw 170 having an axis in the Z-axis direction, and a ball screw A pair of guide rails 171 , a motor 172 connected to the ball screw 170 , and a support member supporting the first processing means 61 , and a nut inside the ball screw 170 is screwed to the ball screw 170 and slidably connected to the guide rail 171 . 173. When the motor 172 rotates the ball screw 170, the support member 173 is guided by the pair of guide rails 171 to move in the Z-axis direction, and accordingly, the first processing means 61 moves in the Z-axis direction.

The first machining means 61 includes a main shaft 610 whose axial direction is the Y-axis direction, a casing 611 fixed to the lower end side of the support member 173 and rotatably supporting the main shaft 610, a motor (not shown) for rotating the main shaft 610, And the cutting edge 613 which is attached to the front end of the main shaft 610 and has an annular outer shape, and the cutting blade 613 rotates as the motor rotates and drives the main shaft 610 .

The second index feeding means 16 includes, for example, a ball screw 160 having an axis in the Y-axis direction, a pair of guide rails 151 arranged in parallel with the ball screw 160, a motor 162 connected to the ball screw 160, and an internal The nut is screwed to the ball screw 160 and the side part is slidably connected to the movable plate 163 of the guide rail 161 . When the ball screw 160 is rotated by the motor 162, the movable plate 163 is guided by the guide rail 151 to move in the Y-axis direction, and the movable plate 163 is provided with the second machining means 62 via the second cutting and feeding means 18. will be indexed in the Y-axis direction.

The second plunge feeding means 18 includes a ball screw 180 having an axis in the Z-axis direction, a pair of guide rails 181 arranged in parallel with the ball screw 180, a motor 182 connected to the ball screw 180, and a second support In the processing means 62 , the inner nut is screwed to the ball screw 180 , and the side part is slidably connected to the support member 183 of the guide rail 181 . When the motor 182 rotates the ball screw 180, the support member 183 is guided by the pair of guide rails 181 to move in the Z-axis direction, and accordingly, the second processing means 62 moves in the Z-axis direction.

The second processing means 62 is arranged to face the first processing means 61 in the Y-axis direction. The above-mentioned first processing means 61 and the second processing means 62 have the same configuration, and therefore the description of the second processing means 62 is omitted.

On the side surface of the casing 611 of the first processing means 61, the imaging unit 20 for imaging the wafer W is disposed. The imaging unit 20 is provided with an imaging element (not shown), an objective lens 200 capable of changing the magnification to a low magnification and a high magnification by a magnification adjustment means, and an objective lens 200 set to a low magnification for imaging. The low-magnification imaging means 201, the high-magnification imaging means 202 for imaging by the objective lens 200 set to high magnification, and the illumination means 203 for irradiating light to the wafer W attracted and held on the holding table 30, The low-magnification imaging means 201 and the high-magnification imaging means 202 can be switched to capture the wafer W from above. The imaging unit 20 and the first processing means 61 are formed integrally, and move in the Y-axis direction and the Z-axis direction in conjunction with each other. In addition, the imaging unit 20 is also arranged on the side surface of the casing of the second processing means 62 .

The processing apparatus 1 is provided with the control means 9 which controls the whole apparatus, for example. The control means 9 is connected to the machining feed means 13 , the first index feed means 15 , the first plunge feed means 17 , the rotation means 32 , and the like by wires not shown, and under the control of the control means 9 , The movement in the X-axis direction of the holding table 30 by the machining feed means 13 , the index feed action in the Y-axis direction of the first machining means 61 by the first index feed means 15 , and the first plunge feed means 17 The cutting feed operation in the Z-axis direction of the first processing means 61 and the rotation operation of the holding table 30 by the rotation means 32 are controlled.

The control means 9 includes a memory unit 90 composed of a memory element such as a memory, and the memory unit 90 stores a processing condition list (device data list) K shown in FIG. 2 . The processing condition list K is a list of multiple processing conditions corresponding to each type of wafer to be processed, and each processing condition can be identified by, for example, a processing condition No. and a processing condition ID. For example, the machining condition of No. 001 shown in the machining condition list K is that the machining condition ID is MM-SAMPLE, and the machining condition of No. 002 is that the machining condition ID is INCH-SAMPLE.

The processing conditions are data that are stored in various settings for performing appropriate cutting processing on the wafers for each type of wafer to be processed, and the various settings are based on the processing feed shown in FIG. 1 . The processing feed rate of the holding table 30 for holding the wafer by means 13, the index feed amount of the first processing means 61 by the first index feeding means 15, and the first processing means 61 by the first plunging feeding means 17 , the number of rotations of the spindle 610 of the first processing means 61, the imaging coordinate position of the X-axis Y-axis plane of the wafer sucked and held on the holding table 30 by the imaging unit 20, and the memory to be described later. Macro images, and memory micro images, which will be described later. The camera coordinate position of the wafer according to the camera unit 20 is determined based on, for example, the center of the holding surface 300a of the holding table 30 that the processing apparatus 1 can always grasp. Under suitable processing conditions, the image obtained by capturing the wafer at the camera coordinate position will be consistent with the memory macroscopic image. The index feed amount of the first machining means 61 based on the first index feed means 15 is, for example, the amount by which the first index feed means 15 moves the first machining means 61 in the Y-axis direction by only one index. The index is the distance from the center line in the width direction of a certain line to be divided S shown in FIG. 3 to the center line of the line to be divided S located beside it.

The wafer W shown in FIG. 1 is, for example, a circular plate-shaped silicon semiconductor wafer, and each device D is formed on the surface Wa of the wafer W in a lattice-shaped region demarcated by lines S to be divided. On the back surface Wb of the wafer W, a dicing tape T having a larger diameter than that of the wafer W is attached. An annular frame F having a circular opening is attached to the outer peripheral region of the adhesive surface of the dicing tape T, and the wafer W is supported by the annular frame F with the dicing tape T interposed therebetween. The state of handling. In addition, let each planned dividing line S extending in the same direction (for example, the X-axis direction in FIGS. 1 and 3 ) on the surface Wa of the wafer W be the dividing planned line S of the first channel, and on the other hand , each of the planned dividing lines S extending on the surface Wa of the wafer W in a direction that intersects with the planned dividing line S of the first scribe line (the Y-axis direction in FIGS. 1 and 3 ) is defined as the second scribe line. The predetermined line S is divided.

As shown in FIG. 3 , the same circuit pattern is formed on the surface of each device D of the wafer W. As shown in FIG. Furthermore, one pattern having a characteristic shape among the circuit patterns is preselected as a macro target PA, and the image of the PA including this macro target (image taken at a low magnification) becomes the image shown in FIG. 4(A). ) shows the memory macroscopic portrait G1. In addition, the PA of the giant optotype is formed at the same position one by one for plural devices D, for example, the corner portion (left corner) of the device D. FIG. In addition, it is preferable that the memory macroscopic image is a simple shape pattern like PA including a circular macroscopic image like the memory macroscopic image G1.

Part of the characteristics of the elements or wirings formed on the surface of the device D shown enlarged in FIG. 3 (the inside of the rectangular frame shown by the dotted line) is preselected as a micro target PC, including the micro The image of the target PC (image captured at high magnification) becomes the memory microscopic image G3 shown in FIG. 5(A). In addition, the size of the micro-target PC is smaller than the size of the macro-target PA. In addition, as the memory microscopic image G3, it is desirable that the image includes a straight line extending in the Y-axis direction and a straight line extending in the X-axis direction.

In this embodiment, for example, in the processing condition AAAA-SAMPLE of No. 003 shown in the processing condition list K shown in FIG. 2, the memory macroscopic image G1 shown in FIG. ) in the memory microscopic portrait G3. In addition, the information of the distance Lx3 in the X-axis direction between the PA of the macroscopic target and the PC of the microscopic target and the information of the distance Ly3 of the Y-axis direction of the PA of the macroscopic target and the PC of the microscopic target shown in FIG. 3 are also stored in the processing condition AAAA-SAMPLE. For example, it is assumed that the processing condition MM-SAMPLE of No. 001 shown in the processing condition list K shown in FIG. The memory macroscopic image G2 of the PB of the rectangular macroscopic target shown, and the memory microscopic image G3 shown in FIG. 5(B) . For example, the processing condition INCH-SAMPLE of No. 002 shown in the processing condition list K shown in FIG. 2 memorizes the memory macro image G1 shown in FIG. The memory microscopic image G4 of the microscopic target PD shown in FIG. 5(B) of another wafer device. In addition, the information on the distance in the X-axis direction between the PA of the macroscopic target shown in FIG. 4(A) and the PD of the microscopic target shown in FIG. 5(B) on the wafers other than the wafer W, and the PA of the macroscopic target and the microscopic target Information on the distance in the Y-axis direction of the PD is also stored in the machining condition INCH-SAMPLE.

Hereinafter, each step of the processing method of the present invention when the wafer W shown in FIGS. 1 and 3 is cut along the line S to be divided by the processing apparatus 1 shown in FIG. 1 will be described. The steps of the processing method of the present invention are carried out in the order shown in the flowchart shown in FIG. 6 .

(1) Holding process In the first holding process, the wafer W supported by the ring-shaped frame F via the dicing tape T shown in FIG. The center of the wafer W is placed on the holding surface 300 a of the holding stage 30 so that the center thereof coincides with the center of the wafer W. Then, the ring frame F is fixed by the fixing clamps 34 arranged around the holding table 30 , and the wafer W is sucked and held by a suction source not shown connected to the holding table 30 by the operation of the holding table 30 . on the holding surface 300a of the platform 30 .

(2-1) The first selection process control means 9 includes a machining condition selection unit 91 that selects one machining condition from the machining condition list K shown in FIG. 2 , and the machining condition selection unit 91 executes the main selection process . The machining condition selection unit 91 sequentially selects, for example, machining conditions with a smaller number of machining condition No. shown in the machining condition list K. Therefore, the machining condition MM-SAMPLE of the machining condition No. 001 shown in the machining condition list K shown in FIG. 2 is first selected by the machining condition selection unit 91 .

(3-1) First first imaging process Next, the holding table 30 that sucks and holds the wafer W shown in FIG. 1 is moved in the X-axis direction by the processing feeding means 13. Then, the imaging unit 20 is moved in the Y-axis direction by the first index feeding means 15 . Then, under the control of the processing feeding means 13 and the first index feeding means 15 according to the control means 9, as shown in FIG. The holding stage 30 is positioned at a predetermined coordinate position on the X-axis and Y-axis plane in such a way as to be directly under the lens 200 . In addition, the positioning of the holding table 30 under the control of the machining feed means 13 and the first index feed means 15 by the control means 9 is based on the machining condition MM-SAMPLE of machining condition No. 001.

Since the magnification of the objective lens 200 is set to a low magnification, and the set light quantity of the illumination means 203 is set to a light quantity suitable for low-magnification imaging, the imaging of the surface Wa of the wafer W by the low-magnification imaging means 201 is formed as follows: possible state. Then, at the position of the X-axis and Y-axis coordinates based on the center of the holding surface 300a of the holding table 30, the approximate center area of the surface Wa of the wafer W is imaged by the low-magnification imaging means 201 to form a relatively wide field of view. 1 image.

In principle, in the holding process performed earlier, the center of the holding surface 300 a of the holding table 30 is aligned with the center of the wafer W, and the wafer W is sucked and held by the holding table 30 . However, for example, the center of the wafer W on the holding table 30 and the center of the holding surface 300a may deviate from the center of the holding surface 300a due to a transfer deviation when the wafer W is transferred to the holding table 30 or the like. When there is a deviation between the center of the holding surface 300a of the holding table 30 and the center of the surface Wa of the wafer W, even if the selected processing conditions are appropriate processing conditions corresponding to the wafer W to be processed, there is a possibility that the If the target does not enter the imaging area of the low-magnification imaging means 201, it is judged that the memory macroscopic image does not match the first image in the first judgment process described later.

Therefore, the imaging of the surface Wa of the wafer W by the low-magnification imaging means 201 is also performed at a plurality of imaging positions other than the imaging position based on the center of the holding surface 300a of the holding table 30, and a plurality of first images are formed. . That is, for example, the first index feeding means 15 moves the imaging unit 20 in the Y-axis direction, and the rotating means 32 (not shown in FIG. 7 ) rotates the holding table 30 at a predetermined rotational speed, and the magnification is low. The imaging means 201 moves so as to draw a spiral trajectory with respect to the wafer W from the center of the holding surface 300 a toward the outside. Then, the surface Wa of the wafer W is captured per unit time by the low-magnification imaging means 201 that moves so as to draw a spiral trajectory, and imaging is performed at a plurality of imaging positions in the vicinity of the center region of the surface Wa of the wafer W, respectively. , forming a plurality of first images. Incidentally, this imaging is performed while the first index feeding means 15 moves the imaging unit 20 in the Y-axis direction for only one index.

(4-1) First judgment process for the first time Information about each first image imaged by the low-magnification imaging means 201 is sent from the imaging unit 20 to the first judgment section 92 of the control means 9. Each of the first images may be subjected to, for example, binarization processing to emphasize edges. As shown in FIG. 8 , the first determination unit 92 compares the memory macroscopic image G2 of the processing condition MM-SAMPLE selected in the first selection process and the image captured by the low-magnification imaging means 201 one by one. The first image of the plural number is judged whether it matches.

In the present embodiment, since the memory macroscopic image G2 memorized in the processing condition MM-SAMPLE does not match the respective first images, the first judgment unit 92 makes a judgment of inconsistency, and moves to the first image shown in the flowchart of FIG. 6 . 1 Repeat the project.

(5-1) The first repetition process In the first judgment process, the first judgment unit 92 judges that the memory macroscopic image G2 of the processing condition MM-SAMPLE does not match the first image, so the program constructed according to the flowchart of FIG. 6 is followed. , and the second selection process described later is carried out in the processing apparatus 1 .

(2-2) Second selection process In the second selection process, the machining condition INCH-SAMPLE of machining condition No. 002 shown in the machining condition list K shown in FIG. The selection unit 91 selects.

(3-2) The first imaging process of the second time Next, according to the machining condition INCH-SAMPLE of machining condition No. 002, in the control of the machining feeding means 13 and the first index feeding means 15 according to the control means 9 Next, the holding table 30 is positioned at a predetermined coordinate position such that the center of the holding surface 300 a of the holding table 30 is positioned directly below the objective lens 200 of the imaging unit 20 . Then, as shown in FIG. 7 , at the position of the X-axis and Y-axis coordinates based on the center of the holding surface 300 a of the holding stage 30 , the approximate center area of the surface Wa of the wafer W is imaged by the low-magnification imaging means 201 . The first image is formed.

Furthermore, the first index feeding means 15 moves the imaging unit 20 in the Y-axis direction, and the rotating means 32 rotates the holding table 30 at a predetermined rotational speed. Then, the low-magnification imaging means 201 moving so as to draw a swirling trajectory on the surface Wa of the wafer W captures images at a plurality of imaging positions near the center region of the surface Wa of the wafer W to form a plurality of first images .

(4-2) Second First Judgment Process Information about each first image imaged by the low-magnification imaging means 201 is sent from the imaging unit 20 to the first judgment part 92 of the control means 9. As shown in FIG. 9 , the first determination unit 92 creates the first image of the memory macroscopic image G1 and the low-magnification imaging means 201 set in the processing condition INCH-SAMPLE selected in the second selection process. For example, a consistent judgment of the portrait.

As described above, in the second first determination process, the first determination unit 92 determines that the memory macroscopic image G1 matches the first image, and, for example, performs the first dicing of the wafer W shown in FIGS. 1 and 3 . The planned dividing line S is aligned roughly parallel to the X-axis direction by a rough θ. The rough θ alignment is performed, for example, by being adjacent to the planned dividing line S (the planned dividing line S extending in the X-axis direction) of one of the first incisions shown in FIGS. 1 and 3 and located at positions separated from each other in the X-axis direction. It is performed by taking the image-captured images of the PA projections of the macroscopic targets of the two devices D. That is, the low-magnification imaging means 201 is used to form a photographic image for rough θ alignment of the PA of the giant optotype of a certain device D, and after the platform 30 is only moved in the X-axis direction by a portion of the device D, the image is carried out according to the low magnification. The imaging of the imaging means 201 forms another imaging image for rough θ alignment of the PA projection of the macroscopic target. Then, the holding stage 30 is rotated by the rotation means 32 by a predetermined angle so that the Y-axis coordinate positions of the respective macroscopic objects of the above-mentioned two rough θ alignment images are approximately the same.

Then, after the holding stage 30 is moved in the X-axis direction by only a few parts of the device D, imaging by the low-magnification imaging means 201 is performed to form an imaging image for rough θ alignment of the PA of the giant optotype of a certain device D. The platform 30 is kept in such a way that the Y-axis coordinate position of the PA of the macroscopic target of the imaged image for rough θ alignment used earlier is approximately the same as the Y-axis coordinate position of the PA of the macroscopic target of the imaged image for rough theta alignment that is formed further. By rotating the rotation means 32 by a predetermined angle, the straight line connecting the PA of the macroscopic target at the separated position in the X-axis direction becomes approximately parallel to the X-axis direction, completing the formation of the planned dividing line S of the first incision and the X-axis. Rough theta alignment with directions roughly parallel. Then, the processing apparatus 1 executes the first second imaging process, which will be described later, in accordance with a program constructed based on the flowchart of FIG. 6 .

(6-1) In the first second imaging process, the magnification of the objective lens 200 shown in FIG. 1 is switched to a high magnification, and the set light amount of the illumination means 203 is set to the light amount suitable for high-magnification imaging, A state in which imaging of the wafer W by the high-magnification imaging means 202 is possible is established. Then, one of the PAs of the macroscopic target detected earlier is located in the center of the imaging area of the high-magnification imaging means 202 .

Next, according to the setting of the processing condition INCH-SAMPLE memorized in Condition No. 002, the holding stage 30 for sucking and holding the wafer W shown in FIG. 1 moves only the giant screen shown in FIG. 3 by the processing and feeding means 13 The distance in the X-axis direction of the target PA and the micro-target PD, and the camera unit 20 moves only the distance in the Y-axis direction of the macro-target PA and the micro-target PD by the first index feeding means 15 . Then, the surface Wa of the wafer W is imaged by the high-magnification imaging means 202 to form a second image with a relatively narrow field of view.

(7-1) The first second judgment process The information on the second image captured by the high-magnification imaging means 202 is sent from the imaging unit 20 to the second judgment section of the control means 9 shown in FIG. 1 93. As shown in FIG. 10 , the second determination unit 93 compares the memory microscopic image G4 set in the processing condition INCH-SAMPLE selected in the second selection process with the second image captured by the high-magnification imaging means 202 , to judge whether it is consistent.

In the present embodiment, since the memory microscopic image G4 memorized in the processing condition INCH-SAMPLE does not match the second image, the second judgment unit 93 makes a judgment of inconsistency, and moves to the second image shown in the flowchart of FIG. 6 . Duplicate engineering.

(8) Second repetition process In the second judgment process, the second judgment unit 93 judges that the memory microscopic image G4 of the machining condition INCH-SAMPLE does not match the second image, so according to the program constructed according to the flowchart of FIG. In the processing apparatus 1, the 3rd selection process mentioned later is implemented.

(2-3) The third selection process In the third selection process, the machining condition AAAA-SAMPLE of the machining condition No. 003 shown in the machining condition list K shown in FIG. 2 is determined by the machining condition The selection unit 91 selects.

(3-3) The first imaging process of the third time Next, according to the setting of the machining condition AAAA-SAMPLE memorized in the machining condition No. 003, the machining feed means 13 and the control means 9 shown in FIG. Under the control of the first index feeding means 15, the holding table 30 and the imaging unit 20 are moved so that the center of the holding surface 300a of the holding table 30 is positioned directly below the objective lens 200 of the imaging unit 20. Then, as shown in FIG. 7 , at the position of the X-axis and Y-axis coordinates based on the center of the holding surface 300 a of the holding stage 30 , the approximate center area of the surface Wa of the wafer W is imaged by the low-magnification imaging means 201 . The first image is formed.

Furthermore, the first index feeding means 15 moves the imaging unit 20 in the Y-axis direction, and the rotating means 32 rotates the holding table 30 at a predetermined rotational speed. Then, the low-magnification imaging means 201 moving so as to draw a swirling trajectory on the surface Wa of the wafer W captures images at a plurality of imaging positions near the center region of the surface Wa of the wafer W to form a plurality of first images .

(4-3) The first judgment process of the third time is the first judgment sent from the imaging unit 20 to the control means 9 shown in FIG. Section 92. As shown in FIG. 9 , the first determination unit 92 is the first determination unit 92 after the memory macroscopic image G1 and the low-magnification imaging means 201 set in the processing condition AAAA-SAMPLE selected in the third selection process are created. For example, a consistent judgment of the image.

As described above, in the third first determination process, the first determination unit 92 determines that the memory macroscopic image G1 matches the first image, and, for example, performs the first dicing of the wafer W shown in FIGS. 1 and 3 . The planned dividing line S is aligned roughly parallel to the X-axis direction by a rough θ. Then, after the rough θ alignment of the planned dividing line S of the first kerf is completed, the processing apparatus 1 executes the second second imaging process, which will be described later, in accordance with a program based on the flowchart of FIG. 6 .

(6-2) In the second imaging process of the second time, the magnification of the objective lens 200 is switched to a high magnification, and the set light quantity of the illumination means 203 is set to a light quantity suitable for high-magnification imaging. The imaging of the wafer W by the imaging means 202 is enabled. Then, one of the PAs of the macroscopic target detected earlier is located in the center of the imaging area of the high-magnification imaging means 202 .

Next, according to the processing condition AAAA-SAMPLE of Condition No. 003, by the processing feeding means 13, the holding stage 30 that attracts and holds the wafer W is moved only by the X-axis of the macroscopic target PA and the microscopic target PC shown in FIG. 3 . For the distance Lx3 in the direction, by the first index feeding means 15, the imaging unit 20 is moved only by the distance Ly3 in the Y-axis direction of the macroscopic target PA and the microscopic target PC. Then, the surface Wa of the wafer W is imaged by the high-magnification imaging means 202 to form a second image.

(7-2) Second second judgment process The information on the second image imaged by the high-magnification imaging means 202 is sent to the second judgment part 93 of the control means 9 shown in FIG. 1 . As shown in FIG. 11 , the second determination unit 93 compares the memory microscopic image G3 set in the processing condition AAAA-SAMPLE with the second image captured by the high-magnification imaging means 202, and determines that they match.

In the second second judgment process as described above, after the second judgment unit 93 judges that the memory microscopic image G3 matches the second image, for example, the first dicing of the wafer W shown in FIGS. 1 and 3 is performed. The planned dividing line S is aligned in parallel with the X-axis direction with high-precision θ alignment. The high-precision θ alignment is, for example, using a line to be divided S (a line to be divided S extending in the X-axis direction) of one of the first cuts shown in FIGS. 1 and 3 adjacent to and separated from each other in the X-axis direction. It is performed by taking the image-captured images of each microscopic target PC of the two devices D in the position. That is, the high-magnification imaging means 202 is used to form a high-precision θ-alignment image of the microscopic target PC of a certain device D, and the holding stage 30 is moved in the X-axis direction by only a few parts of the device D. According to the imaging of the high-magnification imaging means 202 , another imaging image for the θ alignment with high precision is formed on the PC of the microscopic target of a certain device D. FIG.

Then, the holding stage 30 is rotated by a predetermined angle by the rotation means 32 until the deviation of the Y-axis coordinate positions of the micro-target PCs of the above-mentioned two imaged images is within the allowable value, thereby completing the highly accurate θ alignment.

Furthermore, the holding platform 30 will move in the X-axis direction, the center of the surface Wa of the wafer W will be positioned in the imaging area of the high-magnification imaging means 202, and the imaging image will be formed by the high-magnification imaging means 202, and the imaging image will be recognized. Microscopic target PC in . Then, it is determined whether the deviation of the Y-axis coordinate position of the micro-target PC is within the allowable value, and if it is outside the allowable value, the camera unit 20 will use the first The index feeding means 15 is appropriately moved in the Y-axis direction.

When the deviation of the Y-axis coordinate position of the micro-target PC reaches the allowable value, the first index feeding means 15 moves the imaging unit 20 on the Y-axis only by the distance from the micro-target PC to the center line in the width direction of the planned dividing line S. The alignment is performed in which the reference line (hairline) of the imaging unit 20 is superimposed on the line S to be divided. Then, the coordinate position in the Y-axis direction when the line of sight is superimposed on the planned dividing line S is stored in the memory section 90 of the control means 9 as the position of the cutting edge 613 for positioning the first processing means 61 when the wafer W is actually cut. .

As described above, after the coordinate position in the Y-axis direction at the time of actually cutting the planned dividing line S of the first scribe is memorized, the holding table 30 is accurately rotated by 90 degrees by the rotating means 32, and the cutting of the wafer W is performed. Next, when the planned dividing line S of the second kerf is aligned in parallel with the X-axis direction with high accuracy, the Y-axis of the first machining means 61 is positioned when the planned dividing line S of the second kerf is actually cut. The coordinate position is detected and stored in the memory unit 90 .

(9) Processing Process Next, the cutting apparatus 1 shown in FIG. 1 processes the wafer W under the processing conditions AAAA-SAMPLE. For example, firstly, the first processing means 61 is positioned by the first index feeding means 15 at the Y-axis coordinate when actually cutting the planned dividing line S of the first kerf stored in the memory portion 90 of the control means 9 Location. Then, the first cutting feed means 17 lowers the first machining means 61 in the -Z direction, and the first cutting means 61 is positioned at the cutting feed position set in the machining condition AAAA-SAMPLE. Then, the process feeding means 13 processes and feeds the holding table 30 holding the wafer W at the process feed speed set in the process condition AAAA-SAMPLE.

Then, a motor (not shown) rotates the spindle 610 of the first machining means 61 at the number of revolutions set in the machining conditions AAAA-SAMPLE, whereby the cutting edge 613 fixed to the spindle 610 follows the rotation of the spindle 610. While rotating and rotating, the wafer W is cut into the wafer W, and the planned dividing line S is cut.

When the holding table 30 advances to a predetermined position in the X-axis direction of the cutting edge 613 to complete the cutting and dividing line S, the first incision feeding means 17 lifts the first processing means 61 to separate the cutting edge 613 from the wafer W , and secondly, the machining and feeding means 13 will return the holding platform 30 to the machining and feeding start position. Then, the first index feed means 15 moves the first machining means 61 in the Y-axis direction only by the index feed amount set in the machining condition AAAA-SAMPLE, thereby for the cutting line S that is located beside the planned dividing line S to be cut. The planned dividing line S locates the cutting edge 613 . Then, cutting processing is carried out in the same manner as before. Hereinafter, by sequentially performing the same cutting, all the planned dividing lines S of the first kerf are cut. Furthermore, after the holding table 30 is rotated by 90 degrees, the cutting of each planned dividing line S of the second scribe line is performed, whereby all the planned dividing lines S of the wafer W are cut vertically and horizontally.

As described above, the wafer processing method of the present invention includes: the holding process of holding the wafer W by the holding stage 30; the selection process of selecting one processing condition from the processing condition list K in which a plurality of processing conditions are tabulated ; The first judgment process of judging whether the two images are identical is compared with the memory macroscopic image stored in the processing condition selected in the selection process and the first image captured by the low-magnification imaging means 201; In the first judgment process, it is judged as memory When the macroscopic image does not match the first image, return to the first repetition process of the selection process; when the first judgment process determines that the memory macro image is consistent with the first image, compare the memory used for the processing conditions used in the first judgment process. The second judgment process of judging whether the two images are consistent with the second image captured by the microscopic image and the high-magnification imaging means 202; when the second judgment process determines that the memory microscopic image is inconsistent with the second image, return to the second selection process. Repeat the process; and when the second judgment process judges that the memory microscopic image is consistent with the second image, the processing process of processing the wafer W under the processing conditions of the memory microscopic image that is useful in the second judgment process, so it is not necessary to rely on the operation. The selection of the processing conditions by the staff, and the selection of the processing conditions can be carried out even if the QR code is not arranged on the ring frame, etc.

W‧‧‧Wafer Wa‧‧‧Front surface of wafer Wb‧‧‧Back surface of wafer S‧‧‧Division line D‧‧‧Device T‧‧‧Dicing tape F‧‧‧Ring frame 1‧‧ ‧Processing device 10‧‧‧Base 14‧‧‧Gate post 30‧‧‧Holding platform 300‧‧‧Suction part 300a‧‧‧Holding surface 301‧‧‧Frame body 32‧‧‧Rotating means 34‧‧‧ Fixed Clamp 13‧‧‧Machining Feed Means 130‧‧‧Ball Screw 131‧‧‧Guide 132‧‧‧Motor 133‧‧‧Moveable Plate 15‧‧‧First Index Feed Means 150‧‧‧Ball Screw 151 ‧‧‧guide rail 153‧‧‧movable plate 17‧‧‧first incision feeding means 170‧‧‧ball screw 171‧‧‧guideway 172‧‧‧motor 173‧‧‧support member 16‧‧‧second index feed Feeding Means 18‧‧‧Second Plunge Feeding Means 61‧‧‧First Machining Means 610‧‧‧Spindle 611‧‧‧Case 613‧‧‧Cutting Blades 62‧‧‧Second Machining Means 20‧‧‧Camera Unit 200‧‧‧Objective lens 201‧‧‧Low-magnification imaging means 202‧‧‧High-magnification imaging means 203‧‧‧Illumination means 9‧‧‧Control means 90‧‧‧Memory section 91‧‧‧Processing condition selection section 92‧‧‧First Judgment Part 93‧‧‧Second Judgment Part K‧‧‧Processing Condition List

FIG. 1 is a perspective view showing an example of a processing apparatus. Fig. 2 is an example of a machining condition list in which plural machining conditions are tabulated. Fig. 3 is a plan view showing an example of the structure of the surface of the wafer. Figure 4(A) is an example of a memory macroscopic portrait, and Figure 4(B) is another example of a memory macroscopic portrait. Figure 5(A) is an example of the memory microscopic portrait, and Figure 5(B) is another example of the memory microscopic portrait. FIG. 6 is a flowchart illustrating the flow of each process of the wafer processing method. Fig. 7 is a cross-sectional view showing a state in which the holding table is positioned at a predetermined coordinate position on the X-axis Y-axis plane so that the center of the holding surface of the holding table is positioned directly below the imaging unit. Fig. 8 is an explanatory diagram for explaining the judgment of the first judgment unit in the first judgment process for the first time. Fig. 9 is an explanatory diagram explaining the judgment of the first judgment unit in the second (third) first judgment process. Fig. 10 is an explanatory diagram explaining the judgment of the second judgment unit in the first second judgment process. Fig. 11 is an explanatory diagram for explaining the judgment of the second judgment unit in the second judgment process of the second time.

Claims (1)

A method for processing a wafer, which is a wafer processing method using a processing device for processing along the predetermined dividing line of the wafer while holding a wafer on which a device is formed in a region divided by a planned dividing line by means of a holding platform. , characterized in that: the processing apparatus is provided with a low-magnification imaging means and a high-magnification imaging means, and has: a holding process for holding the wafer with the holding platform; and selecting 1 from a processing condition list in which the plurality of processing conditions are tabulated A selection process for a processing condition; a first judgment process for judging whether the two images are consistent by comparing the memory macroscopic image of the processing condition selected in the selection process with the first image captured by the low-magnification imaging means; When the first judging process determines that the memory macroscopic image is inconsistent with the first image, return to the first repeat process of the selection process; When the first judgment process determines that the memory macroscopic image is consistent with the first image, A second judgment process of judging whether the two images are consistent by comparing the memory microscopic image stored in the processing conditions used in the first judgment process with the second image captured by the high-magnification imaging means; In the second judgment process When it is judged that the memory microscopic portrait is inconsistent with the second portrait, return to the second repeated process of the selection process; and when the second judgment process determines that the memory microscopic portrait is consistent with the second portrait, use the memory to be useful in the The second judgment process is a process of processing the wafer under the processing conditions of the memory microscopic image.

Images