TWI788562B - Wafer processing method - Google Patents

Abstract

An object of the present invention is to provide a wafer processing method capable of suppressing the occurrence of processing defects during wafer dicing. The solution is a wafer processing method, including: Calculation process, which is based on the influence of the deviation of heat conduction of the wafer caused by the irradiation of the laser beam, and determines the position of the laser beam irradiated The size of the area that should be guaranteed on both sides of the predetermined division line, when the number of rows of the device contained in the area is set as N (N is a natural number), calculate the conditions that meet N＜ ²ⁿ (n is a natural number) Minimum 2 ⁿ ; 2 ⁿ processing process, which is to irradiate the laser beam to the planned dividing line at intervals of the smallest 2 ⁿ divisions of the planned dividing line, thereby forming processing marks on the wafer; and The process of bisecting is formed on the wafer by irradiating the planned dividing line which bisects the number of columns of the devices respectively included in the plurality of regions partitioned by the processing marks with a laser beam. The processing of the processing trace is repeated until the number of columns of the device included in the area partitioned by the processing trace becomes 1.

Description

Wafer processing method

The present invention relates to a wafer processing method for processing wafers by irradiating laser beams.

By dicing the wafer on which the plurality of devices are formed along the planned dividing lines (dicing lines), a plurality of device wafers each including the device can be obtained. This division of the wafer is performed, for example, using a cutting device having a spindle as a rotation axis. The cutting blade attached to the front end of the main shaft of the cutting device is rotated so that the cutting blade cuts into the wafer along the planned dividing line, whereby the wafer is cut along the planned dividing line.

In the division of wafers using cutting devices, various methods have been proposed in order to prevent processing defects of device wafers and breakage of cutting blades. For example, in Patent Document 1, it is disclosed and repeated: first, the wafer is divided into two wafers with approximately the same area, and then the wafer is divided into two wafers with approximately the same area. This is a method of dividing a wafer into a plurality of device chips.

According to the method described above, the area of the wafers on both sides separated by the planned dividing line cut by the cutting blade becomes approximately the same, and the load acting on the cutting blade during cutting becomes equal on the front side and the back side. . This suppresses the occurrence of processing defects such as chipping and cracking of the device wafer and damage to the cutting blade.

On the other hand, a method of using a laser processing device for dividing a wafer instead of a cutting device has also been proposed. In this method, the wafer is divided by irradiating a laser beam having an absorbing wavelength to the wafer along the dividing line. Patent Document 2 discloses a wafer processing method in which a pulsating laser beam is irradiated on a planned dividing line of a wafer to form a processing mark (groove), and the wafer is divided along the processing mark. [Prior Art Literature] [Patent Document]

[Patent Document 1] Japanese Patent Application Laid-Open No. 4-245663 [Patent Document 2] Japanese Patent Application Laid-Open No. 10-305420

[Problems to be Solved by the Invention]

As mentioned above, when using a laser processing device to process a wafer, due to some reasons, there may be chips or cracks called chipping on the wafer, which may cause poor processing of the wafer, resulting in damage to the device wafer. Yield reduction.

The present invention was developed in view of such problems, and aims to provide a wafer processing method capable of suppressing occurrence of processing defects when dividing a wafer by irradiation of a laser beam. [Means to solve the problem]

According to the present invention, there is provided a wafer processing method for processing a wafer in which devices are respectively formed in a plurality of regions partitioned by a plurality of planned dividing lines by irradiating laser beams along the planned dividing lines. The processing method of a wafer is characterized by including: a calculation process of determining the division plan when the laser beam is irradiated based on the influence of the deviation of the heat conduction of the wafer caused by the irradiation of the laser beam. The size of the area that should be ensured on both sides of the line, when the number of columns of the device contained in the area is set to N (N is a natural number), calculate the smallest 2 that meets N<2 ⁿ (n is a natural number) ⁿ ; ²ⁿ processing process, which is to irradiate the laser beam to the planned dividing line at intervals of the smallest ²ⁿ divisions of the planned dividing line, thereby forming processing marks on the wafer; and bisecting A processing process of forming a processing mark on the wafer by irradiating a laser beam to the planned dividing line that bisects the number of rows of the devices included in the plurality of regions partitioned by the processing mark The process is repeated until the number of columns of the device included in the area partitioned by the processing trace becomes 1.

In addition, in the 2 ⁿ processing process of the present invention, after the laser beam is irradiated along the planned dividing line closest to the end of the wafer to form the processing trace to the depth of cutting the wafer, the The other planned dividing lines are irradiated with the laser beam at intervals of the minimum ²ⁿ division lines.

Moreover, in the present invention, the wafer can also be a GaAs wafer. [Effect of the invention]

The wafer processing method of the present invention is repeated: first, by irradiation of a laser beam, processing marks are formed on the wafer at predetermined intervals, and then, in a plurality of regions that can be divided by the processing marks A process in which the number of rows of devices included is irradiated with a laser beam to form processing marks, thereby processing the wafer. Therefore, the unevenness of heat conduction caused by the irradiation of the laser beam is suppressed, so that processing defects caused by the unevenness of heat conduction can be suppressed, and the yield rate of device chips can be improved.

Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a perspective view showing a configuration example of a wafer according to the present embodiment. As shown in FIG. 1, the wafer 11 is formed in a disk shape having a surface 11a and a back surface 11b. The wafer 11 is, for example, a GaAs wafer.

The wafer 11 is divided into a plurality of regions by a plurality of dividing lines. For example, as shown in FIG. 1, the wafer 11 is divided into a plurality of regions by a plurality of first planned dividing lines 13a and a plurality of second planned dividing lines 13b, and the plurality of first planned dividing lines 13a are defined as The longitudinal direction will be arranged along the first direction (direction shown by arrow A), arranged along the second direction (direction shown by arrow B) intersecting with the first direction, and the plurality of second The planned division lines 13b are arranged so that the longitudinal direction may follow the second direction, and are arranged along the first direction. FIG. 1 shows a wafer 11 in which the first direction is approximately perpendicular to the second direction.

Devices 15 made of IC (Integrated Circuit) etc. are respectively formed on the surface 11a side of the plurality of regions partitioned by the plurality of first planned dividing lines 13a and the plurality of second planned dividing lines 13b.

In addition, the material, shape, structure, size, etc. of the wafer 11 are not limited. For example, the wafer 11 may be a wafer formed of a semiconductor (silicon, InP, GaN, SiC, etc.), ceramics, resin, metal, or other material other than a GaAs wafer. In addition, the types, numbers, shapes, structures, sizes, arrangements, etc. of the devices 15 are not limited.

When the laser beam is irradiated along the plurality of first planned dividing lines 13a and the plurality of second planned dividing lines 13b, processing marks are formed by ablation in the area of the wafer 11 irradiated with the laser beam. Then, when the wafer 11 is cut along the processing lines, the wafer 11 is divided into a plurality of device wafers having the devices 15 formed on the surface. Irradiation of the wafer 11 with a laser beam is performed using a laser processing device.

When processing the wafer 11 by irradiation of a laser beam, first, the wafer 11 is supported by a frame in order to hold the wafer 11 by a chuck table of the laser processing apparatus. FIG. 2 is a perspective view showing the wafer 11 in a state supported by the ring-shaped frame 19 . A ring-shaped frame 19 is attached along the outer periphery of the dicing tape 17, and the rear surface 11b side of the wafer 11 is attached to the dicing tape 17 so that the front surface 11a side of the wafer 11 is exposed upward. Accordingly, the wafer 11 is supported by the frame 19 .

Next, the wafer 11 supported by the frame 19 is supported by a laser processing device. FIG. 3 is a perspective view schematically showing a state in which the wafer 11 is supported by the laser processing apparatus 2 . The laser processing device 2 includes: a chuck table 4 holding a wafer 11 , and a laser processing unit for irradiating a laser beam having an absorbing wavelength to the wafer 11 on the wafer 11 held by the chuck table 4 6.

The chuck table 4 sucks and holds the wafer 11 through the dicing tape 17 . Specifically, the upper surface of chuck table 4 serves as a holding surface for holding wafer 11 , and this holding surface is connected to a suction source (not shown) via a suction path (not shown) formed inside chuck table 4 .

The frame 19 is fixed by a clamping device (not shown) included in the laser processing device 2, and the negative pressure of the suction source is applied to the holding surface in a state where the wafer 11 is supported by the holding surface of the chuck table 4. This is sucked and held while the wafer 11 is in contact with the holding surface. In addition, the chuck table 4 is moved in the machining feeding direction (X-axis direction) and the indexing feeding direction (Y-axis direction) by a moving mechanism (not shown).

The wafer 11 is sucked and held by the chuck table 4 so that the surface 11 a is exposed above. FIG. 3 shows that the wafer 11 is held in such a manner that the first direction of the wafer 11 approximately coincides with the X-axis direction, and the second direction of the wafer 11 approximately coincides with the Y-axis direction. In this state, by irradiating laser beams from the laser processing unit 6 along the plurality of first planned dividing lines 13 a , linear processing marks are formed on the wafer 11 .

The laser processing unit 6 has a cylindrical casing 8 . A pulsed laser beam oscillator (not shown) for oscillating a pulsed laser beam oscillator (not shown) such as a YAG laser oscillator or a _YVO4 laser oscillator equipped in the laser processing device 2 is mounted on the front end of the housing 8. A light collector 10 for collecting light.

Furthermore, an imaging means 12 for capturing a processing area (irradiated area of a laser beam) of the wafer 11 is attached to the housing 8 . The image obtained by the imaging means 12 is used for image processing such as pattern matching for aligning the optical collector 10 with the first planned dividing line 13a or the second planned dividing line 13b. Thereby, the irradiation position of the laser beam can be adjusted.

When the laser beam is irradiated to the wafer 11, the chuck table 4 is moved under the optical collector 10, and the laser beam with an absorbing wavelength to the wafer 11 is collected by the optical collector 10, and is directed toward the wafer 11. While the first planned dividing line 13a is being irradiated, the chuck table 4 is moved in the machining feeding direction (X-axis direction). Thereby, the laser beam is irradiated along the first planned dividing line 13 a, and processing traces composed of linear grooves are formed on the wafer 11 .

In addition, when the wafer 11 is divided by irradiation of a laser beam, processing marks having a depth to cut the wafer 11 are formed along the first planned dividing line 13 a. For example, by irradiating the laser beam a plurality of times along the same first planned dividing line 13a, a processing mark having a depth to cut the wafer 11 can be formed. In addition, the number of times of irradiation of the laser beam in this case is appropriately set so that the wafer 11 can be cut.

Irradiation of the laser beam to the second planned dividing line 13b is also performed in the same manner as the first planned dividing line 13a. Specifically, the first direction of the wafer 11 roughly coincides with the Y-axis direction, and the second direction of the wafer 11 roughly coincides with the X-axis direction. After °rotation, perform the same operation. In this way, once all the first planned dividing lines 13 a and the second planned dividing lines 13 b are cut, the wafer 11 is divided into a plurality of device wafers each including the devices 15 .

Here, consideration is given to the order of irradiating laser beams to the lines to be divided. For example, when the plurality of first planned dividing lines 13a arranged along the second direction are sequentially irradiated with a laser beam from one end portion toward the other end portion of the wafer 11 in the second direction, by The volume of the regions on both sides separated by the first planned dividing line 13a irradiated with the laser beam is biased.

Specifically, first, a laser beam is irradiated along the first planned dividing line 13 a (hereinafter referred to as the outermost first planned dividing line 13 a ) closest to one end of the wafer 11 in the second direction. At this time, among the two regions of the wafer 11 divided by the outermost first planned dividing line 13a, the volume of the region including the one end is smaller than the volume of the other region.

The above-mentioned variation in volume is the cause of variation in conduction of heat generated by laser beam irradiation. That is, when the laser beam is irradiated to the first planned division line 13 a, heat is generated, and this heat is conducted inside the wafer 11 . Here, if there is a volume deviation in the two areas of the wafer 11 divided by the first planned division line 13a irradiated with the laser beam, the heat conduction path will be more blocked in the small volume area. Restricted, so the heat will not diffuse sufficiently and the temperature will rise easily.

Also, when the laser beam is irradiated to the wafer 11 on which the processing marks (grooves) are formed along the plurality of first planned dividing lines 13a, the conduction of heat is hindered by the processing marks, and uneven heat conduction occurs. Specifically, when a laser beam is irradiated along the first planned dividing line 13 a in the region between the two processing marks formed on the wafer 11 , the region is subdivided into two small regions. Here, if the volumes of the two small regions are different, the conduction path of heat will be more restricted in the smaller small region.

When the wafer 11 is divided by irradiation of a laser beam, processing defects such as chips or cracks may occur on the wafer 11. On the side, heat conduction biases the temperature difference as one reason. Therefore, it is desirable to irradiate the wafer 11 with a laser beam under the condition that the unevenness of heat conduction on both sides of the region irradiated with the laser beam is small.

In this embodiment, the order in which the laser beams are irradiated to the planned dividing line is set so that there is no large difference in the volume of the regions located on both sides of the planned dividing line to which the laser beam is irradiated. Thereby, unevenness in heat conduction due to a difference in volume can be suppressed, and processing defects when processing wafer 11 with a laser beam can be suppressed.

In the wafer processing method of the present embodiment, first, the intervals at which the laser beams are irradiated are set based on the influence of uneven conduction of heat generated by the irradiation of the laser beams. Then, a laser beam is irradiated along the first planned division line 13 a of the wafer 11 at such intervals to form linear processing marks on the wafer 11 . Then, the process of re-forming the processing marks by irradiating the first planned dividing line 13a that bisects the number of columns of devices included in the plurality of regions partitioned by the processing marks with a laser beam is repeated, thereby Process marks are formed along all the first planned dividing lines 13a.

By using the wafer processing method described above, the volume difference between the two regions separated by the region irradiated with the laser beam can be reduced, and processing defects can be suppressed. In addition, irradiation of the laser beam to the second planned division line 13b is also performed in the same manner.

Hereinafter, details of the wafer processing method of this embodiment will be described. In addition, the following describes the formation of the wafer 21 formed by irradiating a laser beam with the device 15 in a plurality of regions partitioned by the 25 first planned dividing lines 23 a and the 25 second planned dividing lines 23 b. The case of processing marks (see FIG. 4, FIG. 5, FIG. 6, and FIG. 7) is taken as an example. On the wafer 21 , 24 rows of devices 15 are formed in each of the first direction (the direction indicated by the arrow A) and the second direction (the direction indicated by the arrow B).

<Calculation process> In the wafer processing method of this embodiment, first, a calculation process of calculating the intervals for irradiating laser beams to the first planned division lines 23a is performed in a subsequent process ( ²ⁿ processing process described later).

The intervals at which the laser beams are irradiated are set according to the influence of the deviation of heat conduction on the wafer 21 caused by the irradiation of the laser beams. For example, processing defects (chips, cracks, etc.) caused by uneven heat conduction in the two regions of the wafer 21 partitioned by the first planned dividing line 23a irradiated with the laser beam (chips, cracks, etc.) do not occur or The interval at which the laser beam is irradiated is set so that the frequency of occurrence is equal to or less than a certain value.

Even if there is a volume difference between the two regions located on both sides of the irradiation region of the laser beam, as long as the volumes of these regions are respectively ensured to be a certain amount or more, the conduction of heat generated by the irradiation of the laser beam will be in the two regions. Areas are formed approximately uniformly, and processing defects due to variations in heat conduction are less likely to occur. Therefore, the calculation process is to calculate the intervals at which the laser beams are irradiated so as to secure areas with a volume greater than a certain amount on both sides of the irradiation area of the laser beams.

Specifically, first, the size of the area to be secured on both sides of the first planned dividing line 23a to which the laser beam is irradiated is determined based on the influence of the uneven heat conduction of the wafer 21 caused by the irradiation of the laser beam. . For example, the size of this region is determined so that processing defects (chips, cracks, etc.) due to variations in heat conduction do not occur or occur at a constant frequency or less. In addition, the size of the region may be determined so that the temperature difference between the two regions located on both sides of the first planned dividing line 23a to which the laser beam is irradiated becomes below a certain value.

The size of the above-mentioned regions can also be based on, for example, the material (especially thermal conductivity) of the wafer 21, the thickness of the wafer 21, the size of the device 15, the distance between the devices 15, the wavelength of the laser beam, the intensity of the laser beam It is determined by one or more selected elements. In this case, as long as the relationship between the above-mentioned various elements and the influence of the unevenness of heat conduction (the frequency of processing defects, the degree of temperature difference, etc.) The size of the area to be secured is sufficient.

Next, the number of rows of devices 15 (the number of devices 15 in the second direction) N (N is a natural number) included in the area to be secured on both sides of the first planned dividing line 23a is specified. Then, based on the value of N, the minimum value of 2 ⁿ satisfying N<2 ⁿ (n is a natural number) is calculated. This minimum 2 ⁿ will be the value that should be calculated in the calculation process.

For example, when an area of more than 3 columns of the device 15 should be secured on both sides of the first planned division line 23a to be irradiated with the laser beam, N=3, and the value of ²ⁿ that should be calculated in the calculation process is in line with 3<2 The smallest value of 2 ⁿ for ⁿ becomes 4. And, the minimum 2 ⁿ intervals of the first planned dividing lines 23 a correspond to the intervals between irradiation laser beams in the 2 ⁿ processing steps described later.

< ²ⁿ processing process> Next, based on the minimum ²ⁿ value calculated in the calculation process, the laser beam is irradiated to the wafer 21 at intervals of the minimum ²ⁿ of the first planned division lines 23a. 2 ⁿ processing engineering. For example, when the minimum value of 2 ⁿ is 4, the laser beam is irradiated to the plurality of first planned dividing lines 23 a at four intervals. In addition, when only 2 ⁿ is described in the following description, it represents the minimum 2 ⁿ calculated in the calculation process.

An example of a ²ⁿ processing process will be described using Fig. 4 and Fig. 5 . In the ²ⁿ process, first, a laser beam is irradiated along the first planned dividing line 23a (outermost first planned dividing line 23a ) closest to one end of the wafer 21 in the second direction.

By irradiation of the laser beam, a linear processing mark L1 is formed on the outermost first dividing line 23a, and the wafer 21 is divided into a first region 21a including the one end, and a plurality of The second area 21b of the device 15. FIG. 4 is a plan view showing how the wafer 21 is divided into a first region 21a and a second region 21b.

Next, from the outermost first planned division line 23a, the other first planned division lines 23a are irradiated with laser beams at intervals of 2 ⁿ lines. For example, when the value of 2 ⁿ is 4, a plurality of first planned dividing lines 23 a are irradiated with laser beams at four intervals. As a result, linear processing traces L2 are formed at intervals of ²ⁿ of the plurality of first planned dividing lines 23a.

By forming the processing trace L2, the second region 21b is divided into a third region 21c located on the opposite side to the first region 21a in the second direction, and a plurality of devices 15 arranged in ²ⁿ rows in the second direction. 4th area 21d. Fig. 5 is a plan view showing how the second area 21b is divided into a third area 21c and a plurality of fourth areas 21d.

For example, when a plurality of first planned dividing lines 23 a are irradiated with laser beams at intervals of four, six processing marks L2 are formed in the second region 21 b as shown in FIG. 5 . Thereby, the second area 21b is divided into the third area 21c and the six fourth areas 21d in which the devices 15 are arranged in four rows in the second direction.

Through the above ²ⁿ process, the wafer 21 will be divided into a plurality of regions. The ²ⁿ processing process is to secure an area of N rows or more of the device 15 between the plurality of first planned dividing lines 23a to which laser beams are irradiated. Therefore, even if the conduction of heat is hindered by the processing marks L1 and L2, the heat generated when the laser beam is irradiated to the first planned dividing line 23a can be sufficiently diffused, and the processing caused by the unevenness of heat conduction can be suppressed. Bad happens.

In addition, in the ²ⁿ processing process, the same first planned dividing line 23a may be irradiated with a laser beam a plurality of times to form the processing trace L1 and the processing trace L2 having the depth to cut the wafer 21 . In this case, the order of irradiating the laser beams can be freely set.

For example, it is also possible to irradiate the outermost first planned dividing line 23a with a laser beam a plurality of times to form the processing trace L1 with the depth of cutting the wafer 21, and then irradiate the other first planned dividing lines 23a with a plurality of times of laser beams respectively. beam to form a processing trace L2 of the depth to cut the wafer 21 . Also, by repeating the process of irradiating the outermost first planned dividing line 23a and the other first planned dividing lines 23a with a laser beam once, the processing trace L1 having a depth of cutting the wafer 21 can be formed. And processing marks L2.

<Bisection Processing Process> Next, repeat the bisection processing process of the process of irradiating laser beams to dividing lines that roughly bisect the plurality of fourth regions 21d obtained by ²ⁿ processing processes.

First, the first planned division line 23a that bisects the number of rows of the devices 15 included in the fourth region 21d in the second direction is selected, and the laser beam is irradiated along the first planned division line 23a.

By irradiation of the laser beam, linear processing marks L3 that bisect the number of rows of the device 15 in the second direction are respectively formed in the fourth regions 21d, and the fourth regions 21d are respectively divided into two parts in the second direction. There are 2 fifth regions 21e of 2 ^n-1 rows of devices 15 . FIG. 6 is a plan view showing a fifth region 21e in which the fourth region 21d is divided into two.

When the laser beam is irradiated to the 4th area 21d, the laser beam is irradiated in such a way that the number of columns in the second direction of the devices contained in the 4th area 21d can be halved, so the laser beam is positioned at the position of the irradiated laser beam. The volumes of the fourth regions 21d on both sides of the regions are substantially equal. Therefore, the unevenness of heat conduction at the time of irradiating the laser beam is small, and processing defects due to the unevenness of heat conduction are less likely to occur.

In addition, in the bisecting process, the laser beam may be irradiated a plurality of times on the same first planned dividing line 23a to form the processing mark L3 having a depth that cuts the fourth region 21d. In this case, the order of irradiating the laser beams can be freely set.

For example, one processing mark L3 may be formed by irradiating one first planned dividing line 23 a with a laser beam a plurality of times, and then irradiating another first planned dividing line 23 a with a laser beam to form another processed mark L3 . Moreover, it is also possible to form a plurality of processing marks L3 by repeating the process of irradiating each of the plurality of first planned dividing lines 23a that divide the fourth region 21d in half with a laser beam once.

Also, depending on the number of rows of devices 15, devices 15 may be included in the third region 21c. In this case, the first planned dividing line 23a that bisects the number of rows of the devices 15 included in the third region 21c in the second direction is selected and irradiated with a laser beam. When the number of columns of devices included in the third region 21c is an odd number, the first division plan is selected so that the volume difference between the regions on both sides of the first division plan line 23a to which the laser beam is irradiated can be minimized. Line 23a will suffice.

Next, the laser beam is similarly irradiated to the fifth region 21e, and the fifth region 21e is roughly divided into two halves again. That is, the linear processing trace L4 which bisects the number of columns of the device 15 in the second direction is formed in the fifth region 21e. Thereby, the fifth region 21e is divided into two sixth regions 21f having 2n ⁻² rows of devices 15 in the second direction. FIG. 7 is a plan view showing a sixth region 21f in which the fifth region 21e is divided into two.

When the laser beam is irradiated to the fifth region 21e, the volume of the fifth region 21e located on both sides of the region to which the laser beam is irradiated is the same as when the laser beam is irradiated to the fourth region 21d (see FIG. 6 ). are approximately equal. Therefore, the unevenness of heat conduction at the time of irradiating the laser beam is small, and processing defects due to the unevenness of heat conduction are less likely to occur.

In addition, in the bisecting process, the laser beam may be irradiated a plurality of times on the same first planned dividing line 23a to form the processing mark L4 having the depth to cut the fifth region 21e. In this case, the order of irradiating the laser beams can be freely set.

For example, after a laser beam is irradiated a plurality of times to one first planned division line 23a to form one processing mark L4, another first planned division line 23a may be irradiated with a laser beam. Moreover, it is also possible to form a plurality of processing marks L4 by repeating the process of irradiating each of the plurality of first planned dividing lines 23a that roughly divides the fifth region 21e into two equal parts with a laser beam once.

As above, by repeating: irradiating the first dividing line 23a which bisects the number of rows of devices 15 respectively included in the plurality of regions divided by the processing marks to form the processing marks again During the process, the wafer 21 is divided into a plurality of regions including a row of devices 15 .

In addition, in the ²ⁿ processing process, the wafer 21 is divided into a plurality of fourth regions 21d including ²ⁿ columns of devices 15 by the processing marks L1 and L2 (see FIG. 5 ). Therefore, the laser beam can be irradiated to all the first planned dividing lines 23a included in the fourth region 21d by repeating the process of dividing the number of columns of the device 15 into two equal parts.

After the irradiation of the laser beam to all the first planned division lines 23a is completed, the irradiation of the laser beam to the second planned division lines 23b is performed by the same process. Specifically, first, the chuck table 4 shown in FIG. Direction 90° rotation.

Then, the above-described calculation process, ²ⁿ processing process, and bisecting processing process are performed on the second planned dividing line 23b. In addition, the value of ²ⁿ when the laser beam is irradiated to the second planned division line 23b may be the same as or different from the value of ²ⁿ when the laser beam is irradiated to the first planned division line 23a.

Through the above process, processing marks are formed on all the first planned dividing lines and the second planned dividing lines 23b. In addition, when these processing traces are processing traces having a depth of cutting the wafer 21 , the wafer 21 is diced into a plurality of device wafers each having a device 15 .

As described above, the wafer processing method of the present embodiment repeats the process of irradiating the laser beam to the dividing line that roughly bisects the fourth region 21d partitioned by the processing trace L1 and the processing trace L2. Thereby, the volumes of both sides of the region irradiated with the laser beam become substantially the same, and the uneven conduction of heat generated by the irradiation of the laser beam is suppressed. Therefore, it is possible to suppress processing defects caused by variations in heat conduction, and improve the yield of device wafers.

In addition, when a laser beam is irradiated a plurality of times to the same first planned dividing line 23a to form a processing mark with a depth of cutting the wafer 21, the order of irradiating the laser beam through ²ⁿ processing steps and bisecting processing steps It can be set appropriately. For example, after the wafer 21 is irradiated with a laser beam a plurality of times in the ²ⁿ processing process and the wafer 21 is divided into a plurality of fourth regions 21d, the fourth region 21d may be irradiated with the laser beam in the halving process. bundle.

In addition, according to the procedure shown in FIGS. 4 to 7 , each of the first planned dividing lines 23a may be irradiated once with a laser beam, and then the wafer 21 may be divided by repeating the same operation.

Also, the above-mentioned embodiment is an example in which the laser beam is irradiated to the second planned dividing line 23b by the same method after finishing the process of irradiating the laser beam to all the first planned dividing lines 23a. The sequence of beams is not limited to this. For example, irradiation of the laser beam to the first planned dividing line 23 a and irradiation of the laser beam to the second planned dividing line 23 b may be alternately performed. In this case, an operation of rotating the chuck table 4 shown in FIG. 3 by 90° in the horizontal direction is performed every time the laser beam is irradiated to the planned dividing line.

In addition, the structure, method, etc. of the said embodiment can be changed suitably unless it deviates from the scope of the objective of this invention.

11‧‧‧Wafer 11a‧‧‧surface 11b‧‧‧back side 13a‧‧‧The first dividing line 13b‧‧‧The second division line 15‧‧‧Devices 17‧‧‧Cutting Tape 19‧‧‧Framework 21‧‧‧Wafer 21a‧‧‧1st area 21b‧‧‧Second area 21c‧‧‧The third area 21d‧‧‧4th area 21e‧‧‧5th area 21f‧‧‧6th area 23a‧‧‧The first dividing line 23b‧‧‧The second division line 2‧‧‧Laser processing device 4‧‧‧Suction cup table 6‧‧‧Laser processing unit 8‧‧‧Shell 10‧‧‧Concentrator 12‧‧‧camera means

FIG. 1 is a perspective view showing a configuration example of a wafer. FIG. 2 is a perspective view showing a wafer supported by a frame. 3 is a perspective view schematically showing a state in which a wafer is supported by a laser processing device. 4 is a plan view showing how a wafer is divided into a first region and a second region. 5 is a plan view showing how the second area is divided into a third area and a plurality of fourth areas. FIG. 6 is a plan view showing a fifth area in which the fourth area is divided into two. FIG. 7 is a plan view showing a sixth area in which the fifth area is divided into two.

11‧‧‧Wafer

11a‧‧‧surface

11b‧‧‧back side

13a‧‧‧The first dividing line

13b‧‧‧The second division line

15‧‧‧Devices

Claims (3)

A method of processing a wafer, which is a method of processing a wafer in which devices are respectively formed in a plurality of regions demarcated by a plurality of planned dividing lines by irradiating a laser beam along the planned dividing lines, It is characterized in that it includes: a calculation process, which is based on the influence of the deviation of the heat conduction of the wafer caused by the irradiation of the laser beam, and it is determined that the two sides of the planned division line to which the laser beam is irradiated should be ensured. The size of the region, when the number of columns of the device contained in the region is set as N (N is a natural number), calculate the minimum ²ⁿ that meets N< ²ⁿ (n is a natural number); ²ⁿ processing engineering , which is to irradiate the laser beam to the planned dividing line at intervals of the smallest 2 ⁿ divisions of the planned dividing line, thereby forming processing marks on the wafer; and bisecting the processing process, which is to use The process of forming a processing mark on the wafer by irradiating a laser beam to the planned dividing line that bisects the number of columns of the device included in the plurality of regions partitioned by the processing mark is repeated until the process mark is formed by the process mark The number of columns of the device included in the area partitioned by the processing traces becomes 1. The wafer processing method of claim 1, wherein, in the ²ⁿ processing process, the laser beam is irradiated along the planned division line closest to the end of the wafer to form a cut-off process. After the processing marks as deep as the wafer, the other planned dividing lines are irradiated with the laser beam at intervals of the minimum 2 ⁿ divisions of the planned dividing lines. The wafer processing method as claimed in item 1 or 2 of the scope of the patent application, wherein the wafer is a GaAs wafer.

Images