JP2015103567A - Wafer processing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a wafer processing method capable of further improving transverse intensity.SOLUTION: A method for processing a wafer in which a device is formed in each of the regions partitioned by a plurality of intersecting division schedule lines formed on a surface comprises: a first cutting step of forming a cutting groove having a depth reaching the finishing thickness of the wafer by cutting a surface of the wafer along the division schedule lines with a first cutting blade having a cutting edge composed of abrasive grains and a bond material; a second cutting step of cutting the wafer along the cutting groove with a second cutting blade having a cutting edge composed of a bond material and finer abrasive grains than those of the cutting edge of the first cutting blade and having a thickness thicker than the cutting edge of the first cutting blade, and polishing a side wall of the cutting groove formed in the first cutting step after performing the first cutting step; and a dividing step of thinning the wafer to the finishing thickness by cutting a rear surface of the wafer, exposing the cutting groove on the rear surface of the wafer, and dividing the wafer into individual chips after performing the second cutting step.

Description

  The present invention relates to a method for processing a wafer such as a semiconductor wafer.

  In the semiconductor device manufacturing process, a plurality of regions are defined by dividing lines called streets formed in a lattice shape on the surface of a semiconductor wafer such as a silicon wafer or a gallium arsenide wafer having a substantially disk shape. Devices such as IC and LSI are formed in the region.

  Such a semiconductor wafer (hereinafter sometimes simply referred to as a wafer) is ground into a predetermined thickness by a grinding device and then divided into individual device chips by a cutting device (dicing device). The divided device chips are widely used in various electronic devices such as mobile phones and personal computers.

  In order to improve the bending strength of the device chip, the back surface of the wafer after grinding is polished to remove grinding distortion generated by grinding (for example, see JP 2010-177430 A). .

JP 2010-177430 A

  However, when the wafer is cut along a planned dividing line by a cutting blade and divided into individual device chips, cutting strain is generated on the side surfaces of the device chip. There is an urgent need to further improve the folding strength.

  The present invention has been made in view of the above points, and an object of the present invention is to provide a wafer processing method capable of further improving the bending strength.

  According to the present invention, there is provided a wafer processing method in which a device is formed in each region defined by a plurality of intersecting scheduled lines formed on a surface, and has a cutting blade made of abrasive grains and a bonding material. A first cutting step of cutting the surface of the wafer along the line to be divided by the first cutting blade to form a cutting groove that reaches the finished thickness of the wafer, and after performing the first cutting step, the first cutting step A wafer is cut along the cutting groove with a second cutting blade made of abrasive grains finer than the cutting blade and a bonding material, and having a cutting edge thicker than the first cutting blade, and the first A second cutting step for polishing the side wall of the cutting groove formed in the cutting step; and after performing the second cutting step, grinding the back surface of the wafer to thin the wafer to the finished thickness. To the back surface of the wafer to expose The sections Kezumizo, the wafer processing method of a dividing step of dividing the wafer into individual chips, comprising the is provided.

  In the wafer processing method of the present invention, after forming a cutting groove with a first cutting blade on the surface of the wafer, the cutting groove is formed with a second cutting blade having a finer grain size than the first cutting blade and having a larger thickness. Cut again.

  As a result, since the distortion generated by the first cutting blade on the side wall of the cutting groove that is the outer peripheral side surface of the chip is removed by the second cutting blade, the bending strength of the chip can be further improved.

  After the wafer is ground and thinned, the wafer is divided into individual chips with the first cutting blade, and then the chips move when trying to polish the side of the chip by passing the second cutting blade between the chips. Side polishing is difficult. In the present invention, since the side wall of the half-cut cutting groove is polished by the second cutting blade, such a problem does not occur.

It is a perspective view of the cutting device suitable for implementing the 1st and 2nd cutting step of this invention processing method. It is a surface side perspective view of a semiconductor wafer. It is a perspective view which shows a 1st cutting step. It is an expanded sectional view showing the 1st cutting step. It is an expanded sectional view explaining a 2nd cutting step. It is a perspective view which shows a division | segmentation step.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. Referring to FIG. 1, there is shown a perspective view of a cutting device (dicing device) 2 suitable for performing a cutting step by the wafer processing method of the present invention.

  The cutting device 2 is a two-spindle type cutting device that sucks and holds a workpiece on the chuck table 4, and the chuck table 4 reciprocates in the cutting feed direction (X-axis direction) while indexing feed direction (Y-axis direction). And the work of the first cutting means 6 and the second cutting means 8 moving in the cutting feed direction (Z-axis direction).

  The chuck table 4 is movable in the X-axis direction by the cutting feed means 10, and a first alignment means 12 having a first camera (first imaging means) 13 formed integrally with the first cutting means 6. And a region to be cut of the wafer 11 sucked and held by the chuck table 4 by the second alignment means 14 having the second camera (second imaging means) 15 formed integrally with the second cutting means 8. After the street is detected and the street and the cutting blade are aligned in the Y-axis direction, cutting is performed.

  The cutting feed means 10 includes a pair of X-axis guide rails 16 disposed in the X-axis direction, an X-axis movement base 18 slidably supported by the X-axis guide rails 16, and an X-axis movement base 18. An X-axis ball screw 20 that is screwed into a nut portion (not shown) formed on the X-axis, and an X-axis pulse motor 22 that rotationally drives the X-axis ball screw 20.

  The support base 24 that rotatably supports the chuck table 4 is fixed to the X-axis moving base 18, and is driven by the X-axis pulse motor 22 to rotate the X-axis ball screw 20. It is moved in the X-axis direction.

  On the other hand, the first cutting means 12 and the second cutting means 14 are supported by the guide means 26 so as to be indexable in the Y-axis direction. The guide means 26 includes a vertical column 28 disposed in the Y-axis direction orthogonal to the X-axis so as not to prevent movement of the chuck table 4 and a pair of Y disposed in the Y-axis direction on the side surface of the vertical column 28. A shaft guide rail 30, a first ball screw 32 and a second ball screw 34 arranged in parallel with the Y axis guide rail 30, and a first Y axis pulse motor connected to the first ball screw 32. 36 and a second Y-axis pulse motor 38 connected to the second ball screw 34.

  The Y-axis guide rail 30 supports the first support member 40 and the second support member 42 so as to be slidable in the Y-axis direction, and is provided in the first support member 40 and the second support member 42. Nuts (not shown) are screwed into the first ball screw 32 and the second ball screw 34, respectively.

  Driven by the first Y-axis pulse motor 36 and the second Y-axis pulse motor 38 and the first ball screw 32 and the second ball screw 34 rotate, respectively, the first support member 40 and the second The support members 42 are independently moved in the Y-axis direction.

  The positions of the first support member 40 and the second support member 42 in the Y-axis direction are measured by the linear scale 44 and used for precise control of the Y-axis direction positions. Although it is possible to provide a linear scale separately for each support member, it is better to measure the positions of both the first support member 40 and the second support member 42 with a single linear scale 44. It is possible to precisely control the distance between.

  A first moving member 46 to which the first cutting means 6 is attached is attached to the first support member 40 so as to be slidable in the vertical direction (Z-axis direction). The first Z-axis pulse motor When 48 is driven, the first moving member 46 is moved in the Z-axis direction.

  Similarly, a second moving member 50 to which the second cutting means 8 is attached is attached to the second support member 42 so as to be slidable in the vertical direction (Z-axis direction). By driving the shaft pulse motor 52, the second moving member 50 is moved in the Z-axis direction.

  The first cutting means 6 includes a spindle 69 (see FIG. 3) that is rotationally driven, and a first cutting blade 70 that is attached to the tip of the spindle 69. Similarly, the second cutting means 8 includes a spindle that is rotationally driven, and a second cutting blade 72 (see FIG. 5) attached to the tip of the spindle.

  Referring to FIG. 2, there is shown a front side perspective view of a semiconductor wafer (hereinafter sometimes simply referred to as a wafer) 11 that is an object to be cut by the cutting apparatus 2. The semiconductor wafer 11 is made of, for example, a silicon wafer having a thickness of 700 μm, and a plurality of division lines (streets) 13 are formed in a lattice shape on the surface 11 a and are partitioned by the plurality of division lines 13. A device 15 such as an IC or LSI is formed in each region.

  In the wafer processing method of the present invention, first, the back surface 11b side of the wafer 11 is sucked and held by the chuck table 4, and the wafer is cut along the scheduled dividing line 13 by the first cutting blade 70 as shown in FIGS. The first cutting step of cutting the surface 11a of the eleventh and forming the cutting groove 54 reaching the finished thickness of the wafer 11 is performed.

  In this first cutting step, the first cutting blade 70 that rotates at high speed in the direction of arrow A (for example, 30000 rpm) is fed in the first direction while the chuck table 4 that sucks and holds the wafer 11 is fed in the direction of arrow X1 in FIG. A cut groove 54 that reaches the finished thickness of the wafer 11 is formed by cutting into the wafer 11 along the division line 13 that extends to the end.

  While the first cutting blade 70 is indexed and fed by the pitch of the planned division line 13, all the planned division lines 13 extending in the first direction are cut to form similar cutting grooves 54. Next, after the chuck table 4 is rotated by 90 °, a similar cutting groove 54 is formed along the planned dividing line 13 extending in the second direction orthogonal to the first direction.

  Here, the first cutting blade 70 has a cutting blade in which diamond abrasive grains of # 2000 grain size are hardened by nickel plating, and the width of the cutting blade is, for example, 30 μm. When the first cutting step is performed, a cutting strain is generated on the side wall 54a of the cutting groove 54 as shown in FIG.

  Therefore, the wafer processing method of the present invention has a cutting blade that is made of abrasive grains and a bonding material finer than the cutting blade of the first cutting blade 70 and is thicker than the cutting blade of the first cutting blade 70. As shown in FIG. 5, the second cutting blade 72 cuts the wafer 11 along the cutting groove 54 and performs the second cutting step of polishing the side wall 54a of the cutting groove 54 formed in the first cutting step. .

  Here, it is preferable that the second cutting blade 72 has a cutting blade that is, for example, 5 to 30 μm thicker than the cutting blade of the first cutting blade 70. More preferably, it has a cutting blade that is 10 to 20 μm thicker than the cutting blade of the first cutting blade 70. For example, the cutting blade of the second cutting blade 72 has a cutting blade in which diamond abrasive grains of # 5000 grain size are hardened by nickel plating, and the width of the cutting blade is 50 μm.

  Preferably, after the second cutting step is performed along a predetermined number of cutting grooves 54, the cutting groove 54 is imaged by the second camera 15 and a kerf check is performed to confirm the cutting position of the second cutting step. If correction is necessary, the second cutting blade 72 is moved by a correction amount in the Y-axis direction, and then the second cutting step is continued.

  By performing such a kerf check, the center of the second cutting blade 72 is always positioned at the center of the cutting groove 54, and the side wall 54a of the cutting groove 54 formed in the first cutting step by the second cutting blade 72 is polished. can do.

  In the second cutting step of the present embodiment, the second cutting blade 72 is passed through the half-cut cutting groove 54 in which the wafer 11 is not divided into chips, so that the device 15 does not move and the side wall of the cutting groove 54, i.e., the division. Later, the side surface of the device chip can be polished.

  After the second cutting step is performed, the back surface 11b of the wafer 11 is ground to thin the wafer 11 to a finished thickness, and the cutting grooves 54 are exposed on the back surface 11b of the wafer 11 so that the wafer 11 is separated into individual device chips. A dividing step of dividing into 15 is performed.

  This division step will be described with reference to FIG. In the dividing step, as shown in FIG. 6A, the protective tape 17 is attached to the surface of the wafer 11, and the surface 11a side of the wafer 11 is sucked and held through the protective tape 17 by the chuck table 56 of the grinding apparatus. Then, the back surface 11b side of the wafer 11 is exposed.

  The grinding unit 58 of the grinding apparatus includes a spindle 60 that is rotationally driven by a motor, a wheel mount 62 that is fixed to the tip of the spindle 60, and a grinding wheel 64 that is detachably fixed to the wheel mount 62 with a plurality of screws 63. Is included. The grinding wheel 64 includes an annular wheel base 68 and a plurality of grinding wheels 68 that are annularly fixed to the outer periphery of the lower end of the wheel base 68.

  In the dividing step, while rotating the chuck table 56 in the direction of arrow a at 300 rpm, for example, the grinding wheel 64 is rotated in the same direction as the chuck table 56, that is, in the direction of arrow b at 6000 rpm, for example. In operation, the grinding wheel 68 is brought into contact with the back surface 11 b of the wafer 11.

  Then, the grinding wheel 64 is ground by a predetermined amount at a predetermined grinding feed speed, and the wafer 11 is ground. When the grinding is continued and the wafer 11 is thinned to the finished thickness, as shown in FIG. 6B, the cutting grooves 54 are exposed on the back surface 11b of the wafer 11, and the wafer 11 is turned into individual device chips 15. Divided.

  According to the wafer processing method of the present embodiment, after the cutting groove 54 is formed on the surface 11a of the wafer 11 by the first cutting blade 70, the particle diameter is smaller than the cutting edge of the first cutting blade 70, and the cutting is thick. The cutting groove 54 is cut again with the second cutting blade 70 having a blade.

  Since the distortion generated by the first cutting blade 70 on the side wall 54a of the cutting groove 54 which is the outer peripheral side surface of the device chip 15 is polished and removed by the second cutting blade 72, the further bending strength of the device chip 15 is increased. Can be improved.

4 chuck table 6 first cutting means 8 second cutting means 11 semiconductor wafer 13 scheduled division line 15 device (device chip)
17 Protective Tape 54 Cutting Groove 54a Side Wall 56 Chuck Table 64 Grinding Wheel 68 Grinding Wheel 70 First Cutting Blade 72 Second Cutting Blade

Claims (1)

A wafer processing method in which a device is formed in each region defined by a plurality of intersecting scheduled lines formed on a surface,
A first cutting step of cutting the surface of the wafer along the line to be divided with a first cutting blade having a cutting edge made of abrasive grains and a bond material to form a cutting groove that reaches the finished thickness of the wafer;
After performing the first cutting step, the cutting is performed with a second cutting blade that has a finer abrasive grain and a bond material than the first cutting blade, and has a cutting blade thicker than the first cutting blade. A second cutting step of cutting a wafer along the groove and polishing a side wall of the cutting groove formed in the first cutting step;
After performing the second cutting step, a dividing step of grinding the back surface of the wafer to thin the wafer to the finished thickness, exposing the cutting groove on the back surface of the wafer, and dividing the wafer into individual chips. When,
A wafer processing method characterized by comprising:

Images