JP2007149820A - Laser machining method of wafer - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a laser machining method of a wafer with which the wafer can securely be divided along a street formed in the wafer and deterioration of quality of a divided device can be suppressed. <P>SOLUTION: In the laser machining method of the wafer 20; the wafer 20 is laser-machined along the streets where a function layer is formed on a surface of a substrate, and the device is formed on a plurality of regions formed by a plurality of streets arranged in a lattice shape. Machining regions G and non-machining regions H are alternately established in the respective streets. Pulse laser beams are irradiated along the machining regions G so as to form a laser machining groove 210. Thus, laser machining grooves 210 and non-machining regions H are alternately formed along the streets. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a wafer laser processing method in which a plurality of regions are formed by a plurality of streets formed in a lattice pattern on a surface, and a wafer in which devices are formed in the plurality of regions is laser processed along the streets.

  In the semiconductor device manufacturing process, a plurality of regions are partitioned by dividing lines called streets arranged in a lattice pattern on the surface of a substantially wafer-shaped semiconductor wafer, and devices such as ICs, LSIs, etc. are partitioned in the partitioned regions. Form. Then, the semiconductor wafer is cut along the streets to divide the region in which the device is formed to manufacture individual semiconductor chips. In addition, an optical device wafer in which light-emitting elements such as light-emitting diodes (LEDs) are laminated on the surface of a sapphire substrate is also cut into individual optical devices such as light-emitting diodes (LEDs) by cutting along the streets, and electric equipment Widely used.

  The above-described cutting along the wafer street is usually performed by a cutting device called a dicer. This cutting apparatus includes a chuck table for holding a workpiece such as a wafer, a cutting means for cutting the workpiece held on the chuck table, and a cutting for relatively moving the chuck table and the cutting means. And feeding means. The cutting means includes a cutting tool including a rotating spindle and a grindstone blade mounted on the spindle, and a spindle unit including a drive mechanism for driving the rotating spindle to rotate. In such a cutting apparatus, the cutting tool and the work piece held on the chuck table are relatively cut and fed while rotating the cutting tool at a rotational speed of 20000 to 40000 rpm. However, cutting with a cutting device cannot increase the processing speed, and is not always satisfactory in terms of productivity.

On the other hand, in recent years, as a method of dividing a wafer in which an optical element such as a nitride semiconductor is laminated on the surface of a sapphire substrate along the street, the laser processing groove is formed by irradiating a pulse laser beam along the street formed on the wafer. There has been proposed a method of forming and breaking the wafer along the street by applying an external force along the laser-processed groove. (For example, refer to Patent Document 1.)
JP-A-10-305420

  Thus, according to the laser processing method described above, the laser processing groove can be formed at a relatively high processing speed. However, since the wall surface of the laser processing groove formed along the street is melted and roughened, there is a problem that the luminance is lowered when the device divided individually is a light emitting diode (LED).

  The present invention has been made in view of the above-mentioned facts, and its main technical problem is that it can be surely divided along the street formed on the wafer and suppresses the deterioration of the quality of the divided device. It is to provide a wafer laser processing method that can be used.

In order to solve the above main technical problem, according to the present invention, a wafer in which a device is formed in a plurality of regions formed by a plurality of streets arranged in a lattice shape in which a functional layer is formed on a surface of a substrate is provided. A wafer laser processing method for laser processing along
By alternately setting machining areas and non-machining areas for each street and irradiating a pulse laser beam along the machining areas to form laser machining grooves, the laser machining grooves and non-machining areas are formed along the streets. Alternately forming,
A wafer laser processing method is provided.

The substrate of the wafer is formed of sapphire, the functional layer is made of a nitride semiconductor layer, and the device is a light emitting diode.
The processing region is preferably set to a region including intersection positions of a plurality of streets arranged in a lattice pattern.
The ratio of the length of the processed area to the length of the non-processed area is set to 1: 1 to 2: 1.
When the thickness of the wafer is t, the length L1 of the processed region is set to t ≦ L1 ≦ 3t, and the length L2 of the non-processed region is set to 0.5t ≦ L2 ≦ 1.5t. Yes.
Further, the depth of the laser processed groove is set to 15 to 25 μm.

  According to the wafer laser processing method of the present invention, the processing region and the non-processing region are alternately set on the street, and the laser processing groove is formed by irradiating the pulse laser beam along the processing region. Since the laser-processed groove and the non-processed region are alternately formed along the laser beam, the laser-processed groove becomes a starting point of the break, and this break propagates along the non-processed region, so that the wafer is reliably broken along the street. . In the device thus broken and divided individually, the molten layer produced by irradiating the upper part of the side surface with the laser beam remains, but this molten layer is small relative to the area of the side surface of the device. Therefore, even when the device is a light emitting diode (LED), the decrease in luminance is slight and does not affect the quality.

  Preferred embodiments of a wafer laser processing method according to the present invention will be described below in more detail with reference to the accompanying drawings.

  FIG. 1 is a perspective view of a laser processing apparatus for carrying out a wafer laser processing method according to the present invention. A laser processing apparatus shown in FIG. 1 includes a stationary base 2, a chuck table mechanism 3 that is disposed on the stationary base 2 so as to be movable in a machining feed direction indicated by an arrow X, and holds a workpiece. The laser beam irradiation unit support mechanism 4 is movably disposed in the indexing feed direction indicated by the arrow Y perpendicular to the direction indicated by the arrow X in FIG. 2, and is movable in the direction indicated by the arrow Z to the laser beam irradiation unit support mechanism 4 And a laser beam irradiation unit 5 disposed in the.

  The chuck table mechanism 3 includes a pair of guide rails 31, 31 arranged in parallel along the machining feed direction indicated by the arrow X on the stationary base 2, and the arrow X on the guide rails 31, 31. A first slide block 32 movably disposed in the processing feed direction; and a second slide block 33 disposed on the first slide block 32 movably in the index feed direction indicated by an arrow Y; A cover table 35 supported by a cylindrical member 34 on the second sliding block 33 and a chuck table 36 as a workpiece holding means are provided. The chuck table 36 includes a suction chuck 361 formed of a porous material, and holds, for example, a disk-shaped semiconductor wafer, which is a workpiece, on the suction chuck 361 by suction means (not shown). . The chuck table 36 configured as described above is rotated by a pulse motor (not shown) disposed in the cylindrical member 34. The chuck table 36 is provided with a clamp 362 for fixing an annular frame described later.

  The first sliding block 32 is provided with a pair of guided grooves 321 and 321 fitted to the pair of guide rails 31 and 31 on the lower surface thereof, and in the index feed direction indicated by an arrow Y on the upper surface thereof. A pair of guide rails 322 and 322 formed in parallel with each other are provided. The first sliding block 32 configured in this way is processed by the arrow X along the pair of guide rails 31, 31 when the guided grooves 321, 321 are fitted into the pair of guide rails 31, 31. It is configured to be movable in the feed direction. The chuck table mechanism 3 in the illustrated embodiment includes a machining feed means 37 for moving the first sliding block 32 along the pair of guide rails 31 and 31 in the machining feed direction indicated by the arrow X. The processing feed means 37 includes a male screw rod 371 disposed in parallel between the pair of guide rails 31 and 31, and a drive source such as a pulse motor 372 for rotationally driving the male screw rod 371. One end of the male screw rod 371 is rotatably supported by a bearing block 373 fixed to the stationary base 2, and the other end is connected to the output shaft of the pulse motor 372 by transmission. The male screw rod 371 is screwed into a penetrating female screw hole formed in a female screw block (not shown) provided on the lower surface of the central portion of the first sliding block 32. Therefore, when the male screw rod 371 is driven to rotate forward and backward by the pulse motor 372, the first sliding block 32 is moved along the guide rails 31, 31 in the machining feed direction indicated by the arrow X.

  The laser processing apparatus in the illustrated embodiment includes a processing feed amount detecting means 374 for detecting the processing feed amount of the chuck table 36. The processing feed amount detection means 374 includes a linear scale 374a disposed along the guide rail 31, and a read head disposed along the linear scale 374a along with the first sliding block 32 disposed along the first sliding block 32. 374b. In the illustrated embodiment, the reading head 374b of the feed amount detecting means 374 sends a pulse signal of one pulse every 1 μm to the control means described later. Then, the control means to be described later detects the machining feed amount of the chuck table 36 by counting the input pulse signals. When the pulse motor 372 is used as the drive source of the machining feed means 37, the machining feed amount of the chuck table 36 is counted by counting the drive pulses of the control means to be described later that outputs a drive signal to the pulse motor 372. Can also be detected. When a servo motor is used as a drive source for the machining feed means 37, a pulse signal output from a rotary encoder that detects the rotation speed of the servo motor is sent to a control means described later, and the pulse signal input by the control means. By counting, the machining feed amount of the chuck table 36 can also be detected.

  The second sliding block 33 is provided with a pair of guided grooves 331 and 331 which are fitted to a pair of guide rails 322 and 322 provided on the upper surface of the first sliding block 32 on the lower surface thereof. By fitting the guided grooves 331 and 331 to the pair of guide rails 322 and 322, the guided grooves 331 and 331 are configured to be movable in the indexing and feeding direction indicated by the arrow Y. The chuck table mechanism 3 in the illustrated embodiment is for moving the second slide block 33 along the pair of guide rails 322 and 322 provided in the first slide block 32 in the index feed direction indicated by the arrow Y. First index feeding means 38 is provided. The first index feed means 38 includes a male screw rod 381 disposed in parallel between the pair of guide rails 322 and 322, and a drive source such as a pulse motor 382 for rotationally driving the male screw rod 381. It is out. One end of the male screw rod 381 is rotatably supported by a bearing block 383 fixed to the upper surface of the first sliding block 32, and the other end is connected to the output shaft of the pulse motor 382. The male screw rod 381 is screwed into a penetrating female screw hole formed in a female screw block (not shown) provided on the lower surface of the central portion of the second sliding block 33. Therefore, when the male screw rod 381 is driven to rotate forward and reversely by the pulse motor 382, the second slide block 33 is moved along the guide rails 322 and 322 in the index feed direction indicated by the arrow Y.

  The laser beam irradiation unit support mechanism 4 includes a pair of guide rails 41, 41 arranged in parallel along the indexing feed direction indicated by the arrow Y on the stationary base 2, and the arrow Y on the guide rails 41, 41. The movable support base 42 is provided so as to be movable in the direction indicated by. The movable support base 42 includes a movement support portion 421 that is movably disposed on the guide rails 41, 41, and a mounting portion 422 that is attached to the movement support portion 421. The mounting portion 422 is provided with a pair of guide rails 423 and 423 extending in the direction indicated by the arrow Z on one side surface in parallel. The laser beam irradiation unit support mechanism 4 in the illustrated embodiment includes a second index feed means 43 for moving the movable support base 42 along the pair of guide rails 41, 41 in the index feed direction indicated by the arrow Y. is doing. The second index feed means 43 includes a male screw rod 431 disposed in parallel between the pair of guide rails 41, 41, and a drive source such as a pulse motor 432 for rotationally driving the male screw rod 431. It is out. One end of the male screw rod 431 is rotatably supported by a bearing block (not shown) fixed to the stationary base 2, and the other end is connected to the output shaft of the pulse motor 432. The male screw rod 431 is screwed into a female screw hole formed in a female screw block (not shown) provided on the lower surface of the central portion of the moving support portion 421 constituting the movable support base 42. For this reason, when the male screw rod 431 is driven to rotate forward and backward by the pulse motor 432, the movable support base 42 is moved along the guide rails 41, 41 in the index feed direction indicated by the arrow Y.

  The laser processing apparatus in the illustrated embodiment includes index feed amount detection means 433 for detecting the index feed amount of the movable support base 42 of the laser beam irradiation unit support mechanism 4. The index feed amount detecting means 433 includes a linear scale 433a disposed along the guide rail 41 and a read head 433b disposed on the movable support base 42 and moving along the linear scale 433a. In the illustrated embodiment, the read head 433b of the feed amount detection means 433 sends a pulse signal of one pulse every 1 μm to the control means described later. And the control means mentioned later detects the index feed amount of the laser beam irradiation unit 5 by counting the input pulse signal. In the case where the pulse motor 432 is used as the drive source of the second index sending means 43, the laser beam irradiation unit 5 is counted by counting the drive pulses of the control means described later that outputs a drive signal to the pulse motor 432. It is also possible to detect the index feed amount. Further, when a servo motor is used as the drive source of the second index sending means 43, a pulse signal output from a rotary encoder that detects the rotation speed of the servo motor is sent to the control means described later, and the control means inputs it. The index feed amount of the laser beam irradiation unit 5 can also be detected by counting the number of pulse signals.

  The laser beam irradiation unit 5 in the illustrated embodiment includes a unit holder 51 and laser beam irradiation means 52 attached to the unit holder 51. The unit holder 51 is provided with a pair of guided grooves 511 and 511 that are slidably fitted to a pair of guide rails 423 and 423 provided in the mounting portion 422. By being fitted to the guide rails 423 and 423, the guide rails 423 and 423 are supported so as to be movable in the direction indicated by the arrow Z.

  As shown in FIG. 2, the laser beam irradiation means 52 is oscillated by the pulse laser beam oscillation means 522 disposed at the tip of the casing 521 and the pulse laser beam oscillation means 522 and the transmission optical system 523 disposed in the casing 521. A condenser 53 for irradiating a workpiece held on the chuck table 36 with a pulsed laser beam is provided. The pulse laser beam oscillation means 522 is composed of a pulse laser beam oscillator 522a composed of a YAG laser oscillator or a YVO4 laser oscillator, and a repetition frequency setting means 522b attached thereto. This repetition frequency setting means 522b is controlled by the control means described later. The transmission optical system 523 includes an appropriate optical element such as a beam splitter.

  An imaging means 6 for detecting a processing region to be laser processed by the laser beam irradiation means 52 is disposed at the front end of the casing 521 constituting the laser beam irradiation means 52. The imaging unit 6 includes an illuminating unit that illuminates the workpiece, an optical system that captures an area illuminated by the illuminating unit, an imaging device (CCD) that captures an image captured by the optical system, and the like. The captured image signal is sent to a control means (not shown).

  The laser beam irradiation unit 5 in the illustrated embodiment includes a moving means 53 for moving the unit holder 51 along the pair of guide rails 423 and 423 in the direction indicated by the arrow Z. The moving means 53 includes a male screw rod (not shown) disposed between the pair of guide rails 423 and 423, and a drive source such as a pulse motor 532 for rotationally driving the male screw rod. By driving the male screw rod (not shown) in the forward and reverse directions by the motor 532, the unit holder 51 and the laser beam irradiation means 52 are moved along the guide rails 423 and 423 in the direction indicated by the arrow Z. In the illustrated embodiment, the laser beam irradiation means 52 is moved upward by driving the pulse motor 532 forward, and the laser beam irradiation means 52 is moved downward by driving the pulse motor 532 in reverse. Yes.

  The laser processing apparatus in the illustrated embodiment includes a control means 10. The control means 10 is constituted by a computer, and a central processing unit (CPU) 101 that performs arithmetic processing according to a control program, a read-only memory (ROM) 102 that stores a control program and the like, and a pulse laser beam on a workpiece to be described later. A readable / writable random access memory (RAM) 103 that stores data of X and Y coordinate values of the irradiation start point and end point, calculation results, and the like, a counter 104, an input interface 105, and an output interface 106 are provided. Detection signals from the machining feed amount detection means 374, the index feed amount detection means 433, the imaging means 6 and the like are input to the input interface 105 of the control means 10. A control signal is output from the output interface 106 of the control means 10 to the pulse motor 372, pulse motor 382, pulse motor 432, pulse motor 532, laser beam irradiation means 52, and the like.

The laser processing apparatus in the illustrated embodiment is configured as described above, and a wafer laser processing method performed using the laser processing apparatus will be described below.
FIG. 3 shows a state in which an optical device wafer 20 as a wafer to be laser-processed is attached to a protective tape 22 made of a synthetic resin sheet such as polyolefin mounted on an annular frame 21 with the surface 20a facing upward. It is shown. In the optical device wafer 20, for example, a functional layer made of a nitride semiconductor layer is formed on the surface of a sapphire substrate having a thickness of 90 μm. A plurality of rectangular areas defined by a plurality of first streets 201 and a plurality of second streets 202 arranged in a grid are formed on the surface 20a of the optical device wafer 20, and the plurality of rectangular areas are formed in the plurality of rectangular areas. A device 203 made of a light emitting diode (LED) is formed.

  As shown in FIG. 3, the optical device wafer 20 supported by the annular frame 21 via the protective tape 22 places the protective tape 22 side on the chuck table 36 of the laser processing apparatus shown in FIG. Then, by operating a suction means (not shown), the optical device wafer 20 is sucked and held on the chuck table 36 via the protective tape 22. The annular frame 21 is fixed by a clamp 362.

  As described above, the chuck table 36 that sucks and holds the optical device wafer 20 is positioned directly below the imaging unit 9 by the processing feeding unit 37. When the chuck table 36 is positioned directly below the image pickup means 6, the image pickup means 6 and the control means 10 execute an alignment operation for detecting a processing region to be laser processed of the optical device wafer 20. In other words, the imaging unit 6 and the control unit 10 include a first street 201 formed in a predetermined direction of the optical device wafer 20 and a condenser of the laser beam irradiation unit 52 that irradiates the laser beam along the first street 201. Image processing such as pattern matching for positioning with 53 is executed, and alignment of the laser beam irradiation position is performed. Similarly, the alignment of the laser beam irradiation position is performed on the second street 202 formed on the optical device wafer 20.

  When alignment is performed as described above, the optical device wafer 20 on the chuck table 36 is positioned at the coordinate position shown in FIG. 4B shows a state where the chuck table 36, that is, the optical device wafer 20, is rotated by 90 degrees from the state shown in FIG. 4A.

  As described above, if the first street 201 and the second street 202 formed on the optical device wafer 20 held on the chuck table 36 are detected and the laser beam irradiation position is aligned, the chuck The table 36 is moved, and the uppermost first street 201 in the first street 201 is positioned directly below the imaging means 9 in the state of FIG. Further, as shown in FIG. 5, one end (the left end in FIG. 5) of the first street 201 is positioned directly below the imaging means 9. If the imaging means 9 detects one end (left end in FIG. 5) of the first street 201 in this state, the coordinate value (A1 in FIG. 4A) is used as the processing feed start position coordinate value to the control means 10. send. Next, the chuck table 36 is moved in the direction indicated by the arrow X1 in FIG. 5, and the other end (the right end in FIG. 5) of the first street 201 is positioned directly below the imaging means 6. During this time, as shown in FIG. 4 (a), the imaging means 6 has the coordinate values (E1, E2, E3... En) of the intersection with the second street 202 and the other end (the right end in FIG. 4 (a)). ) Coordinate values (B1) are detected and sent to the control means 10 as intersection coordinate values and machining feed end position coordinate values, respectively. The control means 10 randomly accesses the machining feed start position coordinate value (A1), the intersection coordinate values (E1, E2, E3... En), and the machining feed end position coordinate value (B1) of the input first street 201. Temporarily stored in the memory (RAM) 103 (street detection step).

  When the machining feed start position coordinate value, the intersection coordinate value, and the machining feed end position coordinate value of the uppermost first street 201 in FIG. 4A are detected in this way, the chuck table 36 is moved to the first table 201. Indexing and feeding is performed in the direction indicated by the arrow Y by the interval of the streets 201, and the second first street 201 from the top in FIG. Then, the above-described street detection process is performed on the second first street 201 from the top, and the processing feed start position coordinate value (A2) of the second first street 201 from the top and the intersection The coordinate values (E1, E2, E3... En) and the machining feed end position coordinate value (B2) are detected and temporarily stored in a random access memory (RAM) 103. Thereafter, the above-described index feed and street detection steps are repeatedly executed up to the lowest first street 201 in FIG. 4A, and the processing feed start position coordinate value (A3 to An) of the first street 201 and the intersection. Coordinate values (E1, E2, E3... En) and machining feed end position coordinate values (B3 to Bn) are detected and temporarily stored in a random access memory (RAM) 103.

As described above, when the street detection process is performed on the first street 201, the chuck table 36 and thus the optical device wafer 20 are rotated 90 degrees to be positioned in the state shown in FIG. Next, the street detection process described above is performed on the second streets 202, and the processing feed start position coordinate values (C1 to Cn) and the intersection coordinate values (F1, F2, F3,. Fn) and processing feed end position coordinate values (D1 to Dn) are detected and temporarily stored in a random access memory (RAM) 103.
Note that the machining feed start position coordinate values (A1 to An), the intersection coordinate values (E1, E2, E3... En), and the machining feed end position coordinate values for the first street 201 formed on the optical device wafer 20 ( B1 to Bn) and machining feed start position coordinate values (C1 to Cn), intersection coordinate values (F1, F2, F3... Fn) and machining feed end position coordinate values (D1 to Dn) for the second street 202 are The design value of the optical device wafer 20 may be stored in advance in the read-only memory (ROM) 102 or the random access memory (RAM) 103, and the street detection process described above may be omitted.

Next, a processing region and a non-processing region are alternately set for each street formed on the optical device wafer 20, and a laser beam is formed along the processing region to form a laser processing groove. A fracture line forming process is performed in which laser-processed grooves and non-processed regions are alternately formed along the line.
In the breaking line forming step, first, the chuck table 36 is moved so that the uppermost first street 21 is positioned directly below the condenser 53 of the laser beam irradiation means 52 in FIG. Further, as shown in FIG. 6 (a), the machining feed start position coordinate value (A1) which is one end of the first street 201 (the left end in FIG. 6 (a)) (see FIG. 4 (a)). Is positioned directly below the condenser 53. At this time, the condensing point P of the pulse laser beam is matched with the vicinity of the surface 20 a (upper surface) of the optical device wafer 20. Then, while the chuck table 36, that is, the optical device wafer 20 is processed and sent in the direction indicated by the arrow X1 in FIG. 6, a pulse laser beam having a wavelength of, for example, 360 nm or less having absorption from the condenser 53 to the optical device wafer 20 Repeat irradiation and stopping. When the other end of the first street 201 (the right end in FIG. 6A), that is, the processing feed end position coordinate value (B1) reaches the irradiation position of the condenser 53 of the laser beam irradiation means 52, the pulse laser beam Irradiation is stopped and processing feed of the chuck table 36, that is, the optical device wafer 20, is stopped. As a result, as shown in FIG. 6B, a laser processing groove 210 is formed in the processing region G irradiated with the pulse laser beam along the first street 201. Therefore, on the first street 201, the laser processing grooves 210 and the non-processing regions H formed in the processing region A are alternately formed.

Here, the relationship between the depth of the laser processing groove 210 formed in the length L1 of the processing region G and the length L1 of the processing region G and the length L2 of the non-processing region H will be described.
As the depth of the laser processing groove 210 formed in the length L1 of the processing region G increases, the break along the laser processing groove 210 of the optical device wafer 20 becomes easier, but the brightness of the optical device 203 decreases. Further, as the depth of the laser processing groove 210 is shallower, the brightness of the optical device 203 is less decreased, but it becomes difficult to break the optical device wafer 20 along the laser processing groove 210. Therefore, in order to suppress a decrease in luminance of the optical device 203 and facilitate breakage along the laser processing groove 210 of the optical device wafer 20, when the thickness of the optical device wafer 20 is around 100 μm, the depth of the laser processing groove 210 is increased. A thickness of 15 to 25 μm is appropriate.
Further, in the relationship between the length L1 of the processing region G and the length L2 of the non-processing region H, the break along the laser processing groove 210 of the optical device wafer 20 increases as the ratio of the length L1 of the processing region G increases. Although it is easy, the luminance of the optical device 23 decreases, and as the ratio of the length L2 of the non-processed region H increases, the luminance of the optical device 23 decreases less, but the optical device wafer 20 breaks along the laser processing groove 210. It becomes difficult. Therefore, in order to suppress a decrease in luminance of the optical device 203 and facilitate breakage along the laser processing groove 210 of the optical device wafer 20, the length L1 of the processing region G and the length L2 of the non-processing region H are The length ratio is suitably L1: L2 = 1: 1 to 2: 1.
When the thickness of the optical device wafer 20 is t (μm), the length of the laser processing groove 210 formed in the processing region G is set such that L1 is t ≦ L1 ≦ 3t. The length L2 is desirably set to 0.5t ≦ L2 ≦ 1.5t.

Next, an embodiment of the material and size of the optical device wafer 20 for carrying out the break line forming process described above and the processing conditions of the break line forming process will be described.
(1) Material and size of optical device wafer:
Wafer substrate: Sapphire Laminated functional layer: GaN film Wafer size: 50mm diameter
Wafer thickness: 90 μm
Device size: 300 μm × 300 μm
(2) Processing conditions:
Light source: YAG laser Wavelength: 355nm
Average output: 1W
Repetition frequency: 50 kHz
Condensing spot diameter: φ5μm
Processing feed rate: 50 mm / sec

An embodiment of a processing method when the length L1 of the processing region G is set to 200 μm and the length L2 of the non-processing region H is set to 100 μm according to the processing conditions will be described.
As described above, when the processing feed rate is 50 mm / second, the time required to move the 200 μm processing region G is 0.004 seconds. Further, the time required to move the unprocessed region H of 100 μm is 0.002 seconds. Therefore, after irradiating the pulse laser beam for 0.004 seconds, stopping the irradiation of the pulse laser beam for 0.002 seconds, and repeating this, the laser processing groove 210 having a length of 200 μm and a length of 100 μm along the street An unprocessed region H can be formed. Note that, under the above processing conditions, the depth of the laser processing groove 210 is about 20 μm.

  As described above, when the fracture line forming step is performed on the uppermost first street 201 in FIG. 4A formed on the optical device wafer 20, the chuck table 36 is moved to the arrow shown in FIG. Index feed is performed in the direction indicated by Y by an interval of the first street 201 (index feed process), and the above-described fracture line forming process is performed. When the broken line forming process is performed on all the first streets 201 extending in a predetermined direction of the optical device wafer 20 in this way, the chuck table 36 and thus the optical device wafer 20 held on the first street 201 are moved 90 degrees. All the first streets of the optical device wafer 20 are rotated by performing the broken line forming step along the second streets 202 formed in a direction orthogonal to the first streets 201. The laser processing grooves 210 are formed in the processing region G along the 201 and the second street 202, and the laser processing grooves 210 and the non-processing regions H are alternately formed.

  The laser processing groove 210 formed in the processing region G in the break line forming process described above is preferably formed in a region including the intersection position of the first street 201 and the second street 202 as shown in FIG. . In this way, in order to set the region including the intersection position of the first street 201 and the second street 202 as the processing region G and irradiate the processing region G with the pulsed laser beam, the detection signal from the processing feed amount detection means 374 is detected. The laser beam irradiation means 52 may be controlled based on the intersection coordinate value data stored in the read only memory (ROM) 102 or the random access memory (RAM) 103 of the control means 10. That is, when the processing area G is set to 200 μm and the non-processing area H is set to 100 μm on the optical device wafer having a device size of 300 μm × 300 μm as described above, the detection signal from the processing feed amount detection means 374 A pulse laser beam is irradiated from the position 100 μm before the intersection of the first street 201 and the second street 202 to the position of 100 μm after passing through the intersection. As a result, a laser processing groove 210 having a length of 200 μm is formed along the street in the processing region G including the intersection, and the laser processing groove 210 having a length of 200 μm and the non-processing region H having a length of 100 μm are alternately arranged. It is formed.

  As described above, the optical device wafer 20 on which the fracture line forming process has been performed is transported to the dividing process which is the next process. In the dividing step, an external force is applied to the optical device wafer 20 along the break line formed by the laser processing groove 210 formed in the processing region G of the first street 201 and the second street 202 and the non-processing region H. The As a result, the laser processing groove 210 becomes the starting point of the breakage, and the breakage propagates along the unprocessed region H, so that the optical device wafer 20 is reliably broken along the first street 201 and the second street 202. The

  In this way, in the device 23 in which the optical device wafer 20 is broken along the first street 201 and the second street 202 and individually divided, the broken line is formed at the upper part of the side surface as shown in FIG. The molten layer 210a generated by the laser beam irradiation in the process remains. However, since the thickness of the molten layer 210a is about 25 μm to 20 μm, the thickness of the molten layer 210a is small with respect to the area of the side surface of the optical device 23. There is no impact on quality.

  In addition, in order to make it easier to break along the first street 201 and the second street 202 of the optical device wafer 20 in the division step, the luminance of the device 203 is affected in the unprocessed region H. A laser processing groove having a depth of about 5 μm may be formed.

The perspective view of the laser processing apparatus for enforcing the laser processing method of the wafer by this invention. The block diagram of the laser beam irradiation means with which the laser processing apparatus shown in FIG. 1 is equipped. The perspective view which shows the state which affixed the optical device wafer processed with the laser processing method of the wafer by this invention to the protective tape with which the cyclic | annular flame | frame was mounted | worn. FIG. 4 is an explanatory diagram showing a relationship with a coordinate position in a state where the optical device wafer shown in FIG. 3 is held at a predetermined position of the chuck table of the laser processing apparatus shown in FIG. Explanatory drawing of the street detection process implemented with the laser processing apparatus shown in FIG. Explanatory drawing of the broken line formation process implemented with the laser processing apparatus shown in FIG. Explanatory drawing which shows the state which set the process area | region set to the optical device wafer shown in FIG. 3 to the area | region containing the intersection position of a street. The perspective view of the device which divided | segmented the optical device wafer processed with the laser processing method of the wafer by this invention separately.

Explanation of symbols

2: Stationary base 3: Chuck table mechanism 31: Guide rail 36: Chuck table 37: Work feed means 374: Work feed amount detection means 38: First index feed means
4: Laser beam irradiation unit support mechanism 41: Guide rail 42: Movable support base 43: Second index feed means 433: Index feed amount detection means 5: Laser beam irradiation unit 51: Unit holder 52: Laser beam processing means 53: Light collection Device 6: Imaging means 10: Control means 20: Optical device wafer 201: First street 202: Second street 203: Device 210: Laser processing groove 21: Ring frame 22: Protective tape

Claims (6)

A wafer laser processing method for laser processing a wafer having a device formed in a plurality of regions formed by a plurality of streets in which functional layers are formed on a surface of a substrate and arranged in a grid pattern,
By alternately setting machining areas and non-machining areas for each street and irradiating a pulse laser beam along the machining areas to form laser machining grooves, the laser machining grooves and non-machining areas are formed along the streets. Alternately forming,
A wafer laser processing method characterized by the above.   2. The wafer laser processing method according to claim 1, wherein the wafer substrate is formed of sapphire, the functional layer is formed of a nitride semiconductor layer, and the device is a light emitting diode.   The wafer laser processing method according to claim 1, wherein the processing region is set to a region including intersection positions of a plurality of streets arranged in a lattice pattern.   The wafer laser processing method according to claim 1, wherein a ratio of the length of the processing region to the length of the non-processing region is set to 1: 1 to 2: 1.   When the thickness of the wafer is t, the length L1 of the processing region is set to t ≦ L1 ≦ 3t, and the length L2 of the non-processing region is set to 0.5t ≦ L2 ≦ 1.5t. 5. A wafer laser processing method according to claim 4.   The wafer laser processing method according to claim 1, wherein the depth of the laser processing groove is set to 15 to 25 μm.

Images