JP2023013750A - Laser processing device and control method of the same - Google Patents

Abstract

To provide a laser processing device capable of maintaining a distance of a light focusing point from a front surface side of a workpiece to a constant distance with high accuracy, and provide a control method of the same.SOLUTION: A laser processing device comprises: a first laser light source that emits a processing laser light; a second laser light source that emits a detection laser light; a first light focusing lens that focuses a processing laser light into an inner part of a workpiece, and focuses a detection laser light onto a back surface of the workpiece; a first light detection part that detects a reflection light of the detection laser light from the back surface of the workpiece; a height position detection part that detects a height position of the back surface of the workpiece on the basis of a detection result of the first light detection part; a profile acquisition part that acquires a thickness profile from a cracking detection device; and a light focusing point position adjustment part that adjusts a position of a first light focusing point along a light axial direction of the first light focusing lens on the basis of the detection result of the height position detection part and the thickness profile, and maintains the position of the first light focusing point in a distance that is constant from the front surface of the workpiece.SELECTED DRAWING: Figure 16

Description

The present invention relates to a laser processing apparatus for laser processing a workpiece and a control method thereof.

Conventionally, a silicon wafer (hereinafter abbreviated as wafer), which is a workpiece, is irradiated with a laser beam along a processing line (also referred to as a line to be cut) with a focal point aligned with the inside, and the wafer is cut along the processing line. A laser processing apparatus is known that forms a laser processing region (also referred to as a “modified region” or a “modified layer”) that serves as a starting point for cutting inside. The wafer with the laser-processed regions is then split along the process lines into individual chips by a splitting process such as expanding or breaking. According to such a laser processing apparatus, a laser processing region is formed inside the wafer, and the wafer is cleaved along the processing line starting from the laser processing region. Compared to a general dicing machine that uses a dicing machine, the amount of dust generated is low, and the possibility of dicing scratches, chipping, or cracks on the material surface is low.

In recent years, in the field of semiconductor manufacturing, there is a tendency for wafers to become larger, and wafers are being made thinner in order to further increase packaging density. For such a wafer, the back surface of the wafer is ground (back-grinded) by a back surface grinder to thin the wafer, and then laser processing is performed to form a laser processing region on the wafer by a laser processing device.

In laser processing by such a laser processing apparatus, in order to suppress thermal damage to the device layer formed on the front surface (device formation surface) of the wafer, the laser processing area from the front surface (device formation surface) side of the wafer is It is necessary to maintain the forming position (the position of the focal point of the laser beam) at a constant distance. For this reason, in the laser processing apparatus described in Patent Document 1, a condensing lens for condensing laser light inside the wafer using an autofocus function is placed on the rear surface of the wafer (the surface opposite to the device formation surface). The position is adjusted in the vertical direction so as to follow the unevenness of the surface. At this time, the thickness of the wafer whose back surface is back-grinded by the back-grinding device varies in thickness. Therefore, even if the autofocus function is used to keep the distance between the back surface of the wafer and the focal point of the laser beam focused inside the wafer at a constant distance, the distance between the front surface of the wafer and the focal point is large. changes according to variations in the thickness of the wafer.

Therefore, in the laser processing apparatus described in Patent Document 1, the height profile of the back surface of the wafer along at least one scan line and the height profile of the top surface of a stage (table) that holds the front surface side of the wafer Based on this, a thickness profile is obtained for each processing line of the wafer, which indicates the change in the thickness of the wafer along the processing line. Then, in this laser processing apparatus, the position of the laser processing region (focusing point) from the front surface side of the wafer is maintained at a constant distance based on the thickness profile.

JP 2020-142284 A

By the way, in the laser processing apparatus described in Patent Document 1, the thickness profile is acquired based on the height profile of the back surface of the wafer and the height profile of the top surface of the stage. At this time, since the surface side of the wafer is generally covered with a protective sheet, the protective sheet is sandwiched between the upper surface of the stage and the surface of the wafer. Further, unevenness due to the device layer is formed on the surface of the wafer. For this reason, the method described in Patent Document 1 cannot accurately determine the thickness profile of the wafer. As a result, in the laser processing apparatus described in Patent Document 1, it is difficult to maintain the position of the laser processing area (focusing point) from the front surface side of the wafer at a constant distance with high accuracy.

SUMMARY OF THE INVENTION The present invention has been made in view of such circumstances, and provides a laser processing apparatus and a control method thereof that can maintain the distance of the focal point from the surface side of the workpiece at a constant distance with higher accuracy. for the purpose.

In order to achieve the object of the present invention, a laser processing apparatus is provided in which a processing laser beam is emitted from the back side of the workpiece while a first focal point of the processing laser beam is aligned with the inside of the workpiece. A laser processing apparatus comprising a processing head that irradiates along a processing line of and forms a laser processing area inside the workpiece, and a crack detection device that detects cracks extending from the laser processing area in the workpiece. In, the crack detection device includes an interface position detection unit that detects the interface positions indicating the front and back surfaces of the workpiece at a plurality of predetermined measurement points, and along the processing line based on the detection results of the interface position detection unit a thickness detection unit for detecting a thickness profile indicating a change in the thickness of the workpiece, and the processing head includes a first laser light source that emits a processing laser beam and a second laser that emits a detection laser beam. a laser light source unit including a light source; a processing laser beam emitted from the first laser light source to be focused inside the workpiece; and a detection laser beam emitted from the second laser light source to the back surface of the workpiece a first condenser lens that converges light on the workpiece, a first photodetector that detects the reflected light of the detection laser beam from the back surface of the workpiece, and a workpiece based on the detection result of the first photodetector Based on the height position detection unit that detects the height position of the back surface, the profile acquisition unit that acquires the thickness profile from the thickness detection unit, the detection result of the height position detection unit, and the thickness profile acquired by the profile acquisition unit , adjusting the position of the first condensing point along the optical axis direction of the first condensing lens to maintain the position of the first condensing point at a predetermined constant distance from the surface of the workpiece; and a point position adjuster.

According to this laser processing apparatus, the thickness profile of the workpiece can be detected more accurately than before by the crack detection device, and based on this thickness profile and the detection result of the height position of the back surface of the workpiece, The position of the first focal point can be maintained at a constant distance from the surface of the workpiece.

In the laser processing apparatus according to another aspect of the present invention, the first condensing lens is relatively moved in a direction perpendicular to the optical axis direction of the first condensing lens with respect to the workpiece, and the first condensing point along the processing line, each time the first focal point is moved along the processing line by the first movement mechanism, the first light detection unit, the height position detection unit, and The condensing point position adjustment unit operates repeatedly. Thereby, while laser processing is performed along the processing line, the position of the first focal point can be maintained at a constant distance from the surface of the workpiece.

In the laser processing apparatus according to another aspect of the present invention, the crack detection device includes a light source that emits detection light, a second condenser lens that collects the detection light emitted from the light source on the workpiece, and a second condenser. a condensing point scanning unit for scanning the second condensing point of the detection light along the optical axis direction of the optical lens; and detection from the workpiece while the second condensing point is scanned by the condensing point scanning unit a second light detection unit that detects reflected light of light; and a second moving mechanism that relatively moves the second condenser lens with respect to the workpiece in a direction perpendicular to the optical axis direction of the second condenser lens. , each time the second light-condensing point is sequentially moved to a plurality of measurement points by the second moving mechanism, the light-condensing point scanning unit and the second light detection unit repeatedly operate, and the interface position detection unit The interface position is detected for each of the plurality of measurement points based on the detection results of the second photodetector for each of the plurality of measurement points. As a result, the interface position of the workpiece can be detected with high accuracy, so that the thickness profile can be detected with higher accuracy.

In the laser processing apparatus according to another aspect of the present invention, the interface position detection unit detects the interface position for each of a plurality of measurement points along a scan line passing through the center of the workpiece whose back surface has been ground by the grinding device. . The in-plane thickness distribution of the workpiece can be determined (estimated) based on the results of detecting the thickness of the workpiece at multiple measurement points along the scan line, so the thickness profile can be detected in a short period of time. can do.

In a laser processing apparatus according to another aspect of the present invention, a crack detection device is provided on the processing head.

A control method for a laser processing apparatus for achieving the object of the present invention is a method of controlling a laser beam for processing on the back side of a workpiece with a first focusing lens aligned with the inside of the workpiece. A processing head that forms a laser processing area inside the workpiece and a crack extending from the laser processing area inside the workpiece are detected. and a crack detection device, in which the crack detection device detects the interface position indicating the front surface and the back surface of the workpiece at a plurality of predetermined measurement points. Based on the detection results, the processing line is A thickness profile indicating a change in the thickness of the workpiece along is detected, and the processing head emits a processing laser beam from a first laser light source, a detection laser beam from a second laser light source, and a first The condensing lens converges the processing laser beam emitted from the first laser light source inside the workpiece and converges the detection laser beam emitted from the second laser light source on the back surface of the workpiece. The reflected light of the detection laser beam from the back surface of the workpiece is detected, the height position of the back surface of the workpiece is detected based on the detection result of the reflected light of the detection laser beam, and the thickness profile is obtained from the crack detection device. Based on the height position detection result and the thickness profile, the position of the first condensing point is adjusted along the optical axis direction of the first condensing lens, and the position of the first condensing point is processed. Maintain a predetermined constant distance from the surface side of the object.

According to the present invention, the distance of the focal point from the front surface side of the wafer can be maintained at a constant distance with higher accuracy.

1 is a block diagram showing the configuration of a laser processing system; FIG. It is a block diagram showing a schematic configuration of a laser processing apparatus. It is a block diagram showing an example of a crack detection device mounted in a laser processing device. FIG. 4 is an explanatory diagram showing a state when polarized illumination of detection light is performed on a wafer; FIG. 4 is an explanatory diagram showing a state when polarized illumination of detection light is performed on a wafer; FIG. 4 is an explanatory diagram showing a state when polarized illumination of detection light is performed on a wafer; It is the figure which showed the state of the reflected light received by the light-receiving surface of a photodetector. It is the figure which showed the state of the reflected light received by the light-receiving surface of a photodetector. It is the figure which showed the state of the reflected light received by the light-receiving surface of a photodetector. FIG. 4 is a diagram for explaining a path along which reflected light from a wafer reaches a condenser lens pupil; It is a functional block diagram of the control part of a crack detection apparatus. FIG. 5 is an explanatory diagram for explaining scanning in the Z direction of a focal point of detection light by a drive control unit; 3 is a graph showing the relationship between the amount of movement (μm) of the focal point by the piezo actuator and the detection signal (V) output from the detector body of the photodetector. It is an explanatory view for explaining arrangement of a measurement point on a wafer. 15 is a cross-sectional view of the wafer along the scan line in FIG. 14; FIG. It is a functional block diagram of the control part of a laser processing apparatus. FIG. 4 is an explanatory diagram showing a line scheduled for forming a laser-processed region within a wafer; FIG. 10 is an explanatory diagram for explaining position adjustment of the condensing point in the Z direction by the condensing point position adjusting unit; 5 is a flow chart showing a flow of detection processing of a thickness profile for each wafer processing line by a crack detection device. It is a flowchart which shows the flow of the laser processing process for every processing line by a laser processing apparatus. FIG. 10 is an explanatory view showing a planned line for forming a laser-processed region in a wafer having a non-flat wafer surface; FIG. 21 is an explanatory diagram for explaining the position adjustment of the condensing point in the Z direction by the condensing point position adjusting unit when the laser processing of the wafer shown in FIG. 20 is performed;

[Laser processing system]
FIG. 1 is a block diagram showing the configuration of a laser processing system 100. As shown in FIG.

As shown in FIG. 1, the laser processing system 100 divides a wafer W, which is an object to be processed, along processing lines DL through a plurality of steps to form a plurality of chips (not shown). This laser processing system 100 includes a back surface grinding device 200 , a laser processing device 10 and a dividing device 300 .

In the laser processing system 100, first, a wafer is transported to the back surface grinding device 200, and the back surface of the wafer W is ground in the back surface grinding device 200 (back grinding process). The wafer W whose back surface has been back ground by the back grinding apparatus 200 is thinned to a thickness of 250 μm or less, for example. When performing back grinding, a protective sheet (also referred to as a back grinding tape) is attached in advance to the front surface of the wafer W, and the device (device layer) formed on the front surface of the wafer W is protected by this protective sheet. protected. Here, in this specification, of the two sides of the wafer, the side on which devices are formed is referred to as the "front side" of the wafer, and the side opposite to the front side is referred to as the "back side."

The wafer W that has undergone back grinding by the back grinding device 200 is carried into the laser processing device 10, and in the laser processing device 10, a laser processing region is formed inside the wafer W by a laser beam irradiated from the back side of the wafer W. It is formed along the processing line DL.

A wafer having laser-processed regions formed on all processing lines DL is mounted (mounted) on a dicing frame via a dicing tape on its back surface. In this state, the protective sheet attached to the surface of the wafer is peeled off.

Next, the wafer is loaded into the dividing device 300 along with the dicing frame, and the dicing tape is radially stretched in the dividing device 300 to divide the wafer into individual chips. The flow of wafer processing by the laser processing system 100 has been described above.

Since the back surface grinding device 200 and the dividing device 300 that constitute the laser processing system 100 are well-known technologies, a detailed description thereof will be omitted here.

[Laser processing equipment]
FIG. 2 is a block diagram showing a schematic configuration of the laser processing apparatus 10. As shown in FIG. The XYZ directions in the drawing are orthogonal to each other, the XY directions are horizontal directions, and the Z directions are vertical directions.

As shown in FIG. 2 , the laser processing apparatus 10 forms a laser processing region R inside the wafer W in which the wafer rear surface Wb, which is the rear surface of the wafer W, is back ground by the rear surface grinding apparatus 200 . This laser processing apparatus 10 includes a processing head 16 , a suction stage 18 , a processing head driving section 19 , a table driving section 14 and a control section 12 . Furthermore, the laser processing apparatus 10 includes a crack detection device 500 attached to one side of the processing head 16 . The configuration of crack detection device 500 will be described later.

The suction stage 18 has a holding surface 18a that holds the wafer W by suction. The wafer front surface Wa, which is the surface of the wafer W, is partitioned into grid-like regions by, for example, a plurality of processing lines DL extending in the XY directions. (film)] is formed. The wafer W is placed with the protective sheet 220 covering the holding surface 18 a of the suction stage 18 facing downward, and is held by the suction stage 18 via the protective sheet 220 . Note that the wafer W may be directly sucked and held without the protection sheet 220 interposed therebetween.

The table driving unit 14 corresponds to the first moving mechanism and the second moving mechanism of the present invention, and includes an XY driving mechanism (not shown) for moving the suction stage 18 in the XY directions and a Z-axis movement of the suction stage 18 . and a rotation driving mechanism (not shown) for rotating in the θ direction around the direction. Under the control of the control unit 12, the table drive unit 14 moves the suction stage 18 in the XY directions by the XY drive mechanism and rotates the suction stage 18 in the θ direction around the Z direction by the rotation drive mechanism. As a result, it is possible to perform relative alignment between the wafer W sucked and held by the suction stage 18 and the processing head 16, which will be described later. It becomes possible to perform laser processing by the head 16 .

The processing head 16 is a processing optical system main body (also called a “laser engine” or the like) that accommodates various optical systems for performing laser processing. The processing head 16 includes a laser light source 20 , a dichroic mirror 21 , a condenser lens 22 (also referred to as an objective lens), a piezo actuator 23 and an AF (Automatic Focus) sensor 24 . A laser light source 20, a dichroic mirror 21, a condenser lens 22, and a piezo actuator 23 are provided from top to bottom in the Z direction.

The laser light source 20 corresponds to the first laser light source of the present invention, and constitutes the laser light source section of the present invention together with a laser light source 25 which will be described later. The laser light source 20 emits a processing laser beam LP used for forming a laser processing region R inside the wafer W. As shown in FIG. As the laser light source 20, for example, a semiconductor laser-pumped Nd:YAG (Yttrium Aluminum Garnet) laser is used. The conditions of the processing laser beam LP are, for example, a wavelength of 1.1 μm, a laser beam spot cross-sectional area of 3.14×10 ⁻⁸ cm ² , an oscillation mode of Q-switch pulse, a repetition frequency of 80 to 120 kHz, The pulse width is 180-280ns and the output is 8W.

The dichroic mirror 21 transmits the processing laser beam LP and reflects the detection laser beam LD emitted from the AF sensor 24, which will be described later. As a result, the dichroic mirror 21 transmits the processing laser beam LP incident from the laser light source 20 and emits the processing laser beam LP toward the condenser lens 22 . Also, the dichroic mirror 21 emits the detection laser light LD incident from the AF sensor 24 toward the condenser lens 22 .

The condensing lens 22 corresponds to the first condensing lens of the present invention, and for example, a condenser lens or the like is used. The condensing lens 22 converges the processing laser light LP incident from the dichroic mirror 21 inside the wafer W. As shown in FIG. Thereby, a laser processing region R is formed inside the wafer W. As shown in FIG. Here, the laser processing region R is a region where the physical properties such as density, refractive index, mechanical strength, etc. inside the wafer W become different from the surroundings due to the irradiation of the laser beam L, and the strength is lower than the surroundings. Say. Laser-processed region R includes, for example, a crack region.

Also, the condenser lens 22 converges the detection laser beam LD incident from the dichroic mirror 21 onto the wafer rear surface Wb. Thereby, the condenser lens 22 can simultaneously collect the processing laser beam LP to the inside of the wafer W and the detection laser beam LD to the wafer rear surface Wb.

The piezoelectric actuator 23 expands and contracts in the Z direction (the direction of the optical axis of the condenser lens 22) as the voltage applied from the controller 12 changes, and this expansion and contraction moves the condenser lens 22 in the direction of the optical axis. As a result, the position of the condenser lens 22 in the Z direction can be precisely adjusted, and the condensing point FP1 (corresponding to the first condensing point of the present invention) of the processing laser beam LP condensed inside the wafer W. can be precisely adjusted (finely adjusted) in the Z direction.

The relationship between the amount of movement (extension amount: μm) of the piezo actuator 23 and the voltage applied to the piezo actuator 23 is determined in advance by experiments or the like. The control unit 12 stores a table defining the relationship, and changes the voltage applied to the piezo actuator 23 according to this relationship.

The AF sensor 24 is used to detect the Z-direction height position of the wafer back surface Wb along the processing line DL. This AF sensor 24 comprises a laser light source 25 and a photodetector 26 .

The laser light source 25 corresponds to the second laser light source of the present invention, and emits detection laser light in a wavelength range different from the wavelength range of the processing laser light LP and in a wavelength range that can be reflected by the wafer back surface Wb. The LD is emitted toward the dichroic mirror 21 . The detection laser beam LD is reflected by the dichroic mirror 21 toward the condenser lens 22 and then condensed by the condenser lens 22 onto the back surface Wb of the wafer. Then, the reflected light LR of the detection laser light LD reflected by the wafer back surface Wb travels along the optical path of the detection laser light LD and enters the photodetector 26 of the AF sensor 24 . The symbol RP in the drawing indicates the reflection point of the detection laser beam LD reflected by the wafer rear surface Wb.

The photodetector 26 corresponds to the first photodetector of the present invention, and has a light receiving surface for receiving the reflected light LR. A detection signal corresponding to the position is output to the control unit 12 .

Here, the distribution and light amount of the reflected light LR reflected by the wafer back surface Wb change according to the distance from the condenser lens 22 to the wafer W, that is, the uneven shape (surface displacement) of the wafer back surface Wb. Using this property, the AF sensor 24 detects the laser processing position where the condensing point FP1 is formed within the surface of the wafer W based on the distribution of the reflected light LR reflected by the wafer back surface Wb and the change in the amount of light. Find the height position in the Z direction at (XY position). As a method for measuring the height position of the wafer rear surface Wb in the AF sensor 24, for example, an astigmatism method, a knife edge method, or the like is used. Since these methods are publicly known, detailed description thereof is omitted here.

Measurement of the height position of the wafer rear surface Wb by the AF sensor 24 described above is continuously performed on the wafer W processing line DL. As a result, when the laser processing region R is formed inside the wafer W along the processing line DL, the Z-direction position of the converging point FP1 of the processing laser light LP is adjusted in real time based on the measurement result of the AF sensor 24. control becomes possible.

In this embodiment, the AF sensor 24 is provided in the processing head 16, and the condensing lens 22 simultaneously condenses the processing laser beam LP and the detection laser beam LD. However, it is not limited to this configuration. For example, the AF sensor 24 may be provided independently of and adjacent to the processing head 16 .

The machining head drive unit 19 includes a machining head Z drive mechanism (not shown) for moving the machining head 16 in the Z direction. The machining head Z drive mechanism is, for example, an actuator including a motor and gears. The processing head drive unit 19 moves the processing head 16 in the Z direction by the processing head Z drive mechanism under the control of the control unit 12 . Thereby, the relative distance in the Z direction between the processing head 16 and the wafer W suction-held by the suction stage 18 can be roughly adjusted.

The control unit 12 is implemented by, for example, a personal computer or workstation. The control unit 12 includes a CPU (Central Processing Unit), memory (eg, ROM (Read Only Memory), RAM (Random Access Memory), etc.), storage (eg, HDD (Hard Disk Drive), SSD (Solid State Drive ), etc.) and an input/output circuit unit.

The control unit 12 controls the operation of each unit of the laser processing device 10 . Specifically, operation control of the processing head 16 (emission control of the laser light sources 20 and 25, light reception control of the reflected light by the photodetector 26, position adjustment of the condenser lens 22 by the piezo actuator 23, etc.), processing head drive unit 19 and the table drive unit 14.

When forming the laser processing region R on the wafer W, after adjusting the relative distance (working distance) between the processing head 16 and the wafer W by the processing head driving unit 19 to a desired distance, the above-mentioned AF sensor 24 Measurement of the height position of the wafer rear surface Wb and adjustment of the position of the condenser lens 22 in the Z direction are performed. Next, a processing laser beam LP is emitted from the laser light source 20 and irradiated inside the wafer W through the condenser lens 22 . At this time, although details will be described later, the Z-direction position of the converging point FP1 of the processing laser beam LP irradiated to the inside of the wafer W is adjusted by adjusting the Z-direction position of the condensing lens 22 by the piezo actuator 23. It is adjusted to maintain a constant distance from the wafer surface Wa.

In this state, the suction stage 18 is processed and fed by the table drive unit 14 in the X direction, which is the dicing direction, that is, in the direction perpendicular to the optical axis direction of the condenser lens 22 . As a result, one layer of the laser processed region R is formed along the processing line DL of the wafer W. As shown in FIG. Then, when one layer of the laser-processed region R is formed along the processing line DL, the suction stage 18 is indexed and fed by one pitch in the Y direction by the table drive unit 14, and the laser-processed region R is also formed on the next processing line DL. It is formed. Next, when the laser processing regions R are formed along all the processing lines DL in the X direction, the suction stage 18 is rotated by 90° in the θ direction around the Z direction by the table driving unit 14, A laser processing region R is similarly formed on the processing line DL. Although FIG. 2 shows one layer of the laser processing region R, depending on the thickness of the wafer W, a plurality of layers (for example, two layers, three layers, etc.) of the laser processing region R can be formed inside the wafer W. is sometimes formed.

[Crack detector]
FIG. 3 is a block diagram showing an example of a crack detection device 500 mounted on the laser processing device 10. As shown in FIG. As shown in FIG. 3, the crack detection device 500 has, as a basic function, not only a crack detection function for detecting cracks generated inside the wafer W by the irradiation of the processing laser beam LP, but also a processing laser beam It also has a thickness detection function for detecting the thickness of the wafer W before irradiation with the light LP (that is, after back grinding is performed by the back grinding device 200). Hereinafter, the thickness detection function of the crack detection device 500 will be described after the crack detection function, which is the basic function of the crack detection device 500, will be described. The crack detection device 500 is attached to one side surface of the processing head 16 and can be moved in the Z direction integrally with the processing head 16 by the processing head driving section 19 described above.

[Crack detection function]
When the laser processing region R is formed inside the wafer W by the laser processing apparatus 10, a crack K (crack) extends from the laser processing region R in the thickness direction of the wafer W. As shown in FIG. In order to properly divide the wafer W into chips in the dividing device 300, which is a post-process of the laser processing device 10, a laser processing region R serving as a starting point for cutting the wafer W is properly formed, and the wafer W It is important to accurately detect the crack depth of the crack K formed inside.

Therefore, the crack detection device 500 detects the depth of the crack K formed inside the wafer W by irradiating the wafer W with the detection light L1 and detecting the reflected light L2 from the wafer W. It has a function (crack detection function). In addition, although the crack detection device 500 is used in combination with the laser processing device 10, in the following description, the components related to the crack detection device 500 will be described, and the same members as the configuration of the laser processing device 10 will be the same. , and description thereof may be omitted. In the following description, a three-dimensional orthogonal coordinate system is used in which the chucking stage 18 on which the wafer W is chucked is a plane parallel to the XY plane, and the Z direction is the thickness (depth) direction of the wafer W. FIG.

The crack detection device 500 includes a light source unit 550, an illumination optical system 600, an interface position detection optical system 650, a crack detection optical system 700, a control unit 750, a condensing point position moving mechanism 752, a condensing lens 754, and an operation unit 756. , a display unit 758, and a communication IF 759 that is a communication interface (Interface: IF).

The light source unit 550 emits detection light L1. This detection light L1 is used for detection of the interface position of the wafer W and detection of the crack K formed inside the wafer W. As shown in FIG. Here, when the wafer W is made of silicon, it is desirable to use infrared light with a wavelength of 1,000 nm or more as the detection light L1.

The light source section 550 has light sources 552 A, 552 B, 552 C and a half mirror 554 . Light sources 552A, 552B, 552C and half mirror 554 are arranged along main optical axis AX coaxial with the optical axis of condenser lens 754 .

The light sources 552A, 552B, 552C emit detection light L1 along the main optical axis AX. As the light sources 552A, 552B, 552C, for example, laser light sources (infrared laser light sources, laser diodes) or LED (Light Emitting Diode) light sources can be used.

The light source 552A has a laser aperture capable of illuminating substantially the entire surface of the condensing lens pupil 754a of the condensing lens 754 . This light source 552A is used for detecting the interface position, which will be described later.

The light sources 552B and 552C each have a laser aperture capable of illuminating only a part of the condensing lens pupil 754a of the condensing lens 754 decentered from the main optical axis AX (the optical axis of the condensing lens 754). are doing. These light sources 552B and 552C are used for crack detection, which will be described later.

Half mirror 554 reflects detection light L1 emitted from light source 552A for interface position detection, and transmits detection light L1 emitted from light sources 552B and 552C for crack detection. Hereinafter, although illustration is omitted, the detection light beams L1 emitted from the light sources 552A, 552B, and 552C will be described as L1(A), L1(B), and L1(C), respectively.

The light sources 552A, 552B, and 552C are each connected to a control section 750, and the control section 750 controls emission of the light sources 552A, 552B, and 552C.

The control unit 750 includes a CPU that controls the operation of each unit of the crack detection device 500, a ROM that stores control programs, and an SDRAM (Synchronous Dynamic Random Access Memory) that can be used as a work area for the CPU. The control unit 750 is connected to the control unit 12 of the laser processing apparatus 10 via an interface (not shown), receives an operation input by the operator via an operation unit 756, and detects a crack by detecting a control signal corresponding to the operation input. It transmits to each unit of the device 500 to control the operation of each unit.

In this embodiment, the controller 750 of the crack detection device 500 is shown separately from the controller 12 (see FIG. 3) of the laser processing device 10, but the processes executed by these controllers 750 and 12 are , may be executed by one control unit, or may be executed by a plurality of control units.

The operation unit 756 is means for receiving an operation input by an operator, and is, for example, a keyboard, a mouse, or a pointing device such as a touch panel.

The display unit 758 is a device that displays an operation GUI (Graphical User Interface) for operating the crack detection device 500 and images (for example, crack detection results, etc.). As the display unit 758, for example, a liquid crystal display can be used.

Note that the operation unit 756 and the display unit 758 may be shared with the operation unit and the display unit (both not shown) provided in the laser processing apparatus 10 .

The communication IF 759 is wired or wirelessly connected to the communication IF 28 (see FIG. 16) of the laser processing apparatus 10, which will be described later. Although the details will be described later, the communication IF 759 transmits the thickness profile 42 (see FIG. 11), which is the thickness detection result of the wafer W detected by the thickness detection function described later, under the control of the control unit 750 to the communication IF 28, that is, laser processing. output to device 10;

The illumination optical system 600 guides the detection light L1 emitted from the light source section 550 to the condenser lens 754 . The illumination optical system 600 has relay lenses 602 and 606 and a mirror (for example, a total reflection mirror) 604 . The detection light L1 emitted from the light source unit 550 is transmitted through the relay lens 602, reflected by the mirror 604, and the optical path is bent. The detection light L 1 reflected by the mirror 604 is transmitted through the relay lens 606 , then sequentially reflected by the half mirrors 654 and 652 of the interface position detection optical system 650 and emitted toward the condenser lens 754 . Observation light transmitted through the half mirror 652 (return light from the wafer W) can be observed using the observation optical system 760 . If the observation optical system 760 is not used, a dichroic mirror or a total reflection mirror can be used in place of the half mirror 652. FIG.

The condenser lens 754 corresponds to the second condenser lens of the present invention, and condenses (focuses) the detection light L1 emitted from the illumination optical system 600 onto the wafer W. FIG. The condenser lens 754 is arranged at a position facing the back surface Wb of the wafer, and the optical axis of the condenser lens 754 is arranged coaxially with the main optical axis AX.

The focal point position moving mechanism 752 shifts the position of the focal point FP2 (corresponding to the second focal point of the present invention, see FIG. 12) of the detection light L1 with respect to the wafer W in the Z direction (the optical axis direction of the condenser lens 754). ). The focusing point position moving mechanism 752 includes a piezo actuator 752a that moves the focusing lens 754 in the Z direction.

The piezo actuator 752a expands and contracts in the Z direction (optical axis direction of the condenser lens 754) when the voltage applied from the control unit 750 changes, and this expansion and contraction movement moves the condenser lens 754 in the optical axis direction. As a result, the position of the focal point FP2 (see FIG. 12) of the detection light L1 with respect to the wafer W is changed in the thickness direction of the wafer W (Z direction).

Also, the relationship between the amount of movement (extension amount: μm) of the piezo actuator 752a and the voltage applied to the piezo actuator 752a is determined in advance by experiments or the like. The control unit 750 stores a table that defines the relationship, and changes the voltage applied to the piezo actuator 752a according to this relationship.

The focusing point position moving mechanism 752 moves the focusing lens 754 in the Z direction by driving the piezo actuator 752a under the control of the controller 750 . As a result, the relative distance in the Z direction between the condenser lens 754 and the wafer W changes, and the position of the condensing point FP2 of the detection light L1 with respect to the wafer W changes. can be scanned in the thickness direction (Z direction).

In addition to the piezo actuator 752a, the focal point position moving mechanism 752 may include the machining head Z driving mechanism of the machining head driving section 19 described above. By moving the machining head 16 in the Z direction with the machining head Z drive mechanism, the crack detection device 500 can be moved in the Z direction together with the machining head 16 . This changes the relative distance between the condenser lens 754 and the wafer W, so that the position of the focal point FP2 (see FIG. 12) of the detection light L1 with respect to the wafer W can be changed. Therefore, by combining the position adjustment (coarse adjustment) of the focal point by the processing head Z driving mechanism and the position adjustment (fine adjustment) of the focal point by the piezo actuator 752a, the detection light L1 can be focused only by the piezo actuator 752a. Compared to scanning the light spot FP2 in the thickness direction of the wafer W, the degree of freedom (adjustment range) for adjusting the position of the light spot FP2 of the detection light L1 with respect to the wafer W is increased. Also, crack detection and thickness detection are possible.

The reflected light L2 of the detection light L1 that is condensed by the condensing lens 754 and reflected by the wafer W is guided to the interface position detection optical system 650 and the crack detection optical system 700, and the interface position of the wafer W is detected. It is used for detection of and crack detection.

In the following example, the interface position of the wafer front surface Wa is detected, and then the crack depth is detected with reference to the interface position of the wafer front surface Wa.

In this embodiment, the crack depth is detected using the wafer front surface Wa as a reference, but the present invention is not limited to this. For example, the crack depth is detected using the wafer back surface Wb as a reference. good too. It is also possible to take an average value of the crack depths detected with reference to the interface positions of both the wafer front surface Wa and the wafer back surface Wb.

<Optical system for interface position detection>
The interface position detection optical system 650 is an optical system for detecting the interface of the wafer W (wafer front surface Wa or wafer back surface Wb), and includes a half mirror 652, a half mirror 654, a relay lens 656, a half mirror 658, and a It has a photodetector 660 .

When detecting, for example, the wafer front surface Wa as the interface of the wafer W, the controller 750 causes the light source 552A to emit light to irradiate the wafer W with the detection light L1(A).

The detection light L1(A) from the light source 552A is a laser beam having an opening approximately the same size as the condenser lens pupil 754a of the condenser lens 754, and is sequentially reflected by the half mirrors 654 and 652 to be condensed. The light is guided to lens 754 . The detection light L1(A) irradiates substantially the entire surface of the condensing lens pupil 754a of the condensing lens 754 .

Here, the reflected light of the detection light L1(A) reflected by the wafer W is assumed to be L2(A). Reflected light L2(A) passes through condenser lens 754 , is reflected by half mirror 652 , passes through half mirror 654 , and is guided to relay lens 656 . Reflected light L2(A) transmitted through relay lens 656 is reflected by half mirror 658 and guided to photodetector 660 .

The photodetector 660 is a device for receiving the reflected light L2(A) from the wafer W to detect the wafer surface Wa of the wafer W, and has a detector body 660A and a pinhole panel 660B. there is

As the detector main body 660A, a photodetector (for example, a photodiode) that converts received light into an electrical signal and outputs the electrical signal to the control section 750 can be used.

A pinhole is formed in the pinhole panel 660B for transmitting part of the incident light. The pinhole panel 660B is arranged upstream with respect to the light receiving surface of the detector main body 660A, and is arranged so that the pinhole of the pinhole panel 660B is positioned on the optical axis of the reflected light L2(A). there is The position of the pinhole of the pinhole panel 660B is optically conjugate with the condensing point (front focal position) of the condensing lens 754 (confocal pinhole). Also, the size of the pinholes of the pinhole panel 660B is adjusted to about the diffraction limit of the condenser lens 754 .

The reflected light L2(A) reflected by the wafer W is condensed at the position of the pinhole of the pinhole panel 660B which is optically conjugate with the condensing point of the condensing lens 754 . Then, when the condensing point of the condensing lens 754 coincides with the wafer surface Wa of the wafer W serving as the reflecting surface, the light flux of the detection light L1 (A) is reflected by the wafer surface Wa of the wafer W and becomes a parallel light flux. The light passes through the condensing lens 754 and returns. Therefore, the detection signal output from the detector main body 660A has a sharp peak when the condensing point of the condensing lens 754 coincides with the position of the wafer surface Wa serving as the reflecting surface. The detection signal output from the detector main body 660A has a sharp peak even when the condensing point of the condensing lens 754 coincides with the position of the back surface Wb of the wafer serving as the reflecting surface.

The controller 750 changes the relative distance between the condenser lens 754 and the wafer W using the condenser position moving mechanism 752 while irradiating the wafer W with the detection light L1(A) from the light source 552A. The position of the condensing point of the detection light L1(A) (that is, the front focal position of the condensing lens 754) is moved in the Z direction. Thereby, the focal point of the detection light L1(A) is scanned in the thickness direction of the wafer W. FIG. The controller 750 detects the reflected light L2 (A) from the wafer W when the focal point of the detection light L1 (A) is scanned in the thickness direction (Z direction) of the wafer W by the photodetector 660, By detecting the peak of the detection signal from this photodetector 660, the interface position of the wafer surface Wa in the Z direction can be detected.

In the present embodiment, the confocal method is used to detect the interface position of the wafer W. However, the present invention is not limited to this method. It is also possible to apply focus detection methods.

<Optical system for crack detection>
The crack detection optical system 700 has a relay lens 702 and photodetectors 704 and 706 .

When detecting the crack K formed inside the wafer W, the control unit 750 causes the light sources 552B and 552C to emit light and irradiates the wafer W with the detection lights L1(B) and L1(C). The light sources 552B and 552C each have a laser aperture at a position shifted from the main optical axis AX. As a result, the wafer W is irradiated with detection lights L1(B) and L1(C) that are eccentric with respect to the main optical axis AX.

The reflected lights L2(B) and L2(C) of the detection lights L1(B) and L1(C) respectively reflected by the wafer W pass through the condenser lens 754 and are sequentially reflected by the half mirrors 652 and 654. After passing through the relay lens 656 and the half mirror 658 , the light enters the relay lens 702 . Reflected lights L2(B) and L2(C) transmitted through relay lens 702 are received by photodetectors 704 and 706, respectively.

The photodetectors 704 and 706 are devices for receiving reflected light L2(B) and L2(C) from the wafer W to detect cracks K inside the wafer W. FIG. As the photodetectors 704 and 706, photodetectors (for example, photodiodes) that convert received light into electrical signals and output the electrical signals to the control unit 750 can be used.

The photodetectors 704 and 706 are positioned conjugate with the condenser lens pupil 754a, and are positioned offset from the optical axis of the condenser lens 754 to receive the detected light beams L1(B) and L1(C). ing.

FIGS. 4 to 6 are explanatory diagrams showing the state when the wafer W is illuminated with the detection light L1. 4 shows the case where the crack K exists at the condensing point of the condensing lens 754, FIG. 5 shows the case where the crack K does not exist at the condensing point of the condensing lens 754, and FIG. Each of these shows a case where the crack depth (crack bottom position) of the crack K matches.

7 to 9 are diagrams showing the reflected light L2 received by the light receiving surfaces 704C and 706C of the photodetectors 704 and 706, corresponding to the cases shown in FIGS. 4 to 6, respectively. It is.

Also, FIG. 10 is a diagram for explaining the path along which the reflected light L2 from the wafer W reaches the condenser lens pupil 754a. Here, the case where the detection light L1 passes through the first region G1 on one side (the right side in FIG. 10) of the condenser lens pupil 754a and the wafer W is subjected to polarized illumination will be described.

As shown in FIG. 4, when there is a crack K at the condensing point of the condenser lens 754, the detection light L1 is totally reflected by the crack K, then reflected by the wafer surface Wa, and the reflected light L2 is It is a component that follows the path on the same side as the optical path of the detection light L1 with respect to the main optical axis AX and reaches the area on the same side as the detection light L1 of the condenser lens pupil 754a. That is, as shown in FIG. 10, when the detection light L1 from the light source unit 550 irradiates the wafer W through the condenser lens 754 and the path of the detection light L1 is R1, the crack inside the wafer W is The reflected light L2 totally reflected by K follows the path R2 on the same side (the right side in FIG. 10) with respect to the main optical axis AX as the path R1 of the detection light L1, and passes through the first region G1 of the condenser lens pupil 754a. pass.

As shown in FIG. 5, when there is no crack K at the condensing point of the condensing lens 754, the detection light L1 is reflected by the wafer surface Wa, and the reflected light L2 becomes the detection light L1 of the condensing lens pupil 754a. is the component that reaches the region on the opposite side. That is, as shown in FIG. 10, the reflected light L2 reflected by the wafer surface Wa follows a path R3 on the opposite side (left side in FIG. 10) with respect to the main optical axis AX to the path R1 of the detection light L1 and converges. It passes through the second region G2 of the optical lens pupil 754a.

As shown in FIG. 6, when the condensing point of the condensing lens 754 and the lower end position of the crack K match, the detection light L1 is totally reflected by the crack K, then reflected by the wafer surface Wa, The reflected light component L2a reaching the area on the same side as the detection light L1 of the condenser lens pupil 754a, and the detection light L1 of the condenser lens pupil 754a reflected by the wafer surface Wa of the wafer W without being totally reflected by the crack K. and a non-reflected light component L2b reaching the area on the opposite side. That is, as shown in FIG. 10, of the reflected light L2, the reflected light component L2a (see FIG. 6) totally reflected by the crack K inside the wafer W follows the path R2 to the first light of the condenser lens pupil 754a. The non-reflected light component L2b (see FIG. 6) that passes through the region G1 and is reflected by the wafer surface Wa of the wafer W without being totally reflected by the crack K follows the route R3 to reach the second region of the condenser lens pupil 754a. Pass through G2.

The photodetectors 704 and 706 are arranged so as to be optically conjugate with the first region G1 and the second region G2 of the condenser lens pupil 754a, respectively. As a result, the photodetectors 704 and 706 can selectively receive light that has passed through the first region G1 and the second region G2 of the condenser lens pupil 754a.

Here, in the example shown in FIG. 4 (when there is a crack K at the condensing point of the condensing lens 754), as shown in FIG. Reflected light L2 is incident on 704C. Therefore, the level of the detection signal output from the light receiving surface 704C of the photodetector 704 is higher than the level of the detection signal output from the light receiving surface 706C of the photodetector 706. FIG.

On the other hand, in the example shown in FIG. 5 (when there is no crack K at the condensing point of the condensing lens 754), as shown in FIG. The reflected light is incident on the Therefore, the level of the detection signal output from the light receiving surface 706C of the photodetector 706 is higher than the level of the detection signal output from the light receiving surface 704C of the photodetector 704. FIG.

Further, in the example shown in FIG. 6 (when the condensing point of the condensing lens 754 coincides with the lower end position of the crack K), as shown in FIG. Since the respective components L2a and L2b of the reflected light L2 are incident, the levels of the detection signals output from the light receiving surfaces 704C and 706C of the photodetectors 704 and 706 are approximately equal.

Thus, the amount of light received by the light receiving surfaces 704C and 706C of the photodetectors 704 and 706 changes depending on whether or not there is a crack K at the condensing point of the condensing lens 754. FIG. In this embodiment, using such properties, the crack depth (crack bottom end position or crack top end position) of the crack K formed inside the wafer W is detected.

Specifically, when the output of the detection signals output from the light receiving surfaces 704C and 706C of the photodetectors 704 and 706 are respectively D1 and D2, the existence of the crack K at the condensing point of the condensing lens 754 is determined. The evaluation value S for can be expressed by the following equation.

S=(D1-D2)/(D1+D2) (1)
In equation (1), when the condition of S=0 is satisfied, that is, when the amounts of light received by the light receiving surfaces 704C and 706C of the photodetectors 704 and 706 match each other, the condensing point of the condensing lens 754 and the bottom end of the crack It shows a state in which the position (or crack top position) matches.

The controller 750 controls the focal point position moving mechanism 752 to move the position of the focal point FP2 (see FIG. 12) of the detection light L1 in the Z direction (optical axis direction of the condenser lens 754). Further, the controller 750 sequentially changes the position of the focal point FP2 from the interface position of the wafer front surface Wa in the thickness direction (Z direction) of the wafer W, while The output detection signals are acquired sequentially. Then, the control unit 750 calculates the evaluation value S represented by the formula (1) based on this detection signal, and evaluates this evaluation value S and the condensing point position information to evaluate the crack depth of the crack K (crack bottom position or crack top position) can be detected. The above is the crack detection function of the crack detection device 500 .

[Thickness detection function and external output function of crack detection device]
Next, the function of detecting the thickness of the wafer W by the crack detection device 500 and the function of outputting the thickness detection result (thickness profile 42, see FIG. 11) of the wafer W to the back grinding device 200 will be described.

As described above, the crack detection device 500 has a thickness detection function and an external output function in addition to the crack detection function. The thickness detection function of the crack detection device 500 utilizes each component of the crack detection device 500 after the back grinding is performed by the back grinding device 200 and before the laser processing region R is formed by the laser processing device 10. , the thickness of the wafer W (especially silicon layer) is detected. Moreover, the external output function of the crack detection device 500 outputs the thickness detection result of the wafer W to the back surface grinding device 200 . The thickness detection function and the external output function will be described below.

The thickness detection function is performed by the table driving unit 14, the suction stage 18, the light source 552A, the illumination optical system 600, the interface position detection optical system 650, the control unit 750, and the focal point position moving mechanism, which are components of the crack detection device 500. 752 (piezo actuator 752a), condenser lens 754, operation unit 756 and display unit 758 are used to detect the thickness of wafer W. FIG.

FIG. 11 is a functional block diagram of the control section 750 of the crack detection device 500. As shown in FIG. In addition, in FIG. 11, the illustration of the functions other than the thickness detection function of the crack detection device 500 is omitted as appropriate.

<Thickness detection function>
As shown in FIG. 11 and FIG. 3 already described, when executing the thickness detection function, the control unit 750 executes a control program read from a storage unit (not shown) to control the light source control unit 800 and drive control. It functions as a portion 802 , an interface position detection portion 803 and a thickness detection portion 804 .

The light source control unit 800 causes the light source 552A to emit the detection light L1(A) when executing the thickness detection function. As a result, the detection light L1 (A) is irradiated onto the wafer W via the illumination optical system 600 and the condenser lens 754, and the reflected light L2 (A) from the wafer W is reflected by the condenser lens 754 and the interface position detection optical system. It is detected by the photodetector 660 (corresponding to the second photodetector of the present invention) via the system 650 .

The drive control unit 802 controls the table drive unit 14 to relatively move the suction stage 18 (wafer W) with respect to the condenser lens 754 in the XY directions (corresponding to the vertical direction of the present invention). The optical axis of the condensing lens 754 is aligned with the measuring point MP (see FIG. 12) for detecting the thickness of . In this case, the drive control section 802 functions together with the table drive section 14 as the second moving mechanism of the present invention.

A plurality of measurement points MP (see FIG. 12) are arranged along a scan line CL (see FIG. 14) passing through the center of the wafer W. As shown in FIG. A plurality of measurement points MP may be arranged on the processing line DL of the wafer W, and a plurality of measurement points MP are arranged on the processing line DL at predetermined intervals (for example, at intersections of the vertical and horizontal processing lines DL). You may

FIG. 12 is an explanatory diagram for explaining scanning in the Z direction of the focal point FP2 of the detection light L1(A) by the drive control unit 802. As shown in FIG. As shown in FIG. 12, after the condenser lens 754 is aligned with the measurement point MP, the drive control unit 802 drives at least one of the piezo actuator 752a and the processing head Z driving mechanism to move the condenser lens 754 and the wafer W together. By adjusting the relative distance of , the position of the condensing point FP2 of the condensing lens 754 [detection light L1 (A)] is set to a position spaced upward in the Z direction from the back surface Wb of the wafer.

Next, the drive control unit 802 drives the piezo actuator 752a of the focal point position moving mechanism 752 to move the condensing lens 754 downward in the Z direction (that is, toward the wafer W). The position of the focal point FP2 of the detection light L1(A) with respect to W is changed in the Z direction, that is, the focal point FP2 is scanned in the Z direction. As a result, the photodetector 660 of the interface position detection optical system 650 detects the reflected light L2(A) from the wafer W for each position of the condenser lens 754 in the Z direction, and the reflected light L2(A) is A detection signal is output to the thickness detection unit 804 . In this case, the drive control section 802 and the piezo actuator 752a correspond to the focal point scanning section of the present invention.

The reflected light L2(A) is condensed at the position of the pinhole of the pinhole panel 660B that is optically conjugate with the condensing point of the condensing lens 754 . When the condensing point FP2 of the condensing lens 754 coincides with the back surface Wb of the wafer due to the downward movement of the condensing lens 754, the detection signal output from the detector main body 660A has a sharp peak. Then, the downward movement of the condenser lens 754 by the piezo actuator 752a is continued, and when the condensing point FP2 of the condenser lens 754 coincides with the position of the wafer surface Wa, the detection signal output from the detector body 660A is also a sharp peak.

FIG. 13 is a graph showing the relationship between the movement amount (μm) of the focal point FP2 by the piezo actuator 752a and the detection signal (V) output from the detector main body 660A of the photodetector 660. FIG.

According to the graph shown in FIG. 13, there are peaks P1 and P2 of the detection signal (V) when the amount of movement (μm) of the piezoelectric actuator 752a, that is, the amount of movement of the condenser lens 754 is around 20 μm and around 230 μm. indicates the interface position on the wafer back surface Wb, and the peak P2 indicates the interface position on the wafer front surface Wa. Thereby, the interface positions indicating the wafer front surface Wa and the wafer back surface Wb can be detected, and the thickness of the wafer W can be detected based on the interface positions of both the wafer front surface Wa and the wafer back surface Wb.

Hereinafter, when there are a plurality of measurement points MP, the drive control unit 802 operates repeatedly for each measurement point MP to align the condensing lens 754 with respect to the measurement point MP and move the condensing point FP2 in the Z direction. Scanning and are performed repeatedly. This makes it possible to scan the focal point FP2 for each measurement point MP while changing the measurement point MP multiple times.

Returning to FIG. 11, the interface position detection unit 803 detects the detection signal ( 13), the interface position of the wafer W at the measurement point MP is detected. Specifically, the interface position detector 803 detects the interface position indicating the wafer front surface Wa and the wafer back surface Wb by detecting the peaks P1 and P2 of the detection signal output from the photodetector 660. do. Thereby, the interface position of the silicon layer of the wafer W can be detected.

The thickness detector 804 detects the thickness of the wafer W at the measurement point MP by the interface position detector 803 . For example, when the interface position detector 803 detects the wafer front surface Wa and the wafer back surface Wb from the detection signals shown in FIG. 20 μm corresponding to the interface position of is subtracted. That is, the thickness detector 804 detects the thickness of the wafer W based on the amount of movement of the condensing lens 754 when the interface position indicating the wafer front surface Wa and the wafer back surface Wb is detected. Thereby, the thickness (210 μm) of the wafer W at the measurement point MP can be detected.

FIG. 14 is an explanatory diagram for explaining the arrangement of measurement points MP on the wafer W. As shown in FIG. 15 is a cross-sectional view of the wafer W along the scan line CL in FIG. 14. FIG. 15 indicates the thickness of the wafer W at each measurement point MP. The measurement points MP shown in FIGS. 14 and 15 are part of the plurality of measurement points MP arranged along the scan line CL, and the number and arrangement of the measurement points MP are shown in FIGS. It is not limited to the example shown in 15.

As shown in FIGS. 14 and 15, a plurality of measurement points MP are arranged along a scan line CL passing through the center of the wafer W. As shown in FIGS. Therefore, every time the optical axis of the condensing lens 754 is sequentially moved to each measurement point MP by the table driving unit 14, the Z-direction scanning of the condensing point FP2 by the piezo actuator 752a and the reflected light by the photodetector 660 are performed. Detection of L2(A), detection of the interface position by the interface position detection unit 803, and thickness detection by the thickness detection unit 804 are repeatedly executed. Thereby, the thickness of the wafer W is detected for each of the plurality of measurement points MP along the scan line CL.

Here, the shape and thickness of the wafer W back-grinded by the back-grinding device 200 are substantially point-symmetrical with respect to its center (the center during grinding) (see Patent Document 1 above). In this case, the thickness of the wafer W varies concentrically from its center and smoothly in the planar direction. Therefore, the in-plane thickness distribution of the wafer W can be determined (estimated) based on the detection results of the thickness of the wafer W at each of the plurality of measurement points MP along the scan line CL.

Returning to FIG. 11, the thickness detection unit 804 detects the thickness of the wafer W along the processing line DL for each processing line DL of the wafer W based on the detection results of the thickness of the wafer W at each of the plurality of measurement points MP along the scan line CL. A thickness profile 42 showing the variation (distribution) of the thickness of W is detected.

The thickness profile 42 for each processing line DL detected by the thickness detection unit 804 is output to the external output unit 806 and displayed on the display unit 758 . The above is the thickness detection function of the crack detection device 500 .

<External output function>
Next, the external output function of the crack detection device 500 will be specifically described. This external output function externally outputs the thickness profile 42 for each processing line DL to the laser processing apparatus 10 using the control unit 750 and the communication IF 759 that are components of the crack detection apparatus 500 .

When executing the external output function, the control unit 750 functions as the external output unit 806 by executing a control program read from a storage unit (not shown).

The external output unit 806 externally outputs the thickness profile 42 for each processing line DL detected by the thickness detection unit 804 to the communication IF 28 (see FIG. 16) of the laser processing apparatus 10 via the communication IF 759 . Thereby, in the laser processing apparatus 10, laser processing can be performed with reference to the thickness profile 42 for each processing line DL. The above is the external output function of the crack detection device 500 .

[Laser processing by laser processing equipment]
FIG. 16 is a functional block diagram of the controller 12 of the laser processing apparatus 10. As shown in FIG. Based on the thickness profile 42 detected by the crack detection device 500 and the detection result of the height position of the wafer back surface Wb by the AF sensor 24, the laser processing device 10 moves the wafer along the processing line DL for each processing line DL. A laser processing is performed to form a laser processing region R within W and at a position at a certain distance from the wafer surface Wa.

As shown in FIG. 16, the control unit 12 of the laser processing apparatus 10 includes the table driving unit 14, the laser light sources 20 and 25, and the photodetector 26, as well as a storage unit 27 and a communication IF 28. is connected. The communication IF 28 is wired or wirelessly connected to the communication IF 759 of the crack detection device 500 described above, and the thickness profile 42 is input from the communication IF 759 . The storage unit 27 stores the thickness profile 42 input from the crack detection device 500 in addition to the control program (not shown) of the control unit 12 .

By executing the control unit program read from the storage unit 27, the control unit 12 controls the light source control unit 32, the drive control unit 34, the height position detection unit 36, the profile acquisition unit 38, and the condensing point position adjustment unit 40. function as

The light source control unit 32 causes the laser light source 25 of the AF sensor 24 to continuously emit the detection laser light LD. As a result, the detection laser beam LD is condensed onto the wafer back surface Wb via the dichroic mirror 21 and the condenser lens 22, and the reflected light LR from the wafer back surface Wb is condensed via the condenser lens 22 and the dichroic mirror 21 for AF detection. It continuously impinges on the photodetector 26 of the sensor 24 . Further, the light source control unit 32 causes the laser light source 20 to continuously emit the processing laser light LP. As a result, the processing laser beam LP is condensed at the condensing point FP1 inside the wafer W via the dichroic mirror 21 and the condensing lens 22 .

The drive control unit 34 controls the table drive unit 14 to relatively move the suction stage 18 (wafer W) with respect to the condenser lens 22 in the XY directions (corresponding to the vertical direction of the present invention). The optical axis of the condenser lens 22 is aligned with the DL processing start position. In this case, the drive control section 34 functions together with the table drive section 14 as the first moving mechanism of the present invention. Then, when the laser machining process is started, the drive control unit 34 controls the table drive unit 14 to execute the above-described processing feed and 1-pitch index feed.

The height position detection unit 36 performs laser processing of the wafer W using a known method such as an astigmatism method and a knife edge method based on the detection signal of the reflected light LR continuously input from the photodetector 26. The height position in the Z direction of the back surface Wb of the wafer at the position is continuously detected, and the detection result of the height position is continuously output to the condensing point position adjustment unit 40 .

The profile acquisition unit 38 acquires the thickness profile 42 for each processing line DL input from the communication IF 759 of the crack detection device 500 via the communication IF 28 and stores it in the storage unit 27 .

FIG. 17 is an explanatory diagram showing a formation scheduled line RP of the laser processing region R within the wafer W. As shown in FIG. As shown in FIG. 17 and FIG. 16 already described, the condensing point position adjusting unit 40 moves the condensing point FP1 along the planned formation line RP separated from the wafer surface Wa by a predetermined constant distance. The piezo actuator 23 is driven to adjust the position of the condensing point FP1 (condensing lens 22) in the Z direction so that the laser processing region R is formed along the planned formation line RP.

FIG. 18 is an explanatory diagram for explaining the position adjustment of the condensing point FP1 (condensing lens 22) in the Z direction by the condensing point position adjusting unit 40. As shown in FIG.

As shown in FIG. 18, the focal point position adjusting unit 40 adjusts the Z-direction height position of the wafer back surface Wb at the laser processing position based on the detection result of the height position input from the height position detection unit 36. discriminate. Here, in the focal point position adjusting unit 40, there are a processing depth design value, which is a design value of the Z-direction distance from the wafer back surface Wb to the line to be formed RP, and a design value (reference value) of the thickness of the wafer W. A thickness design value is set in advance.

The focal point position adjusting unit 40 detects the thickness of the wafer W at the laser processing position by referring to the thickness profile 42 corresponding to the processing line DL in the storage unit 27 . Next, the focal point position adjusting unit 40 adds an offset Δh to the processing depth design value based on the thickness of the wafer W at the laser processing position.

For example, if the actual thickness of the wafer W at the laser processing position is larger than the thickness design value, a positive offset Δh is added to the processing depth design value according to the increase, and conversely, the actual thickness of the wafer W is becomes smaller than the thickness design value, a negative offset .DELTA.h is added to the machining depth design value according to the decrease. The offset Δh is determined so that the distance from the wafer surface Wa to the line to be formed RP is constant. Then, the focal point position adjusting unit 40 drives the piezo actuator 23 based on the Z-direction height position of the wafer back surface Wb at the previously determined laser processing position and the offset processing depth design value. The Z-direction position of the condensing point FP1 is adjusted to align the Z-direction position of the condensing point FP1 with the line to be formed RP. Reference symbol SD indicates the position of the focal point of the detection laser beam LD after position adjustment by the focal point position adjusting unit 40 .

Hereinafter, while the drive control unit 34 controls the table drive unit 14 and executes processing feed, detection of the reflected light LR by the photodetector 26, detection of the height position by the height position detection unit 36, Z-direction position adjustment of the condensing point FP1 by the condensing point position adjusting unit 40 is repeatedly performed. As a result, the laser-processed region R is formed inside the wafer W along the planned formation line RP.

[Method for Controlling Laser Processing Apparatus (Action)]
FIG. 19 is a flow chart showing the flow of detection processing of the thickness profile 42 for each processing line DL of the wafer W by the crack detection device 500 having the above configuration.

As shown in FIG. 19, the wafer W whose back surface Wb has been back ground by the back surface grinding device 200 is set on the suction stage 18 of the crack detection device 500 (laser processing device 10) (step S1).

Next, when the operator inputs an operation for measuring the thickness profile 42 of the wafer W to the operation unit 756, the control unit 750 controls the light source control unit 800, the drive control unit 802, the interface position detection unit 803, the thickness detection unit 804, and the It functions as an external output unit 806 .

Then, the light source control unit 800 emits the detection light L1(A) from the light source 552A (step S2). Thereby, the wafer W is irradiated with the detection light L<b>1 (A) through the illumination optical system 600 and the condenser lens 754 .

Further, the drive control unit 802 controls the table drive unit 14 to align the optical axis of the condenser lens 754 with the measurement point MP on the scan line CL of the wafer W. FIG. Next, the drive control unit 802 drives at least one of the piezo actuator 752a and the processing head Z drive mechanism to adjust the relative distance between the condenser lens 754 and the wafer W, and then drives the piezo actuator 752a to detect light L1. The position of the focal point FP2 in (A) is changed in the Z direction (step S3). As a result, the focal point FP2 is scanned in the Z direction.

While the condensing point FP2 is scanned in the Z direction, the reflected light L2 (A) from the wafer W is detected by the photodetector 660 via the condensing lens 754 and the interface position detection optical system 650. 660 repeatedly outputs a detection signal to the thickness detector 804 (NO in steps S4 and S5).

When the scanning of the condensing point FP2 in the Z direction is completed (YES in step S5), the interface position detection unit 803, as shown in FIG. Based on the input detection signals, interface positions indicating the wafer front surface Wa and the wafer back surface Wb are detected. Then, the thickness detector 804 detects the thickness of the wafer W at the measurement point MP based on the amount of movement of the condenser lens 754 when the interface positions indicating the wafer front surface Wa and the wafer back surface Wb are detected by the interface position detector 803. Detect (step S6).

Subsequently, the alignment of the optical axis of the condenser lens 754 with respect to the next measurement point MP and the processing from step S3 to step S6 are repeatedly executed for each of the remaining measurement points MP along the scan line CL ( YES in step S7, step S8). As a result, the thickness detection unit 804 detects the thickness of all the plurality of measurement points MP along the scan line CL, and the thickness profile 42 for each processing line DL is detected based on the thickness of each measurement point MP (step S7 NO, step S9). As described above, in this embodiment, by using the interface detection function of the crack detection device 500, when the wafer surface Wa is covered with the protective sheet 220 or the device layer of the wafer surface Wa is uneven, However, the thickness of the silicon layer of the wafer W can be accurately detected. As a result, the thickness profile 42 can be accurately detected.

When the thickness detection unit 804 completes the thickness profile 42 for each processing line DL, the external output unit 806 outputs each thickness profile 42 to the communication IF 28 of the laser processing apparatus 10 via the communication IF 759 (step S10).

When the external output of each thickness profile 42 by the external output unit 806 is completed, the process proceeds to laser processing by the laser processing apparatus 10 .

FIG. 20 is a flow chart showing the flow of laser processing for each processing line DL by the laser processing apparatus 10 configured as described above. As shown in FIG. 20, after the thickness profile 42 is detected, when the operator inputs an operation to start laser processing of the wafer W to an operation unit (not shown) of the laser processing apparatus 10 (can be used also as the operation unit 756), control The unit 12 functions as a light source control unit 32 , a drive control unit 34 , a height position detection unit 36 , a profile acquisition unit 38 and a condensing point position adjustment unit 40 .

The profile acquisition unit 38 acquires the thickness profile 42 for each processing line DL input from the crack detection device 500 via the communication IF 28, and stores these thickness profiles 42 in the storage unit 27 (step S11). Note that step S11 is not particularly limited as long as it is performed before step S15, which will be described later.

Further, the drive control unit 802 controls the table drive unit 14 to align the optical axis of the condenser lens 22 with the processing start position of the processing line DL.

When this alignment is completed, the light source control unit 32 causes the laser light source 25 to continuously emit the detection laser beam LD (step S12), and causes the laser light source 20 to continuously emit the processing laser beam LP ( step S12A). As a result, the detection laser beam LD is condensed onto the wafer back surface Wb by the condensing lens 22, and the reflected light LR from the wafer back surface Wb is continuously incident on the photodetector 26, and the photodetector 26 is reflected. A detection signal of the light LR is continuously output to the height position detection unit 36 (step S13). In addition, the processing laser beam LP is condensed at the condensing point FP1 inside the wafer W via the condensing lens 22 .

When the detection of the reflected light LR by the photodetector 26 is started, the height position detector 36 detects the laser processing position of the wafer W based on the detection signal of the reflected light LR continuously input from the photodetector 26. The height position of the wafer back surface Wb in the Z direction is continuously detected, and the detection result of the height position is continuously output to the condensing point position adjusting unit 40 (step S14).

Next, the focal point position adjusting unit 40 determines the Z-direction height position of the wafer back surface Wb at the laser processing position based on the detection results of the height position continuously input from the height position detecting unit 36. . Further, the focal point position adjusting unit 40 detects the thickness of the wafer W at the laser processing position by referring to the thickness profile 42 corresponding to the processing line DL in the storage unit 27, and based on the detection result, the above-mentioned figure As shown in 18, the offset Δh is added to the machining depth design value (step S15).

Then, the focal point position adjusting unit 40 drives the piezo actuator 23 based on the Z-direction height position of the wafer back surface Wb at the laser processing position and the offset processing depth design value to shift the focal point FP1. Position adjustment in the Z direction is performed to align the position of the focal point FP1 in the Z direction with the line to be formed RP (step S16). As a result, the laser processing region R is formed in the formation scheduled line RP.

Thereafter, while the drive control unit 34 controls the table drive unit 14 to move the suction stage 18 (wafer W) relative to the condenser lens 22 in the X direction (NO in step S17). , step S18), the above-described processing from step S12 to step S18 is repeatedly executed. As a result, the laser-processed region R is formed along the planned formation line RP, and the formation position of the laser-processed region R from the wafer front surface Wa side is maintained at a constant distance.

Then, when the laser processing of one processing line DL is completed (YES in step S17), the drive control unit 34 controls the table drive unit 14 to move the suction stage 18 in the XY directions with respect to the condenser lens 22. By relatively moving, the optical axis of the condenser lens 22 is aligned with the processing start position of the next processing line DL (NO in step S19, step S20). When this alignment is completed, the processing from step S12 to step S18 described above is repeated again, and laser processing is performed on the next processing line DL. Thereafter, the laser processing of the remaining processing lines DL is similarly performed (YES in step S19).

As described above, in the present embodiment, the thickness profile 42 of the wafer W can be detected more accurately than before by using the interface detection function of the crack detection device 500. The distance can be maintained at a constant distance with higher accuracy.

[others]
FIG. 21 is an explanatory view showing a formation scheduled line RP of a laser processing region R in a wafer W having a non-flat wafer surface Wa. FIG. 21 is an explanatory diagram for explaining the position adjustment of the focal point FP1 in the Z direction by the focal point position adjusting unit 40 when laser processing is performed on the wafer W shown in FIG.

In the above embodiment, the case where the wafer surface Wa is flat has been described as an example. However, as shown in FIG. 18a) and the condenser lens 22 are fluctuating due to disturbances. In this case, the line to be formed RP also has a nonlinear shape (for example, a wavy shape).

Even in such a case, as shown in FIG. 22, the focal point position adjusting unit 40 is offset based on the Z-direction height position of the wafer back surface Wb at the laser processing position and the thickness of the wafer W at the laser processing position. The piezo actuator 23 is driven to align the Z-direction position of the focal point FP1 with the planned formation line RP. Then, while the processing feed is being executed, similarly to the above embodiment, the photodetector 26 detects the reflected light LR, the height position detection unit 36 detects the height position, and the condensing point position adjustment unit 40 is repeatedly performed to adjust the position of the focal point FP1 in the Z direction. In this case, the piezo actuator 23 expands and contracts in the Z direction while the wafer is being fed to maintain the distance between the wafer surface Wa and the focal point FP1 at a constant distance. As a result, the laser-processed region R is formed inside the wafer W along the planned formation line RP.

The laser processing apparatus 10 and the crack detection apparatus 500 of the above embodiment employ a configuration in which the processing head 16 is movable in the Z direction and the suction stage 18 is movable (rotatable) in the XYθ directions. As long as the relative position in the XYZθ direction between the head 16 and the suction stage 18 can be changed, the configuration is not limited to the embodiment. For example, the suction stage 18 may be configured to be movable (rotatable) in the XYZθ directions, or a configuration in which the processing head 16 is movable in the YZ directions and the suction stage 18 is movable (rotatable) in the Xθ directions. may be

The laser processing apparatus 10 and the crack detection apparatus 500 of the above-described embodiment employ a configuration in which the processing head 16 and the crack detection apparatus 500 are integrated and movable in the Z direction. A configuration in which the crack detection device 500 and the crack detection device 500 can move independently of each other in the Z direction may be adopted. Further, the direction in which the processing head 16 and the crack detection device 500 can move independently of each other is not limited to the Z direction, and may be configured to move in other directions (eg, the X direction and the Y direction).

In the above embodiment, the thickness profile 42 for each processing line DL is detected based on the detection results of the thickness of the wafer W at each of the plurality of measurement points MP along the scan line CL. may be actually measured.

In the above embodiments, the wafer W (silicon wafer) was taken as an example of the workpiece of the present invention, but the present invention can also be applied to laser processing of glass substrates, piezoelectric ceramic substrates, glass substrates, and the like. can be done.

10 laser processing device 12 control unit 14 table driving unit 16 processing head 18 suction stage 18a holding surface 19 processing head driving unit 20 laser light source 21 dichroic mirror 22 condenser lens 23 piezo actuator 24 AF sensor 25 laser light source 26 photodetector 27 storage Unit 32 Light source control unit 34 Drive control unit 36 Height position detection unit 38 Profile acquisition unit 40 Condensing point position adjustment unit 42 Thickness profile 100 Laser processing system 200 Back surface grinding device 220 Protection sheet 300 Division device 500 Crack detection device 550 Light source unit 552A Light source 552B Light source 552C Light source 554 Half mirror 600 Illumination optical system 602 Relay lens 604 Mirror 606 Relay lens 650 Interface position detection optical system 652 Half mirror 654 Half mirror 656 Relay lens 658 Half mirror 660 Photodetector 660A Detector body 660B Pin Hole panel 700 Crack detection optical system 702 Relay lens 704 Photodetector 704C Light receiving surface 706 Photodetector 706C Light receiving surface 750 Control unit 752 Condensing point position moving mechanism 752a Piezo actuator 754 Condensing lens 754a Condensing lens pupil 756 Operation unit 758 display unit 760 observation optical system 800 light source control unit 802 drive control unit 803 interface position detection unit 804 thickness detection unit 806 external output unit AX main optical axis CL scan line DL processing line FP1 condensing point FP2 condensing point G1 first area G2 Second area GUI Operation IF 28 Communication IF 759 Communication K Crack L Laser light L1 Detection light L2 Reflected light L2a Reflected light component L2b Non-reflected light component LD Detection laser light LP Processing laser light LR Reflected light MP Measuring point Nd Semiconductor laser excitation P1 Peak P2 Peak R Laser processing area R1 Path R2 Path R3 Path RP Planned formation line S Evaluation value W Wafer Wa Front surface Wb Wafer back surface Z Processing head Δh Offset

Claims (6)

irradiating the processing laser light from the back side of the work piece along a processing line of the work piece with the first focal point of the processing laser light aligned with the inside of the work piece; A laser processing apparatus comprising a processing head that forms a laser processing area inside a workpiece, and a crack detection device that detects a crack extending from the laser processing area in the workpiece,
The crack detection device is
an interface position detection unit that detects interface positions indicating the front surface and the back surface of the workpiece at a plurality of predetermined measurement points;
a thickness detection unit that detects a thickness profile indicating a change in thickness of the workpiece along the processing line based on the detection result of the interface position detection unit;
with
The processing head is
a laser light source unit including a first laser light source that emits the processing laser light and a second laser light source that emits the detection laser light;
The processing laser beam emitted from the first laser light source is focused inside the workpiece, and the detection laser beam emitted from the second laser light source is focused on the back surface of the workpiece. a first condenser lens;
a first light detection unit that detects reflected light of the detection laser light from the back surface of the workpiece;
a height position detection unit that detects the height position of the back surface of the workpiece based on the detection result of the first light detection unit;
a profile acquisition unit that acquires the thickness profile from the thickness detection unit;
Adjusting the position of the first condensing point along the optical axis direction of the first condensing lens based on the detection result of the height position detection unit and the thickness profile acquired by the profile acquisition unit , a condensing point position adjusting unit that maintains the position of the first condensing point at a predetermined constant distance from the surface of the workpiece;
A laser processing device comprising: The first condensing lens is moved relative to the workpiece in a direction perpendicular to the optical axis direction of the first condensing lens to move the first condensing point along the processing line. A first movement mechanism is provided,
Each time the first focal point is moved along the processing line by the first moving mechanism, the first light detection section, the height position detection section, and the focal point position adjustment section operate repeatedly. The laser processing apparatus according to claim 1. The crack detection device is
a light source that emits detection light;
a second condenser lens for condensing the detection light emitted from the light source onto the workpiece;
a condensing point scanning unit that scans a second condensing point of the detection light along the optical axis direction of the second condensing lens;
a second photodetector that detects reflected light of the detection light from the workpiece while the second focal point is scanned by the focal point scanning unit;
a second movement mechanism that relatively moves the second condenser lens with respect to the workpiece in a direction perpendicular to the optical axis direction of the second condenser lens;
with
each time the second moving mechanism sequentially moves the second light-condensing point to the plurality of measurement points, the light-condensing point scanning unit and the second light detection unit repeatedly operate;
3. The laser processing apparatus according to claim 1, wherein the interface position detection section detects the interface position for each of the plurality of measurement points based on detection results of the second light detection section for each of the plurality of measurement points. . 4. Any one of claims 1 to 3, wherein the interface position detection unit detects the interface position for each of the plurality of measurement points along a scan line passing through the center of the workpiece whose back surface has been ground by the grinding device. The laser processing device according to the item. The laser processing apparatus according to any one of claims 1 to 4, wherein the crack detection device is provided on the processing head. The laser beam for processing is directed along the processing line of the workpiece from the back side of the workpiece while the first condensing point of the laser beam for machining is aligned with the inside of the workpiece by the first condenser lens. and a processing head for forming a laser processing area inside the workpiece, and a crack detection device for detecting a crack extending from the laser processing area inside the workpiece. In the control method,
A change in the thickness of the workpiece along the processing line is detected based on the detection results obtained by detecting interface positions indicating the front surface and the back surface of the workpiece at a plurality of predetermined measurement points by the crack detection device. Detect the thickness profile showing
The processing head is
emitting the processing laser beam from a first laser light source;
A detection laser beam is emitted from the second laser light source,
The processing laser light emitted from the first laser light source is condensed inside the workpiece by a first condenser lens, and the detection laser light emitted from the second laser light source is focused on the workpiece. Focus the light on the back of the
detecting reflected light of the detection laser light from the back surface of the workpiece;
detecting the height position of the back surface of the workpiece based on the detection result of the reflected light of the detection laser beam;
obtaining the thickness profile from the crack detection device;
Adjusting the position of the first condensing point along the optical axis direction of the first condensing lens based on the detection result of the height position and the thickness profile, and adjusting the position of the first condensing point. A control method for a laser processing apparatus that maintains a predetermined constant distance from the surface side of the workpiece.

Images