JP2013132664A - Method for processing workpiece - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a laser processing method capable of reliably forming a division starting point with high energy utilization efficiency.SOLUTION: The method for forming division starting points on the workpiece includes: an ejecting step of ejecting an amplified light from a light source by amplifying an oscillation pulse light oscillated by an oscillator in an amplifier with a CPA method; and an irradiation step of creating a cleavage or dehiscence of the workpiece between the irradiated areas by applying a pulse laser light to the workpiece so that the irradiated area for the individual unit pulse light is discretely formed on the processed surface, thereby forming the division starting point on the workpiece. In the ejecting step, the pulse laser light is ejected from the light source to the workpiece as two pulse lights in which the irradiated areas are substantially the same in the irradiation step while delaying at the time interval shorter than the ejecting period of the unit pulse light by regulating the timing of taking out the pulse laser light from the amplifier.

Description

  The present invention relates to a processing method for processing a workpiece by irradiating a laser beam.

  As a technique for processing a workpiece by irradiating pulsed laser light (hereinafter also simply referred to as laser light) (hereinafter also simply referred to as laser processing or laser processing technique), an ultrashort pulse laser having a pulse width of the order of psec. By irradiating the upper surface of the workpiece while scanning the light, the workpiece is sequentially cleaved or cleaved between the irradiated regions for each unit pulsed light, and is formed in each. A method of forming a starting point (division starting point) for division as a cleavage plane or a continuous surface of a cleavage plane is already known (see, for example, Patent Document 1).

JP 2011-131256 A

  Specifically, the processing method (cleavage / dehiscence processing) disclosed in Patent Document 1 irradiates discretely with unit pulse light having a pulse width of 100 psec or less at intervals of about 4 μm to 50 μm. Cleavage / cleavage along the crystal plane between the irradiated regions (or more macroscopic than that) by causing alteration, melting, evaporative removal, etc. of the substance in the central portion of the irradiated region This is a technique for developing cracks. Therefore, it is not necessary to perform processing more than necessary in the irradiated region. Rather, it is required that the cleaving / dehiscence progresses reliably from the irradiated region.

  For example, when unit pulse light with a high peak power density and a small pulse width is irradiated, the energy applied to the irradiated region becomes excessive and damages the irradiated region more than necessary, while cleavage / dehiscence is preferred. It is possible that things will not progress. This is because the energy of the irradiated unit pulse light cannot be sufficiently directed to the progress of cleavage / dehiscence.

  The present invention has been made in view of the above problems, and an object of the present invention is to provide a laser processing method that has high energy utilization efficiency and can form the division starting point more reliably.

  In order to solve the above-mentioned problem, the invention of claim 1 is a processing method for forming a division starting point in a workpiece, and amplifies an oscillation pulse light oscillated by an oscillator provided in a light source in an amplifier provided in the light source. Then, an emission step of emitting pulsed laser light, which is amplified light, from the light source, and an irradiation region for each unit pulse light of the pulsed laser light are discretely formed on the processing surface of the workpiece. By irradiating the workpiece with the pulsed laser light as described above, causing the workpiece to be cleaved or cleaved between the irradiated regions, the starting point for dividing the workpiece And in the emission step, the pulse laser beam is adjusted by adjusting the extraction timing of the pulse laser beam from the amplifier. The workpiece is irradiated as two pulsed light beams that are delayed by a time interval shorter than the emission period of the unit pulsed light while the irradiated region is substantially the same in the irradiation step. The light is emitted from the light source.

  According to a second aspect of the present invention, there is provided a processing method for a workpiece according to the first aspect, wherein in the emission step, the amplified light is extracted from the amplifier, and only a part of the amplified light is the amplifier. The two pulses are divided into a first time extracted from the first time and a second time after the remaining of the amplified light is amplified in the amplifier after the first time has elapsed. Light is emitted from the light source.

  Invention of Claim 3 is a processing method of the workpiece of Claim 1, Comprising: In the said output process, when the said oscillation pulse light is oscillated from the said oscillator and k is made into a natural number, Emitting the k-order or k + 1-order amplified light of the oscillation pulse light oscillated first and the k-order amplification light of the pulse light oscillated later from the light source in this order as the two pulse lights; It is characterized by.

  According to a fourth aspect of the present invention, there is provided a processing method for a workpiece according to any one of the first to third aspects, wherein the two lasers are adjusted by adjusting a timing at which the pulsed laser light is extracted from the amplifier. The ratio of the pulse energy of the pulsed light is adjusted.

  According to the first to fourth aspects of the present invention, the energy utilization efficiency is improved by performing the cleavage / cleavage processing along the planned processing line on the workpiece with the double pulsed laser beam. High cleavage / cleavage processing becomes possible. That is, the division starting point can be more reliably formed by cleavage / cleavage.

It is a figure which shows typically the process aspect by cleavage / cleaving process. 2 is a schematic diagram schematically showing a configuration of a laser processing apparatus 50. FIG. It is a figure which shows the structure of laser light source SL with which the laser processing apparatus 50 is provided. It is a figure which shows typically the profile of the unit pulsed light PL of the laser beam LB made into the double pulse. 6 is a diagram for explaining amplification of pulsed light in the output amplifier 105 and extraction of pulsed light from the output amplifier 105. FIG. It is a figure for demonstrating the 2nd aspect of double pulsing. It is an optical microscope image of the C surface sapphire substrate surface after a process about an Example and a comparative example.

<Processing principle>
The basic principle of processing realized in the embodiment of the present invention is the same as the principle of processing disclosed in Patent Document 1. Therefore, only the outline will be described below. In general, the processing performed in the present invention is performed by irradiating the upper surface (processing surface) of a workpiece while scanning with a pulsed laser beam (hereinafter, also simply referred to as laser beam). The workpiece is cleaved or cleaved in sequence between the irradiated areas, and the starting point (dividing starting point) for splitting is formed as a continuous surface of the cleaved surface or cleaved surface formed in each. Is.

  Note that in this embodiment mode, cleavage refers to a phenomenon in which a workpiece is cracked substantially regularly along a crystal plane other than the cleavage plane, and the crystal plane is referred to as a cleavage plane. In addition to cleaving and cleaving that are microscopic phenomena completely along the crystal plane, cracks that are macroscopic cracks may occur along a substantially constant crystal orientation. Depending on the substance, only one of cleavage, cleaving, or cracking mainly occurs, but in the following, in order to avoid complicated explanation, cleavage / cleavage is not distinguished from each other without distinguishing cleavage, cleaving, and cracking. Collectively called open. Further, the above-described processing may be simply referred to as cleavage / dehiscence processing or the like.

  In the following description, the workpiece is a hexagonal single crystal substance, and each axis of the a1, a2, and a3 axes that are symmetric with each other at an angle of 120 ° in the C plane. An example will be described in which the direction is the cleavage / easy cleavage direction and the planned processing line is perpendicular to any of the a1 axis direction, the a2 axis direction, and the a3 axis direction. More generally speaking, this is equivalent to two different cleavage / cleavage easy directions (the direction of the symmetry axis of the two cleavage / cleavage easy directions) is the direction of the planned machining line. Is the case. In the following, the laser light irradiated for each individual pulse is referred to as unit pulse light.

  FIG. 1 is a diagram schematically showing a processing mode by cleavage / cleavage processing. FIG. 1 illustrates the case where the a1 axis direction and the planned machining line L are orthogonal to each other. FIG. 1A is a diagram illustrating an azimuth relationship between the a1 axis direction, the a2 axis direction, the a3 axis direction, and the planned processing line L in such a case. FIG. 1B shows a state in which the unit pulse light of the first pulse of the laser light is irradiated to the irradiated region RE1 at the end of the processing line L.

  In general, irradiation with unit pulse light gives high energy to a very small region of the workpiece, and therefore, such irradiation is equivalent to the irradiation region of the unit pulse light (laser light) on the surface to be irradiated or the target. It causes alteration, melting, evaporative removal, etc. of substances in a range wider than the irradiation area.

  However, if the irradiation time of the unit pulse light, that is, the pulse width is set to be extremely short, a substance that is narrower than the spot size of the laser light and exists in the substantially central region of the irradiated region RE1 obtains kinetic energy from the irradiated laser light. In this way, it is transformed into a plasma or is heated to a gas state, etc., and further changes in the direction perpendicular to the surface to be irradiated. On the other hand, by irradiation with unit pulse light including reaction force caused by the scattering. The generated impact and stress act around the irradiated region, particularly in the a1 axis direction, the a2 axis direction, and the a3 axis direction, which are easy cleavage / cleavage directions. As a result, micro-cleavage or cleaving partially occurs along the direction while maintaining an apparent contact state, or thermal distortion is inherent even without cleaving or cleaving. A state occurs. In other words, it can be said that the irradiation with the ultra-short pulse unit pulse light acts as a driving force for forming a weak intensity portion that is substantially linear in a top view toward the cleavage / cleavage easy direction.

  In FIG. 1B, among the weak strength portions formed in each of the above cleavage / cleavage easy directions, the weak strength portions W11a and W12a in the −a2 direction and the + a3 direction close to the extending direction of the processing line L are shown. This is schematically indicated by a broken arrow.

  Subsequently, as shown in FIG. 1C, the second unit pulse light of the laser light is irradiated, and the irradiated region RE2 is located on the processing planned line L at a position away from the irradiated region RE1 by a predetermined distance. As in the case of the first pulse, a weak intensity portion is formed in the second pulse along the easy cleavage / cleavage direction. For example, a weak strength portion W11b is formed in the -a3 direction, a weak strength portion W12b is formed in the + a2 direction, a weak strength portion W11c is formed in the + a3 direction, and a weak strength portion W12c is formed in the -a2 direction. Will be.

  However, at this time, the weak intensity portions W11a and W12a formed by the irradiation of the unit pulse light of the first pulse exist in the extending direction of the weak intensity portions W11b and W12b, respectively. That is, the extending direction of the weak strength portions W11b and W12b is a location where cleavage or cleavage (energy absorption rate is high) can occur with less energy than other locations. Therefore, actually, when the unit pulse light of the second pulse is irradiated, the impact or stress generated at that time propagates to the easy-cleavage / cleavage direction and the weak intensity part existing ahead, and the weak intensity part Complete cleavage or cleavage occurs from W11b to the weak strength portion 11a and from the weak strength portion W11b to the weak strength portion W11a at almost the instant of irradiation. Thereby, the cleavage / cleavage surfaces C11a and C11b shown in FIG. 1 (d) are formed. The cleavage / cleavage surfaces C11a and C11b can be formed to a depth of about several μm to several tens of μm in a direction perpendicular to the drawing of the workpiece. In the cleavage / cleavage surfaces C11a and C11b, the crystal plane slips as a result of receiving a strong impact or stress, and undulations occur in the depth direction.

  Then, as shown in FIG. 1 (e), the irradiated regions RE11, RE12, RE13, RE14,... Are sequentially irradiated with unit pulse light by scanning the laser light along the planned processing line L. Then, due to the impact and stress generated during the irradiation, the cleavage / cleavage surfaces C11a and C11b, C12a and C12b, C13a and C13b, C14a and C14b. It will be formed sequentially along. In this aspect, the cleavage / dehissing surface is continuously formed by the cleavage / dehissing process in the present embodiment.

  From another point of view, the surface layer portion of the workpiece is expanded by applying thermal energy by irradiation of unit pulse light, and approximately the center of each of the irradiated regions RE11, RE12, RE13, RE14,. It can be said that cleavage / dehiscence is progressing by applying a tensile stress perpendicular to the cleavage / dehiscence planes C11a and C11b, C12a and C12b, C13a and C13b, C14a and C14b,. .

  That is, in the case shown in FIG. 1, a plurality of irradiated regions that exist discretely along the planned processing line L, and a cleavage / cleavage surface formed between the plurality of irradiated regions, As a whole, it becomes a division starting point when the workpiece is divided along the planned machining line L. After the formation of the division starting point, the workpiece can be divided in a mode generally along the planned processing line L by performing division using a predetermined jig or apparatus.

  In the case shown in FIG. 1, the unit pulse light is irradiated so that the planned processing line is perpendicular to any of the a1 axis direction, the a2 axis direction, and the a3 axis direction. A mode in which the unit pulse light is irradiated so that the processing line is parallel to any of the a1 axis direction, the a2 axis direction, and the a3 axis direction may be used. A mode in which unit pulse light that forms each irradiated region is irradiated so as to be formed in a zigzag manner (zigzag) in a mode along two cleavage / leaving easy directions sandwiching L. Also good.

  In order to realize the above cleavage / cleavage processing, it is necessary to irradiate a short pulse laser beam with a short pulse width. Specifically, it is necessary to use laser light having a pulse width of 100 psec or less. For example, it is preferable to use laser light having a pulse width of about 1 psec to 50 psec.

  On the other hand, the irradiation pitch of unit pulse light (center distance of irradiated spots) may be determined in the range of 4 μm to 50 μm. If the irradiation pitch is larger than this, the formation of the weak strength portion in the cleavage / cleavage easy direction may not progress to such an extent that a cleavage / cleavage surface can be formed. From the viewpoint of reliably forming the division starting point consisting of In view of scanning speed, processing efficiency, and product quality, it is preferable that the irradiation pitch is large. However, in order to make the formation of the cleavage / cleavage surface more reliable, it is determined within the range of 4 to 30 μm. Desirably, it is more preferably about 4 μm to 20 μm.

  Now, when the repetition frequency of laser light is R (kHz), unit pulse light is emitted from the laser light source every 1 / R (msec). When the laser beam moves relative to the workpiece at a speed V (mm / sec), the irradiation pitch Δ (μm) is determined by Δ = V / R. Therefore, the scanning speed V and the repetition frequency of the laser beam are determined so that Δ is about several μm. For example, the scanning speed V is preferably about 50 mm / sec to 3000 mm / sec, and the repetition frequency R is preferably about 1 kHz to 200 kHz, particularly about 10 kHz to 200 kHz. Specific values of V and R may be appropriately determined in consideration of the material of the workpiece, the absorption rate, the thermal conductivity, the melting point, and the like.

The laser beam is preferably irradiated with a beam diameter of about 1 μm to 10 μm. In such a case, the peak power density upon laser light irradiation is approximately 0.1 TW / cm ² to several tens TW / cm ² .

  The laser beam irradiation energy (pulse energy) may be appropriately determined within a range of 0.1 μJ to 50 μJ. However, in the present embodiment, since the irradiation energy is efficiently used in the manner described later, sufficiently suitable processing is possible in the range of 0.1 μJ to 10 μJ.

<Overview of laser processing equipment>
Next, the outline of the laser processing apparatus capable of realizing the above-described cleavage / dehiscence processing will be described.

  FIG. 2 is a schematic diagram schematically showing a configuration of a laser processing apparatus 50 capable of realizing such processing. The laser processing apparatus 50 includes a stage 7 on which a workpiece (hereinafter also referred to as a substrate) 10 is placed, and a controller that performs various operations (observation operation, alignment operation, processing operation, etc.) of the laser processing apparatus 50. 1 is configured such that the workpiece 10 can be processed by irradiating the workpiece 10 placed on the stage 7 with the laser beam LB.

  The stage 7 can be moved in the horizontal direction by a moving mechanism 7m. The moving mechanism 7m moves the stage 7 in a predetermined XY 2-axis direction within a horizontal plane by the action of a driving unit (not shown). Thereby, the movement of the laser beam irradiation position and the like are realized. As for the moving mechanism 7m, a rotation (θ rotation) operation in a horizontal plane around a predetermined rotation axis can be performed independently of horizontal driving.

  Further, in the laser processing apparatus 50, surface observation for directly observing the workpiece 10 from the side irradiated with the laser light (referred to as a surface) through an imaging unit (not shown) or the stage 10 is placed on the stage 7. Backside observation in which observation is performed through the stage 7 from the other side (referred to as the backside) can be performed.

  As described above, the stage 7 is formed of a transparent member such as quartz, and a suction pipe (not shown) serving as an intake passage for adsorbing and fixing the workpiece 10 is provided therein. . The suction pipe is provided, for example, by drilling a predetermined position of the stage 7 by machining.

  With the workpiece 10 placed on the stage 7, suction is performed on the suction pipe by the suction means 11 such as a suction pump, for example, and the suction provided at the stage 7 mounting surface side tip of the suction pipe. The workpiece 10 (and the transparent substrate protection sheet 4) is fixed to the stage 7 by applying a negative pressure to the holes. In addition, in FIG. 2, although the case where the to-be-processed object 10 is affixed on the transparent substrate protection sheet 4 is illustrated, sticking of the transparent substrate protection sheet 4 is not essential.

  More specifically, in the laser processing apparatus 50, the laser light LB is emitted from the laser light source SL, reflected by a dichroic mirror 51 provided in a lens barrel (not shown), and then the laser light LB is supplied to the stage 7 Then, the light is condensed by the condenser lens 52 so as to be focused on the part to be processed of the workpiece 10 placed on the workpiece 10 and irradiated onto the workpiece 10. By combining the irradiation of the laser beam LB and the movement of the stage 7, the workpiece 10 can be processed while the laser beam LB is scanned relative to the workpiece 10. For example, in order to divide the workpiece 10, processing such as grooving (scribing) can be performed on the surface of the workpiece 10.

As the laser light source SL, an Nd: YAG laser is preferably used. Alternatively, an embodiment using an Nd: YVO ₄ laser or other solid-state laser may be used. Details of the configuration of the laser light source SL will be described later.

  Further, the emission control of the laser beam LB emitted from the laser light source SL, adjustment of the pulse repetition frequency and pulse width, and the like are realized by the irradiation control unit 23 of the controller 1. When a predetermined setting signal according to the processing mode setting data D2 is issued from the processing unit 25 to the irradiation control unit 23, the irradiation control unit 23 sets the irradiation condition of the laser beam LB according to the setting signal.

In the present embodiment, the laser light source SL having a wavelength of 500 nm to 1600 nm is used. Further, in order to realize the processing with the processing pattern described above, the pulse width of the laser beam LB needs to be about 1 psec to 50 psec. The repetition frequency R is preferably about 10 kHz to 200 kHz, and the laser beam irradiation energy (pulse energy) is preferably about 0.1 μJ to 50 μJ. In such a case, the peak power density in the irradiation with the laser beam LB is approximately 0.1 TW / cm ² to several tens TW / cm ² .

  The controller 1 controls the operation of each of the above-described units, and implements a control unit 2 that realizes processing of the workpiece 10 in various modes to be described later, a program 3p that controls the operation of the laser processing apparatus 50, and processing processing. It further includes a storage unit 3 for storing various data referred to at the time.

  The control unit 2 is realized by a general-purpose computer such as a personal computer or a microcomputer, for example, and various components can be obtained by reading and executing the program 3p stored in the storage unit 3 into the computer. Is realized as a functional component of the control unit 2.

  Specifically, the control unit 2 includes a drive control unit 21 that controls operations of various drive parts related to processing such as driving of the stage 7 by the moving mechanism 7m and focusing operation of the condenser lens 52. An imaging control unit 22 that controls the imaging of the workpiece 10 by the imaging unit that does not, an irradiation control unit 23 that controls the irradiation of the laser beam LB from the laser light source SL, and the workpiece 10 on the stage 7 by the suction unit 11. Are mainly provided with a suction control unit 24 for controlling the suction fixing operation and a processing unit 25 for executing processing on the processing target position in accordance with the given processing position data D1 and processing mode setting data D2.

  The storage unit 3 is realized by a storage medium such as a ROM, a RAM, and a hard disk. The storage unit 3 may be implemented by a computer component that implements the control unit 2, or may be provided separately from the computer, such as a hard disk.

  Various input instructions given by the operator to the laser processing apparatus 50 are preferably performed using a GUI realized in the controller 1. For example, a processing menu is provided on the GUI by the operation of the processing unit 25.

  Further, the laser processing apparatus 50 having the above configuration differs in the combination of the irradiation condition of the laser beam LB from the laser light source SL and the scanning condition of the laser beam LB on the workpiece 10 by moving the stage 7. Thus, it is preferable that various processing modes can be selectively performed. For example, it is preferable to select a mode in which processing is performed under processing conditions suitable for the above-described cleavage / dehissing processing, as well as processing under processing conditions in which a region to be processed is continuously formed. It is.

  It is preferable that the processing mode can be selected according to a processing menu provided to the operator in the controller 1 by the operation of the processing unit 25, for example. The storage unit 3 of the controller 1 stores processing position data D1 describing a processing position to be processed in the workpiece, and for each parameter of the laser beam according to the mode of laser processing in each processing mode. Processing mode setting data D2 in which conditions and driving conditions of stage 7 (or their settable ranges) are described is stored. The processing unit 25 acquires the processing position data D1 and the conditions corresponding to the selected processing mode from the processing mode setting data D2, and the drive control unit 21 or the like so that the operation according to the conditions is executed. The operation of each corresponding unit is controlled through the irradiation control unit 23 and others.

<Laser light source>
FIG. 3 is a diagram illustrating a configuration of the laser light source SL provided in the laser processing apparatus 50. The laser light source SL used in the present embodiment is capable of high-intensity and ultrashort pulse laser output by the chirp pulse amplification (CPA) method. The laser light source SL includes an oscillator 101, a pulse stretcher 102, a first polarizer 103, a Faraday rotator 104, an output amplifier 105, and a pulse compressor 106.

  As shown in FIG. 3, the oscillator 101, the pulse stretcher 102, the first polarizer 103, the Faraday rotator 104, and the output amplifier 105 are arranged in this order on the first optical path OP1. The pulse compressor 106 is arranged on the second optical path OP2 branched from the first polarizer 103.

  The oscillator 101 is a laser oscillator that outputs pulsed light (first pulsed light) PL1 having a pulse width of femtosecond (fsec) order or picosecond (psec) order with an output period that is a reciprocal of a preset repetition frequency R. It is. The oscillator 101 is preferably an example of a mode-locked laser oscillator. In the following description, the pulse oscillation period inside the oscillator 101 is assumed to be X (nsec).

The pulse expander 102 expands the pulse width of the first pulse light PL1 output from the oscillator 101 by using the spectrum width of the first pulse light PL1, and is 10 ³ times to 10 ⁵ times the first pulse light PL1. The light is emitted as the second pulsed light PL2 having a pulse width of about twice. The pulse stretcher 102 is made of, for example, a bulk type diffraction grating pair (grating pair) or a fiber.

  The first polarizer 103 is provided between the pulse stretcher 102 and the Faraday rotator 104, and directs the second pulse light PL2 emitted from the pulse stretcher 102 toward the Faraday rotator 104 on the first optical path OP1. While transmitting, the third pulse light PL3 emitted from the output amplifier 105 and passing through the Faraday rotator 104 is reflected to the second optical path OP2.

  The Faraday rotator 104 polarizes the second pulsed light PL2 incident from the pulse stretcher 102 via the first polarizer 103 and emits it as the third pulsed light PL3 toward the output amplifier 105, while entering from the output amplifier 105. The fourth pulsed light PL4 is polarized and emitted toward the first polarizer 103 as the fifth pulsed light PL5.

  The output amplifier 105 internally amplifies the third pulse light PL3 incident from the Faraday rotator 104, and emits the fourth pulse light PL4, which is the amplified light obtained thereby, at a predetermined timing. More specifically, the output amplifier 105 includes a first amplification mirror 105a, a second amplification mirror 105b, an amplification medium 105c, a Pockels cell (Pockels crystal) 105d, a second polarizer 105e, and an intermediate mirror 105f. Prepare.

  In the output amplifier 105, the second polarizer 105e, the Pockels cell 105d, and the first amplification mirror 105a are arranged in this order after the Faraday rotator 104 on the first optical path OP1. Further, the intermediate mirror 105f, the amplification medium 105c, and the second amplification mirror 105b are arranged in this order on the third optical path OP3 branched from the second polarizer 105e. Hereinafter, for the sake of convenience, the third optical path OP3 from the first amplification mirror 105a to the second amplification mirror 105b is also included for the sake of convenience, including the portion from the first amplification mirror 105a overlapping the first optical path OP1 to the second polarizer 105e. It shall be called.

In the output amplifier 105 having such a configuration, the third pulsed light PL3 that is incident from the Faraday rotator 104 and passes through the second polarizer 105e and the Pockels cell 105d is roughly expressed by the first amplification mirror 105a as indicated by an arrow AR1. Each time the light is repeatedly reflected between the second amplification mirror 105b and the second amplification mirror 105b, it is amplified in the amplification medium 105c. For example, the third pulsed light PL3 is amplified so that the pulse energy becomes about 10 ¹⁰ times. In the following description, the reciprocating cycle of the pulsed light (the time required for reciprocating the third optical path OP3) between the first amplifying mirror 105a and the second amplifying mirror 105b is Y (nsec).

The amplification medium 105c may be selected according to the wavelength of the laser beam LB to be emitted from the laser light source SL. For example, when the laser light source SL is configured as an Nd: YAG laser as described above, Y ₃ Al ₅ O ₁₂ crystal (Nd: YAG crystal) added with neodymium (Nd) is used as the amplification medium 105c. There may be. Note that amplification in the amplification medium 105c is performed by excitation with an excitation source (not shown) under the control of the irradiation control unit 23.

The Pockels cell 105d and the second polarizer 105e are provided to control the emission of the fourth pulsed light PL4 from the output amplifier 105. The Pockels cell 105d is made of a KDP (KH ₂ PO ₄ ) crystal or the like and has a function of changing the polarization state of light according to an applied voltage. The voltage application to the Pockels cell 105d is controlled by the irradiation control unit 23. The second polarizer 105e transmits the third pulsed light PL3 emitted from the Faraday rotator 104 toward the first amplification mirror 105a, while being reflected by the first amplification mirror 105a and passed through the Pockels cell 105d. The three-pulse light PL3 is transmitted or reflected according to the polarization state given by the Pockels cell 105d.

  More specifically, the Pockels cell 105d is emitted from the Faraday rotator 104, reflected by the third pulse light PL3 transmitted through the second polarizer 105e, and the second amplification mirror 105b, and travels on the third optical path OP3. At the timing when the generated third pulsed light PL3 is incident, the applied voltage is controlled so that these lights are transmitted as they are. On the other hand, at the timing when the third pulse light PL3 reflected by the first amplification mirror 105a or the fourth pulse light PL4 that is the amplified light is incident, if the light is not emitted to the outside but is further amplified, Pockels The voltage applied to the cell 105d is controlled so that the third pulsed light PL3 or the fourth pulsed light PL4 is in a polarization state reflected by the second polarizer 105e. On the other hand, if the fourth pulse light PL4 is emitted to the outside of the output amplifier 105, the voltage applied to the Pockels cell 105d is in a polarization state in which the fourth pulse light PL4 can pass through the second polarizer 105e. To be controlled. Hereinafter, emitting the fourth pulsed light PL4 to the outside of the output amplifier 105 is also referred to as “extracting” the pulsed light. Also, the voltage applied to the Pockels cell 105 d when the fourth pulse light PL 4 is extracted from the output amplifier 105 is referred to as “extraction voltage”, and the Pockels cell when the fourth pulse light PL 4 is amplified in the output amplifier 105. The voltage applied to 105d is referred to as “amplified voltage”.

  The intermediate mirror 105f is provided to change the direction of the third optical path OP3 due to restrictions on the size of the output amplifier 105 and the like. It is not an essential aspect that the output amplifier 105 includes the intermediate mirror 105f. Alternatively, FIG. 3 illustrates an example in which one intermediate mirror 105f is provided, but in the output amplifier 105, more intermediate mirrors 105f are provided on the first optical path OP1 or the third optical path OP3. There may be.

  The fourth pulse light PL4 emitted (extracted) from the output amplifier 105 is again polarized by the Faraday rotator 104 as described above to become the fifth pulse light PL5. The fifth pulsed light PL5 is reflected by the first polarizer 103, travels on the second optical path OP2, and enters the pulse compressor 106.

The pulse compressor 106 compresses the pulse width of the incident fifth pulse light PL5 and has an ultrashort sixth pulse light having a pulse width of about 1/10 to 1/10 ⁵ times the fifth pulse light PL5. It emits as PL6. Similar to the pulse stretcher 102, the pulse compressor 106 includes, for example, a bulk-type diffraction grating pair (grating pair).

  In the laser light source SL having the above-described configuration, the ultrashort pulse light with a small pulse energy output from the oscillator 101 is roughly expanded in pulse width by the pulse expander 102 and amplified by the output amplifier 105. In addition, after the pulse width is compressed by the pulse compressor 106, the laser beam LB is emitted to the outside. For example, pulse light emitted from the oscillator 101 with a pulse width of 100 fsec and a pulse energy of 100 nJ is expanded by the pulse expander 102 to 1 nsec, and then amplified by the output amplifier 105 to 10 J. Furthermore, if the pulse width is again compressed to 100 fsec by the pulse compressor 106, it becomes possible to emit a laser beam LB with an ultrashort pulse and a very large pulse energy.

  In the case where the laser beam LB is output as a harmonic such as the third harmonic of the Nd: YAG laser, the light emitted from the pulse compressor 106 is incident on a nonlinear optical crystal (not shown), and its wavelength Should be converted.

<Improvement of pulse energy utilization efficiency by double pulse>
As described above, the cleavage / cleaving process realized by the above-described principle is performed between the irradiated regions by causing alteration, melting, evaporation removal, etc. of the substance in the central portion of each irradiated region. / This is a technique for developing dehiscence. Therefore, it is not necessary to perform more than necessary processing in the irradiated region. Rather, it is required to make sure that the cleavage / dehiscence progresses from the irradiated region in the easy cleavage / dehiscence direction.

  For example, when unit pulse light with a large pulse energy and a small pulse width is irradiated, the energy applied to the irradiated region becomes excessive and damages the irradiated region more than necessary, but cleavage / dehiscence progresses favorably. It can happen that not. This is because the energy of the irradiated unit pulse light cannot be sufficiently directed to the progress of cleavage / dehiscence. More specifically, it is considered that the transition from the energy absorption of the electronic system to the vibration of the molecular system by the energy requires about 10 psec. Therefore, if the energy of the irradiated unit pulsed light can be appropriately distributed to the formation of weak parts due to alteration / melting / evaporation removal, etc. of the substance and the subsequent cleavage / dehiscence progress, the use of laser light energy Efficiency will increase.

  In the present embodiment, such an improvement in energy utilization efficiency is devised in the manner of extracting pulsed light from the laser light source SL, and the original emission period of individual unit pulsed light that is the reciprocal of the repetition frequency R has arrived. Each time, two ultrashort pulses are continuously emitted in an extremely short time compared to the emission period, and these two ultrashort pulses are irradiated onto substantially the same irradiated region. Thereby, cleavage / dehiscence can be more reliably caused to the workpiece.

  Hereinafter, the emission mode of the laser beam LB in this mode is referred to as double pulse. FIG. 4 is a diagram schematically showing a profile (time change) of the unit pulse light PL of the laser light LB that has been double-pulsed. In the case shown in FIG. 4, the unit pulsed light PL is emitted at a certain time t1 and the subsequent pulsed light emitted at a certain time t2 after the laser light LB is double-pulsed. And PLβ.

  For example, since the repetition frequency R of the laser beam LB in the cleavage / cleavage processing is 1 kHz to 200 kHz as described above, the original emission period T of the unit pulse light PL, which is the reciprocal number thereof, is about 5 μsec to 1 msec. The emission interval (t2-t1) between the pre-pulse light PLα and the post-pulse light PLβ is on the order of several nanoseconds to several tens of nanoseconds, which is much shorter than this.

  More specifically, the pre-pulse light PLα causes alteration, melting, evaporation and the like of the substance to form a weak intensity portion, and the post-pulse light PLβ is irradiated at a timing when the weak intensity portion is in an instantaneous high temperature state. By doing so, cleavage / dehiscence progresses from the weak strength portion. In such a case, cleavage with high energy utilization efficiency is realized by irradiating the pre-pulse light PLα and the post-pulse light PLβ with the pulse energy necessary and sufficient for each role. Furthermore, more efficient cleavage / cleavage processing is possible when the pulse energy of the post-pulse light PLβ is larger than the pulse energy of the pre-pulse light PLα.

  For confirmation, in the present embodiment, as described above, processing is performed while relatively scanning the laser beam at a scanning speed of about 50 mm / sec to 3000 mm / sec. The first pulsed light PLα and the postpulsed light PLβ cannot be irradiated to the same irradiated region. This is because, since the laser light source SL and the workpiece are relatively moved after the irradiation with the pre-pulse light PLα and before the irradiation with the post-pulse light PLβ, the pre-pulse light PLα and the post-pulse light are emitted. This is because the position of the irradiated region formed by PLβ should be different. However, for example, assuming that the scanning speed of the laser beam LB is 3000 mm / sec (= 3 m / sec) and the emission interval between the pre-pulse light PLα and the post-pulse light PLβ is 100 nsec, which is slightly larger, The calculated shift in both positions is only 300 nm. On the other hand, the beam diameter of the laser beam is about 1 μm to 10 μm, and the distance between irradiated regions formed on the workpiece 10 during the cleavage / cleavage processing is 4 μm to 50 μm. The value of 300 nm is about 1/100 to 1/1000 of the latter, and can be regarded as being sufficiently within the error range. Therefore, there is no problem even when the pre-pulse light PLα and the post-pulse light PLβ are irradiated to substantially the same irradiated region in processing. This means that the emission of the pre-pulse light PLα and the post-pulse light PLβ can be practically handled as the emission of one unit pulse light PL.

<First Mode of Double Pulse>
Next, a specific mode of double pulsing will be described in detail. In this embodiment, two modes will be described.

  FIG. 5 is a diagram for explaining amplification of pulsed light in the output amplifier 105 and extraction of pulsed light from the output amplifier 105. FIG. 5A shows a normal case where double pulsing is not performed. FIG. 5B schematically shows the state of the first mode of double pulsing. More specifically, the profiles shown in FIGS. 5A and 5B are obtained from the third pulse light PL3 incident on the Pockels cell 105d from the first amplification mirror 105a side or the fourth pulse light PL4 that is the amplified light. The time change of pulse energy is shown.

  As described above, the third pulsed light PL3 obtained by extending and polarizing the first pulsed light PL1 output from the oscillator 101 is incident on the output amplifier 105 every period determined by the repetition frequency R. Each incident third pulse light PL3 is amplified inside the output amplifier 105. Specifically, by applying an amplification voltage to the Pockels cell 105d, the third pulsed light PL3 is repeatedly reciprocated in the reciprocating period Y (nsec) through the third optical path OP3 and is amplified every time it passes through the amplification medium 105c. The Therefore, as shown in FIG. 5A, the third pulsed light PL3 or the fourth pulsed light PL4 that is the amplified light is incident on the Pockels cell 105d every Y (nsec).

  When double pulsing is not performed, the third pulse light PL3 is amplified until it becomes the fourth pulse light PL4 having a predetermined pulse energy (in the case of FIG. 5A, when the fourth-order amplification is received). As shown in FIG. 5A, during the time Δt when the fourth pulsed light PL4 passes through the Pockels cell 105d, the voltage applied to the Pockels cell 105d is switched from the amplified voltage to the extraction voltage. As a result, the fourth pulsed light PL4 passes through the second polarizer 105e and is emitted from the output amplifier 105.

  The fourth pulse light PL4 is emitted in this manner every time the third pulse light PL3 enters the output amplifier 105 at a timing determined by the repetition frequency R. Since the emitted fourth pulse light PL4 is compressed as described above and finally emitted as the sixth pulse light PL6, as a result, the laser light source SL has an irradiation condition of repetition frequency R. The laser beam LB is emitted.

  On the other hand, in the case of double pulse conversion, as shown in FIG. 5B, the third pulse light PL3 is amplified in the same manner as described above, but the time during which the extraction voltage is applied to the Pockels cell 105d. Δt1 is determined to be shorter than the time Δt when double pulsing is not performed. Then, the fourth pulse light PL4 is extracted from the output amplifier 105 for the part of the total energy component that has passed through the Pockels cell 105d within this time Δt1. After passing through the pulse compressor 106, it is emitted from the laser light source SL as the first pulsed light PLα.

  On the other hand, among the total energy components of the fourth pulse light PL4, the remainder that has not passed through the Pockels cell 105d within the time Δt1 cannot be transmitted through the second polarizer 105e, and thus is amplified again by reciprocating through the third optical path OP3. Will receive. After the amplification, the applied voltage to the Pockels cell 105d is switched from the amplified voltage to the extraction voltage for a time Δt2 during which the remaining component including the amplified component passes through the Pockels cell 105d again. Thereby, the remaining components are also emitted from the output amplifier 105. After passing through the pulse compressor 106, it is emitted from the laser light source SL as post-pulse light PLβ.

  In other words, the first aspect of double pulsing is roughly as follows. One pulse light oscillated from the oscillator 101 is expanded and then separated in the output amplifier 105 to cause a time delay. It can be said that it is what realizes pulsing. In the first aspect, the emission interval (t2-t1) between the first pulse light PLα and the second pulse light PLβ is approximately Y (nsec).

  In FIG. 5B, for the sake of illustration, the extraction voltage is set for the Pockels cell 105d only for the time Δt1, and after a part of the fourth pulsed light PL4 is extracted, in the next cycle. The extraction voltage is set for the time Δt2, but this is not an essential aspect. That is, the amplification after the previous extraction may be repeated many more times.

  In the case of the first aspect, by appropriately determining the value of Δt1 and the timing of subsequent extraction, the energy distribution between the pre-pulse light PLα and the post-pulse light PLβ can be adjusted, and such adjustment is preferable. It is possible to perform cleaving / cleavage processing with high energy utilization efficiency.

<Second form of double pulsing>
In the first aspect of the double pulse conversion described above, the single pulse light oscillated from the oscillator 101 is separated in the output amplifier 105 and a double delay is realized by causing a time delay. The second mode described below realizes double pulsing based on a principle different from this.

  FIG. 6 is a diagram for explaining a second mode of double pulsing. FIG. 6A schematically shows the state of pulse oscillation in the oscillator 101. FIG. 6B schematically shows the state of the amplification of the pulsed light in the output amplifier 105 and the extraction of the pulsed light from the output amplifier 105. More specifically, the profile shown in FIG. 6B is the time of the pulse energy of the third pulse light PL3 incident on the Pockels cell 105d from the first amplification mirror 105a side or the fourth pulse light PL4 that is the amplified light. It shows a change.

  As described above, a pulse is oscillated within the oscillator 101 at the oscillation period X (nsec). In the case of not performing double pulsing or in the case of the first mode described above, the first pulsed light PL1 is output from the oscillator 101 for each emission period T that is the reciprocal of the repetition frequency R. On the other hand, in the second mode, as shown in FIG. 6A, the oscillator 101 continuously oscillates at the time interval X (nsec) for each emission period T that is the reciprocal of the repetition frequency R. One pulse light is regarded as the first pulse light PL1 and is taken out. Hereinafter, of the two pulse lights, the one oscillated first in time is referred to as a first oscillation light PL1α, and the one oscillated later is also referred to as a rear oscillation light PL1β.

  In this manner, the first pulsed light PL1 output from the oscillator 101 travels on the first optical path OP1, undergoes expansion, etc., and eventually enters the output amplifier 105 as the third pulsed light PL3. More specifically, as shown in FIGS. 6B and 6C, the first incident light PL3α derived from the first oscillation light PL1α and the second incidence light PL3β derived from the second oscillation light PL1β are expressed by X ( nsec) at an interval of nsec). In FIGS. 6B and 6C, for convenience of illustration, the profiles of the base line and the first incident light PL3α are indicated by solid lines, and the profile of the rear incident light PL3β is indicated by a broken line.

  The first incident light PL3α and the rear incident light PL3β are respectively amplified by the output amplifier 105. The behavior after the amplification starts is that the delay time X (nsec) of the rear incident light PL3β with respect to the first incident light PL3α and the output amplifier. It differs depending on the magnitude relationship with the reciprocation period Y (nsec) of the pulsed light in 105.

  In the case of X> Y, as shown in FIG. 6B, after the time XY (nsec) elapses after the first incident light PL3α is subjected to k + 1 order amplification, the rear incident light PL3β is kth order. The relationship of receiving amplification is repeated.

  On the other hand, when X <Y, as shown in FIG. 6C, after the time X (nsec) has elapsed after the first incident light PL3α is subjected to the kth order amplification, the last incident light PL3β is the kth order. Is repeated (k is a natural number).

  In the second aspect, this relationship is used to realize double pulsing of the laser light LB emitted from the laser light source SL. Specifically, after satisfying X> Y, the Pockels are only during the time Δt3 during which the k + 1st order amplified light of the first incident light PL3α and the kth order amplified light of the rear incident light PL3β pass through the Pockels cell 105d. The voltage applied to the cell 105d is switched from the amplified voltage to the extracted voltage. Then, two pulse lights are continuously emitted at an emission interval XY (nsec). If the former of these two pulse lights is the first extracted light PL4α and the latter is the rear extracted light PL4β, the first extracted light PL4α is emitted from the laser light source SL as the first pulsed light PLα after passing through the pulse compressor 106. The post-extraction light PL4β passes through the pulse compressor 106 and is then emitted from the laser light source SL as post-pulse light PLβ. That is, as a result, a double pulse is realized at an emission interval of t2−t1 = XY. On the other hand, X <Y is satisfied, and as shown in FIG. 6C, the time Δt4 during which the k-th order amplified light of the first incident light PL3α and the k-th order amplified light of the rear incident light PL3β pass through the Pockels cell 105d. When the voltage applied to the Pockels cell 105d is switched from the amplified voltage to the extraction voltage only for the interval, the first extraction light PL4α and the rear extraction light PL4β having the same pulse energy are emitted at the emission interval X (nsec). Double pulsing is realized.

  As described above, generally speaking, the second mode of double pulsing is to adjust the extraction timing from the output amplifier 105 after amplifying the two pulse lights oscillated from the oscillator 101 respectively. In other words, it can be said that the double pulse is realized.

  In addition, when the pulse energy ratio between the first oscillation light PL1α oscillated from the oscillator 101 and the second oscillation light PL1β is different, the first extraction light PL4α and the rear extraction light regardless of whether X> Y or X <Y. It becomes possible to arbitrarily change the pulse energy ratio of the light PL4β. Thereby, the ratio of the pulse energy of the first extracted light PL4α and the pulse energy of the second extracted light PL4β can be adjusted to a desired value. For example, when the pulse energy of the post-oscillation light PL1β is larger than the pulse energy of the pre-oscillation light PL1α, the pulse energy of the post-extraction light PL4β is higher than the pulse energy of the pre-extraction light PL4α even when X <Y. It can be enlarged.

  Accordingly, by appropriately adjusting the pulse energy ratio between the pre-oscillation light PL1α and the post-oscillation light PL1β, similarly to the first aspect, in the second aspect, cleavage / cleavage processing with high energy utilization efficiency can be performed. It becomes possible.

  As described above, according to the present embodiment, energy is utilized by performing cleavage / cleavage processing along a planned processing line on a workpiece with a double-pulsed laser beam. Highly efficient cleavage / cleavage processing is possible. That is, the division starting point can be more reliably formed by cleavage / cleavage.

  As an example, cleaving / dehiscence processing was performed on the surface of a C-plane sapphire substrate using the laser beam LB double-pulsed in the first mode. Further, as a comparative example, cleaving / cleavage processing was performed under the same conditions as in the example except that double pulsing was not performed.

  Specifically, the repetition frequency R was set to 13.3 kHz and the scanning speed V was set to 200 mm / sec so that the irradiation pitch Δ of each unit pulse light of the laser light LB was 15 μm. Further, the sum of the pulse energies of the pre-pulse light PLα and the post-pulse light PLβ in the embodiment is set to 5 μJ, and in the comparative example, the pulse energy of one pulse light is made the same. Furthermore, Δt1 and Δt2 are set so that the power ratio of the first pulse light PLα and the second pulse light PLβ in the embodiment is 1: 2.

  FIG. 7: is an optical microscope image by epi-illumination and transmitted illumination of the C surface sapphire substrate surface after a process about an Example and a comparative example. It seems that there is no difference between the example and the comparative example in the image by the epi-illumination. From the image by the transmission illumination, only in the case of the example, the cleavage / dehiscence is linear in the drawing view shown in FIG. The same processing marks as those on the surfaces C11a, C11b, C12a, C12b, C13a, C13b, C14a, C14b,... Are clearly observed. Such a result indicates that double pulsing has an effect of more reliably causing cleavage / dehiscence in cleaving / dehiscence processing.

DESCRIPTION OF SYMBOLS 1 Controller 7 Stage 7m Movement mechanism 10 Workpiece 50 Laser processing apparatus 51 Dichroic mirror 52 Condensing lens 101 Oscillator 102 Pulse stretcher 103 1st polarizer 104 Faraday rotator 105 Output amplifier 105a 1st amplification mirror 105b 2nd amplification mirror 105c Amplifying medium 105d Pockels cell 105e Second polarizer 105f Intermediate mirror 106 Pulse compressor C11a, C11b, C12a, C12b, C13a, C13b, C14a, C14b Cleaved / cleaved surface L Processing line LB Laser light PLα Prior pulse light PLβ Post-pulse light PL Unit pulse light PL1α Post-oscillation light PL1β Pre-oscillation light PL3α Post-incident light PL3β Pre-incident light PL4α Pre-extraction light PL4β Post-extraction light RE11, RE12, RE13, R 14 the irradiated region

Claims (4)

A processing method for forming a division starting point on a workpiece,
An emission step of amplifying the oscillating pulsed light oscillated by an oscillator provided in the light source in an amplifier provided in the light source, and emitting a pulsed laser light as amplified light from the light source;
By irradiating the workpiece with the pulsed laser light such that the irradiated region for each unit pulsed light of the pulsed laser light is discretely formed on the workpiece surface of the workpiece, An irradiation step of forming a starting point for division in the workpiece by causing cleavage or cleavage of the workpiece between the irradiation regions;
With
In the emission step, by adjusting the extraction timing of the pulse laser light from the amplifier, the pulse laser light is delayed at a time interval shorter than the emission period of the unit pulse light while the irradiation is performed. Emitting from the light source so that the workpiece is irradiated as two pulsed light in which the irradiated region is substantially the same in the process;
A processing method of a workpiece characterized by the above. A processing method for a workpiece according to claim 1,
In the emission step, the amplified light is extracted from the amplifier by a first time when only a part of the amplified light is extracted from the amplifier, and after the first time, the amplified light is extracted. The remaining is divided into a second time after being amplified in the amplifier, thereby emitting the two pulsed lights from the light source.
A processing method of a workpiece characterized by the above. A processing method for a workpiece according to claim 1,
In the emitting step, when the two oscillation pulse lights are oscillated from the oscillator and k is a natural number, the oscillation light is oscillated later with k-order or k + 1-order amplification light of the oscillation pulse light oscillated first. The k-th order amplified light of the pulsed light is emitted from the light source as the two pulsed lights in this order,
A processing method of a workpiece characterized by the above. A processing method for a workpiece according to any one of claims 1 to 3,
Adjusting the ratio of the pulse energy of the two pulse lights by adjusting the extraction timing of the pulse laser light from the amplifier;
A processing method of a workpiece characterized by the above.