Crack detection device

Abstract

The present invention provides a crack detection device that can non-destructively and accurately detect the three-dimensional location of cracks formed inside a workpiece, and can accurately predict the quality of the chip's machining before the workpiece fractures. [Solution] The crack detection device (1, 1A) includes a crack plane position detection means (200) that detects the planar position of a crack formed inside a workpiece by focusing a first detection light onto the workpiece and imaging the first reflected light from the workpiece, and a crack depth detection means (100) that detects the depth position of a crack by focusing a second detection light into the workpiece and imaging the second reflected light from the workpiece, and is movable integrally with the crack plane position detection means along the machining direction of the workpiece, and is positioned so that when moved in the machining direction, the measurement point where the second detection light is focused falls within the field of view of the crack plane position detection means, thereby enabling the detection of the planar position of a crack and the detection of the depth position of a crack in parallel.

Description

Technical Field

[0001] The present invention relates to a crack detection device, and more particularly to a crack detection device that detects cracks formed inside a workpiece.

Background Art

[0002] Conventionally, laser dicing apparatuses are known that focus a condensing point inside a wafer of silicon or the like, irradiate laser light along a scheduled cutting line, and form a laser processed region serving as a starting point for cutting inside the wafer along the processing line. The wafer on which the laser processed region is formed is then cut along the scheduled cutting line by a cleaving process such as expanding or braking to be divided into individual chips.

[0003] Incidentally, when a laser processed region is formed on a wafer by a laser dicing apparatus, a crack extends from the laser processed region in the thickness direction of the wafer.

[0004] After the laser processed region is formed by the laser dicing apparatus and before the cleaving process, by detecting whether the laser processed region serving as the starting point for dividing the wafer is properly formed, that is, the depth position of the crack formed inside the wafer, it becomes possible to accurately predict the quality of division into chips in the cleaving process.

[0005] Patent Document 1 proposes a technique for non-destructively inspecting cracks formed inside a workpiece. In the technique disclosed in Patent Document 1, light is made incident from one side of the crack, light transmitted through the region including the crack is detected, and the depth position of the crack is inspected by utilizing the decrease in the amount of detected light due to scattering at the crack.

Prior Art Documents

Patent Documents

[0006]

Patent Document 1

[0007] The technology disclosed in Patent Document 1 involves aligning the inspection position of the crack inspection means with the planned cutting line to inspect the crack depth dimension, but does not disclose the measurement of the crack depth position or the measurement of the crack's position in the plane direction (hereinafter also referred to as dicing line position confirmation). Furthermore, in conventional crack detection devices, dicing line position confirmation had to be performed as a separate process at a different time from the inspection of the crack depth dimension. As a result, the measurement time during laser processing is increased, leading to a problem of reduced processing throughput.

[0008] Furthermore, measuring only the depth of a crack does not reveal the exact location of the crack on the workpiece's street, and simply checking the dicing line position does not reveal how far the crack extends in the thickness direction of the workpiece. Thus, conventional crack detection devices require separate measurements of crack depth and crack position in the planar direction, making it difficult to efficiently and accurately measure the three-dimensional position of a crack. Consequently, it was difficult to accurately predict the quality of the chip's machining before the workpiece was fractured.

[0009] This invention has been made in view of these circumstances, and aims to provide a crack detection device that can non-destructively and accurately detect the three-dimensional position of a crack formed inside a workpiece, and can accurately predict the quality of the chip's machining before the workpiece fractures. [Means for solving the problem]

[0010] To achieve the above objective, a crack detection device according to a first aspect of the present invention includes: a crack plane position detection means for detecting the planar position of a crack formed inside a workpiece by focusing a first detection light onto the workpiece and imaging the first reflected light from the workpiece; and a crack depth detection means for detecting the depth position of a crack by focusing a second detection light into the workpiece and imaging the second reflected light from the workpiece, wherein the crack depth detection means is movable integrally with the crack plane position detection means along the machining direction of the workpiece, and is arranged such that when moved in the machining direction, the measurement point where the second detection light is focused falls within the field of view of the crack plane position detection means.

[0011] In the crack detection device according to the second aspect of the present invention, in the first aspect, when the crack depth detection means is moved in the direction of machining, the measurement point where the second detection light is focused is arranged coaxially with respect to the center of the field of view of the crack plane position detection means along the direction of machining.

[0012] In the third aspect of the present invention, the crack detection device is such that, in the first or second aspect, an optical element included in the crack plane position detection means and a portion of the optical elements included in the crack depth detection means are shared.

[0013] In the crack detection device according to the fourth aspect of the present invention, in the third aspect, a focusing lens for focusing a first detection light onto the workpiece and a focusing lens for focusing a second detection light into the interior of the workpiece are shared. [Effects of the Invention]

[0014] According to the present invention, the depth position of a crack and the position of the crack in the planar direction can be measured almost simultaneously (in parallel). This makes it possible to non-destructively and accurately detect the three-dimensional position of a crack formed inside a workpiece, and to accurately predict the quality of the chip's machining before the workpiece fractures. [Brief explanation of the drawing]

[0015] [Figure 1]Figure 1 is a block diagram showing a crack detection apparatus according to the first embodiment of the present invention. [Figure 2] Figure 2 is a block diagram showing the control flow in the crack detection apparatus according to the first embodiment of the present invention. [Figure 3] Figure 3 is a plan view for explaining the arrangement of a crack depth detection device and a crack planar position detection device (optical system for dicing line position confirmation). [Figure 4] Figure 4 is a block diagram showing the crack planar position detection device (optical system for dicing line position confirmation). [Figure 5] Figure 5 is a plan view showing an observed image of a workpiece obtained by the crack planar position detection device (optical system for dicing line position confirmation). [Figure 6] Figure 6 is a block diagram showing the crack depth detection device. [Figure 7] Figure 7 is an explanatory diagram showing a state when oblique illumination with detection light is performed on a workpiece. [Figure 8] Figure 8 is an explanatory diagram showing a state when oblique illumination with detection light is performed on a workpiece. [Figure 9] Figure 9 is an explanatory diagram showing a state when oblique illumination with detection light is performed on a workpiece. [Figure 10] Figure 10 is a diagram showing the state of reflected light received by a photodetector (corresponding to Figure 7). [Figure 11] Figure 11 is a diagram showing the state of reflected light received by a photodetector (corresponding to Figure 8). [Figure 12] Figure 12 is a diagram showing the state of reflected light received by a photodetector (corresponding to Figure 9). [Figure 13] Figure 13 is a diagram for explaining the path through which reflected light from a workpiece reaches the pupil of a condenser lens. [Figure 14] Figure 14 is a flowchart showing the operation of the crack detection apparatus according to the first embodiment of the present invention. [Figure 15] Figure 15 is a block diagram showing a crack detection apparatus according to the second embodiment of the present invention. [Figure 16]FIG. 16 is a block diagram showing a crack detection apparatus according to the second embodiment of the present invention. [Mode for Carrying Out the Invention]

[0016] Hereinafter, embodiments of the crack detection apparatus according to the present invention will be described with reference to the accompanying drawings.

[0017] [First Embodiment] First, the first embodiment of the present invention will be described. FIG. 1 is a block diagram showing a crack detection apparatus according to the first embodiment of the present invention. In the following description, a three-dimensional orthogonal coordinate system is used in which the stage ST(XYθ) on which the workpiece W is placed is a plane parallel to the XY plane, and the Z direction is the depth direction of the workpiece W.

[0018] The crack detection apparatus 1 is used in combination with a laser dicing apparatus (not shown) that forms a laser processing region inside a workpiece (for example, a silicon wafer) W, but in the following description, components related to the crack detection apparatus 1 will be described, and descriptions of the configuration of the laser dicing apparatus will be omitted.

[0019] The crack detection apparatus 1 according to the present embodiment includes a crack depth detection device 100 that detects the depth position of a crack K formed inside the workpiece W, and a crack plane position detection device (hereinafter referred to as an optical system for dicing line position confirmation) 200 that detects the position of the crack K on the XY plane. Here, the crack depth detection device 100 and the crack plane position detection device 200 are examples of crack depth detection means and crack plane position detection means, respectively.

[0020] The crack depth detection device 100 and the optical system 200 for dicing line position confirmation are attached to the Z stage ST(Z), and are integrally movable in the Z direction.

[0021] The control device 10 includes a processor such as a CPU (Central Processing Unit) that controls the operation of each part of the crack detection device 1, a storage device (e.g., an HDD (Hard Disk Drive) or SSD (Solid State Drive)) that stores control programs and various data, and RAM (Random Access Memory) that can be used as a working area for the CPU. A general-purpose computer (e.g., a personal computer) can also be used as the control device 10.

[0022] The control device 10 includes an operating unit (e.g., a keyboard, and pointing devices such as a mouse and touch panel) for receiving operation input from the operator, and transmits control signals corresponding to the operation input to each part of the crack detection device 1 to control the operation of each part. The control device 10 also includes a display unit (e.g., a liquid crystal display) for displaying a GUI (Graphical User Interface), images, and various calculation results (e.g., crack detection results).

[0023] The crack measurement control unit 12 controls the crack depth detection device 100 and the dicing line position confirmation optical system 200 according to the control signal from the control device 10. As shown in Figure 2, the crack measurement control unit 12 receives the measurement result (received light data DAT1) from the crack depth detection device 100 and transmits the measurement result (observation result DAT2) from the dicing line position confirmation optical system 200 to the control device 10.

[0024] The stage control unit 14 outputs a movement instruction to the stage ST(XYθ) according to a control signal from the control device 10, enabling it to move the workpiece W placed on the stage ST(XYθ) in the XY direction and rotate it around the Z axis on the XY plane. The stage control unit 14 also outputs a movement instruction to the Z stage ST(Z) according to a control signal from the control device 10, enabling it to move the crack depth detection device 100 and the dicing line position confirmation optical system 200 in the Z direction. The stage control unit 14 controls the stage ST(XYθ) and the Z stage ST(Z) according to a control signal from the control device 10 to adjust the relative positions of the crack depth detection device 100 and the dicing line position confirmation optical system 200 and the workpiece W.

[0025] Figure 3 is a plan view illustrating the arrangement of the crack depth detection device 100 and the crack plane position detection device (optical system for confirming the position of the dicing line) 200.

[0026] In Figure 3, the symbol C1 indicates the position of the measurement point of the crack depth detection device 100, and the symbol C2 indicates the position of the center of the observation position (field of view) V2 of the dicing line position confirmation optical system 200.

[0027] Furthermore, the white arrows in Figure 3 indicate the direction in which the laser dicing device forms the laser-processed area (hereinafter referred to as the processing direction). As shown in Figure 3, the crack K is formed along the processing direction. In the example shown in Figure 3, the processing direction is the X direction.

[0028] As shown in Figure 3, when detecting a crack K, the relative positions of the crack depth detection device 100, the dicing line position confirmation optical system 200, and the workpiece W are adjusted so that the measurement point C1 of the crack depth detection device 100 and the field of view center C2 of the dicing line position confirmation optical system 200 are coaxial along the machining direction. Then, the measurement of the planar position of the crack K (dicing line position confirmation) and the measurement of the depth position are performed simultaneously or in parallel. This makes it possible to obtain the planar position (position in the XY direction) and the depth direction (Z direction) of the crack K together, and to measure the formation state of the crack K as three-dimensional information.

[0029] Furthermore, during dicing line position verification, it is possible to acquire position data of the crack K in the direction perpendicular to the machining direction (X direction) (Y direction). Therefore, it is possible to correct the positional deviation in the Y direction when measuring the depth of the crack K, and to easily adjust the position in the Y direction when measuring the depth of the crack K. As a result, the measurement throughput is also improved.

[0030] According to this embodiment, the formation position of the crack K on the plane of the workpiece W and its position in the depth direction (the propagation position of the crack K) can be obtained, making it possible to identify processing defects and determine whether the processing conditions are appropriate.

[0031] In the example shown in Figure 3, the measurement point C1 of the crack depth detection device 100 and the field of view center C2 of the optical system 200 for confirming the dicing line position are positioned coaxially along the machining direction, but the present invention is not limited to this. For example, if the measurement point C1 of the crack depth detection device 100 is moved in the machining direction, it is possible to shift the measurement point C1 of the crack depth detection device 100 and the field of view center C2 of the optical system 200 for confirming the dicing line position in a direction perpendicular to the machining direction (Y direction), as long as they remain within the field of view V2 of the optical system 200 for confirming the dicing line position.

[0032] Furthermore, in the example shown in Figure 3, the optical system 200 for confirming the dicing line position is positioned in front of the crack depth detection device 100 along the machining direction, and the planar position of the crack K is measured before the depth position of the crack K is measured. However, the present invention is not limited to this. It is also possible to reverse the arrangement of the optical system 200 for confirming the dicing line position and the crack depth detection device 100 along the machining direction from the example in Figure 3, and measure the depth position of the crack K first.

[0033] The crack plane position detection device (optical system for confirming dicing line position) 200 and the crack depth detection device 100 will be described below.

[0034] (Crack plane position detection device (optical system for confirming dicing line position)) Figure 4 is a block diagram showing the crack plane position detection device (optical system for confirming dicing line position) 200.

[0035] As shown in Figure 4, the dicing line position confirmation optical system 200 detects the position of the crack K in the XY direction by irradiating the workpiece W with illumination light (first detection light) L21 emitted from the light source 210, and capturing the reflected light L22 from the workpiece W with the camera 218.

[0036] The light source 210 emits illumination light L21 toward the half mirror 214 of the observation optical system 212. As the light source 210, for example, an infrared light source that emits infrared light that can penetrate the inside of the workpiece (silicon wafer) W (e.g., an infrared laser light source, an IR LED (Infrared Emitting Diode)) can be used.

[0037] Furthermore, under processing conditions where the crack K propagates to the surface of the workpiece W (the surface on the incident side of the illumination light L21), and the observation direction is towards the pattern side, it is also possible to use a light source other than an infrared light source (for example, a visible light source) as the light source 210.

[0038] Furthermore, as the light source 210, in order to evaluate the reflectance of the detected light L1 (reflected light L2) when detecting the depth position of the crack K, light in the same wavelength range (including approximately the same wavelength) as the detected light L1 (reflected light L2) may be selected.

[0039] The illumination light L21 is reflected by the half mirror 214, passes through the observation optical system 212 (including, for example, a relay lens), and is focused onto the surface or interior of the workpiece W by the focusing lens 216. The reflected light (first reflected light) L22 reflected by the workpiece W then passes through the focusing lens 216, the observation optical system 212, and the half mirror 214 and is imaged by the camera 218. The observation image of the workpiece W captured by the camera 218 (observation result DAT2 in Figure 2) is transmitted to the control device 10 via the crack measurement control unit 12.

[0040] Here, the camera 218 used is one that is highly sensitive to the illumination light L21 emitted from the light source 210. In other words, if the light source 210 is an infrared light source, the camera 218 used is an infrared camera that is highly sensitive to infrared light.

[0041] Figure 5 is a plan view showing an observed image of the workpiece W. In the figure, the symbols DP indicate the device pattern of the semiconductor device, and the symbols SR indicate the streets (lines to be cut) formed between the semiconductor devices on the workpiece (silicon wafer) W.

[0042] As shown in Figure 5, the location of the crack K formed in street SR can be measured from the observation image captured by camera 218.

[0043] Next, the control system of the dicing line position confirmation optical system 200 will be described. As shown in Figure 4, the dicing line position confirmation optical system 200 includes an optical system control unit 202, an area storage unit 204, an image recording unit 206, and an image analysis unit 208.

[0044] The optical system control unit 202 includes a processor such as a CPU, a control program, a memory (e.g., ROM (Read Only Memory)) for storing various data, and RAM that can be used as a working area for the CPU.

[0045] The image recording unit 206 is a memory for recording the observation image captured by the camera 218. The image recording unit 206 may also be a buffer memory for temporarily recording the observation image.

[0046] The image analysis unit 208 detects the position of the crack K in the XY direction from the observation image captured by the camera 218.

[0047] Furthermore, the image analysis unit 208 may detect the location of a measurable area for detecting the depth position of the crack K from the observation image captured by the camera 218. In this case, the measurable area refers to a small area on the street SR suitable for detecting the depth position of the crack K. If the detection light L1 (see Figure 6) is shone on an area where regions with different reflectivity to the detection light L1 (see Figure 6) of the crack K depth position are mixed, the detection light L1 will be focused into the inside of the workpiece W across these multiple areas, reducing the accuracy of crack K depth position detection. The measurable area is determined by avoiding areas where regions with different reflectivity to the detection light L1 (see Figure 6) of the crack K depth position are mixed.

[0048] Specifically, the measurable area is determined based on the reflectance distribution of each small area within the observed image (field of view V2). Specifically, it is defined as a small area where the change in reflectance along the width direction (Y direction) of the street SR and its gradient are small, i.e., a small area where the reflectance distribution is approximately flat. For example, the image analysis unit 208 determines one or more small areas within each small area where the change in reflectance along the width direction of the street SR (excluding both ends in the width direction) is less than or equal to a predetermined reference value (e.g., 10%) as the measurable area. Alternatively, the image analysis unit 208 determines the small area within each small area where the change in reflectance along the width direction of the street SR is smallest as the measurable area.

[0049] The position of the measurable area of ​​the workpiece W, determined in this way, is stored in the area storage unit 204 and used to detect the depth position of the crack K.

[0050] (Crack depth detection device) Next, the crack depth detection device 100 will be described with reference to Figure 6.

[0051] As shown in Figure 6, the crack depth detection device 100 according to this embodiment includes a light source unit 120, an illumination optical system 140, an interface detection optical system 160, a crack detection optical system 180, a detection signal processing unit 102, an alignment mechanism 104, and a focusing lens 106.

[0052] The light source unit 120 emits detection light L1, which is used to detect the interface of the workpiece W and to detect cracks K formed inside the workpiece W. Here, if the workpiece W is a silicon wafer, it is desirable to use infrared light with a wavelength of 1,000 nm or more as the detection light L1.

[0053] The light source unit 120 includes a light source 122, a collimating lens 124, and a limiting member 126. The light source 122, collimating lens 124, and limiting member 126 are arranged along the principal optical axis AX, which is coaxial with the lens optical axis of the focusing lens 106.

[0054] The light source 122 emits detection light L1 along the main optical axis AX. As the light source 122, for example, an infrared light source (e.g., an infrared laser light source, an IR LED (Infrared Emitting Diode)), or an LED (Light Emitting Diode) light source can be used. The light source 122 is connected to the detection signal processing unit 102, and the detection signal processing unit 102 controls the emission of light from the light source 122.

[0055] The collimating lens 124 is adjusted so that the detected light L1 emitted from the light source 122 becomes parallel light.

[0056] The limiting member 126 has two selectively openable and closable openings 126A and 126B. The limiting member 126 is a light-shielding member that blocks a portion of the detection light L1 from the collimating lens 124. The limiting member 126 can be extended and retracted relative to the optical path of the detection light L1 by a drive mechanism (not shown). The detection signal processing unit 102 controls the extension and retraction of the limiting member 126 by controlling an actuator (not shown). When the crack detection optical system 180 detects a crack K, the limiting member 126 blocks a portion of the detection light L1, causing the detection light L1 to be eccentric with respect to the main optical axis AX. The limiting member 126 will be described later.

[0057] The illumination optical system 140 guides the detection light L1 emitted from the light source unit 120 to the focusing lens 106. The illumination optical system 140 includes relay lenses 142 and 146 and a mirror 144. The detection light L1 emitted from the light source unit 120 passes through the relay lens 142 and is reflected by the mirror (e.g., a total internal reflection mirror) 144, causing the optical path to be bent. The detection light L1 reflected by the mirror 144 passes through the relay lens 146 and is emitted towards the focusing lens 106.

[0058] The focusing lens 106 focuses (concentrates) the detection light L1 emitted from the illumination optical system 140 onto the workpiece W. The focusing lens 106 is positioned opposite the workpiece W and is coaxial with the principal optical axis AX.

[0059] The alignment mechanism 104 adjusts the focusing position of the detection light L1 on the workpiece W. The alignment mechanism 104 includes a drive unit that moves at least one of the stage ST(Z) (focusing lens 106) and the stage ST(XYθ) on which the workpiece W is placed. The alignment mechanism 104 can move the focusing position of the detection light L1 in the XYZ direction by adjusting the relative position in the XYZ direction between the focusing lens 106 and the stage ST(XYθ).

[0060] The reflected light L2, focused by the condensing lens 106 and reflected by the workpiece W, is guided to the interface detection optical system 160 and the crack detection optical system 180, and used for interface detection and crack detection of the workpiece W, respectively.

[0061] The detection signal processing unit 102 includes a CPU that controls the operation of each part of the crack depth detection device 100, a memory (e.g., ROM) that stores the control program, and RAM that can be used as a working area for the CPU.

[0062] (Optical system for interface detection) Next, we will explain the detection of the interface of the workpiece W. In the following explanation, we will describe the case in which the interface of the back surface of the workpiece W (the surface in contact with the stage ST(XYθ)) is detected and the depth position of the crack K is detected based on the interface position on the back surface of the workpiece W. Note that in detecting the depth position of the crack K, the surface of the workpiece W may be used as the reference, or the average value of the depth positions of the crack K detected based on both the front and back surfaces of the workpiece W may be taken.

[0063] The interface detection optical system 160 is an optical system for detecting the interface (front or back surface) of the workpiece W, and includes a half mirror 162, a relay lens 164, a half mirror 166, and a photodetector 168.

[0064] When detecting the interface of the workpiece W, the detection signal processing unit 102 irradiates the workpiece W with the detection light L1 while the limiting member 126 is moved out of the optical path of the detection light L1.

[0065] The half mirror 162 transmits the detection light L1 incident from the illumination optical system 140 to the focusing lens 106 side and reflects the reflected light L2 from the workpiece W. The reflected light L2 from the workpiece W is reflected by the half mirror 162, bending its optical path and guiding it to the relay lens 164. The reflected light L2 that has passed through the relay lens 164 is reflected by the half mirror 166 and guiding it to the photodetector 168.

[0066] The photodetector 168 is a device for detecting the interface of the workpiece W by receiving reflected light L2 from the workpiece W, and includes a detector body 168A and a pinhole panel 168B.

[0067] As the detector body 168A, a photodetector (for example, a photodiode) that converts the received light into an electrical signal and outputs it to the detection signal processing unit 102 can be used.

[0068] The pinhole panel 168B has pinholes formed in it to allow a portion of the incident light to pass through. The pinhole panel 168B is positioned on the light-receiving surface side of the detector body 168A so that the pinholes are on the optical axis of the reflected light L2.

[0069] The interface detection optical system 160 is an optical system in which the position of the pinhole in the pinhole panel 168B is optically conjugate to the position of the focal point of the condensing lens 106.

[0070] The detection signal processing unit 102 moves the limiting member 126 out of the optical path of the detection light L1 and irradiates the workpiece W with the detection light L1 while relatively moving the focusing lens 106 and the stage ST(XYθ) in the Z direction. The detection signal processing unit 102 detects the reflected light L2 when the detection light L1 is focused on the interface (front or back surface) of the workpiece W and calculates the position of the interface of the workpiece W in the Z direction from the position of the focusing point at that time.

[0071] (Optical system for crack detection) Next, we will explain how to detect a crack K formed inside the workpiece W.

[0072] The crack detection optical system 180 includes a relay lens 182, a half mirror 184, and photodetectors 186 and 188.

[0073] When detecting a crack K formed inside the workpiece W, the detection signal processing unit 102 inserts a limiting member 126 into the optical path of the detection light L1. Here, the detection signal processing unit 102, the limiting member 126, and the crack detection optical system 180 each function as part of the crack detection means. The limiting member 126 is provided with two selectively openable and closable openings 126A and 126B at positions offset from the principal optical axis AX, and the light from the detection light L1 that passes through either opening 126A or 126B is irradiated onto the workpiece W. In the example shown in Figure 6, the two openings 126A and 126B are, for example, positioned symmetrically to each other with respect to the principal optical axis AX. As a result, the detection light (second detection light) L1, which is eccentric with respect to the principal optical axis AX, is irradiated onto the workpiece W.

[0074] Here, Figure 6 shows the detection light transmitted through openings 126A and 126B superimposed when detecting the depth position of crack K. In reality, the depth position of crack K is detected twice by opening one of the openings 126A and 126B and closing the other. The detection signal processing unit 102 then calculates the average value of the depth positions of crack K detected using the detection light L1 transmitted through openings 126A and 126B as the depth position of crack K.

[0075] The detection of the depth position of the crack K, performed by opening one of the openings 126A and 126B and closing the other, will be explained below with reference to Figures 6 to 13.

[0076] When detection light L1, which has passed through one of the apertures 126A and 126B, is irradiated onto the workpiece W, the reflected light (second reflected light) L2 from the workpiece W is reflected by the half mirror 162, then sequentially passes through the relay lens 164 and the half mirror 166 before entering the relay lens 182. The reflected light L2 that has passed through the relay lens 182 is received by the photodetectors 186 and 188 via the half mirror 184. Here, the transmittance and reflectance of the light incident on the half mirror 184 are assumed to be approximately 50% each.

[0077] Photodetectors 186 and 188 are devices for detecting cracks K inside the workpiece W by receiving reflected light L2 from the workpiece W. Photodetector 186 includes a detector body 186A and a pinhole panel 186B, and photodetector 188 includes a detector body 188A and a pinhole panel 188B.

[0078] As the detector bodies 186A and 188A, photodetectors (for example, photodiodes) that convert received light into electrical signals and output them to the detection signal processing unit 102 can be used.

[0079] Pinhole panels 186B and 188B have pinholes formed in them to allow a portion of the incident light to pass through. Pinhole panels 186B and 188B are positioned on the light-receiving side of the detector bodies 186A and 188A, respectively. Pinhole panels 186B and 188B are positioned such that the pinholes are offset from the optical axis of the reflected light L2.

[0080] Figures 7 to 9 are explanatory diagrams showing the situation when the workpiece W is illuminated by the detected light L1. Figure 7 shows the case when a crack K exists at the focal point of the focusing lens 106, Figure 8 shows the case when there is no crack K at the focal point of the focusing lens 106, and Figure 9 shows the case when the focal point of the focusing lens 106 coincides with the depth position (lower end position) of the crack K. Figures 10 to 12 show the reflected light L2 received by the photodetectors 186 and 188, respectively, corresponding to the cases shown in Figures 7 to 9. Figure 13 is a diagram to explain the path of the reflected light L2 from the workpiece W to the pupil 106a of the focusing lens. Here, we will explain the case when the detected light L1 passes through the first region G1 on one side of the pupil 106a of the focusing lens (right side in Figure 13) and illuminates the workpiece W.

[0081] As shown in Figure 7, if a crack K exists at the focal point of the focusing lens 106, the detected light L1 undergoes total internal reflection at the crack K, and the reflected light L2 follows the same path as the optical path of the detected light L1 with respect to the principal optical axis AX, becoming a component that reaches the same region of the focusing lens pupil 106a as the detected light L1. That is, as shown in Figure 8, when the detected light L1 from the light source 122 is irradiated onto the workpiece W via the focusing lens 106, if the path of the detected light L1 is R1, the reflected light L2 that undergoes total internal reflection at the crack K inside the workpiece W follows a path R2 that is on the same side (right side in Figure 8) as the path R1 of the detected light L1 with respect to the principal optical axis AX, and passes through the first region G1 of the focusing lens pupil 106a.

[0082] As shown in Figure 8, if there is no crack K at the focal point of the focusing lens 106, the detected light L1 is reflected from the back surface of the workpiece W, and the reflected light L2 becomes a component that reaches the region of the focusing lens pupil 106a opposite to the detected light L1. That is, as shown in Figure 13, the reflected light L2 reflected from the back surface of the workpiece W follows a path R3 that is opposite to the path R1 of the detected light L1 with respect to the principal optical axis AX (left side in Figure 13) and passes through the second region G2 of the focusing lens pupil 106a.

[0083] As shown in Figure 9, when the focal point of the focusing lens 106 coincides with the lower end position of the crack K, the detected light L1 is divided into a reflected light component L2a that undergoes total internal reflection at the crack K and reaches the same region of the focusing lens pupil 106a as the detected light L1, and a non-reflected light component L2b that is not totally reflected at the crack K but is reflected off the back surface of the workpiece W and reaches the region of the focusing lens pupil 106a opposite to the detected light L1. In other words, as shown in Figure 13, of the reflected light L2, the reflected light component L2a that undergoes total internal reflection at the crack K inside the workpiece W follows path R2, which is on the same side as the path R1 of the detected light L1 with respect to the principal optical axis AX (right side in Figure 13), and passes through the first region G1 of the focusing lens pupil 106a. The unreflected light component L2b that is not totally internally reflected at the crack K and is reflected on the back surface of the workpiece W follows path R3, which is on the opposite side of the path R1 of the detected light L1 with respect to the principal optical axis AX (left side in Figure 13), and passes through the second region G2 of the focusing lens pupil 106a.

[0084] The pinhole panels 186B and 188B are positioned such that the pinholes are optically conjugate to the first region G1 and the second region G2 of the focusing lens pupil 106a, respectively. This allows the detector body 186A and the detector body 188A to selectively receive light that has passed through the first region G1 and the second region G2 of the focusing lens pupil 106a, respectively.

[0085] In the example shown in Figure 7 (where a crack K exists at the focal point of the focusing lens 106), as shown in Figure 10, among the detector bodies 186A and 188A, reflected light L2 is received by the light-receiving surface 186C of detector body 186A, and the level of the detection signal output from the light-receiving surface 186C becomes higher than the level of the detection signal output from the light-receiving surface 188C of detector body 188A.

[0086] On the other hand, in the example shown in Figure 8 (where there is no crack K at the focal point of the focusing lens 106), as shown in Figure 11, among the detector body 186A and the detector body 188A, reflected light is received by the light-receiving surface 188C of detector body 188A, and the level of the detection signal output from the light-receiving surface 188C becomes higher than the level of the detection signal output from the light-receiving surface 186C.

[0087] Furthermore, in the example shown in Figure 9 (where the focal point of the condensing lens 106 coincides with the lower end position of the crack K), as shown in Figure 12, the respective components L2a and L2b of the reflected light L2 are received by the light-receiving surfaces 186C and 188C, respectively, and the levels of the detection signals output from the light-receiving surfaces 186C and 188C become approximately equal.

[0088] Thus, the amount of light received by the light-receiving surfaces 186C and 188C changes depending on whether or not a crack K exists at the focusing point of the focusing lens 106. In this embodiment, this property can be used to detect the depth position (lower end position of the crack or upper end position of the crack) of a crack K formed inside the workpiece W.

[0089] Specifically, when the detection signals output from the light-receiving surfaces 186C and 188C are denoted as D1 and D2, respectively, the evaluation value S, which shows the relationship between the focal point of the condensing lens 106 and the depth position of the crack K, can be expressed by the following equation.

[0090] S = (D1 - D2) / (D1 + D2) ... (1) In equation (1), when the condition S=0 is met, that is, when the amount of light received by the light-receiving surfaces 186C and 188C is the same, it indicates that the focusing point of the focusing lens 106 coincides with the lower end position of the crack (or the upper end position of the crack).

[0091] The detection signal processing unit 102 controls the alignment mechanism 104 to sequentially change the focusing point of the focusing lens 106 from the interface position Z(0) on the back surface to the thickness direction (Z direction) of the workpiece W, while sequentially acquiring detection signals output from the light-receiving surfaces 186C and 188C. Based on these detection signals, it calculates the evaluation value S shown in equation (1), and by evaluating this evaluation value S, it is possible to detect the depth position of the crack K (crack lower end position or crack upper end position).

[0092] Next, the opening to be opened is changed, and the depth position of crack K is detected. The average of the two detected depth positions of crack K is then calculated as the depth position of crack K.

[0093] In this embodiment, even if there is a misalignment (a misalignment in the direction perpendicular to the processing direction (Y direction)) between the optical axis (principal optical axis AX) of the focusing lens 106 and the crack K, the effect of this misalignment can be canceled out and the depth position of the crack K can be accurately detected by using two detection lights L1 irradiated from symmetrical positions (eccentric positions) on either side of the principal optical axis AX. Furthermore, since the planar position of the crack K can be measured simultaneously (in parallel) with the depth position of the crack K by confirming the dicing line position, it is possible to accurately acquire three-dimensional information of the crack K in combination with the effect of detecting the depth position of the crack K using the two detection lights L1 described above.

[0094] In this embodiment, the depth position of the crack K is detected twice using the detection light L1 transmitted through the openings 126A and 126B. However, the depth position of the crack K may be detected only once using only one of the openings.

[0095] Furthermore, by using a polarizer, the depth position of the crack K may be detected simultaneously using the detection light L1 transmitted through apertures 126A and 126B (see, for example, Japanese Patent Application Publication No. 2021-143873).

[0096] In this embodiment, photodetectors were used as the detector bodies 168A, 186A, and 188A, but the present invention is not limited thereto. For example, an infrared camera or the like may be used instead of a photodetector.

[0097] Furthermore, when detecting cracks K, the half-mirror 166 may be moved out of the optical path of the reflected light L2. In this case, the half-mirror 166 of the interface detection optical system 160 may be replaced with a total reflection mirror or a dichroic mirror, etc.

[0098] Furthermore, in this embodiment, the transmittance and reflectance of the light incident on the half mirror 184 are set to approximately 50%, but the present invention is not limited thereto. If the transmittance and reflectance of the half mirror 184 are different, when calculating the evaluation value S, the outputs D1 and D2 of the detection signals from the photodetectors 186 and 188 may be multiplied by weighting coefficients corresponding to the transmittance and reflectance of the half mirror 184.

[0099] In this embodiment, two openable and closable openings 126A and 126B are provided, but the configuration for biased illumination is not limited to the above. For example, three or more openings may be formed in the limiting member 126, and there may be three or more methods of biased illumination. Alternatively, for example, only one opening may be provided in the limiting member 126, and the method of biased illumination may be changed by rotating this limiting member around the principal optical axis AX.

[0100] Furthermore, in this embodiment, biased illumination was performed by blocking a portion of the detected light L1 using a limiting member 126, but the present invention is not limited thereto. For example, biased illumination may be performed by arranging the light source 122 of the light source unit 120 offset from the principal optical axis AX. In this case, it is preferable that the optical axis of the light source 122 is parallel to the principal optical axis AX. In the second embodiment, the light source 122 is provided at multiple positions offset from the principal optical axis AX. That's good too.

[0101] Furthermore, although the crack detection optical system 180 in this embodiment is equipped with two photodetectors, the present invention is not limited thereto. For example, instead of the half-mirror 184 and photodetectors 186 and 188, a two-part photodetector with a light-receiving surface divided into two may be provided, and the evaluation value S may be calculated from the output of each light-receiving surface. Alternatively, a photodetector in which the light-receiving surface is divided into a number of parts corresponding to the type of illumination method may be used.

[0102] Furthermore, the configuration for guiding the detected light L1 and reflected light L2 is merely an example and is not limited to this embodiment. For example, it is possible to arrange the light source 122, the focusing lens 106, and the stage ST(XYθ) on which the workpiece W is placed in a straight line without providing the mirror 144.

[0103] (Crack detection method) Figure 14 is a flowchart showing the operation of a crack detection device according to the first embodiment of the present invention.

[0104] First, the stage control unit 14 controls the stage ST(XYθ) and the Z stage S(Z) to move the crack depth detection device 100 and the crack plane position detection device (optical system for confirming the dicing line position) 200 to the crack measurement position (step S10).

[0105] Next, at the crack measurement location, an observation image DAT2 of the crack plane position and received light data DAT1 of the crack depth are acquired (step S12) and transmitted to the control device 10 (step S14).

[0106] Next, the control device 10 creates three-dimensional information of the crack (three-dimensional data indicating the position of crack K) using the observation image DAT2 of the crack plane position and the received light data DAT1 of the crack depth (step S16).

[0107] By using the 3D data indicating the location of the crack K created as described above, it is possible to accurately predict the quality of the chip's machining before the workpiece is fractured.

[0108] [Second Embodiment] Next, a second embodiment of the present invention will be described. In the following description, components similar to those in the first embodiment will be denoted by the same reference numerals and their descriptions will be omitted.

[0109] The crack detection device 1A according to this embodiment shares some of the optical elements of the crack depth detection device 100 and the dicing line position confirmation optical system 200.

[0110] Figures 15 and 16 are block diagrams showing a crack detection device according to a second embodiment of the present invention. Figure 15 shows the optical path when detecting the depth of a crack K, and Figure 16 shows the optical path when detecting the planar position of a crack K. For simplicity, the interface detection optical system 160 is omitted in Figures 15 and 16.

[0111] The crack detection device 1A according to this embodiment includes an illumination optical system 212A for confirming the position of the dicing line and an imaging optical system 212B for confirming the position of the dicing line.

[0112] The illumination optical system 212A for dicing line position confirmation includes a relay lens 220 and a half mirror 222. The half mirror 222 is positioned between the relay lens 142 and the half mirror 144A of the illumination optical system 140 for crack depth detection.

[0113] The imaging optical system 212B for confirming the position of the dicing line includes a half mirror 144A, a relay lens 146, and a focusing lens 106. The half mirror 144A, relay lens 146, and focusing lens 106 of the illumination optical system 140 for detecting crack depth are shared with the imaging optical system 212B for confirming the position of the dicing line.

[0114] During dicing line position confirmation, as shown in Figure 16, illumination light L21 emitted from light source 210 reaches half mirror 222 via relay lens 220, is sequentially reflected by half mirror 222 and half mirror 144A, and focused onto workpiece W by condensing lens 106. The reflected light L22 from workpiece W then passes through condensing lens 106, half mirror 162, relay lens 146, and half mirror 144A, and can be imaged by camera 218. This allows the planar position of crack K to be detected from the observation image captured by camera 218.

[0115] On the other hand, when detecting the depth of the crack K, as shown in Figure 15, the detection light L1 passes through the half mirror 222 of the illumination optical system 212A for confirming the dicing line position, is reflected by the half mirror 144A, and reaches the workpiece W. This makes it possible to detect the depth of the crack K, similar to the first embodiment.

[0116] According to this embodiment, by sharing some of the optical elements that constitute the crack detection device, the number of components can be reduced and the device can be made smaller.

[0117] Furthermore, according to this embodiment, since the planar position and depth position of the crack K can be detected using the same focusing lens 106, three-dimensional information of the crack K can be obtained more easily.

[0118] In this embodiment, a light source 122 for detecting the depth of the crack K and a light source 210 for detecting the planar position of the crack K are provided separately, but the present invention is not limited thereto. For example, the light source 210 for detecting the planar position of the crack K may be omitted and shared with the light source 122 for detecting the depth of the crack K. In this case, when detecting the planar position of the crack K (when confirming the dicing line position), the limiting member 126 can be moved out of the optical path. [Explanation of Symbols]

[0119] 1, 1A…Crack detection device, 10…Control device, 12…Crack measurement control unit, 14…Stage control unit, 100…Crack depth detection device, 200…Crack plane position detection device (optical system for dicing line position confirmation)

Claims

1. A crack plane position detection means for detecting the planar position of a crack formed inside a workpiece by focusing a first detection light onto the workpiece and imaging the first reflected light from the workpiece, A crack depth detection means for detecting the depth position of a crack by focusing a second detection light into the workpiece and imaging the second reflected light from the workpiece, wherein the means is movable integrally with the crack plane position detection means along the machining direction of the workpiece, and is positioned such that when moved in the machining direction, the measurement point where the second detection light is focused falls within the field of view of the crack plane position detection means, thereby enabling the detection of the plane position of the crack and the detection of the depth position of the crack in parallel. A crack detection device equipped with the following features.

2. The crack detection device according to claim 1, wherein when the crack depth detection means is moved in the direction of machining, the measurement point where the second detection light is focused is arranged coaxially with respect to the center of the field of view of the crack plane position detection means along the direction of machining.

3. The crack detection device according to claim 1 or 2, wherein an optical element included in the crack plane position detection means and a portion of the optical elements included in the crack depth detection means are shared.

4. The crack detection device according to claim 3, wherein the focusing lens for focusing the first detection light onto the workpiece and the focusing lens for focusing the second detection light into the interior of the workpiece are shared.

Images