Laser Processing Equipment

Abstract

To provide a laser processing device that can suppress influence of dullness of a phase modulation quantity in a modulation pattern and uneven distribution of turn-around points of the phase modulation quantity in the modulation pattern.SOLUTION: A laser processing device 1 comprises a support part 2, a light source 3, a spatial light modulator 5 and a light condensing part 6. A control part 10 inputs an image signal to the spatial light modulator 5 so as to make the spatial light modulator 5 display a modulation pattern corresponding to the image signal. The control part 10 executes processing for irradiating a wafer 20 with laser light L while relatively moving a light condensing point C of the laser light L along an X-direction. In the processing, the control part 10 inputs, to the spatial light modulator 5, the image signal including a first signal whose gradation values vary at a constant inclination from the minimum value to the maximum value, from a region of an image signal corresponding to one end of the modulation pattern in a first direction toward a region of an image signal corresponding to the other end of the modulation pattern in the first direction.SELECTED DRAWING: Figure 13

Description

[Technical field]

[0001] The present invention relates to a laser processing apparatus. [Background technology]

[0002] Patent Document 1 describes a laser processing device. This laser processing device includes a spatial light modulator for modulating laser light emitted from a light source. The spatial light modulator modulates the laser light by displaying a modulation pattern on a liquid crystal layer based on a voltage applied to the liquid crystal layer. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2015-223620 A Summary of the Invention [Problem to be solved by the invention]

[0004] In Patent Document 1, the spatial light modulator is configured by stacking a drive circuit layer, a pixel electrode layer, a reflective film, an alignment film, a liquid crystal layer, an alignment film, a transparent conductive film, and a transparent substrate in this order on a semiconductor substrate. In such a spatial light modulator, when a signal indicating a modulation pattern is input to the drive circuit layer, a voltage corresponding to the signal is applied to each pixel electrode, and an electric field is formed between each pixel electrode and the transparent conductive film. When the electric field is formed, the alignment direction of the liquid crystal molecules in the liquid crystal layer changes for each region (hereinafter sometimes referred to as "pixel") corresponding to each pixel electrode, and the refractive index changes for each region corresponding to each pixel electrode. This state is a state in which a modulation pattern is displayed on the liquid crystal layer.

[0005] With a modulation pattern displayed on the liquid crystal layer, when laser light is incident from the outside onto the liquid crystal layer through the transparent substrate and the transparent conductive film, reflected by the reflective film, and emitted from the liquid crystal layer to the outside via the transparent conductive film and the transparent substrate, the laser light is modulated according to the modulation pattern displayed on the liquid crystal layer. In this way, the spatial light modulator allows modulation of the laser light by appropriately setting the modulation pattern to be displayed on the liquid crystal layer.

[0006] An image signal can be used as a signal for displaying the modulation pattern on the spatial light modulator. In this case, a voltage corresponding to the gradation value of the image signal input to the spatial light modulator is applied to the liquid crystal layer, and a refractive index change corresponding to the voltage value occurs, thereby displaying the modulation pattern. Then, the laser light incident on the spatial light modulator is phase-modulated according to this modulation pattern. Therefore, in this case, the image signal and the modulation pattern, and the gradation value of the image signal and the phase modulation amount imparted to the laser light by the modulation pattern correspond to each other. In the following, the phase modulation amount imparted to the laser light by the modulation pattern may be abbreviated to simply be referred to as the phase modulation amount of the modulation pattern.

[0007] However, the actual phase modulation amount of the modulation pattern may deviate from the ideal state. The present inventor has found that this deviating phase modulation amount may cause the following problems.

[0008] That is, when a spatial light modulator having a phase modulation capability of an upper limit of 2π is to display a modulation pattern including a region with a phase modulation amount larger than 2π, the modulation pattern is reproduced by folding back the region exceeding 2π. In this case, due to the dulling of the phase modulation amount, the modulation state of the laser light may differ between a region in the modulation pattern where folding back occurs (inverted region) and a region in the modulation pattern where folding back does not occur (normal region). Therefore, if there is a bias in the ratio of the inverted region and the normal region in the modulation pattern, the change in the ratio becomes large when the modulation pattern changes, and as a result, the change in the modulation state of the laser light also becomes large. Note that, as an example, the modulation state of the laser light is the diffraction efficiency of the laser light when the modulation pattern is a diffraction grating pattern.

[0009] On the other hand, the present inventor has found that the following problem may occur due to the fact that the folded portion of the phase modulation amount is unevenly present in a part of the modulation pattern. That is, when the folded portion of the phase modulation amount is unevenly present in a part of the modulation pattern, the focusing state of the laser light may vary between the part of the laser light modulated through the part where the folded portion is unevenly present and the part of the laser light modulated without passing through the part where the folded portion is unevenly present. Therefore, it is required to suppress the influence of the dulling of the phase modulation amount of the modulation pattern and the uneven distribution of the folded portion of the phase modulation amount in the modulation pattern.

[0010] Therefore, an object of the present invention is to provide a laser processing apparatus capable of suppressing the influence of the dulling of the phase modulation amount of the modulation pattern and the uneven distribution of the folding back portion of the phase modulation amount in the modulation pattern. [Means for solving the problem]

[0011] The laser processing apparatus according to the present invention is a laser processing apparatus for forming a modified region in an object by irradiating the object with laser light, and includes a support section for supporting the object, a light source for emitting laser light, a spatial light modulator for modulating the laser light emitted from the light source in accordance with a modulation pattern and emitting the modulated laser light, a focusing section including a focusing lens for focusing the laser light emitted from the spatial light modulator toward the object, and a control section for inputting an image signal to the spatial light modulator to cause the spatial light modulator to display a modulation pattern corresponding to the image signal, and the control section controls movement of at least one of the support section and the focusing section to form a modified region in the object. a processing process for irradiating the object with the laser light while moving the focal point of the laser light relative to the object along the X direction along the light incident surface, and a gradation value corresponding to the phase modulation amount in the modulation pattern is set in each of the regions constituting the image signal, and in the processing process, the control unit inputs an image signal including a first signal whose gradation value changes at a constant gradient from a minimum value to a maximum value from a region corresponding to one end of the modulation pattern in the first direction to a region corresponding to the other end of the modulation pattern in the first direction to the spatial light modulator, thereby causing the spatial light modulator to display a modulation pattern including the first pattern corresponding to the first signal.

[0012] In this laser processing device, the control unit inputs an image signal to the spatial light modulator, thereby causing the spatial light modulator to display a modulation pattern, and performs processing while modulating the laser light according to the modulation pattern. In the image signal, a gradation value corresponding to the phase modulation amount at each position of the modulation pattern is set for each region constituting the image signal. That is, in the spatial light modulator, the phase modulation amount at each position of the modulation pattern is set to an amount corresponding to the gradation value of each region of the image signal, thereby displaying a desired pattern as a whole. The image signal includes a first signal in which the gradation value changes at a constant gradient from a region corresponding to one end of the modulation pattern to a region corresponding to the other end of the modulation pattern. As a result, the modulation pattern displayed on the spatial light modulator includes a first pattern in which the phase modulation amount changes in one direction at a constant gradient according to the first signal. According to the present inventor, when the modulation pattern includes such a first pattern, the influence of the dulling of the phase modulation amount of the modulation pattern and the uneven distribution of the folding back portion of the phase modulation amount in the modulation pattern are suppressed.

[0013] In the laser processing device according to the present invention, in the processing process, the control unit may input an image signal including a second signal corresponding to the second pattern to the spatial light modulator so as to display a modulation pattern including a second pattern for splitting the laser light into a plurality of processing lights on the spatial light modulator. In this case, the second pattern is superimposed on the modulation pattern. In such a case, the variation among the plurality of processing lights is suppressed.

[0014] In the laser processing device according to the present invention, the first direction may be a direction corresponding to a direction intersecting with the branching direction of the laser light, in which case variation between the processing lights can be more effectively suppressed.

[0015] In the laser processing device according to the present invention, in the processing, the control unit may input an image signal including a third signal corresponding to the third pattern to the spatial light modulator so as to display a modulation pattern including a third pattern for changing the focusing position of the laser light according to the radial position of the focusing lens on the spatial light modulator. In this case, it is possible to suppress uneven distribution of the folded portion of the phase modulation amount in the third pattern so as to correspond to a specific radial position of the focusing lens, and suppress variation in the focusing state of the laser light.

[0016] In the laser processing device according to the present invention, the control unit may change the gradation value in the first signal at a constant gradient from the minimum value to the maximum value from a region corresponding to one end of the modulation pattern in a second direction intersecting with the first direction to a region corresponding to the other end of the modulation pattern in the second direction. In this case, the gradation value of the image signal changes at a constant gradient in two directions intersecting with each other. Therefore, it is possible to suppress the influence of the slowing of the phase modulation amount and the uneven distribution of the folding back portion of the phase modulation amount within the plane of the spatial light modulator corresponding to the two directions.

[0017] In the laser processing device according to the present invention, the control unit may change the gradation value in the first signal so that the first signal has a plurality of periods in at least one direction. In this case, it is possible to more reliably suppress the influence of the dulling of the phase modulation amount and the uneven distribution of the folding back portion of the phase modulation amount. Effect of the Invention

[0018] According to the present invention, it is possible to provide a laser processing device capable of suppressing the influence of the dulling of the phase modulation amount of the modulation pattern and the uneven distribution of the folding back portion of the phase modulation amount in the modulation pattern. [Brief description of the drawings]

[0019] [Figure 1] FIG. 1 is a schematic diagram showing a laser processing device according to the present embodiment. [Diagram 2] FIG. 2 is a cross-sectional view of a portion of the spatial light modulator shown in FIG. [Diagram 3] FIG. 3 is a plan view of a wafer which is an object of one embodiment. [Figure 4] FIG. 4 is a cross-sectional view of a portion of the wafer shown in FIG. [Diagram 5] FIG. 5 is a cross-sectional view of a wafer for explaining an example of the operation of the laser processing device in the case of three-point branching. [Figure 6] FIG. 6 is a schematic diagram showing an example of a modulation pattern. [Figure 7] FIG. 7 is a schematic diagram showing an example of a modulation pattern. [Figure 8] FIG. 8 is a schematic diagram showing an example of a modulation pattern including a diffraction grating pattern. [Figure 9] FIG. 9 is a diagram showing an example of a diffraction grating pattern. [Figure 10] FIG. 10 is a diagram showing a state in which a distortion correction pattern is superimposed on a diffraction grating pattern. [Figure 11] FIG. 11 is a diagram showing an example of a diffraction grating pattern. [Figure 12] FIG. 12 is a diagram showing an example of an averaging pattern. [Figure 13] FIG. 13 is a diagram showing an example of a modulation pattern formed by superimposing an averaged pattern on a diffraction grating pattern. [Figure 14] FIG. 14 is a diagram showing an example of a modulation pattern formed by superimposing an averaging pattern on a distortion correction pattern. [Figure 15] FIG. 15 is a schematic diagram showing the relationship between the condenser lens and the laser light. [Figure 16] FIG. 16 is a diagram showing an example of a shaping pattern. [Figure 17] FIG. 17 is a cross-sectional photograph showing the processing result when the shaping pattern shown in FIG. 16 is used. [Figure 18] FIG. 18 is a diagram showing a modulation pattern formed by superimposing the averaging pattern shown in FIG. 12 on the shaping pattern shown in FIG. [Figure 19]FIG. 19 is a cross-sectional photograph showing the processing result when the shaping pattern shown in FIG. 16 and the modulation pattern shown in FIG. 18 are used. [Figure 20] FIG. 20 is a diagram showing a modified example of the averaging pattern. [Figure 21] FIG. 21 is a diagram for explaining a modified example of the shaping pattern. [Figure 22] FIG. 22 is a diagram for explaining a modified example. [Figure 23] FIG. 23 is a diagram for explaining a modified example. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0020] Hereinafter, an embodiment will be described in detail with reference to the drawings. In each drawing, the same or corresponding parts are denoted by the same reference numerals, and duplicated explanations will be omitted. [Laser processing equipment configuration]

[0021] FIG. 1 is a schematic diagram showing a laser processing apparatus according to the present embodiment. As shown in FIG. 1, the laser processing apparatus 1 includes a support unit 2, a light source 3, an optical axis adjustment unit 4, a spatial light modulator 5, a focusing unit 6, an optical axis monitor unit 7, a visible imaging unit 8, an infrared imaging unit 9, and a control unit 10. The laser processing apparatus 1 is an apparatus for forming a modified region 12 in an object 11 by irradiating the object 11 with laser light L. In the following description, the three directions perpendicular to each other are referred to as the X direction, the Y direction, and the Z direction, respectively. In this embodiment, the X direction is the first horizontal direction, the Y direction is the second horizontal direction perpendicular to the first horizontal direction, and the Z direction is the vertical direction.

[0022] The support unit 2 supports the object 11 so that the surface 11a of the object 11 is perpendicular to the Z direction, for example by adsorbing a film (not shown) attached to the object 11. The support unit 2 is movable along each of the X direction and the Y direction, and is rotatable about an axis parallel to the Z direction.

[0023] The light source 3 emits laser light L, for example, by a pulse oscillation method. The laser light L is transparent to the object 11.

[0024] The optical axis adjustment unit 4 adjusts the optical axis of the laser light L emitted from the light source 3. In this embodiment, the optical axis adjustment unit 4 adjusts the optical axis of the laser light L while changing the traveling direction of the laser light L emitted from the light source 3 to be along the Z direction. The optical axis adjustment unit 4 is composed of, for example, a plurality of reflecting mirrors whose positions and angles can be adjusted.

[0025] The spatial light modulator 5 is disposed in the housing H. The spatial light modulator 5 modulates the laser light L emitted from the light source 3. In this embodiment, the laser light L traveling downward from the optical axis adjustment unit 4 along the Z direction enters the housing H, the laser light L entering the housing H is reflected horizontally by the mirror M1 so as to form an angle with respect to the Y direction, and the laser light L reflected by the mirror M1 enters the spatial light modulator 5. The spatial light modulator 5 modulates the laser light L thus entered while reflecting it horizontally along the Y direction.

[0026] The focusing unit 6 is attached to the bottom wall of the housing H. The focusing unit 6 focuses the laser light L modulated by the spatial light modulator 5 on the object 11 supported by the support unit 2 from the surface 11a side along the Z direction. In this embodiment, the laser light L reflected horizontally along the Y direction by the spatial light modulator 5 is reflected downward along the Z direction by the dichroic mirror M2, and the laser light L reflected by the dichroic mirror M2 enters the focusing unit 6. The focusing unit 6 focuses the laser light L thus incident on the object 11. In this embodiment, the focusing unit 6 is configured by attaching the focusing lens unit 61 to the bottom wall of the housing H via a driving mechanism 62. The driving mechanism 62 moves the focusing lens unit 61 along the Z direction, for example, by the driving force of a piezoelectric element.

[0027] An imaging optical system (not shown) is disposed between the spatial light modulator 5 and the light collecting unit 6 inside the housing H. The imaging optical system constitutes a double-telecentric optical system in which the reflecting surface of the spatial light modulator 5 and the entrance pupil plane of the light collecting unit 6 (collecting lens 61a described later) are in an imaging relationship. As a result, the image of the laser light L on the reflecting surface of the spatial light modulator 5 (the image of the laser light L modulated by the spatial light modulator 5) is transferred (imaged) on the entrance pupil plane of the light collecting unit 6.

[0028] A pair of distance measuring sensors S1, S2 are attached to the bottom wall of the housing H so as to be located on both sides of the condenser lens unit 61 in the X direction. Each of the distance measuring sensors S1, S2 emits distance measuring light (e.g., laser light) to the surface 11a of the target object 11 and detects the distance measuring light reflected by the surface 11a to obtain displacement data of the surface 11a.

[0029] The optical axis monitor unit 7 is disposed in the housing H. The optical axis monitor unit 7 detects a portion of the laser light L that has passed through the dichroic mirror M2 (for example, 0.5 to 5% of the laser light L that has entered the dichroic mirror M2). The detection result by the optical axis monitor unit 7 indicates, for example, the relationship between the optical axis of the laser light L that enters the condenser lens unit 61 and the optical axis of the condenser lens unit 61.

[0030] The visible imaging unit 8 is disposed in the housing H. The visible imaging unit 8 emits visible light V and acquires an image of the object 11 by the visible light V. In this embodiment, the visible light V emitted from the visible imaging unit 8 is irradiated onto the surface 11a of the object 11 via the dichroic mirror M2 and the light collecting unit 6, and the visible light V reflected by the surface 11a is detected by the visible imaging unit 8 via the light collecting unit 6 and the dichroic mirror M2.

[0031] The infrared imaging unit 9 is attached to a side wall of the housing H. The infrared imaging unit 9 emits infrared light and obtains an image of the object 11 by the infrared light as an image. In this embodiment, the housing H and the infrared imaging unit 9 are movable together along the Z direction.

[0032] The control unit 10 controls the operation of each part of the laser processing device 1. The control unit 10 has a processing unit 101, a storage unit 102, and an input receiving unit 103. The processing unit 101 is configured as a computer device including a processor, a memory, a storage, a communication device, etc. In the processing unit 101, the processor executes software (programs) loaded into the memory, etc., and controls reading and writing of data in the memory and the storage, as well as communication by the communication device. The storage unit 102 is, for example, a hard disk, etc., and stores various data. The input receiving unit 103 is an interface unit that receives input of various data from an operator. In this embodiment, the input receiving unit 103 configures a GUI (Graphical User Interface).

[0033] In the laser processing device 1 configured as described above, when the laser light L is focused inside the object 11, the laser light L is absorbed in a portion corresponding to the focusing point C of the laser light L, and a modified region 12 is formed inside the object 11. The modified region 12 is a region having density, refractive index, mechanical strength, and other physical properties different from the surrounding non-modified region. Examples of the modified region 12 include a melting treatment region, a crack region, an insulation breakdown region, and a refractive index change region. The modified region 12 has a property that cracks tend to extend from the modified region 12 to the incident side of the laser light L and the opposite side. Such a property of the modified region 12 is utilized for cutting the object 11.

[0034] As an example, the operation of the laser processing apparatus 1 in the case of forming a modified region 12 inside the object 11 along a line 15 for cutting the object 11 will be described.

[0035] First, the laser processing device 1 rotates the support part 2 about an axis parallel to the Z direction as a center line so that the line 15 set on the object 11 is parallel to the X direction. Next, the laser processing device 1 moves the support part 2 along each of the X direction and the Y direction based on an image (e.g., an image of a functional element layer of the object 11) acquired by the infrared imaging unit 9 so that the focal point C of the laser light L is located on the line 15 when viewed from the Z direction. Hereinafter, such "alignment of the focusing part 6 with respect to the processing start position on the line 15" will be referred to as "alignment".

[0036] Next, based on an image (e.g., an image of the surface 11a of the target 11) acquired by the visible imaging unit 8, the laser processing device 1 moves the housing H (i.e., the focusing unit 6) along the Z direction so that the focusing point C of the laser light L is located on the surface 11a. Hereinafter, such "alignment of the focusing unit 6 with respect to the surface 11a" is referred to as "height setting." Next, using that position as a reference, the laser processing device 1 moves the housing H (i.e., the focusing unit 6) along the Z direction so that the focusing point C of the laser light L is located at a predetermined depth from the surface 11a.

[0037] Next, the laser processing apparatus 1 emits the laser light L from the light source 3 and moves the support part 2 along the X direction so that the focal point C of the laser light L moves relatively along the line 15. Hereinafter, the "relative movement direction of the laser light L with respect to the object 11" is referred to as the "relative movement direction A of the laser light L." At this time, the laser processing apparatus 1 operates the drive mechanism 62 of the focusing part 6 so that the focal point C of the laser light L is located at a predetermined depth from the surface 11a based on the displacement data of the surface 11a acquired by the distance measuring sensor located in front in the relative movement direction A of the laser light L out of the pair of distance measuring sensors S1, S2.

[0038] As a result of the above, a row of modified regions 12 is formed along the line 15 and at a certain depth from the surface 11a of the object 11. When the laser light L is emitted from the light source 3 using the pulse oscillation method, multiple modified spots 12s are formed so as to be lined up in a row along the X direction. One modified spot 12s is formed by irradiation with one pulse of laser light L. A row of modified regions 12 is a collection of multiple modified spots 12s lined up in a row. Adjacent modified spots 12s may be connected to each other or separated from each other depending on the pulse pitch of the laser light L (the value obtained by dividing the relative moving speed of the focal point C with respect to the object 11 by the repetition frequency of the laser light L). [Configuration of spatial light modulator]

[0039] The spatial light modulator 5 of this embodiment is a reflective liquid crystal (LCOS: Liquid Crystal on Silicon) spatial light modulator (SLM: Spatial Light Modulator). Fig. 2 is a cross-sectional view of a portion of the spatial light modulator shown in Fig. 1. As shown in Fig. 2, the spatial light modulator 5 is configured by laminating a drive circuit layer 52, a pixel electrode layer 53, a reflective film 54, an alignment film 55, a liquid crystal layer 56, an alignment film 57, a transparent conductive film 58, and a transparent substrate 59 in this order on a semiconductor substrate 51.

[0040] The semiconductor substrate 51 is, for example, a silicon substrate. The drive circuit layer 52 configures an active matrix circuit on the semiconductor substrate 51. The pixel electrode layer 53 includes a plurality of pixel electrodes 53a arranged in a matrix along the surface of the semiconductor substrate 51. Each pixel electrode 53a is formed of, for example, a metal material such as aluminum. A voltage is applied to each pixel electrode 53a by the drive circuit layer 52.

[0041] The reflective film 54 is, for example, a dielectric multilayer film. The alignment film 55 is provided on the surface of the liquid crystal layer 56 on the reflective film 54 side, and the alignment film 57 is provided on the surface of the liquid crystal layer 56 opposite the reflective film 54. Each alignment film 55, 57 is formed of, for example, a polymer material such as polyimide, and the contact surface of each alignment film 55, 57 with the liquid crystal layer 56 is subjected to, for example, a rubbing treatment. The alignment films 55, 57 align the liquid crystal molecules 56a contained in the liquid crystal layer 56 in a fixed direction.

[0042] The transparent conductive film 58 is provided on the surface of the transparent substrate 59 on the alignment film 57 side, and faces the pixel electrode layer 53 with the liquid crystal layer 56 and the like sandwiched therebetween. The transparent substrate 59 is, for example, a glass substrate. The transparent conductive film 58 is formed of, for example, a light-transmitting and conductive material such as ITO. The transparent substrate 59 and the transparent conductive film 58 transmit the laser light L.

[0043] In the spatial light modulator 5 configured as above, when an image signal indicating a modulation pattern is input from the control unit 10 to the drive circuit layer 52, a voltage corresponding to the image signal is applied to each pixel electrode 53a, and an electric field is formed between each pixel electrode 53a and the transparent conductive film 58. When this electric field is formed, the arrangement direction of the liquid crystal molecules 216a in the liquid crystal layer 56 changes for each region (pixel 56p) corresponding to each pixel electrode 53a, and the refractive index changes for each region corresponding to each pixel electrode 53a. This state is a state in which the modulation pattern is displayed on the liquid crystal layer 56.

[0044] With the modulation pattern displayed on the liquid crystal layer 56, when laser light L enters the liquid crystal layer 56 from the outside through the transparent substrate 59 and the transparent conductive film 58, is reflected by the reflective film 54, and is emitted from the liquid crystal layer 56 to the outside through the transparent conductive film 58 and the transparent substrate 59, the laser light L is modulated according to the modulation pattern displayed on the liquid crystal layer 56. In this way, with the spatial light modulator 5, by appropriately setting the modulation pattern to be displayed on the liquid crystal layer 56, it is possible to modulate the laser light L (for example, modulate the intensity, amplitude, phase, polarization, etc. of the laser light L). [Object composition]

[0045] Fig. 3 is a plan view of a wafer which is an object of an embodiment. Fig. 4 is a cross-sectional view of a portion of the wafer shown in Fig. 3. The object 11 of this embodiment is a wafer 20 as shown in Figs. 3 and 4. The wafer 20 has a first surface 20a and a second surface (laser light incident surface) 20b opposite to the first surface 20a. The wafer 20 is configured by stacking a functional element layer 22 on a semiconductor substrate 21.

[0046] The semiconductor substrate 21 is, for example, a silicon substrate. The semiconductor substrate 21 has a first surface 21a and a second surface 21b opposite to the first surface 21a. The second surface 21b is the second surface 20b of the wafer 20. The semiconductor substrate 21 is provided with a notch 21c indicating a crystal orientation. Note that the semiconductor substrate 21 may be provided with an orientation flat instead of the notch 21c.

[0047] The functional element layer 22 is provided on the first surface 21a of the semiconductor substrate 21. The functional element layer 22 includes a plurality of functional elements 22a arranged in a matrix along the first surface 21a of the semiconductor substrate 21. Each functional element 22a is, for example, a light receiving element such as a photodiode, a light emitting element such as a laser diode, a circuit element such as a memory, etc. Each functional element 22a may be configured three-dimensionally by stacking a plurality of layers.

[0048] The wafer 20 is cut into individual functional elements 22a along each of the multiple lines 15. The multiple lines 15 extend in a lattice pattern along the second surface 21b of the wafer 20 so as to pass between each of the multiple functional elements 22a (more specifically, through the center of the street region 23 extending between adjacent functional elements 22a) when viewed in the thickness direction of the wafer 20. Each line 15 is a virtual line set on the wafer 20 by the laser processing device 1. Note that each line 15 may be a line actually drawn on the wafer 20. [An example of laser processing equipment operation]

[0049] FIG. 5 is a cross-sectional view of a wafer for explaining an example of the operation of the laser processing apparatus in the case of three-point branching. As shown in FIG. 5, in the laser processing apparatus 1, the wafer 20 is supported by the support unit 2 so that the second surface 20b of the wafer 20 is perpendicular to the Z direction. Then, the control unit 10 controls the spatial light modulator 5, and a predetermined modulation pattern (for example, a modulation pattern including a diffraction pattern) is displayed on the liquid crystal layer 56 of the spatial light modulator 5. In this state, the laser light L is emitted from the light source 3, and the laser light L is focused on the wafer 20 from the second surface 20b side by the focusing unit 6. That is, the laser light L is modulated by the spatial light modulator 5, and the modulated laser light L is focused on the wafer 20 from the second surface 20b side by the focusing unit 6.

[0050] As a result, the laser light L is branched (diffracted) into a plurality of processing lights L1, L2, and L3 including a zero-order light, and a plurality of focus points C1, C2, and C3 of the plurality of processing lights L1, L2, and L3 are positioned at different positions in the Z direction and the X direction. In this embodiment, the processing light L2 is the zero-order light. The focus point C1 of the processing light L1 is located forward of the focus point C2 of the processing light L2, which is the zero-order light, in the relative movement direction A of the laser light L, and the focus point C3 of the processing light L3 is located rearward of the focus point C2 of the processing light L2, which is the zero-order light, in the relative movement direction A of the laser light L. As an example, the processing light L1 is a +1-order light, and the processing light L3 is a -1-order light.

[0051] In this embodiment, the laser light L is modulated by the spatial light modulator 5 so that the multiple focal points C1, C2, and C3 have a positional relationship in which the focal point C2 is located closer to the first surface 20a of the wafer 20 in the Z direction than the focal point C3 in the Z direction, and the focal point C1 is located closer to the first surface 20a of the wafer 20 in the Z direction than the focal point C2. Furthermore, in this embodiment, the laser light L is modulated by the spatial light modulator 5 so that the processing light L1 branched to the most forward side in the relative movement direction A of the laser light L among the multiple processing lights L1, L2, and L3 has the largest output (energy, intensity). In addition, even if there is a processing light among the multiple processing lights L1, L2, and L3 that has an output equal to the output of processing light L1, if the output of the other processing lights is less than the output of processing light L1, processing light L1 is included in the case where it has the greatest output.

[0052] In a state where the laser light L is split into a plurality of processing lights L1, L2, and L3 including a zero-order light, the control unit 10 controls the support unit 2 so that the X direction coincides with the extending direction of the line 15 and the plurality of focusing points C1, C2, and C3 move relatively along the line 15. As a result, three rows of modified regions 12 are formed along one line 15. The distances from the second surface 20b of the wafer 20 to the respective modified regions 12 are different from each other, and the depth at which each modified region 12 is formed corresponds to the depth at which each focusing point C1, C2, and C3 is aligned. The above is an example of the operation of the laser processing device 1, and an example of the processing process performed by the control unit 10. In the above example, an example in which the laser light L is split into three processing lights L1, L2, and L3 has been described, but the processing process is not limited to this, and the laser light L may be split into two or four or more processing lights, or the laser light L may be used without being split. [Issues surrounding spatial light modulators and knowledge regarding solutions]

[0053] Next, a first problem when using the spatial light modulator 5 as described above and the inventor's findings related to a method for solving the first problem will be described. FIG. 6 is a diagram showing an example of a modulation pattern displayed on a spatial light modulator. FIG. 6(a) shows an ideal modulation pattern Pi, and FIG. 6(b) shows an actual modulation pattern Pr. As shown in FIG. 6, for example, between adjacent pixels 56p at a pixel pitch D, dulling may occur in the actual modulation pattern Pr with respect to the phase modulation amount of the ideal modulation pattern Pi. The inventor has found that this dulling in the phase modulation amount may cause the following problems.

[0054] That is, as shown in Fig. 7, when a modulation pattern Pl including a region with a phase modulation amount larger than 2π (for example, 2π to 6π) is to be displayed on a spatial light modulator 5 having a phase modulation ability of an upper limit of 2π (see Fig. 7(a)), the modulation pattern Pl is reproduced by folding back the region exceeding 2π (see Fig. 7(b)). In this case, due to the slowing of the phase modulation amount, the modulation state of the laser light L may differ between a region where folding back occurs (inverted region) and a region where folding back does not occur (normal region) in the modulation pattern Pl. This point will be explained in more detail.

[0055] First, an image signal can be used as a signal for displaying a modulation pattern on the spatial light modulator 5. In this case, a voltage corresponding to the gradation value of the image signal input to the spatial light modulator 5 is applied to the liquid crystal layer 56, and a refractive index change corresponding to the voltage value occurs, thereby displaying the modulation pattern. Then, the laser light incident on the spatial light modulator 5 is phase-modulated according to this modulation pattern. Therefore, in this case, the image signal and the modulation pattern, and the gradation value of the image signal and the phase modulation amount given to the laser light L by the modulation pattern correspond to each other. For this reason, hereinafter, the gradation value and the phase modulation amount may be described as having the same meaning.

[0056] Here, the modulation pattern includes a diffraction grating pattern for splitting the laser light L into three processing lights L1, L2, and L3. Fig. 8 is a schematic diagram showing an example of a modulation pattern including a diffraction grating pattern. In each example of Fig. 8, the spatial light modulator 5 has a phase modulation capability of 2π, and the phase modulation amount (phase difference) is set to 0 [rad] when the gradation value is 0, and 2π [rad] when the gradation value is 256.

[0057] 8(a) shows only the diffraction grating pattern Pg whose gradation value T1 is 82. In this case, in the diffraction grating pattern Pg, the gradation value T1 does not exceed 256, that is, the phase modulation amount does not exceed 2π, so that no inverted region occurs.

[0058] 8(b), a modulation pattern Pc1 is shown, which is configured by superimposing an offset pattern Pb1 having a constant offset value Q1 of 100 on a diffraction grating pattern Pg having a gradation value T1 of 82. In this case, too, the total gradation value T2 (gradation value T1+offset value Q1) in the modulation pattern Pc1 is 182, which does not exceed 256, that is, the phase modulation amount does not exceed 2π, so no inversion region occurs. Note that the offset pattern is intended to diagrammatically show a case in which a certain modulation pattern (for example, a diffraction grating pattern) is superimposed on another modulation pattern (for example, a distortion correction pattern or an aberration correction pattern of the spatial light modulator 5).

[0059] On the other hand, in the example shown in FIG. 8(c), a modulation pattern Pc2 is shown which is configured by superimposing an offset pattern Pb2 whose offset value Q2 is constant at 200 on a diffraction grating pattern Pg whose gradation value T1 is 82. In this case, the total gradation value (gradation value T1+offset value Q2) in the modulation pattern Pc2 is 282, and an excess T3 that exceeds 256 (i.e., a portion where the phase modulation amount exceeds 2π) occurs. This excess T3 is calculated as "offset value Q2+gradation value T1-256". For this reason, in this modulation pattern Pc2, aliasing of the phase modulation amount occurs, resulting in an inversion region.

[0060] The gradation value T4 of the modulation pattern Pc2 in the inversion region is calculated by subtracting the excess T3 from the offset value Q2. That is, the gradation value T4 of the modulation pattern Pc2 is calculated as "offset value Q2-(offset value Q2+gradation value T1-256)". According to this, the gradation value T4 here is 174. Note that hereinafter, the gradation value T1 may be referred to as the gradation value of the normal region, and the gradation value T4 may be referred to as the gradation value of the inversion region.

[0061] In the case where the laser light L is split into three processing lights L1, L2, and L3, and there is no blunting of the phase modulation amount, when the gradation value T1 in the normal rotation area of ​​the diffraction grating pattern Pg is 82 (when the phase modulation amount is 0.64π), the diffraction efficiency of the processing lights L1 and L3, which are ±1st order light, and the diffraction efficiency of the processing light L2, which is the 0th order light, match, and the output balance of the processing lights L1, L2, and L3 is achieved. In a similar case, when the gradation value T4 in the inversion area is 174 (when the phase modulation amount is 1.36π), the diffraction efficiencies of the processing lights L1, L2, and L3 also match, and the output balance is achieved.

[0062] On the other hand, in the actual case where the phase modulation amount is blunted, when the gradation value T1 in the normal rotation area is 89 (when the phase modulation amount is 0.70π), the diffraction efficiency of the processing light L1, L3, which is the ±1st order light, and the diffraction efficiency of the processing light L2, which is the 0th order light, match, and the output balance of the processing light L1, L2, and L3 is achieved. When the gradation value T1 in the normal rotation area is 89, the gradation value T4 in the inverted area corresponding to this gradation value T1 is 167 (see the above calculation formula). And, when the gradation value T4 in the inverted area is 167 (when the phase modulation amount is 1.30π), the output balance of the three processing light L1, L2, and L3 is greatly disrupted, and the 0th order light becomes relatively weak. In other words, when there is blunting in the phase modulation amount, the diffraction efficiency varies between the normal rotation area and the inverted area.

[0063] Here, in the case of a modulation pattern Pc3 (see FIG. 10) configured by superimposing a distortion correction pattern Pd that acts at least partially as the offset patterns Pb1 and Pb2 (adding offset values ​​Q1 and Q2 to the diffraction grating pattern Pg) on ​​the diffraction grating pattern Pg shown in FIG. 9, the normal region R1 and the inverted region R2 are mixed. Even if the diffraction efficiency varies between the normal region R1 and the inverted region R2 as described above, when the diffraction grating pattern Pg is used alone or the gradation value is constant throughout the modulation pattern, the gradation value can be recalculated based on actual measurement and correction can be performed.

[0064] However, as shown in Fig. 10, in practice, multiple modulation patterns are combined or various modulation patterns are switched between, making such correction difficult. When the laser light L is split into two processing lights at an equal ratio, the gradation value T1 in the normal rotation region R1 and the gradation value T4 in the inverted region R2 are both 128, and no problem occurs. However, when the laser light L is split into two processing lights at different ratios, a similar problem occurs.

[0065] In particular, in the modulation pattern Pc1 to which a relatively small offset value is applied to the diffraction grating pattern Pg shown in FIG. 11(a) (see FIG. 11(b)), the ratio of the normal region R1 to the inverted region R2 is 1:0 (the entire region is the normal region R1), whereas in the modulation pattern Pc2 to which a relatively large offset value is applied (see FIG. 11(c)), the ratio of the normal region R1 to the inverted region R2 is 0:1 (the entire region is the inverted region R2), resulting in large variations in diffraction efficiency depending on the patterns and regions superimposed on the diffraction grating pattern Pg. Note that the horizontal axis of each graph in FIG. 11 indicates the number (number) of pixels 56p of the liquid crystal layer 56 as the position of the modulation pattern. That is, 256 on the horizontal axis indicates the position of the modulation pattern corresponding to the 256th pixel 56p from one end of the liquid crystal layer 56.

[0066] From this viewpoint, if the bias in the ratio between the normal region R1 and the reversed region R2 is suppressed, the variation in the diffraction efficiency according to the type of the modulation pattern to be synthesized and the variation in the diffraction efficiency according to each region of the modulation pattern are suppressed, and the influence of the dulling of the phase modulation amount is mitigated. Therefore, here, an averaging pattern is used to average the ratio between the normal region R1 and the reversed region R2.

[0067] Fig. 12 is a diagram showing an example of an averaging pattern. As shown in Fig. 12, the averaging pattern Ps is a modulation pattern in which the gradation value changes at a constant gradient from the minimum value (0) to the maximum value (256) (the phase modulation amount changes from 0 to 2π) from one end of the averaging pattern Ps to the other end.

[0068] Fig. 13 is a diagram showing an example of a modulation pattern formed by superimposing an averaging pattern on a diffraction grating pattern. Fig. 13(a) shows a modulation pattern Pc4 formed by superimposing an averaging pattern Ps on a diffraction grating pattern Pg, Fig. 13(b) shows a modulation pattern Pc5 formed by superimposing an offset pattern Pb1 and an averaging pattern Ps on the diffraction grating pattern Pg, and Fig. 13(c) shows a modulation pattern Pc6 formed by superimposing an offset pattern Pb2 and an averaging pattern Ps on the diffraction grating pattern Pg.

[0069] In the modulation pattern Pc4, the ratio of the normal region R1 to the inverted region R2 is 0.67 to 0.33, in the modulation pattern Pc5, the ratio of the normal region R1 to the inverted region R2 is 0.67 to 0.33, and in the modulation pattern Pc6, the ratio of the normal region R1 to the inverted region R2 is 0.73 to 0.27. In either case, the ratio of the normal region R1 to the inverted region R2 is averaged compared to when the averaging pattern Ps is not used. As a result, the bias in the ratio of the normal region R1 to the inverted region R2 is suppressed, and the variation in the diffraction efficiency according to the type of the modulation pattern to be synthesized and the variation in the diffraction efficiency according to each region of the modulation pattern are suppressed, and the influence of the dulling of the phase modulation amount is mitigated.

[0070] Here, it is possible to adjust the degree of averaging by changing the gradient (folding period) of the gradation values ​​in the averaging pattern Ps. The example in Fig. 12 is an example in which the gradient is 0.5 and the period is 1, so that the gradation values ​​increase from 0 to 256 as the position of the averaging pattern Ps changes from 0 to 512. On the other hand, when the period is 2, the gradient of the gradation values ​​in the averaging pattern Ps doubles, so that the gradation values ​​increase from 0 to 256 as the position of the averaging pattern Ps changes from 0 to 256, and then increase again from 0 to 256 as the position of the averaging pattern Ps changes from 257 to 512.

[0071] FIG. 14 is a diagram showing an example of a modulation pattern formed by superimposing an averaging pattern on a distortion correction pattern. FIG. 14(a) shows a case where an averaging pattern Ps is not used, FIG. 14(b) shows a case where an averaging pattern Ps with a period of 1 is used, FIG. 14(c) shows a case where an averaging pattern Ps with a period of 2 is used, FIG. 14(d) shows a case where an averaging pattern Ps with a period of 4 is used, and FIG. 14(e) shows a case where an averaging pattern Ps with a period of 8 is used. As shown in FIG. 14, the larger the period of the averaging pattern Ps (the larger the gradient of the gradation value), the more the aliasing of the phase modulation amount increases, and the bias in the ratio between the normal rotation region R1 and the inversion region R2 is dispersed and averaged. The averaging pattern Ps is, for example, a shift pattern having a function of shifting the focal point C of the laser light L.

[0072] Next, a second problem in the case where the spatial light modulator 5 is used and the inventor's findings related to a method for solving the second problem will be described. FIG. 15 is a schematic diagram showing the relationship between a condenser lens and a laser beam. As shown in FIG. 15, the condenser lens unit 61 of the focusing section 6 includes a condenser lens 61a for focusing the laser beam L toward the wafer 20. In the laser processing apparatus 1, the spatial light modulator 5 modulates the laser beam L, thereby changing the focusing position of the laser beam L according to the NA region (the position in the radial direction Kd of the condenser lens 61a) into which the laser beam L is incident on the condenser lens 61a, and the focusing point C of the laser beam L can be shaped into an elongated shape.

[0073] Here, by modulating the laser light L using the shaping pattern Pf shown in Fig. 16, the condensing position of the laser light L is changed in the Z direction so that the condensing position of the laser light L becomes deeper in the Z direction (farther away from the condensing lens 61a) as the incident position of the laser light L in the condensing lens 61a moves away from the center in the radial direction Kd, and the condensing point C of the laser light L is shaped to be elongated in the Z direction. In this shaping pattern Pf, the turning-back portion B of the phase modulation amount is formed concentrically with respect to the region where the laser light L is incident in the shaping pattern Pf (the region corresponding to the entrance pupil surface of the condensing lens 61a). That is, in such a shaping pattern Pf, the turning-back portion B of the phase modulation amount is unevenly distributed in a part of the shaping pattern Pf.

[0074] For this reason, there may be a variation in the focusing state of the laser light L between a portion of the laser light L modulated through a portion of the shaping pattern Pf where the turning-back portion B is concentrated and a portion of the laser light L modulated without passing through a portion of the shaping pattern Pf where the turning-back portion B is concentrated. As a result, when another modulation pattern is superimposed on the shaping pattern Pf and an offset is applied to the phase modulation amount of the shaping pattern Pf, the processing result may change depending on the magnitude of the offset value.

[0075] Fig. 17(a) is an example showing the processing result when the offset value is 64, and Fig. 17(b) is an example showing the processing result when the offset value is 96. As shown in Fig. 17, the processing result changes depending on the magnitude of the offset value. This is considered to be because the laser light L modulated at the portion where the turning portion B in the shaping pattern Pf is unevenly distributed is brought into an unintended focused state.

[0076] In contrast to this, Fig. 18 is a diagram showing a modulation pattern formed by superimposing the averaging pattern shown in Fig. 12 on the shaping pattern shown in Fig. 16. As shown in Fig. 18, in the modulation pattern Pe formed by superimposing the averaging pattern Ps on the shaping pattern Pf, the turning-back portion B of the phase modulation amount is dispersed over the entire region in which the laser light L is incident in the modulation pattern Pe, and the positions where the turning-back portion B is formed are averaged, compared to the shaping pattern Pf.

[0077] Fig. 19 is a cross-sectional photograph showing the processing results when using the shaping pattern shown in Fig. 16 and the modulation pattern shown in Fig. 18. Fig. 19(a) shows the processing result when an offset value of 64 is applied to the shaping pattern Pf, and Fig. 19(b) shows the processing result when an offset value of 96 is applied to the shaping pattern Pf. Fig. 19(c) shows the processing result when an offset value of 64 is applied to the modulation pattern Pe, and Fig. 19(d) shows the processing result when an offset value of 96 is applied to the modulation pattern Pe.

[0078] As shown in (a) and (b) of FIG. 19, when the averaging pattern Ps is not superimposed on the shaping pattern Pf, the processing result also changes when the offset value changes. In contrast, as shown in (c) and (d) of FIG. 19, when the averaging pattern Ps is superimposed on the shaping pattern Pf, the change in the processing result with respect to the change in the offset value is small. In this way, by using the averaging pattern Ps, it is possible to suppress the uneven distribution of the folding back portion B of the phase modulation amount in the modulation pattern, and as a result, the variation in the processing result is suppressed. With the laser processing device 1 according to this embodiment, the following laser processing can be performed by utilizing the above knowledge. [One example of laser processing according to this embodiment]

[0079] First, as described above, the laser processing apparatus 1 according to this embodiment includes the support unit 2 for supporting the object 11 (here, the wafer 20), the light source 3 for emitting laser light L, the spatial light modulator 5 for modulating the laser light L emitted from the light source 3 according to a modulation pattern and emitting the modulated laser light L, the focusing unit 6 including a focusing lens 61a for focusing the laser light L emitted from the spatial light modulator 5 toward the wafer 20, and the control unit 10 for controlling each unit of the laser processing apparatus 1. The control unit 10 inputs an image signal to the spatial light modulator 5, thereby causing the spatial light modulator 5 to display a modulation pattern corresponding to the image signal.

[0080] In such a laser processing apparatus 1, at the time of laser processing, first, the wafer 20 is supported by the support part 2 so that the second surface 20b of the wafer 20 is perpendicular to the Z direction. Next, the control part 10 inputs an image signal to the spatial light modulator 5 to display a predetermined modulation pattern on the liquid crystal layer 56 of the spatial light modulator 5. In this state, the control part 10 controls the light source 3 to emit the laser light L from the light source 3, and the laser light L is condensed toward the wafer 20 from the second surface 20b side by the condenser lens 61a. That is, the laser light L is modulated by the spatial light modulator 5, and the modulated laser light L is condensed on the wafer 20 from the second surface 20b side by the condenser lens 61a. As a result, a condensing point C of the laser light L is formed inside the wafer 20.

[0081] At the same time, the control unit 10 controls the movement of the support unit 2 here so that the focal point C of the laser light L moves relative to the wafer 20 along a line 15 along the X direction. As a result, the wafer 20 is irradiated with the laser light L (processing is executed) while the focal point C of the laser light L moves relative to the wafer 20 along the X direction.

[0082] When processing by splitting the laser light L is desired, the control unit 10 can input an image signal including a second signal corresponding to the diffraction grating pattern Pg to the spatial light modulator 5 so as to display a modulation pattern including a diffraction grating pattern Pg (second pattern) for splitting the laser light L into a plurality of processing lights L1, L2, and L3 on the spatial light modulator 5 in this processing process. The branching direction of the laser light L at this time is, for example, the X direction which is the processing proceeding direction. Note that the image signal corresponding to a certain modulation pattern is an image signal in which a gradation value corresponding to the phase modulation amount at each pixel 56p is set in each region of the image signal corresponding to each pixel 56p of the liquid crystal layer 56 of the spatial light modulator 5.

[0083] On the other hand, when it is desired to process the focal point C of the laser light L into a long shape, the control unit 10 can input an image signal including a third signal corresponding to the shaping pattern Pf to the spatial light modulator 5 so as to display a modulation pattern including a shaping pattern Pf (third pattern) on the spatial light modulator 5 for changing the focal position of the laser light L depending on the position in the radial direction Kd of the focusing lens 61a during the processing.

[0084] In either case, in the processing, the control unit 10 inputs an image signal including a first signal whose gradation value changes at a constant gradient from a minimum value to a maximum value from a region corresponding to one end of the modulation pattern in the first direction to a region corresponding to the other end of the modulation pattern in the first direction to the spatial light modulator, thereby causing the spatial light modulator 5 to display a modulation pattern including an averaged pattern Ps (first pattern) according to the first signal. Therefore, at this time, the modulation pattern displayed on the spatial light modulator 5 is a modulation pattern configured by superimposing at least the diffraction grating pattern Pg and / or the shaping pattern Pf and the averaged pattern Ps. Similarly, at this time, in the image signal input to the spatial light modulator 5, a gradation value is set in which at least the gradation value of the second signal and / or the third signal and the gradation value of the first signal are superimposed.

[0085] In particular, when the laser light L is branched into a plurality of processing lights L1, L2, and L3 in the processing process, the control unit 10 can set the first direction, which is the direction in which the gradation value of the averaged pattern Ps changes, to a direction corresponding to a direction intersecting the branching direction of the laser light L (or a direction corresponding to a direction parallel to the branching direction). In other words, when the branching direction of the laser light L is the X direction, the first direction can be set to a direction corresponding to the Y direction intersecting the X direction (or a direction corresponding to the X direction).

[0086] In addition, the control unit 10 may change the gradation value in the first signal at a constant gradient from the minimum value to the maximum value from a region corresponding to one end of the averaged pattern Ps in a second direction intersecting the first direction to a region corresponding to the other end of the averaged pattern Ps in the second direction. When the first direction corresponds to the Y direction as described above, the second direction corresponds to the X direction, for example. In this case, the spatial light modulator 5 can display the averaged pattern Ps in which the phase modulation amount changes at a constant gradient in at least two directions. Note that the gradient of the gradation value in the first direction and the gradient of the gradation value in the second direction may be the same or different. For example, when the number of pixels 56p of the liquid crystal layer 56 of the spatial light modulator 5 is different in two directions, when the phase modulation amount is changed at a constant gradient (at the same period) from the minimum value to the maximum value from one end to the other end in each direction, the gradient of the gradation value in the first direction and the gradient of the gradation value in the second direction will be different in the image signal.

[0087] Furthermore, in the processing, the control unit 10 can generate and input an image signal so that another modulation pattern, such as a distortion correction pattern or an aberration correction pattern, is superimposed on the image signal of the spatial light modulator 5. In this case, in the spatial light modulator 5, an offset corresponding to the other modulation pattern is imparted to the diffraction grating pattern Pg and the shaping pattern Pf, and further, an averaging pattern Ps is superimposed.

[0088] In addition, in the processing, when the averaged pattern Ps, which is a shift pattern, is superimposed on the modulation pattern displayed on the spatial light modulator 5, the focal point C of the laser light L may be shifted from the line 15 within the wafer 20 as a result. Therefore, in this case, the control unit 10 can control the optical axis adjustment unit 4 so that the focal point C of the laser light L is positioned on the line 15. At this time, the control unit 10 can refer to an image acquired by the visible imaging unit 8 (for example, an image of the second surface 20b of the wafer 20).

[0089] As described above, in the laser processing device 1, the control unit 10 inputs an image signal to the spatial light modulator 5, thereby causing the spatial light modulator 5 to display a modulation pattern, and executes processing while modulating the laser light L according to the modulation pattern. In the image signal, a gradation value corresponding to the phase modulation amount at each position of the modulation pattern is set for each region constituting the image signal. That is, in the spatial light modulator 5, the phase modulation amount at each position of the modulation pattern is set to an amount corresponding to the gradation value of each region of the image signal, so that a desired modulation pattern is displayed as a whole. The image signal includes a first signal in which the gradation value changes at a constant gradient from a region corresponding to one end of the modulation pattern to a region corresponding to the other end of the modulation pattern. As a result, the modulation pattern displayed on the spatial light modulator 5 includes an averaged pattern Ps (first pattern) in which the phase modulation amount changes in one direction at a constant gradient according to the first signal. As described above, when the modulation pattern includes such an averaged pattern Ps, the influence of the dulling of the phase modulation amount of the modulation pattern and the uneven distribution of the folding back portion B of the phase modulation amount in the modulation pattern are suppressed.

[0090] In addition, in the laser processing apparatus 1, in the processing process, the control unit 10 may input an image signal including a second signal corresponding to the diffraction grating pattern Pg (second pattern) to the spatial light modulator 5 so as to display a modulation pattern including the diffraction grating pattern Pg for splitting the laser light L into a plurality of processing lights on the spatial light modulator 5. In this case, the diffraction grating pattern Pg is superimposed on the modulation pattern. In such a case, the variation among the plurality of processing lights L1, L2, and L3 is suppressed.

[0091] Furthermore, in the laser processing apparatus 1, the first direction may be a direction corresponding to a direction intersecting the branching direction of the laser light L. In this case, it is possible to more effectively suppress the variation among the processing light L1, L2, and L3.

[0092] Furthermore, in the laser processing apparatus 1, in the processing, the control unit 10 may input an image signal including a third signal corresponding to the shaping pattern Pf to the spatial light modulator 5 so as to display a modulation pattern including a shaping pattern Pf (third pattern) for changing the focusing position of the laser light L depending on the position in the radial direction Kd of the focusing lens 61a on the spatial light modulator 5. In this case, it is possible to suppress uneven distribution of the folded portion B of the phase modulation amount in the shaping pattern Pf so as to correspond to a specific position in the radial direction Kd of the focusing lens 61a, and suppress variation in the focusing state of the laser light L.

[0093] Furthermore, in the laser processing apparatus 1, the control unit 10 may change the gradation value in the first signal from a region corresponding to one end of the modulation pattern in a second direction intersecting with the first direction to a region corresponding to the other end of the modulation pattern in the second direction at a constant gradient from the minimum value to the maximum value. In this case, the gradation value of the image signal changes at a constant gradient in two directions intersecting with each other. Therefore, it is possible to suppress the influence of the slowing of the phase modulation amount and the uneven distribution of the folding back portion of the phase modulation amount within the plane of the spatial light modulator 5 corresponding to the two directions.

[0094] Furthermore, in the laser processing apparatus 1, the control unit 10 may change the gradation value in the first signal so that the first signal has a plurality of periods in at least one direction. In this case, it is possible to more reliably suppress the influence of the dulling of the phase modulation amount and the uneven distribution of the folding back portion of the phase modulation amount.

[0095] The above embodiment describes one aspect of the present invention. Therefore, the present invention can be modified without being limited to the above-mentioned aspect.

[0096] For example, in the above embodiment, an example in which the gradient of the gradation value is 0.5 and the cycle is 1 is given in FIG. 12. However, the averaging pattern Ps is not limited to this. FIG. 20 is a diagram showing a modified example of the averaging pattern. The averaging pattern Ps according to the modified example of FIG. 20(a) is an example in which the gradient is 0.25 and the cycle is 1 such that the gradient value increases from 0 to 256 while the position of the averaging pattern Ps changes from 0 to 1024. The averaging pattern Ps according to the modified example of FIG. 20(b) is an example in which the gradient is 1 and the cycle is 2 such that the gradient value increases from 0 to 256 while the position of the averaging pattern Ps changes from 0 to 256, and the gradient value increases again from 0 to 256 while the position of the averaging pattern Ps changes from 257 to 512. In this example, one folding part is generated. Similarly, FIG. 20(c) shows an example in which the gradient is 2 and the cycle is 4. In this way, the gradient and cycle of the averaging pattern Ps can be set arbitrarily. Furthermore, the direction in which the period exists is not limited to one direction, but may be two directions.

[0097] In the above embodiment, the shaping pattern Pf for shaping the focal point C of the laser light L into an elongated shape in the Z direction is illustrated in Fig. 16. However, the shaping pattern Pf is not limited to this, and can be arbitrarily modified to shape the focal point C as shown in each drawing of Fig. 21.

[0098] Here, when the laser light L is split into a plurality of processing lights for processing as described above, the above-mentioned measure (first measure) against the variation in the output balance between each processing light is effective. On the other hand, when splitting processing in this way, a measure (second measure) against the influence of non-modulated light of the laser light L that has not been modulated by the spatial light modulator 5 on the focusing state on the zero-order light (e.g., processing light L2) of each processing light generated by splitting becomes effective.

[0099] The averaging pattern Ps has a function of shifting the focal point C of the laser light L. In a processing process, when the wafer 20 is arranged so that the street region 23 of the wafer 20 is aligned along the X direction and the focal point C of the laser light L is relatively moved in the X direction within the street region 23, it is possible to deal with both of the above-mentioned first and second measures by making the averaging pattern Ps to shift the focal point C in the Y direction.

[0100] However, while the first measure requires that the averaging pattern Ps shifts the light-focus point C in the Y direction by a relatively small amount, such as about 2 μm, the second measure requires that the averaging pattern Ps shifts the light-focus point C in the Y direction by a relatively large amount, such as about 5 μm or more. The second measure is only necessary when the zero-order light is used, from the viewpoint of suppressing interference between the zero-order light and non-modulated light.

[0101] Therefore, the required shift amount in the Y direction may be different in the first case where no splitting is performed, the second case where splitting is performed but the zero-order light is not used, and the third case where splitting is performed while using the zero-order light. More specifically, in the first case, it is not necessary to shift the focal point C in the Y direction to accommodate at least the first and second measures, and in the second case, the second measure is not required, so the shift amount in the Y direction can be suppressed to about 2 μm for the first measure, and in the third case, it is desirable to further consider the second measure and set the shift amount in the Y direction to about 5 μm or more. Therefore, the laser processing device 1 is configured to be able to accommodate such first to third cases.

[0102] That is, in the laser processing device 1, by adjusting the optical axis as a default state using the averaged pattern Ps with a shift amount in the Y direction of 5 μm, the state where all the focusing points of the laser light L are shifted in the Y direction by 5 μm is set as the reference position. This makes it possible to deal with the first and second measures that are necessary in all of the first to third cases.

[0103] However, in this case, the following new problem may arise: In other words, when the width of the street region 23 in the Y direction is narrow, as shown in Fig. 22, if the focal points C1, C2, and C3 of the multiple processing beams L1, L2, and L3 of the laser beam L are shifted in the Y direction by a shift amount Ds of about 6 µm, the focal point C0 of the unmodulated beam will be positioned on the functional element 22a (active area) facing the street region 23, and there is a risk of light leakage becoming a problem.

[0104] As a measure against such a new problem, the laser processing apparatus 1 can be adapted to the first measure by first adjusting the optical axis to a state in which the averaged pattern Ps with a shift amount in the Y direction of 2 μm is used as a default. This makes it possible to suitably handle the first case in which no branching process is performed and the second case in which the zeroth-order light is not used. On the other hand, in the laser processing apparatus 1, in the third case in which branching process is performed using the zeroth-order light, the gradation value is set so that the default averaged pattern Ps that generates a shift amount in the Y direction of about 2 μm is further shifted in the X direction by about 5 μm, thereby shifting the focusing points C1, C2, and C3 in both the X direction and the Y direction as shown in FIG. 23. In this case, the total shift amount Ds in the X direction and the Y direction is sufficiently increased while suppressing the shift amount in the Y direction, making it possible to address the second problem.

[0105] In the laser processing device, the shift direction of the focal point C in the X direction can be switched by, for example, inverting the sign of the gradient of the gradation value of the averaged pattern Ps depending on whether the moving direction of the focal point C (processing proceeding direction) is the X positive direction or the X negative direction. As an example, when the processing proceeding direction is the X positive direction, the shift direction of the focal point C can also be the X positive direction, and when the processing proceeding direction is the X negative direction, the shift direction of the focal point C can also be the X negative direction.

[0106] Furthermore, in the above embodiment, the averaging pattern Ps has been described as having a function of shifting the focal point C of the laser light L along the X direction or the Y direction, but the averaging pattern Ps may have a function of shifting the focal point C of the laser light L along the Z direction. [Explanation of symbols]

[0107] 1...laser processing device, 2...support portion, 5...spatial light modulator, 6...focusing portion, 10...control portion, 61a...focusing lens.

Claims

1. A laser processing apparatus for forming a modified region in an object by irradiating the object with laser light, comprising: A support portion for supporting the object; a light source for emitting the laser light; a spatial light modulator for modulating the laser light emitted from the light source in accordance with a modulation pattern and emitting the modulated laser light; a focusing unit including a focusing lens for focusing the laser light emitted from the spatial light modulator toward the target; a control unit that inputs an image signal to the spatial light modulator to cause the spatial light modulator to display the modulation pattern corresponding to the image signal; Equipped with the control unit controls movement of at least one of the support unit and the focusing unit to move a focusing point of the laser light relative to the object along an X direction along a laser light incident surface of the object, while executing a processing process of irradiating the object with the laser light; A gradation value corresponding to a phase modulation amount in the modulation pattern is set in each of the regions constituting the image signal, In the processing, the control unit inputs, to the spatial light modulator, the image signal including a first signal whose gradation value changes at a constant gradient from a minimum value to a maximum value from the region corresponding to one end of the modulation pattern in the first direction toward the region corresponding to the other end of the modulation pattern in the first direction, thereby causing the spatial light modulator to display the modulation pattern including a first pattern corresponding to the first signal. Laser processing equipment.

2. In the processing, the control unit inputs the image signal including a second signal corresponding to the second pattern to the spatial light modulator so as to display the modulation pattern including a second pattern for splitting the laser light into a plurality of processing lights on the spatial light modulator. The laser processing device according to claim 1.

3. The first direction is a direction corresponding to a direction intersecting a branching direction of the laser light. The laser processing device according to claim 2.

4. In the processing, the control unit inputs the image signal including a third signal corresponding to the third pattern to the spatial light modulator so as to display the modulation pattern including a third pattern for changing a focusing position of the laser light depending on a position in a radial direction of the focusing lens on the spatial light modulator. The laser processing device according to any one of claims 1 to 3.

5. the control unit changes the gradation value of the first signal at a constant gradient from a minimum value to a maximum value from the region corresponding to one end of the modulation pattern in a second direction intersecting the first direction to the region corresponding to the other end of the modulation pattern in the second direction. The laser processing device according to any one of claims 1 to 4.

6. the control unit changes the gradation value in the first signal so as to have a plurality of periods in at least one direction; The laser processing device according to any one of claims 1 to 5.

Images