processing device

Abstract

The processing device includes: a processing laser irradiation device that irradiates a processing laser while moving in a traveling direction along a plurality of scribe lanes extending in at least one direction of a wafer; a coaxial height detection section that has the same optical axis as the processing laser, irradiates a detection laser to a surface of the wafer, and detects a height position of the surface of the wafer based on reflected light of the detection laser; a front height detection section located in front of the processing laser irradiation device; and a rear height detection section located in the rear of the processing laser irradiation device opposite to the front; and the front height detection section and the rear height detection section are movable in a direction in which the plurality of scribe lanes are arranged in parallel, and detect a height position of a scribe lane.

Description

Technical Field

[0001] This invention relates to a processing apparatus. This application claims priority based on Japanese Patent Application No. 2024-045193, filed on March 21, 2024, the contents of which are incorporated herein by reference. Background Technology

[0002] In the past, in order to divide wafers with semiconductor devices or electronic components formed on their surface into individual chips, a dicing processing device was used. Dicing is a process of forming grinding grooves on the wafer by using a thin whetstone containing fine diamond abrasive grains with a thickness of about 30 μm to cut the wafer.

[0003] In this processing apparatus, a dicing blade is rotated at a high speed of, for example, 30,000 to 60,000 rpm to grind the wafer, so as to completely cut off (full cut) or incompletely cut off (half cut or partial cut).

[0004] In recent years, in order to replace the previous cutting method using dicing blades, a processing apparatus has been proposed that aligns a focal point with a laser and incident a laser beam into the interior of the wafer, forming a modified region inside the wafer, thereby dividing it into individual chips (for example, see Patent Document 1). In this technology, in order to ensure that the modified region formed inside the wafer is formed at a certain depth from the wafer surface, an autofocusing mechanism is required to detect the height position (position in the thickness direction) of the wafer surface, thereby controlling the focal point position of the laser with high precision.

[0005] In the technology disclosed in Patent Document 1, in order to uniformly form a modified region (modified layer) at a specified depth inside the wafer, a detection laser (autofocus (AF) laser) is irradiated onto the wafer surface, and the height position of the wafer surface is detected based on its reflected light, thereby performing processing while maintaining a certain processing position.

[0006] Patent Document 1: Japanese Patent Application Publication No. 2016-139726

[0007] Because the AF laser travels along the same optical axis as the processing laser, the surface height position of the wafer can be detected approximately simultaneously with the wafer's processing point, enabling accurate processing at the designated processing point. However, when the processing speed is increased, or when the height position of the wafer surface varies significantly and the AF laser's tracking performance is insufficient, the detector that detects the AF laser will experience a delay in responding to this variation, potentially failing to detect the accurate height position of the wafer surface.

[0008] Here, by placing a detector, such as a displacement sensor, ahead of the processing travel direction, the surface height position of the wafer can be detected before processing. Therefore, by detecting the unevenness of the wafer surface beforehand and utilizing the detection results in height position detection using an AF laser, the response delay of the detector used for AF laser detection can be reduced. However, in this case, since the wafer surface height position is not detected at the time of wafer processing, it becomes difficult to process the wafer in a manner that corresponds to changes in the wafer's state or other characteristics during the period from the initial wafer surface detection to the wafer processing time. Summary of the Invention

[0009] The present invention was made in view of this situation, and its object is to provide a processing apparatus capable of accurately processing wafers.

[0010] <1> An embodiment of the present invention relates to a processing apparatus comprising: a processing laser irradiation device that irradiates a processing laser while moving in a traveling direction along a plurality of scribe lines extending in at least one direction of a wafer; a coaxial height detection unit having the same optical axis as the processing laser that irradiates a detection laser onto the surface of the wafer and detects the height position of the wafer surface based on the reflected light of the detection laser; a front height detection unit located in front of the processing laser irradiation device; and a rear height detection unit located behind the processing laser irradiation device on the opposite side from the front. The front height detection unit is movable in a parallel direction of the plurality of scribe lines to detect the height position of the scribe lines before processing by the processing laser irradiation device, and the rear height detection unit is movable in a parallel direction of the plurality of scribe lines to detect the height position of the scribe lines after processing by the processing laser irradiation device.

[0011] <2> According to the above <1> The processing apparatus described above can use the height information obtained through the aforementioned front height detection unit to correct the height position of the wafer surface.

[0012] <3> According to the above <1> or <2> The processing apparatus can determine the surface condition of the dicing track after processing based on the height information obtained by the front height detection unit and the height information obtained by the rear height detection unit.

[0013] <4> According to the above <1> to <3> The processing apparatus described in any one of the above descriptions can perform anomaly detection on the surface condition of the wafer.

[0014] The processing apparatus according to the present invention is capable of accurately processing wafers. Attached Figure Description

[0015] Figure 1 This is a schematic structural diagram of the processing apparatus involved in the embodiments of the present invention.

[0016] Figure 2 This is a schematic structural diagram of the processing apparatus involved in the same embodiment.

[0017] Figure 3 It is a 3D diagram of a wafer used to illustrate the dicing paths and processing directions on the wafer.

[0018] Figure 4 This is a functional block diagram of the control unit.

[0019] Figure 5 This is a graph showing the output characteristics of the AF signal.

[0020] Figure 6 This is a schematic diagram showing the situation where the front height detection unit performs anomaly detection.

[0021] Figure 7 This is a schematic diagram illustrating an example of the positional relationship between a coaxial AF device, a front height detection unit, and a rear height detection unit when observing a wafer from the positive Z-axis direction.

[0022] Figure 8 This is a schematic diagram illustrating an example of the positional relationship between a coaxial AF device, a front height detection unit, and a rear height detection unit when viewed from a direction perpendicular to the extension direction of the dicing track on the wafer surface.

[0023] Figure 9 This is a schematic diagram illustrating an example of the positional relationship between the coaxial AF device, the front height detection unit, and the rear height detection unit when viewed from the direction of the grading track's extension.

[0024] Figure 10 This is a schematic diagram illustrating an example of the positional relationship between the coaxial AF device, the front height detection unit, and the rear height detection unit.

[0025] Figure 11 This is a schematic diagram illustrating an example of wafer surface height position detection using a coaxial AF device.

[0026] Figure 12 It is a timing diagram showing the operation flow of the processing device.

[0027] Figure 13 It is a schematic diagram used to illustrate the operation flow of the processing device, showing the predetermined processing point and the processing point.

[0028] Figure 14 This is an example of the wafer height position detection result of the coaxial AF device in this embodiment.

[0029] Figure 15 This is an example of the wafer height position detection results from an existing coaxial AF device. Detailed Implementation

[0030] Hereinafter, a processing apparatus according to an embodiment of the present invention will be described with reference to the accompanying drawings. In the XYZ directions shown in each figure, the direction indicated by the + direction is the positive direction, and the direction indicated by the - direction is the negative direction.

[0031] Figure 1 This is a schematic structural diagram of the processing apparatus according to an embodiment of the present invention. Figure 1 As shown, the processing device 1 includes a stage 2, a laser irradiation device 3 for processing, a coaxial height detection unit 4, a front height detection unit 5, a rear height detection unit 6, and a control unit 7.

[0032] The stage 2 is movable in the XY direction and holds the wafer W. During wafer W processing, the stage 2 moves in the XY direction while the wafer W is mounted.

[0033] The processing laser irradiation device 3 irradiates the wafer W with a processing laser while moving along multiple dicing tracks extending in at least one direction in the travel direction. For example... Figure 2 As shown, the processing laser irradiation device 3 includes a processing laser source 31, a dichroic mirror 32, a condenser lens 33, and a condenser lens driving unit 34. On the optical path of the processing laser L1, i.e., the first optical path OP1, the processing laser source 31, the dichroic mirror 32, and the condenser lens 33 are arranged in the order of these components, with the condenser lens 33 positioned closest to the wafer W.

[0034] The processing laser source 31 emits a processing laser L1 for forming a modified region inside the wafer W. For example, the processing laser source 31 emits a pulse width of less than 1 μs and a peak power density of 1 × 10⁻⁶ at the focal point. 8 (W / cm) 2 The laser pulse width and peak power density of the laser described above are merely examples, and are not particularly limited as long as the desired modified region can be formed. The processing laser L1 emitted from the processing laser source 31 is transmitted through the dichroic mirror 32 and focused into the interior of the wafer W by the focusing lens 33. The Z-direction position of the focusing point of the processing laser L1 (the thickness direction position of the wafer W) is adjusted by slightly moving the focusing lens 33 in the Z-direction (the optical axis direction of the processing laser L1) by the focusing lens driving unit 34. In addition, the dichroic mirror 32 reflects the AF detection light L2 emitted from the coaxial AF device 4 described later.

[0035] The condenser lens driving unit 34 is driven based on the processing conditions set by the control unit 7, so that the distance between the condenser lens 33 and the surface of the wafer W is constant. As a result, the condenser lens 33 moves slightly in the Z direction (the thickness direction of the wafer) to follow the displacement of the wafer W surface, so that the focusing point of the processing laser L1 is located at a certain distance (depth) from the surface of the wafer W. Consequently, a modified region R can be formed at a desired location inside the wafer W. The condenser lens driving unit 34 is, for example, an actuator.

[0036] In this embodiment, "processing" refers to the formation of a modified region. A modified region R is formed at a location within the dicing channel where a processing laser L1 is irradiated.

[0037] The coaxial AF device (coaxial height detection unit) 4 has the same optical axis as the processing laser, irradiates the surface of the wafer W with the detection laser, and detects the height position of the wafer surface based on the reflected light from the detection laser. Figure 3 As shown, the coaxial AF device 4 illuminates the surface of wafer W with a detection light (AF detection light) L2, and detects the height position of the wafer W surface based on its reflected light. The optical path of the AF detection light L2, i.e., the second optical path OP2, is deflected by a dichroic mirror 32 and shares a portion of the optical path with the first optical path OP1 of the processing laser L1. A focusing lens 33 is arranged on this shared optical path. Therefore, before the detection light L2 reaches wafer W, the optical axis of the AF detection light L2 is the same as the optical axis of the processing laser L1. The coaxial AF device 4 moves together with the processing laser irradiation device 3 in the processing travel direction to detect the height position of the wafer W surface.

[0038] Here, the so-called "processing direction" within a single lane refers to, for example, Figure 3 As indicated by arrow Dx, with reference to the processing point P having the modified region R already formed on wafer W, the processing apparatus 1 travels in the direction of the unprocessed region (towards the predetermined processing point P'). The processing direction within a single dicing track is referred to as the X-direction travel direction. When processing multiple dicing tracks parallel to each other on the surface of wafer W, when processing wafer W across these multiple dicing tracks, as shown... Figure 3 As indicated by arrow Dy, the direction in which the processing device 1 travels towards another dicing lane, where the processing point P has already been formed, is sometimes referred to as the "processing travel direction," relative to the dicing lane where the processing point P has already been formed. In this case, the travel direction between adjacent dicing lanes is called the Y-direction travel direction. That is, Figure 3 The direction of the middle arrow Dx is the X-direction of travel, and the direction of the arrow Dy is the Y-direction of travel.

[0039] like Figure 2As shown, the coaxial AF device 4 includes an AF light source 41, a collimating lens 42, a knife edge 43, a focusing optical system 44, a moving lens drive unit 45, a semi-transparent mirror 46, an imaging lens 47, and a detector 48. The coaxial AF device 4 uses the so-called knife edge method to detect the height position of the wafer W surface.

[0040] The AF light source 41 is the light source that emits the AF detection light L2, and is composed of a laser source such as a laser diode (LD) light source or a super luminescent diode (SLD) light source. The AF detection light L2 has a wavelength that is different from the processing laser L1 and can be reflected by the surface of the wafer W.

[0041] like Figure 2 As shown, the AF detection light L2 emitted from the AF light source 41 is collimated by the collimating lens 42, and a portion of it is blocked by the knife edge 43. Furthermore, the light that is not blocked by the knife edge 43 is reflected by the semi-transparent mirror 46, then reflected by the dichroic mirror 32, and focused by the condenser lens 33 to illuminate the wafer W.

[0042] The focusing optical system 44 is positioned independently of the shared optical path OP2 of the AF detection light L2 and the first optical path OP1 of the processing laser L1. Specifically, it is disposed between the dichroic mirror 32 and the semi-transparent mirror 46 on the AF light reflection path 244. The AF light reflection path 244 is used to guide the reflected light of the AF detection light L2, reflected from the surface of the wafer W, to the detector 48.

[0043] The focusing optical system 44 comprises at least a plurality of lenses, including a movable lens 442 configured to move along the second optical path OP2, and adjusts the focusing point of the AF detection light L2 independently of the focusing point of the processing laser L1 in the Z direction. Starting from the wafer W side, the focusing optical system 44 sequentially includes a fixed lens 441 that is not movable along the second optical path OP2 and a movable lens 442 that is movable along the second optical path. In this embodiment, the fixed lens 441 is a positive lens, and the movable lens 442 is a negative lens.

[0044] The moving lens driving unit 45 moves the moving lens 442 along the second optical path OP2. When the moving lens 442 moves along the second optical path OP2, with the Z-direction position of the focusing point of the processing laser L1 fixed, the Z-direction position of the focusing point of the AF detection light L2 changes according to the moving direction and amount of the moving lens 442. That is, the relative distance between the focusing point of the processing laser L1 and the focusing point of the AF detection light L2 changes.

[0045] The reflected light from the surface of wafer W, the detection light L2 of AF, is refracted by the condenser lens 33, reflected by the dichroic mirror 32, and transmitted through the focusing optical system 44 and the semi-transparent mirror 46. Then, the reflected light is focused by the imaging lens 47 and illuminates the detector 48, forming a focused image on the light-receiving surface of the detector 48.

[0046] The detector 48 is composed of a photodiode divided into two parts, which has a light-receiving element (photoelectric conversion element) divided into two parts. In detail, the detector 48 divides and receives the focused image of the reflected light of the AF detection light L2, and outputs an output signal (electrical signal) corresponding to the light amount of each part to the AF signal processing unit 74 described later.

[0047] The wafer W surface displacement detection method using a photodiode divided into two parts can be performed using known methods.

[0048] As described above, the coaxial AF device 4 is structured such that a focusing lens 33 is arranged on the shared optical path of the first optical path OP1 of the processing laser L1 and the second optical path OP2 of the AF detection light L2. In this structure, when the relative distance between the focusing lens 33 and the wafer W changes in order to change the processing depth of the modified region, not only does the focusing point of the processing laser L1 change in the Z-direction position relative to the wafer W, but the focusing point of the AF detection light L2 also changes in the Z-direction position relative to the wafer W.

[0049] In this way, even if the relative distance between the condenser lens 33 and the wafer W changes, by moving the movable lens 442, the condenser point of the processing laser L1 can be made to align with the upper surface Wb of the wafer W while keeping the Z-direction position of the condenser point of the AF detection light L2 fixed.

[0050] The forward height detection unit 5 is a displacement sensor that detects the surface height position of the wafer W, specifically the area on the wafer W to be processed. The forward height detection unit 5 is located ahead of the processing travel direction (X-direction and / or Y-direction travel direction). The forward height detection unit 5 detects the surface height position of the wafer W earlier than the processing of the wafer W. In particular, when the forward height detection unit 5 is located away from the coaxial AF device 4 in the Y-direction travel direction, it can detect the surface height position of the dicing track to be processed in advance, i.e., the surface shape of the dicing track.

[0051] Furthermore, the front height detection unit 5 is movable relative to the coaxial AF device 4 in at least either the X or Y direction, in front of the coaxial height detection unit 4. Details will be described later. Because it is movable relative to the coaxial AF device 4 in at least either the X or Y direction, the surface height position of the wafer W before processing can be detected at a position corresponding to the user's requirements.

[0052] The specific structure of the front height detection unit 5 is not particularly limited. For example, the front height detection unit 5 can be a displacement sensor using a triangulation method or a white confocal sensor. In particular, when the front height detection unit 5 is a white confocal sensor, it is less susceptible to the influence of films or the like formed on the surface of the wafer W, thereby enabling stable detection of the height position on the surface of the wafer W. Furthermore, the front height detection unit 5 can also be a displacement detection device that uses an AF mechanism similar to the coaxial AF device 4. In this case, the front height detection unit 5 has an optical axis different from that of the processing laser L1 and is located in front in the processing travel direction of the wafer W. In addition, the AF detection light provided by the front height detection unit 5 can be subjected to averaging processing, etc., and can also have a focusing spot diameter wider than that of the AF light source of the coaxial AF device 4.

[0053] The rear height detection unit 6 is a displacement sensor that detects the surface height position of the wafer W, specifically the surface height position of the processed area within the wafer W. The rear height detection unit 6 differs from the front height detection unit 5 in that it is located behind the processing travel direction on the opposite side from the front.

[0054] Furthermore, the rear height detection unit 6 is movable relative to the coaxial AF device 4 in at least either the X or Y direction behind the coaxial height detection unit 4. Details will be described later. Because it is movable relative to the coaxial AF device 4 in at least either the X or Y direction, the surface height position of the processed wafer W can be detected at a position corresponding to the user's requirements.

[0055] The specific structure of the rear height detection unit 6 is not particularly limited. For example, the rear height detection unit 6 can be a displacement sensor using a triangulation method or a white confocal sensor. Furthermore, the rear height detection unit 6 can also be a displacement detection device that uses an AF mechanism similar to the coaxial AF device 4. In this case, the rear height detection unit 6 has an optical axis different from the processing laser L1 and is located rearward in the processing direction of the wafer W. Furthermore, the AF detection light provided by the rear height detection unit 6 can be subjected to averaging processing, and can also have a wider focusing spot diameter than the AF light source of the coaxial AF device 4. Since the length measurement method or state can be changed independently, it can be configured to have a different length measurement method or state than the front height detection unit 5. For example, in the rear height detection unit 6, by narrowing the beam spot to measure the length of a small area and making it sensitive to small changes, it is possible to determine whether a half-cut crack has occurred after processing.

[0056] The control unit 7, composed of a CPU, memory, input / output circuits, etc., controls various operations of each part of the machining apparatus 1. For example, the control unit 7 controls the operations of the stage 2, the machining laser irradiation device 3, the coaxial height detection unit 4, the front height detection unit 5, and the rear height detection unit 6. The control unit 7 includes a stage control unit 71, a machining condition setting unit 72, a coaxial AF setting unit 73, an AF signal processing unit 74, a detection condition setting unit 75, a status determination unit 76, and an abnormality detection unit 77.

[0057] The stage control unit 71 moves the stage 2, which is in the state of holding the wafer W, in the XY direction. The stage 2 moves in conjunction with the operation of the processing laser irradiation device 3, the coaxial height detection unit 4, the front height detection unit 5, and the rear height detection unit 6.

[0058] The processing condition setting unit 72 sets the processing conditions when the processing laser irradiation device 3 processes the wafer W. The processing conditions are conditions set for the processing device 1 to process the wafer W, and include, for example, the number and position of the dicing traces, the formation position (depth) of the modified region, the output intensity and pulse width of the processing laser source, and the stage movement speed.

[0059] The coaxial AF setting unit 73 sets parameters such as the output intensity of the laser emitted from the AF light source 41. Furthermore, the coaxial AF setting unit 73 sets detection conditions for the detection light L2 used for AF detection by the detector 48. These detection conditions include at least one of the AF application range on the wafer surface and the AF gain setting value. The AF gain setting value refers to the value of the AF gain set for the height position of the wafer W surface at each sampling point of the detection light L2 detected by the detector 48. By appropriately setting the AF gain value, the detector 48 can appropriately detect the height position of the wafer W surface for a continuously moving wafer W, thereby suppressing detection delays.

[0060] In addition, the coaxial AF setting unit 73 controls the drive of the moving lens drive unit 45, so that the moving lens 442 of the focusing optical system 44 moves along the second optical path OP2, thereby making the focusing point of the AF detection light L2 coincide with the upper surface Wb of the wafer W (specifically, making the output of the AF signal zero).

[0061] The AF signal processing unit 74 generates an AF signal E, AF data, or AF information, which are displacement signals (detection signals), based on the output signals from each light-receiving element of the detector 48. This displacement signal represents the displacement (defocus distance) of the surface of the wafer W relative to a reference position in the Z direction. For example, the AF signal corresponds to the height position of the upper surface Wb of the wafer W at a predetermined position. In other words, the AF signal processing unit 74 detects the height position of the upper surface Wb of the wafer W based on the output signals from each light-receiving element of the detector 48. The AF signal E is sent to the processing condition setting unit 72, and a portion of the processing conditions is output to the condenser lens driving unit 34.

[0062] Figure 5 This is a graph showing the output characteristics of the AF signal E. Figure 5 In the diagram, the horizontal axis represents the displacement (defocus distance) of the upper surface Wb of wafer W relative to the reference position in the Z direction (wafer thickness direction), and the vertical axis represents the output value of the AF signal E. It is assumed that the focusing point of the AF detection light L2 has been pre-adjusted to coincide with the reference position (origin) of the upper surface Wb of wafer W.

[0063] like Figure 5 As shown, the output characteristic of the AF signal E is an S-shaped curve with the reference position (origin) of the upper surface Wb of wafer W as the zero-crossing point. Furthermore, when the position of the upper surface Wb of wafer W is located at... Figure 5When the range indicated by the middle arrow is within the measurement range (capture range) that can detect the displacement of the upper surface Wb of wafer W, the relationship between the displacement of the upper surface Wb of wafer W and the output of the AF signal E is a monotonically increasing curve (or monotonically decreasing curve) passing through the origin. This curve largely shows a roughly linear change. That is, if the output of the AF signal E is zero, it can be known that the upper surface Wb of wafer W is located at the focal point corresponding to the focusing point of the detection light L2 used in the AF. On the other hand, if the output of the AF signal E is not zero, the displacement direction and amount of the upper surface Wb of wafer W can be determined.

[0064] The AF signal processing unit 74 determines whether the AF signal E (the height position of the upper surface Wb of the wafer W) is within a specified range. Alternatively, the AF signal processing unit 74 may not determine whether the height position of the upper surface Wb of the wafer W is within a specified range. For example, the AF signal processing unit 74 determines whether the position (actuator height position) of the condenser lens driving unit 34, obtained by converting the height position of the upper surface Wb of the wafer W obtained based on the AF signal E into the position of the condenser lens driving unit 34, is within a specified range, corresponding to the position of the condenser lens 33 when the focusing point of the AF detection light L2 is located on the upper surface Wb of the wafer W.

[0065] Here, the specified range is, for example, the distance obtained by subtracting the distance the condenser lens driving unit 34 needs to move to measure the overall thickness of wafer W from the distance or range (full stroke) that the condenser lens driving unit 34 can move, or the range that the condenser lens driving unit 34 can move. For example, when the thickness of wafer W is 400 μm, the actual refractive index is 4, and the full stroke of the thickness detection range is 120 μm (with the downward or upward direction in the Z direction as positive), in order to measure the overall thickness of wafer W, the condenser lens driving unit 34 needs to move 100 μm in the Z direction (= wafer thickness / actual refractive index = 400 μm / 4). Therefore, if the height position of the condenser lens driving unit 34 is in the range of 0 μm (e.g., the initial position or reference position of the condenser lens driving unit 34 in the Z direction) to 20 μm, the thickness of wafer W can be measured throughout the entire wafer by controlling the movement of the condenser lens 33 by controlling the condenser lens driving unit 34. This determination can improve the detection accuracy of the height position of the upper surface Wb.

[0066] The height position of the wafer W surface detected by the front height detection unit 5 is sent to the coaxial AF setting unit 73. Therefore, the coaxial AF device 4 can detect the height position of the wafer W based on the height position of the wafer W surface detected by the front height detection unit 5. For example, by detecting the height position of the wafer W based on the height position of the wafer W surface detected by the front height detection unit 5, the coaxial AF device 4 can correct errors from the surface shape tracked during processing. Furthermore, the coaxial AF device 4 can correct for vibrations of the laser head, etc., that may occur during processing based on the height position of the wafer W surface detected by the front height detection unit 5. Therefore, the coaxial AF device 4 can detect an accurate height position of the wafer W, improving the height position detection sensitivity.

[0067] The detection condition setting unit 75 sets the distances in the X and Y directions between the front height detection unit 5 and the coaxial AF device 4, and the distances in the X and Y directions between the rear height detection unit 6 and the coaxial AF device 4. In other words, regarding the front height detection unit 5, the detection condition setting unit 75 allows the front height detection unit 5 to move to any position in the X and Y directions in front of the coaxial height detection unit 4. Furthermore, regarding the rear height detection unit 6, the detection condition setting unit 75 allows the rear height detection unit 6 to move to any position in the X and Y directions behind the coaxial height detection unit 4. The spacing of the scribe lines (INDEX distance I) (e.g., reference) Figure 7 The specific method varies depending on the type of wafer W. The detection condition setting unit 75 can detect the height position of the dicing trace at an appropriate location in the Y direction by setting the INDEX distance I.

[0068] The control unit 7 may also include a state determination unit 76 that determines the surface condition of the processed wafer W. The state determination unit 76 can determine the surface condition of the processed wafer W based on height information obtained from the front height detection unit 5 and height information obtained from the rear height detection unit 6. For example, if there is a significant difference between the length measurement data obtained from the coaxial AF device 4 or the front height detection unit 5 and the length measurement data obtained from the rear height detection unit 6, there is a possibility that surface damage caused by processing has occurred. Therefore, a threshold can be set for the difference between the length measurement data value obtained from the rear height detection unit 6 and the length measurement data obtained from the coaxial AF device 4 or the front height detection unit 5; when the difference is above the threshold, the surface condition is determined to be poor.

[0069] Furthermore, by using a high-resolution sensor in the rear height detection unit 6, it is possible to determine whether a half-cut exists. For example, by applying the same structure as the coaxial AF device 4 to the rear height detection unit 6 and setting the laser spot diameter to 10 μm or less, it is possible to detect whether a half-cut exists. Specifically, for example, the height position of the wafer W surface in a non-half-cut state is measured in advance using the front height detection unit 5 and the rear height detection unit 6, and the difference is calculated. A threshold is set based on this difference. The difference in height position obtained by the front height detection unit 5 and the rear height detection unit 6 through actual measurement can be compared with the threshold to determine whether a half-cut exists. When using only existing coaxial AF devices, if the beam spot diameter is too small, minute stains or changes on the wafer surface can be detected. Therefore, when using only existing coaxial AF devices, the spot diameter needs to be set to a certain size, for example, tens of μm to 100 μm. With a spot diameter of tens of μm to 100 μm, it is difficult to detect half-cuts with high precision. On the other hand, as described above, by using a high-resolution sensor or the like in the rear height detection unit 6, it is possible to determine whether there is a half-cut.

[0070] In addition, the control unit 7 may also include an anomaly detection unit 77 that detects surface anomalies of the wafer W by either the coaxial AF device 4 or the front height detection unit 5. Figure 6 This is a schematic diagram illustrating anomaly detection performed by the anomaly detection unit 77. Two dashed lines represent the upper and lower limits of the set detection threshold, while solid lines represent the surface height positions of single dicing traces detected or anticipated by the front height detection unit 5. The front height detection unit 5 moves relative to the stage 2 in the negative X-axis direction as the stage 2 moves in the positive X-axis direction, detecting the height position of the wafer W surface. When the front height detection unit 5 detects a surface height position at each processing point of the wafer W that exceeds the detection threshold, as shown by the dashed lines, the anomaly detection unit 77 detects the surface height position at each processing point of the wafer W exceeding the detection threshold as an anomaly value.

[0071] Next, the positional relationship between the coaxial AF device 4, the front height detection unit 5, and the rear height detection unit 6 will be explained. Figure 7 This is a schematic diagram showing an example of the positional relationship between the coaxial AF device 4, the front height detection unit 5, and the rear height detection unit 6 when viewed from the positive Z-axis direction relative to the wafer W. Figure 7 The diagram shows a dicing lane S with processed processing points P and a dicing lane S' with unprocessed predetermined processing points P'.

[0072] For example, such as Figure 7As shown, the front height detection unit 5 is located in the negative X-axis direction and the positive Y-axis direction relative to the coaxial AF device 4. That is, the front height detection unit 5 is located in front of the coaxial AF device 4. The front height detection unit 5 detects the height position of the predetermined processing point P' arranged along the scribe line S'.

[0073] Figure 8 This is a schematic diagram showing the positional relationship between the coaxial AF device, the front height detection unit, and the rear height detection unit when viewed from the wafer surface in a direction perpendicular to the extension direction of the scribe line (Y direction). Furthermore, Figure 9 This is a schematic diagram showing the positional relationship between the coaxial AF device, the front height detection unit, and the rear height detection unit when viewed from a direction parallel to the grading track (X direction). For example... Figure 8 and Figure 9 As shown, the forward height detection unit 5 is located closer to the front of the processing travel direction than the processing laser L1 and the AF laser L2. When the processing laser L1 irradiates the wafer W, a modified layer is formed on the dicing track with the processing focus f of the processing laser L1 as the starting point, and an upper crack is formed in the positive Z-axis direction of the modified layer, and a lower crack is formed in the negative Z-axis direction of the modified layer.

[0074] The distance between the front height detection unit 5 and the coaxial AF device 4 in the X-axis direction is preferably as small as possible. By minimizing the distance between the front height detection unit 5 and the coaxial AF device 4 in the X-axis direction, the X-axis movement distance not used for machining can be reduced. More specifically, by minimizing the distance between the front height detection unit 5 and the coaxial AF device 4 in the X-axis direction, the increase in machining stroke can be suppressed, and the machining time required can be shortened.

[0075] As described above, the front height detection unit 5 and the rear height detection unit 6 can be moved to any position in the X direction (the extension direction of the marking track) and the Y direction (the parallel direction of the marking tracks). The front height detection unit 5 and the rear height detection unit 6 will function while maintaining the positional relationship between the coaxial AF device 4, the front height detection unit 5, and the rear height detection unit 6 after the movement.

[0076] Figure 10In (a), relative to the coaxial AF device 4, the front height detection unit 5 is located in the negative X-axis direction and the positive Y-axis direction, while the rear height detection unit 6 is located in the positive X-axis direction and the negative Y-axis direction. That is, relative to the coaxial AF device 4, the front height detection unit 5 is located at a position offset forward in the Y-axis direction, and the rear height detection unit 6 is located at a position offset backward in the Y-axis direction. If this positional relationship is maintained, the front height detection unit 5 can measure the profile of the entire length of the dicing track before processing, while the coaxial AF device 4 detects the height position of the wafer W surface, the processing laser irradiation device 3 processes the wafer W, and the rear height detection unit 6 measures the profile after processing for status determination.

[0077] Figure 10 In (b), relative to the coaxial AF device 4, the front height detection unit 5 is located at the same position in the negative X-axis direction and the Y-axis direction (on the same processing line), and the rear height detection unit 6 is located in the positive X-axis direction and the negative Y-axis direction. That is, relative to the coaxial AF device 4, the front height detection unit 5 is located on the same processing line, and the rear height detection unit 6 is located at a position offset backward in the Y-axis direction. If this positional relationship is maintained, the time interval between the contour measurement performed by the front height detection unit 5 and the wafer W surface height position detection performed by the coaxial AF device 4 can be shortened. As a result, the time-dependent changes in the state of the wafer W after the contour measurement performed by the front height detection unit 5 can be reduced. Consequently, the contour measurement result obtained by the front height detection unit 5 can be reflected, thereby more appropriately detecting the wafer W surface height position obtained by the coaxial AF device 4.

[0078] Figure 10 In (c), relative to the coaxial AF device 4, the front height detection unit 5 is located in the negative X-axis direction and the positive Y-axis direction, while the rear height detection unit 6 is located at the same position (on the same processing line) in the positive X-axis direction and the Y-axis direction. That is, relative to the coaxial AF device 4, the front height detection unit 5 is located at a position offset forward in the Y-axis direction, and the rear height detection unit 6 is located on the same processing line. If this positional relationship is maintained, the front height detection unit 5 can measure the profile of the entire length of the dicing track before processing, while the coaxial AF device 4 detects the height position of the wafer W surface, the processing laser irradiation device 3 processes the wafer W, and the rear height detection unit 6 measures the profile after processing in real time for status determination.

[0079] Figure 10In (d), the front height detection unit 5 and the rear height detection unit 6 are located on the same processing line as the coaxial AF device 4. This positional relationship shortens the time interval between the contour measurement performed by the front height detection unit 5 and the surface height position detection of the wafer W performed by the coaxial AF device 4. Furthermore, the rear height detection unit 6 can measure the processed contour in real time for status determination.

[0080] Since the front height detection unit 5 and the rear height detection unit 6 can be moved to any position relative to the coaxial AF device 4 in the X and Y directions, they can be set to positions corresponding to the user's requirements.

[0081] The positional relationship between the coaxial AF device 4 and the front height detection unit 5 in the XY plane is not limited to... Figure 10 The positional relationship shown is acceptable as long as the front height detection unit 5 is located behind the machining travel direction. The positional relationship between the coaxial AF device 4 and the rear height detection unit 6 in the XY plane is not limited to the positional relationship described above, as long as the rear height detection unit 6 is located behind the machining travel direction. In addition, the front height detection unit 5 may also be located in the Y direction of the coaxial AF device 4, and the rear height detection unit 6 may also be located in the Y direction of the coaxial AF device 4.

[0082] Next, an example of the processing method for wafer W will be described in detail below.

[0083] When the processing apparatus only has a coaxial AF device as a unit for detecting the height position of the wafer W surface, the processing apparatus and the processing time of wafer W simultaneously detect the height position of the wafer W surface. Therefore, when the coaxial AF device can track the surface of wafer W, such as... Figure 11 As shown in (a), the height position of the wafer W surface can be accurately detected. However, when the processing speed is increased, or when the height position of the wafer W surface changes significantly and the tracking performance of the coaxial AF device for the wafer W surface is insufficient, the response of the detector that detects the AF using the detection light L2 will be delayed in response to this change. Figure 11 (b) It may be impossible to detect the accurate height position of the wafer W surface. To address this, for example, one could consider reducing the response delay by increasing the AF gain, but... Figure 11 As shown in (c), when the coaxial AF device fails to follow the unevenness of the wafer W surface and generates signal overshoot or oscillation, it cannot detect the accurate height position of the wafer W surface.

[0084] To address the aforementioned issues, the front height detection unit 5 sends the detected height information (height information) to the detection condition setting unit 75 of the coaxial AF device 4. This optimizes the wafer W surface height position detection conditions of the coaxial AF device 4. Therefore, through the coaxial AF device 4 and the front height detection unit 5, an accurate wafer W surface height position can be detected.

[0085] The operation flow of processing device 1 is explained. Figure 12 This is a timing diagram showing the operation flow of the processing device 1. Figure 13 This is a schematic diagram showing multiple predetermined processing points P'. Figure 13 The image shows the l-th slab track S. l With the (l+1)th lane S l+1 The subscript of the predetermined processing point P' or processing point P indicates the location of the predetermined processing point P' or processing point P. For example, the predetermined processing point P... l,m 'Indicates that the scheduled processing point is on the l-th slicing lane S' l It is positioned at the m-th location. Here, it is assumed that the front height detection unit 5 is positioned at the predetermined processing point P. l,m The coaxial AF device 4 and the laser irradiation device 3 for processing are located at the processing point P. 1,m-1 The rear height detection unit 6 is located at processing point P. 1+1,m-2 The location is used to explain the processing flow.

[0086] The front height detection unit 5 detects the predetermined processing point P of wafer W. l,m The height position of 'will determine the predetermined processing point P' l,m The height position information is sent to the detection condition setting unit 75 of the coaxial AF device 4 (step Sa1). The front height detection unit 5 then sets the predetermined machining point P. l,m The height position information is sent to the detection condition setting unit 75 of the rear height detection unit 6 (step Sa2).

[0087] Next, the control unit 7 controls the stage 2 (step Sb1) to move it from the predetermined processing point P. l,m 'To the next scheduled processing point P' 1,m+1 The distance (step Sb2). Therefore, the forward height detection unit 5 detects the distance from the predetermined processing point P. l,m Move to the next scheduled processing point P 1,m+1 ', Coaxial AF device 4 and processing laser irradiation device 3 from the previous processing point P 1,m-1 Move to the designated processing point P l,m ', The rear height detection unit 6 detects the height from processing point P. 1+1,m-2 Move to processing point P 1+1,m-1 .

[0088] Next, the control unit 7 sets the parameters for the predetermined processing point P based on the height position information. l,m The detection conditions (height position detection conditions) of the reflected light of the detection light L2 used in the AF are determined (step Sc1). Next, the coaxial AF device 4 detects the predetermined machining point P. l,m 'Height position (step Sc2). Specifically, to the predetermined processing point P l,m After irradiating the AF detection light L2, it is sent to the predetermined processing point P. l,m The reflected light from the surface of the AF detection light L2 is detected by the detector 48. The detector 48 divides and receives the focused image of the reflected light from the AF detection light L2, and outputs an output signal corresponding to the respective light amount to the AF signal processing unit 74. The AF signal processing unit 74 generates an AF signal E based on the output signal. Then, the AF signal processing unit 74 sends the AF signal E to the processing condition setting unit 72 of the processing laser irradiation device 3 (step Sc3). Here, the coaxial AF device 4 can detect a predetermined processing point P based on appropriate detection conditions. l,m 'Height position'.

[0089] Next, the machining condition setting unit 72 sets the machining conditions based on the AF signal E, and the machining laser source 31 irradiates the predetermined machining point P with laser L1. l,m 'Proceed with processing (step Sd1). Therefore, the predetermined processing point P...' l,m 'Become a processing point P' 1,m At this time, the drive of the moving lens drive unit 45 is controlled by the coaxial AF setting unit 73 to make the output of the AF signal E zero. Furthermore, by controlling the drive of the condenser lens drive unit 34 based on the processing conditions set by the processing condition setting unit 72, the Z-direction position of the focusing point of the processing laser L1 is adjusted. As a result, a modified region R is formed at the processing point P.

[0090] Located at processing point P 1+1,m-1 The rear height detection unit has 6 pairs of processing points P 1+1,m-1 The surface height position is detected (step Se1). Next, the rear height detection unit 6 measures the machining point P. 1+1,m-1 Perform status determination and anomaly determination (step Se2). Specifically, the predetermined processing point P obtained from the front height detection unit 5... 1+1,m-1 '(Processing point P)' 1+1,m-1 The height position of ) and the machining point P obtained by the rear height detection unit 6 1+1,m-1 The height position is compared, and if the difference is below a desired threshold, the surface condition is determined to be good. The threshold is set based on the height position of the wafer W surface detected by the front height detection unit 5. Then, the condition determination unit 76 sends a request to the processing point P. 1+1,m-1The surface condition is determined (step Se3). When the determination result is unsatisfactory or an abnormality is detected, the control unit 7 can stop the operation of the processing device 1.

[0091] Repeat steps Sa1 to Sd3 above to process all predetermined processing points P'. Whether all predetermined processing points P' have been processed is determined by comparing the information of predetermined processing points P' registered in the process formula with the processing results. For example, it is determined whether the number of processing points P' processed by the processing device 1 stored in the storage unit of the control unit 7 matches the number of predetermined processing points P' in the process formula. The same determination is made regarding whether the surface condition of all processing points P has been inspected and judged.

[0092] Figure 14 This is a graph illustrating an example of the wafer height position detection results from a coaxial AF device. Figure 14 An example of the wafer surface height position detection result in an arbitrary dicing track is shown when the AF signal processing unit 74 optimizes the detection conditions based on the generated AF signal E. Figure 14 The diagram shows region A, which has steep and uneven surfaces, and region B, which has a gentler surface compared to region A. Here, by optimizing the detection conditions as described above by the AF signal processing unit 74, the height position of the scribe line can be accurately detected in either region A or B.

[0093] Figure 15 This is an example of the wafer height position detection results using existing methods. Figure 15 The results show the height position detection of the BG tape and the W surface of the wafer using only a coaxial AF device without using a front height detection unit. Figure 15 In this diagram, region C corresponds to the end of wafer W, and region D corresponds to the region further central than the end of wafer W. When wafer W is ground by back-side grinding, the height position of wafer W is easily disrupted at its end due to over-grinding or non-uniformity of the patterned devices on the surface of wafer W. Therefore, it is difficult to detect the accurate height position using only existing coaxial AF devices. On the other hand, by utilizing the height position information pre-acquired by the front height detection unit 5, the detection accuracy of the height position is improved.

[0094] As described above, even in the presence of steep unevenness such as dicing tape or dicing frame, the accurate height position of the wafer W surface can be detected by pre-detecting the height position of the wafer W surface and pre-setting the optimal detection conditions for the height position of the wafer W surface of the coaxial AF device 4. Furthermore, as described above, the detection condition setting unit 75 can, for example, set the AF gain setting value as the detection condition.

[0095] The embodiments of the present invention have been described above. Furthermore, the technical scope of the present invention is not limited to the above embodiments, and various modifications can be made without departing from the spirit of the present invention.

[0096] The coaxial AF device 4 is a structure that uses the knife-edge method to detect the height position of the wafer W surface, but the coaxial AF device is not limited to this. The coaxial AF device can also be a structure that applies commonly known methods such as astigmatism.

[0097] Furthermore, in the above embodiment, the processing apparatus 1 is fully controlled by a control unit, but it is also possible to control the processing apparatus 1 through multiple control units. Additionally, the control unit may be integrated with the processing apparatus or may be a separate unit.

[0098] The control unit 7 may also consist of a wafer transport unit, an operation panel, a television monitor, and indicator lights.

[0099] The control panel can be equipped with switches and displays for operating various parts of the processing unit. The television monitor can display wafer images captured by the CCD camera, or program content and various messages. Indicator lights indicate the operating status of processing unit 1, including processing in progress, processing completion, and emergency stop.

[0100] For example, without departing from the spirit of the invention, it is appropriate to replace the constituent elements in the above embodiments with known constituent elements. In addition, the above variations can be appropriately combined.

[0101] Symbol Explanation

[0102] 1. Processing equipment

[0103] 2 platforms

[0104] 3. Laser irradiation device for processing

[0105] 4. Coaxial height detection unit

[0106] 5. Forward Height Detection Unit

[0107] 6. Rear Height Detection Unit

[0108] 7. Control Department

Claims

1. A processing apparatus comprising: A laser irradiation device for processing irradiates a processing laser while moving along multiple dicing tracks extending in at least one direction of the wafer in a traveling direction; The coaxial height detection unit has the same optical axis as the processing laser, irradiates the detection laser onto the surface of the wafer, and detects the height position of the wafer surface based on the reflected light of the detection laser. The front height detection unit is located in front of the processing laser irradiation device; and The rear height detection unit is located behind the processing laser irradiation device on the opposite side from the front. The forward height detection unit is capable of moving in the parallel direction of the multiple dicing tracks to detect the height position of the dicing tracks before processing by the processing laser irradiation device. The rear height detection unit can move in the parallel direction of the multiple dicing tracks to detect the height position of the dicing tracks after processing by the processing laser irradiation device.

2. The processing apparatus according to claim 1, wherein, The height information obtained through the front height detection unit is used to correct the height position of the wafer surface.

3. The processing apparatus according to claim 1 or 2, wherein, The surface condition of the dicing track after processing is determined based on the height information obtained by the front height detection unit and the height information obtained by the rear height detection unit.

4. The processing apparatus according to claim 1 or 2, wherein, Anomaly detection is performed on the surface condition of the wafer.

Images