Shape measurement system, processing apparatus, shape measurement method, processing method, and chip manufacturing method

Abstract

This technology shortens the measurement time when measuring the three-dimensional shape of an object using interferometric light. [Solution] The present invention provides a shape measuring system comprising a holding unit, a measuring optical system positioned away from the holding unit in a first direction, a moving unit for relatively moving the measuring optical system and the holding unit along a second direction, and a controller, wherein the measuring optical system includes a light source emitting low-coherence light, a branching unit that splits the light from the light source into reference light for irradiating a reference surface which is a reflective surface and measuring light for irradiating an object, an objective lens for focusing the measuring light onto the object, and a photoelectric conversion device having a plurality of photoelectric conversion elements that receive interference light resulting from the interference of the reference light reflected from the reference surface and the measuring light reflected from the object, wherein the optical axis of the objective lens is tilted at a predetermined angle with respect to the second direction when scanning an object with the measuring light by relatively moving the measuring optical system and the holding unit along a second direction using the moving unit.

Description

Technical Field

[0001] The present invention relates to a shape measurement system for measuring the three-dimensional shape of an object, a processing apparatus for processing a workpiece, a shape measurement method for measuring the three-dimensional shape of an object, a processing method for processing a workpiece, and a chip manufacturing method for manufacturing chips from a workpiece.

Background Art

[0002] In a wafer on the surface of which an insulating film (i.e., a Low-k film) formed of a Low-k material (that is, a material having a lower dielectric constant than silicon dioxide) is provided, the wafer may be divided into a plurality of chips along a plurality of planned division lines set on the surface of the wafer.

[0003] For division, a method has been proposed in which, first, a laser beam is irradiated along each planned division line to form narrow processing grooves (i.e., grooves) from which the Low-k film has been removed by ablation processing, and then a cutting blade is cut into the processing grooves to divide the wafer into a plurality of chips (see, for example, Patent Document 1).

[0004] A laser processing apparatus that removes the Low-k film to form grooves may include a white light interference microscope in order to determine the processing quality of ablation processing by measuring the three-dimensional shape of the processing grooves (see, for example, Patent Document 2).

[0005] When measuring the three-dimensional shape of a processing groove formed on one surface (that is, the surface to be measured) of a workpiece such as a wafer with a white light interference microscope, the white light interference microscope is moved along the thickness direction (that is, the height direction) of the workpiece, and a predetermined range is imaged using interference light at each height position.

[0006] After measuring the three-dimensional shape in the predetermined range of the surface to be measured in this manner, the workpiece and the white light interference microscope are slightly translated in parallel along the surface to be measured. Similarly, the white light interference microscope is moved along the height direction, and a predetermined range is imaged with interference light at each height position, thereby measuring the three-dimensional shape in this predetermined range (i.e., the imaged region).

[0007] In this way, by repeating the movement of the white light interference microscope along the height direction of the workpiece and the parallel translation of the workpiece and the white light interference microscope along the surface to be measured, the entire machining groove can be imaged, and the three-dimensional shape of the entire machining groove can be measured.

[0008] However, since the predetermined range is imaged at each height position, the moving speed of the white light interference microscope along the height direction is limited by the frame rate of the image sensor in the white light interference microscope (i.e., the number of still images that can be captured by the image sensor per unit time). Therefore, it is difficult to simply increase the moving speed in the height direction.

[0009] On the other hand, a large number of machining grooves may be formed on a single workpiece, and it may be necessary to sequentially measure a plurality of workpieces each having a large number of machining grooves. Therefore, in the measurement of the three-dimensional shape of an object using interference light, improvement in measurement efficiency is required.

Prior Art Document

Patent Document

[0010]

Patent Document 1

Patent Document 2

Summary of the Invention

Problem to be Solved by the Invention

[0011] This invention has been made in view of the aforementioned problems, and aims to shorten the measurement time when measuring the three-dimensional shape of an object using interference light. [Means for solving the problem]

[0012] According to one aspect of the present invention, a shape measuring system for measuring the three-dimensional shape of an object is provided, comprising: a holding unit for holding the object; a measuring optical system positioned at a distance from the holding unit in a first direction; a moving unit for moving the measuring optical system and the holding unit relatively along a second direction intersecting the first direction; and a controller for controlling the moving unit, wherein the measuring optical system comprises: a light source emitting low-coherence light; a branching unit for splitting the light from the light source into reference light for irradiating a reference surface which is a reflective surface and measuring light for irradiating the object; an objective lens for focusing the measuring light onto the object; and a photoelectric conversion device having a plurality of photoelectric conversion elements for receiving interference light resulting from the interference of the reference light reflected from the reference surface and the measuring light reflected from the object, wherein when the measuring optical system and the holding unit are moved relatively along the second direction by the moving unit and the object is scanned with the measuring light, the optical axis of the objective lens is inclined at a predetermined angle with respect to the second direction.

[0013] Preferably, the controller controls the moving unit to change the relative position of the measuring optical system and the object, causing the photoelectric converter to image the object. At each relative position, the interference light of the measurement light reflected from the surface of the object to be measured and the reference light is received by a plurality of photoelectric converters, and the three-dimensional shape is calculated based on the information of each relative position and the amount of light received by each photoelectric converter at each relative position.

[0014] According to another aspect of the present invention, a processing apparatus for processing a workpiece comprises: a holding unit for holding the workpiece; a processing unit for processing the workpiece; a measuring optical system positioned away from the holding unit in a first direction; a moving unit for moving the measuring optical system and the holding unit relative to each other along a second direction intersecting the first direction; and a controller for controlling the moving unit, wherein the measuring optical system includes a light source emitting low-coherence light and a reference light source that irradiates a reference surface, which is a reflective surface, with light from the light source. A processing apparatus is provided, comprising: a measuring light irradiated onto the workpiece; a branching unit that splits the measuring light into two; an objective lens that focuses the measuring light onto the workpiece; and a photoelectric conversion device having a plurality of photoelectric conversion elements that receive interference light resulting from the interference of the reference light reflected from the reference surface and the measuring light reflected from the workpiece, wherein when the measuring optical system and the holding unit are moved relative to each other along the second direction by the moving unit to scan the workpiece with the measuring light, the optical axis of the objective lens is inclined at a predetermined angle with respect to the second direction.

[0015] According to yet another aspect of the present invention, a shape measuring method for measuring the three-dimensional shape of an object, comprising: low-coherence light split into a reference light irradiated onto a reference surface and a measuring light irradiated onto the object, the measuring light being irradiated onto the object held in a holding unit via an objective lens; interference light resulting from the interference of the reference light reflected from the reference surface and the measuring light reflected from the object being received by a photoelectric converter; and the relative positions of the photoelectric converter and the object held in the holding unit, which are positioned apart from each other in a first direction, intersecting the first direction. A shape measurement method is provided which measures the three-dimensional shape of an object by changing the relative position of the photoelectric converter and the object in the second direction, wherein the optical axis of the objective lens is tilted at a predetermined angle with respect to the second direction, and the photoelectric converter, the objective lens, and the holding unit are moved relatively along the second direction to scan the object with the measurement light.

[0016] According to yet another aspect of the present invention, a processing method for processing a workpiece, comprising: processing the workpiece with a processing unit to form a processing mark; after the formation of the processing mark, irradiating the processing mark of the workpiece, which is held in a holding unit, with the measurement light from the low-coherence light split into a reference light irradiated onto a reference surface and a measurement light irradiated onto the processing mark, via an objective lens; and receiving the interference light, which is the result of the interference between the reference light reflected from the reference surface and the measurement light reflected from the processing mark, with a photoelectric converter; and after the formation of the processing mark, the photoelectric converter and the workpiece held in the holding unit are arranged apart from each other in a first direction. A processing method is provided which includes changing the relative position of the processing mark on the workpiece along a second direction intersecting the first direction, receiving the interference light with the photoelectric converter, and changing the relative position of the photoelectric converter and the processing mark in the second direction to measure the three-dimensional shape of the processing mark, and by changing the relative position of the photoelectric converter and the processing mark along the first direction, the optical axis of the objective lens is tilted at a predetermined angle with respect to the second direction, and the photoelectric converter, the objective lens and the holding unit are moved relatively along the second direction to scan the processing mark with the measurement light.

[0017] According to yet another aspect of the present invention, a method for manufacturing a chip from a workpiece, comprising: processing the workpiece with a processing unit to form a processing groove having a depth that does not completely cut the workpiece; measuring the three-dimensional shape of the processing groove after its formation; and cutting the workpiece so as to pass through the bottom of the processing groove to divide it into chips, wherein the three-dimensional shape of the processing groove is measured by irradiating the processing groove of the workpiece, held in a holding unit, with the measurement light from low-coherence light that has been split into a reference light irradiated onto a reference surface and a measurement light irradiated onto the processing groove, and the measurement light being irradiated onto the processing groove is irradiated onto the processing groove of the workpiece held in a holding unit via an objective lens, and the measurement light reflected from the reference surface A chip manufacturing method is provided, comprising: receiving interference light, which is formed when light and the measurement light reflected from the processing groove, interfere with each other, using a photoelectric converter; and changing the relative position of the photoelectric converter and the processing groove of the workpiece held by the holding unit, which are arranged apart in a first direction, along a second direction intersecting the first direction, wherein the relative position of the photoelectric converter and the processing groove of the workpiece held by the holding unit along the second direction is such that the optical axis of the objective lens is tilted at a predetermined angle with respect to the second direction, and the photoelectric converter, the objective lens, and the holding unit are moved relative to each other along the second direction to scan the processing groove with the measurement light.

[0018] According to yet another aspect of the present invention, a method for manufacturing a chip from a workpiece comprises: processing the workpiece with a processing unit to form a processing groove that completely cuts the workpiece and divides it into chips; and measuring the three-dimensional shape of the processing groove after its formation, wherein the three-dimensional shape of the processing groove is measured by irradiating the processing groove of the workpiece, which is held by a holding unit, with the measurement light from low-coherence light that has been split into a reference light irradiated onto a reference surface and a measurement light irradiated onto the processing groove, via an objective lens, and measuring the interference between the reference light reflected from the reference surface and the measurement light reflected from the processing groove. A chip manufacturing method is provided, comprising receiving light with a photoelectric converter, and changing the relative position of the photoelectric converter and the processed groove portion of the workpiece held by the holding unit, which are arranged apart from each other in a first direction, along a second direction intersecting the first direction, wherein the optical axis of the objective lens is tilted at a predetermined angle with respect to the second direction, and the photoelectric converter, the objective lens, and the holding unit are moved relative to each other along the second direction to scan the processed groove portion with the measuring light. [Effects of the Invention]

[0019] A shape measuring system according to one aspect of the present invention comprises a holding unit for holding an object, a measuring optical system positioned at a distance from the holding unit in a first direction, and a moving unit for moving the measuring optical system and the holding unit relative to each other along a second direction intersecting the first direction.

[0020] The measurement optical system includes a light source that emits low-coherence light, a branching unit that splits the light from the light source into reference light and measurement light, an objective lens that focuses the measurement light onto the object, and a photoelectric conversion device having a plurality of photoelectric conversion elements that receive interference light.

[0021] Furthermore, when the measuring optical system and the holding unit are moved relative to each other along the second direction by the moving device and the object is scanned with the measuring light, the optical axis of the objective lens of the measuring optical system is tilted at a predetermined angle with respect to the second direction (for example, the X-axis direction).

[0022] In this way, by tilting the optical axis of the objective lens at a predetermined angle with respect to the second direction, and moving the measuring optical system and holding unit relative to each other along the second direction, height information of the object can be acquired in the area scanned by the measurement light.

[0023] In contrast, conventionally, in order to acquire height information of an object in a band-shaped region along the second direction, the optical axis of the objective lens is first aligned parallel to the first direction (for example, the Z-axis direction) which is perpendicular to the second direction, and the measuring optical system and holding unit are moved relatively along the first direction (so-called vertical scanning) to acquire height information in a region of the object that is illuminated in a spot shape by the measurement light.

[0024] Next, the measuring optical system and the holding unit are shifted relative to each other by a predetermined distance in a second direction, thereby moving the measurement light to other areas of the object adjacent to the area in question in the second direction. Then, with the measurement light illuminating in a spot shape, the measuring optical system and the holding unit are moved relative to each other again along the first direction, thereby acquiring height information in the other areas.

[0025] In this manner, by repeatedly moving the measuring optical system and holding unit relatively in a first direction and then shifting them relatively in a second direction by a predetermined distance (so-called vertical scanning low-coherence interferometry), height information of the object was acquired in a band-shaped region along the second direction.

[0026] In contrast, by tilting the optical axis of the objective lens at a predetermined angle with respect to the second direction, and moving the measuring optical system and holding unit relative to the second direction, height information of the object can be acquired in the area scanned by the measurement light, thus significantly simplifying the aforementioned repetitive operation. Therefore, the measurement time when measuring three-dimensional shapes can be shortened.

[0027] A processing apparatus according to another aspect of the present invention also includes the shape measuring system described above. Therefore, the measurement time when measuring a three-dimensional shape can be reduced. Similarly, the shape measuring method, processing method, and chip manufacturing method according to further other aspects of the present invention can also reduce the measurement time when measuring a three-dimensional shape. [Brief explanation of the drawing]

[0028] [Figure 1] This is a schematic diagram of the processing equipment. [Figure 2] This is a schematic diagram of the shape measurement system. [Figure 3] This is the result of a simulation demonstrating the interference of low-coherence light. [Figure 4] Figure 4(A) shows how the measurement light reflected at point A is received by pixel P0 at time t1, and Figure 4(B) shows how the measurement light reflected at point A is received by pixel PN at time t2. [Figure 5] Figure 5(A) is a graph showing the time change of the optical path length when the measurement light reflected at point A is received at any of the pixels P, and Figure 5(B) is a graph showing the optical path difference ΔL(t) adjusted to be zero at pixel PM. [Figure 6] Figure 6(A) shows the image taken at time tA when the optical axis passes through point A, Figure 6(B) shows the image taken at time tB, and Figure 6(C) shows the image taken at time tC. [Figure 7] This is a flowchart showing the processing method for a workpiece. [Figure 8] This is a perspective view of the workpiece. [Figure 9]Figure 9(A) shows the ablation process applied to a workpiece, and Figure 9(B) is a cross-sectional view of the workpiece including the processing marks. [Figure 10] Figure 10(A) shows the start of imaging for one processing mark, and Figure 10(B) shows the end of imaging for one processing mark. [Figure 11] This is a flowchart showing an example of a device chip manufacturing method. [Figure 12] Figure 12(A) shows the modification region formation process, Figure 12(B) shows the back surface grinding process, and Figure 12(C) is a perspective view of the device chip. [Figure 13] This flowchart shows a modified example of a device chip manufacturing method. [Figure 14] This is a cross-sectional view of a workpiece cut by ablation. [Modes for carrying out the invention]

[0029] (First Embodiment) An embodiment according to one aspect of the present invention will be described with reference to the attached drawings. Figure 1 is a schematic diagram of a processing apparatus 2 for processing a workpiece (i.e., object) 11. In Figure 1, some of the components of the processing apparatus 2 are simply shown as functional blocks. The X axis, Y axis and Z axis shown in Figure 1 are orthogonal to each other (i.e., intersect).

[0030] In Figure 1, the positive direction of the X-axis is defined as the +X direction, and the negative direction of the X-axis is defined as the -X direction. The +X and -X directions are opposite directions. In this specification, the +X and -X directions are sometimes collectively referred to as the X-axis direction (or second direction or machining feed direction).

[0031] In Figure 1, the positive direction of the Y-axis is defined as the +Y direction, and the negative direction of the Y-axis is defined as the -Y direction. The +Y and -Y directions are opposite directions. In this specification, the +Y and -Y directions are sometimes collectively referred to as the Y-axis direction (or indexing feed direction).

[0032] Furthermore, in Figure 1, the positive direction of the Z-axis is defined as the +Z direction, and the negative direction of the Z-axis is defined as the -Z direction. The +Z and -Z directions are opposite to each other. For example, the -Z direction is the vertical direction (i.e., the direction of gravity). In this case, the -Z direction is downward, and the +Z direction is upward. In this specification, the +Z and -Z directions are sometimes collectively referred to as the Z-axis direction (or the first direction or height direction).

[0033] The processing apparatus 2 includes a disc-shaped chuck table (holding unit) 4 that holds the workpiece 11 by suction using negative pressure. The chuck table 4 has a disc-shaped frame made of a metal such as stainless steel. A disc-shaped recess with a diameter smaller than the outer diameter of the frame is formed in the radial center of the frame.

[0034] A disc-shaped porous plate (not shown) made of porous ceramics is placed in the recess of the frame, and the porous plate is fixed to the frame using an adhesive or the like. The upper surfaces of the frame and the porous plate are substantially flush and exposed facing upward.

[0035] A suction source (not shown), such as an ejector or vacuum pump, is connected to the frame, and the negative pressure from the suction source is transmitted to the upper surface of the porous plate via the frame. The upper surface of the frame and the upper surface of the porous plate form a holding surface 4a that holds the workpiece 11 by suction.

[0036] The workpiece 11 is held by suction on the holding surface 4a in the form of a workpiece unit 21 integrated with the frame 17 via a dicing tape 19, as shown in Figure 8, for example. However, for convenience, the dicing tape 19 and the frame 17 are omitted from the illustration in Figure 1 (and similarly in Figures 2, 9(A) to 10(B)).

[0037] By the way, instead of the chuck table 4, an electrostatic chuck that holds the workpiece 11 by electrostatic attraction may be used. Alternatively, the workpiece 11 may be fixed directly or indirectly to the holding surface by a holding unit having a pressing mechanism including a clamp that can be opened and closed by an actuator, instead of or in combination with negative pressure or electrostatic force.

[0038] In this embodiment, the holding surface 4a of the chuck table 4 is exposed upwards, but it may be reversed so that the holding surface 4a is exposed downwards. Similarly, when employing an electrostatic chuck, a pressing mechanism, etc., the holding surface may also be exposed downwards.

[0039] The chuck table 4 is supported by a moving mechanism (i.e., a moving unit) 6. The moving mechanism 6 has a θ-table (not shown) that rotates the chuck table 4 around a rotation axis arranged along the Z-axis. Furthermore, the moving mechanism 6 has a first moving plate (not shown) that supports the chuck table 4 and the θ-table.

[0040] The first movable plate is slidably supported by a pair of X-axis guide rails (not shown), each of which has its longitudinal portion aligned along the X-axis. Between the pair of X-axis guide rails is a first screw shaft (not shown), whose longitudinal portion is aligned along the X-axis. A first nut portion (not shown) is provided on the lower surface of the first movable plate.

[0041] A first screw shaft, whose longitudinal portion is arranged along the X-axis, is rotatably connected to the first nut portion via a plurality of balls (not shown). A first drive source (not shown), such as a servo motor, is connected to one end of the first screw shaft. When the first drive source is operated, the first movable plate moves along the X-axis.

[0042] The first moving plate, a pair of X-axis guide rails, a first screw shaft, a first nut section, a first drive source, etc. constitute the X-axis movement mechanism. The pair of X-axis guide rails, the first screw shaft, etc. are supported by a second moving plate (not shown).

[0043] The second movable plate is slidably supported by a pair of Y-axis guide rails (not shown), each of which has its longitudinal portion aligned along the Y-axis. Between the pair of Y-axis guide rails is a second screw shaft (not shown), whose longitudinal portion is aligned along the Y-axis.

[0044] A second nut portion (not shown) is provided on the lower surface of the second movable plate, and a second screw shaft, whose longitudinal portion is arranged along the Y-axis, is rotatably connected to the second nut portion via a plurality of balls (not shown).

[0045] A second drive source (not shown), such as a servo motor, is connected to one end of the second screw shaft. When the second drive source is activated, the second moving plate moves along the Y-axis direction. The second moving plate, a pair of Y-axis direction guide rails, the second screw shaft, the second nut section, the second drive source, etc., constitute a Y-axis direction movement mechanism.

[0046] A light concentrator 10, which constitutes the laser beam irradiation unit (i.e., processing unit) 8, is positioned above the chuck table 4. The laser beam irradiation unit 8 has a laser oscillator 12. The laser oscillator 12 has a laser medium, which is a crystal such as Nd:YAG.

[0047] By irradiating the laser medium with excitation light from a light source such as a flash lamp or laser diode, the laser oscillator 12 emits a pulsed laser beam having a predetermined wavelength (for example, 1064 nm).

[0048] The laser beam emitted from the laser oscillator 12 is converted to a wavelength absorbed by the workpiece 11 (for example, 355 nm) by a wavelength conversion device (not shown) that includes a nonlinear optical crystal such as an LBO (i.e., LiB3O5) crystal, and then enters the light concentrator 10.

[0049] Then, after the direction of travel is changed by the mirror 10a located inside the light concentrator 10, the light passes through the focusing lens 10b located inside the light concentrator 10 and is irradiated toward the holding surface 4a (i.e., along the Z-axis direction) approximately perpendicular to it.

[0050] With the focal point of the pulsed laser beam L positioned near the surface 11a of the workpiece 11 held by the holding surface 4a, the focal point and the chuck table 4 are moved relative to each other along the X-axis direction by the moving mechanism 6, thereby enabling ablation processing on the workpiece 11 (i.e., the workpiece 11 can be processed).

[0051] The laser beam L emitted from the laser oscillator 12 may also be irradiated from the light concentrator 10 toward the holding surface 4a after passing through an optical modulator such as an AOM (Acousto-Optic Modulator), EOM (Electro-Optic Modulator), or LCOS-SLM (Liquid Crystal on Silicon - Spatial Light Modulator).

[0052] The optical modulator has various functions. For example, the optical modulator adjusts the power of the laser beam L (i.e., average output, peak power, etc.). The optical modulator may also control whether or not to irradiate the laser beam L onto the focuser 10.

[0053] Within the processing apparatus 2, the position of the light concentrator 10 is substantially fixed, but the focusing lens 10b is configured to be adjustable along the Z-axis direction by a piezoelectric actuator. Therefore, the position of the focusing point of the laser beam L can be adjusted along the Z-axis direction.

[0054] A measuring optical system 20 is provided near the light concentrator 10. The measuring optical system 20 is positioned away from the chuck table 4 in the Z-axis direction. The measuring optical system 20 will now be described with reference to Figure 2.

[0055] Figure 2 is a schematic diagram of a shape measurement system 42 including a measurement optical system 20. The measurement optical system 20 has a housing 22, and a white light source (i.e., light source) 24 is fixed inside the housing 22. The white light source 24 has a light source such as an SLD (Superluminescent Diode) or an LED (Light Emitting Diode).

[0056] The white light source 24 emits light containing a variety of wavelengths continuously across a relatively wide wavelength band. In this example, the white light source 24 emits white light with a central wavelength of 465 nm, but this white light does not have to be perfectly white in the strict sense.

[0057] The white light source 24 of this embodiment emits low-coherence light 24a. Low-coherence light 24a is characterized by a short coherence distance 24b (see Figure 3). In this specification, low-coherence light 24a means so-called white light in which the coherence distance 24b is greater than 0 μm and 36 μm or less (more preferably greater than 0 μm and 10 μm or less).

[0058] A first beam splitter 26 is provided at a predetermined distance from the white light source 24 in the -X direction. A second beam splitter 28 is provided below the first beam splitter 26. Furthermore, an objective lens 30 is provided below the second beam splitter (i.e., the branching section) 28.

[0059] In this example, the first beam splitter 26 and the second beam splitter 28 are half-mirrors. Near the second beam splitter 28, a mirror 32 is provided to which light reflected by the second beam splitter 28 is incident. The mirror 32 has a reference surface 32a which is a reflective surface.

[0060] A portion of the low-coherence light 24a emitted from the white light source 24 sequentially passes through the second beam splitter 28 and the objective lens 30, becoming the measurement light 24a1 that is irradiated onto the surface 11a of the workpiece 11.

[0061] The objective lens 30 has the function of focusing the measurement light 24a1 onto the surface 11a of the workpiece 11. In this embodiment, the optical axis 30a of the objective lens 30 is inclined by a predetermined angle φ with respect to the X-axis direction in the XZ plane.

[0062] The predetermined angle φ corresponds to the acute angle formed between the optical axis 30a and the X-axis direction, and is, for example, greater than 0 degrees and less than 90 degrees. However, the closer φ is to 90 degrees, the narrower the measurable range of the measurement optical system 20 in the Z-axis direction becomes.

[0063] In contrast, the closer φ is to 0 degrees, the greater the distance between the objective lens 30 and the workpiece 11, as the objective lens 30 must be positioned away from the workpiece 11 to prevent contact between the objective lens 30 and the housing 22. This increases the distance from the objective lens 30 to the workpiece 11 (i.e., WD: working distance).

[0064] However, increasing the WD in this way reduces the numerical aperture (i.e., NA), thus decreasing the resolution of the measurement optical system 20. In other words, φ is determined appropriately, taking into account the trade-off between the measurable range in the Z-axis direction and the resolution.

[0065] A preferred range for φ is, for example, 75 degrees to 89 degrees, more preferably 78 degrees to 88 degrees, and even more preferably 80 degrees to 86 degrees. An example of the most preferred φ is 85 degrees.

[0066] After passing through the objective lens 30 and irradiating the surface 11a, the measurement light 24a1 reflected from the surface 11a proceeds sequentially through the second beam splitter 28 and the first beam splitter 26 to the imaging lens 34 located above the first beam splitter 26.

[0067] In contrast, the remaining portion of the low-coherence light 24a irradiated from the white light source 24 is sequentially reflected by the first beam splitter 26 and the second beam splitter 28 to become the reference light 24a2 irradiated on the reference surface 32a.

[0068] Thus, the second beam splitter 28 has the function of splitting the low-coherence light 24a into measurement light 24a1 and reference light 24a2. The reference light 24a2 reflected by the reference surface 32a proceeds sequentially through the second beam splitter 28 and the first beam splitter 26 to the imaging lens 34.

[0069] Above the imaging lens 34, a camera (i.e., a photoelectric converter) 38 including a solid-state image sensor 36 is provided. The camera 38 receives interference light resulting from the interference of the measurement light 24a1 and the reference light 24a2.

[0070] The measuring optical system 20 in this embodiment is a so-called Michelson type, but it may also be a Mirau type. In the case of a Mirau type, a mirror 32 having a reference surface 32a is positioned between the objective lens 30 and the holding surface 4a on the optical axis of the objective lens 30.

[0071] The camera 38 has a general solid-state image sensor 36, such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor. Multiple photoelectric conversion elements 36a capable of receiving interference light are regularly arranged (for example, in a grid pattern) on the imaging surface of the solid-state image sensor 36.

[0072] Each photoelectric conversion element 36a includes, for example, a photodiode and a current-voltage conversion circuit, and generates a voltage signal corresponding to the brightness of the received light. The voltage signal generated by each photoelectric conversion element 36a is ultimately output as brightness information to the controller 40, which will be described later.

[0073] As shown in Figure 2, the moving mechanism 6, the measuring optical system 20, and the controller 40 constitute a shape measuring system 42 for measuring the three-dimensional shape of the workpiece 11. When imaging the surface (i.e., the surface to be measured) 11a of the workpiece 11 held by the holding surface 4a, the moving mechanism 6 moves the measuring optical system 20 and the workpiece 11 relative to each other along the X-axis.

[0074] When scanning the surface 11a of the workpiece 11 with the measuring light 24a1 due to this relative movement, the optical axis 30a is tilted at a predetermined angle φ with respect to the X-axis, so that the three-dimensional shape of the surface 11a (i.e., the area to be measured 11c) can be measured along the movement trajectory of the measuring light 24a1 (i.e., an elongated strip-shaped region).

[0075] One of the features of this embodiment is that by tilting the optical axis 30a with respect to the X-axis direction (or, more precisely, tilting it with respect to the Z-axis direction), the three-dimensional shape of the surface 11a can be measured simply by moving the measuring optical system 20 (i.e., the objective lens 30) in the X-axis direction relative to the holding surface 4a while keeping the Z-axis position of the measuring optical system 20 fixed.

[0076] Figure 3 shows the simulation results illustrating the interference of low-coherence light 24a. The horizontal axis of Figure 3 represents the optical path difference (μm) between the measurement light 24a1 and the reference light 24a2 from the camera 38 to the measurement area 11c, and the vertical axis of Figure 3 represents the luminance (au) of the interference light received by the camera 38.

[0077] When the optical path difference is 0 (μm), the measurement light 24a1 and the reference light 24a2 reinforce each other, and the brightness of the interference light reaches its peak value. Even when the optical path difference is not 0 (μm), although weaker than when the optical path difference is 0 (μm), the brightness and darkness of the interference light are still present on the imaging surface of the solid-state image sensor 36.

[0078] In the example shown in Figure 3, outside the range of ±1 μm for the optical path difference (i.e., the coherence distance 24b), the measurement light 24a1 and the reference light 24a2 are canceled out by interference, and virtually no brightness or darkness of interference occurs on the imaging surface of the solid-state image sensor 36.

[0079] In this example, the low-coherence light 24a is so-called white light with a central wavelength of 465 nm, and interference between the measurement light 24a1 and the reference light 24a2 occurs only in the range where the optical path difference is 1.395 μm, which corresponds to one path of 2.79 μm (= 465 nm × 6 wavelengths).

[0080] Next, referring to Figures 4(A) to 6(C), the mechanism by which the measurement optical system 20 measures the three-dimensional shape of the surface 11a will be explained. For convenience, the explanation will describe multiple photoelectric conversion elements 36a arranged one-dimensionally on a specific XZ plane where the optical axis 30a exists, but a similar explanation also applies to multiple photoelectric conversion elements 36a arranged in a grid on a two-dimensional plane orthogonal to the optical axis 30a.

[0081] Furthermore, in the explanation of Figures 4(A) to 6(C), for convenience, each photoelectric conversion element 36a is referred to as a pixel and represented by the letter P, and a sign from 0 to N is placed to the lower right of each pixel P along the one-dimensional arrangement direction (where N is typically a natural number between 1000 and 2000).

[0082] Note that Pixel P M This refers to a pixel P0 located at one end of a one-dimensional array of multiple pixels P, and a pixel P located at the opposite end. N This is pixel P, located in the center of the line segment connecting and intersecting the optical axis 30a.

[0083] Figure 4(A) shows how the measurement light 24a1 reflected at point A on surface 11a is received by pixel P0 at time t1. Note that each pixel P also receives the reference light 24a2 simultaneously, but for the sake of explanation, the reference light 24a2 is omitted in Figure 4(A).

[0084] Furthermore, although the objective lens 30 is omitted for convenience in Figures 4(A), 4(B), and 6(A) to 6(C), this objective lens 30 is a telecentric lens, and the measurement light 24a1 irradiated onto the surface 11a via the objective lens 30 is irradiated approximately parallel to the optical axis of the objective lens 30.

[0085] In this embodiment, when imaging the workpiece 11, the chuck table 4 is moved in the +X direction at a constant speed while imaging is performed. Therefore, at time t2, a predetermined time has elapsed from time t1, point A moves relative to the +X direction by ΔX (see Figure 4(B)).

[0086] FIG. 4(B) is a diagram showing a state in which the measurement light 24a1 reflected at point A on the surface 11a at time t2 is received by pixel P N . Note that the reference light 24a2 is also omitted in FIG. 4(B).

[0087] As shown in FIG. 4(B), due to the inclination of the optical axis 30a (that is, the predetermined angle φ), the optical path length L0 when the measurement light 24a1 reflected at point A is received by pixel P0 at time t1 is greater than the optical path length when the measurement light 24a1 reflected at point A is received by pixel P at time t2 N L N that is N L < L0).

[0088] As can be seen from FIG. 4(A) and FIG. 4(B), due to the inclination of the optical axis 30a, the measurement light 24a1 reflected at point A and received by the solid-state image sensor 36 has different optical path lengths according to the relative position in the X-axis direction between point A and the measurement optical system 20.

[0089] FIG. 5(A) is a graph showing the temporal change of the optical path length (that is, L(t)) when the measurement light 24a1 reflected at point A is received by any one of the pixels P, and corresponds to the period from FIG. 4(A) (t=t1) to FIG. 4(B) (t=t2).

[0090] In the present embodiment, point A is imaged using interference light. The interference light is represented by the difference between the measurement light 24a1 and the reference light 24a2, and the optical path length L of the reference light 24a2 REF (Note that L REF is always kept constant), the optical path difference ΔL(t) obtained by subtracting L from L(t) is expressed as ΔL(t)=L(t)-L REF (see FIG. 5(B)).

[0091] For example, if the measuring optical system 20 is fixed in the XY plane relative to the workpiece 11 and then moved along the Z-axis direction, the optical path length of the measuring light 24a1 changes according to the position of the measuring optical system 20 in the Z-axis direction. In other words, the optical path difference ΔL(t) moves up and down according to the position of the measuring optical system 20 in the Z-axis direction.

[0092] Therefore, in this embodiment, pixel P M In this process, the position of the measurement optical system 20 in the Z-axis direction is adjusted so that the optical path difference ΔL(t) becomes zero at a predetermined timing (i.e., the conditions for the strongest interference (e.g., light conditions) are met) (see Figure 5(B)).

[0093] Figure 5(B) shows pixel P M This graph shows the optical path difference ΔL(t) adjusted to be zero at pixel P. M The optical path difference ΔL(t) = 0 occurs when the optical axis 30a passes through point A, and the height position of point A in the Z-axis direction at this time is defined as the reference height Z0. Note that point A does not necessarily have to be a point on a flat plane, but may be any point on surface 11a.

[0094] After setting the reference height Z0, how to measure the height of the surface 11a when there are irregularities on the surface 11a will be explained with reference to Figures 6(A) to 6(C). Figure 6(A) shows the time t when the optical axis 30a passes through point A. A This figure shows an image of the surface 11a of the workpiece 11.

[0095] time t A Next, the measurement light 24a1 reflected at point A is taken to pixel P M Light is received at the following point: Point A is at reference height Z0 relative to the measurement optical system 20, and pixel P M At this point, the optical path difference ΔL(t) becomes zero, resulting in the strongest interference.

[0096] Figure 6(A) shows that the interference is strongest at point A and pixel P. M The intermediate points are indicated by relatively thick solid lines. Also, pixel P MThen the conditions for the strongest interference (for example, the light condition) are met, so point A and pixel P M The optical path length L between them A The length can be calculated precisely.

[0097] Figure 6(B) shows the chuck table 4 and the workpiece 11 being moved at a predetermined speed in the +X direction, at time t A Later time t B (that is, t B >t A This is a diagram showing the imaging of surface 11a at time t. B Therefore, the intersection of the optical axis 30a and the surface 11a lies between point B and point A. Note that point B is located at the vertex of a convex portion that protrudes further in the +Z direction than point A.

[0098] Here, pixel P0 and pixel P M Pixel P located between the two B In this case, we assume that the interference between the measurement light 24a1 reflected at point B and the reference light 24a2 is strongest. In Figure 6(B), the strongest interference is shown at pixel P. B And shown by a relatively thick solid line between point B.

[0099] Next, the height Z of point B B This involves considering the height relative to the reference height Z0. In this embodiment, we use ΔL(t) shown in Figure 5(B) to determine the position of pixel P B The optical path difference ΔL when interference is strongest. B Convert this to height in the Z-axis direction.

[0100] In this embodiment, ΔL(t) is a function of time t and also a function of the position of pixel P. Therefore, if we express this as ΔL(P), then pixel P B By specifying this, the graph of ΔL(t) shown in Figure 5(B) can be converted to ΔL(P B )(that is, optical path difference ΔL B ) was revealed.

[0101] Specifically, the optical path difference ΔL BConsidering that this includes the outgoing and return lengths of the measurement light 24a1, the optical path difference ΔL B Half of (=ΔL) B Consider a right triangle with the hypotenuse ( / 2) and the base at position Z=Z0, where one acute angle tangent to the base is the predetermined angle φ (see the right side of Figure 6(B)). As is clear from this right triangle, the height of point B is Z B This is expressed by the formula shown in (1). Note that the calculations for the formula shown in (1) are performed automatically by the controller 40.

[0102]

number

[0103] In the examples shown in Figures 5(B) and 6(B), ΔL(P B ) is positive (i.e., 0 < ΔL(P B )) is (that is, Z B (It protrudes above Z0), but if ΔL(P B ) is negative (that is, ΔL(P B )<0) (that is, Z B Similarly, if (the height of point B is lower than Z0), then the height of point B is Z B This can be calculated. Specifically, in Figure 5(B), ΔL(P B If ) is negative, the height Z of point B B This is expressed by the formula shown in (2).

[0104]

number

[0105] In Figure 6(B), the measurement light 24a1 reflected at point A is at pixel P C It is received by the pixel P C Therefore, the conditions for the strongest interference (e.g., the light condition) are not met, and instead, for example, the dark condition is met. This is called pixel P C And shown by a relatively thin solid line between point A.

[0106] Figure 6(C) shows the chuck table 4 and workpiece 11 being moved at a predetermined speed in the +X direction, at time t C (>t B This is a diagram showing the imaging of surface 11a at time t. C Therefore, the optical axis 30a passes through point B. Pixel P M Therefore, the condition for the strongest interference is not met, and for example, the dark condition is satisfied. This is called pixel P M And shown by a relatively thin solid line between point B.

[0107] Also, time t C Then, the measurement light 24a1 reflected at point A reaches pixel P D It is received by the pixel P D Therefore, the condition for the strongest interference is not met (for example, the dark condition is met). Pixel P is defined as the condition for the strongest interference not being met. D And shown by a relatively thin solid line between point A.

[0108] To give a slightly more general explanation, in this embodiment, the line passes through point A and through pixel P0 and pixel P N The measurement light 24a1 reflected at an arbitrary point Q (not shown) located on a line segment parallel to the line segment connecting the two points (a hypothetical line segment Q (not shown)) satisfies the condition for the strongest interference at pixel P. By identifying the pixel P that satisfies the condition for the strongest interference, the height of point Q in the Z-axis direction can be determined.

[0109] By moving the measurement optical system 20 and the chuck table 4 relative to each other along the X-axis, the pixel P that satisfies the condition for the strongest interference and the height of point Q in the Z-axis direction can be identified in a one-to-one correspondence along the moving trajectory of the focal point of the measurement light 24a1. In this way, the uneven shape of the surface 11a can be identified along the moving trajectory of the focal point of the measurement light 24a1.

[0110] Note that the X and Y coordinates of points A, B, etc. (i.e., point Q) are determined by the reference pixel P (for example, pixel P MThe controller 40 can determine the position of each pixel P relative to the optical system, the initial positions of the measuring optical system 20 and the workpiece 11, the moving speed, the tilt angle of the optical axis 30a, etc.

[0111] If the moving mechanism 6 has a position information detection mechanism such as a linear encoder, the controller 40 can also determine the X and Y coordinates of points A, B, etc. (i.e., point Q) based on the position information detected by the position information detection mechanism.

[0112] In this embodiment, by tilting the optical axis 30a of the objective lens 30 at a predetermined angle φ with respect to the X-axis direction, and moving the measuring optical system 20 and the chuck table 4 relatively along the X-axis direction, height information of the workpiece 11 can be obtained in the area scanned by the measuring light 24a1.

[0113] Therefore, compared to conventional vertical scanning low-coherence interferometry, the repeated operations of moving the measurement optical system 20 in the Z-axis direction and shifting the chuck table 4 by a predetermined distance in the X-axis direction can be significantly simplified, thus shortening the measurement time when measuring three-dimensional shapes.

[0114] Furthermore, as a method for determining when the interference is strongest, it is conceivable to use the rate of change in brightness in data or a graph that includes the relative position of the measuring optical system 20 and the workpiece 11 (in this example, where the relative movement speed along the X-axis is constant, the relative position is also a function of time) and brightness.

[0115] For example, in a graph with relative position on the horizontal axis and luminance on the vertical axis, the controller 40 monitors the rate of change in luminance (i.e., the derivative of luminance) associated with the change in the relative position of the measurement optical system 20, and based on the optical path difference between the measurement light 24a1 and the reference light 24a2 when the rate of change in luminance is largest, the height Z of point B is determined. B To decide.

[0116] This makes it possible to identify pixels P that satisfy the brightness condition, even when it is difficult to determine whether the brightness condition is met based solely on the brightness level due to differences in reflectivity of the surface 11a, etc. In other words, a shape measurement system 42 with high robustness and noise immunity can be realized.

[0117] Now, let's return to Figure 1. The controller 40 controls the moving mechanism 6, the laser beam irradiation unit 8, the measurement optical system 20, etc. The controller 40 is composed of a computer including, for example, a processor 40a represented by a CPU (Central Processing Unit) and memory 40b.

[0118] Memory 40b includes a main memory such as DRAM (Dynamic Random Access Memory) and an auxiliary memory such as flash memory, a hard disk drive, or a solid-state drive.

[0119] The auxiliary storage device stores software, including a predetermined program. The controller 40's functions are realized by operating the processor 40a and other components according to this software.

[0120] This software changes the relative position between the measuring optical system 20 and the workpiece 11 along the X-axis while having the camera 38 image the surface 11a of the workpiece 11. Interfering light is received by multiple photoelectric conversion elements 36a at each relative position, and the three-dimensional shape is calculated based on the information of each relative position and the amount of light received by each photoelectric conversion element 36a at each relative position.

[0121] Furthermore, the amount of light received by each photoelectric conversion element 36a (i.e., pixel) corresponds to the amount of charge generated by the photodiode in the photoelectric conversion element 36a according to the brightness of the light it receives, and also corresponds to the voltage signal generated by the current-voltage conversion circuit in the photoelectric conversion element 36a.

[0122] Of course, the controller 40 grasps the relative positions of the measuring optical system 20 and the workpiece 11 in the X-axis and Y-axis directions in real time, and this relative position information can be used by the software to calculate the position information of the surface 11a.

[0123] After scanning the surface 11a from one end to the other in the X-axis direction with the measuring light 24a1, the controller 40 moves the measuring optical system 20 and the workpiece 11 in the Y-axis direction by a predetermined amount (for example, the spot diameter of the measuring light 24a1 on the surface 11a) using the moving mechanism 6 (i.e., performs index feeding).

[0124] By repeatedly scanning the surface 11a with the measurement light 24a1 in the X-axis direction and feeding the index in the Y-axis direction, the measurement light 24a1 scans approximately the entire surface 11a. Alternatively, instead of scanning approximately the entire surface 11a, only the area of ​​the three-dimensional shape to be measured may be scanned with the measurement light 24a1.

[0125] The software sequentially calculates the surface irregularities of the surface 11a based on the above-described (1) or (2), according to the brightness measured by a photoelectric conversion element 36a that receives interference light between the measurement light 24a1 reflected from the surface 11a of the workpiece 11 and the reference light 24a2.

[0126] In other words, the software calculates the three-dimensional shape of the surface 11a based on the information of the relative positions of the measuring optical system 20 and the workpiece 11, and the amount of light received by each photoelectric conversion element 36a at each relative position.

[0127] The calculated three-dimensional shape of the surface 11a is displayed on the display device 44 by the controller 40. The display device 44 is, for example, a touch panel display that also functions as an input device. The operator can set processing conditions and perform other operations through the display device 44.

[0128] Next, with reference to Figures 7 to 10, a processing method for performing laser processing on a workpiece 11 using the processing apparatus 2 (i.e., processing the workpiece 11) will be described. This processing method includes a shape measurement method for measuring the three-dimensional shape of the workpiece 11 (see S20, S30, and S40 in Figure 7).

[0129] Figure 7 is a flowchart showing the processing method for the workpiece 11. In this embodiment, each step is performed in the order of S10 to S40. First, the workpiece 11 is subjected to ablation processing with the laser beam irradiation unit 8 (i.e., the workpiece 11 is processed) to form a processing mark 11d (see Figure 9(B)) on the workpiece 11 (processing step S10).

[0130] Figure 8 is a perspective view of the workpiece 11 before processing. The workpiece 11 includes a disc-shaped wafer made of single-crystal silicon, a semiconductor material. The workpiece 11 has a circular front surface 11a and a circular back surface 11b. Multiple mutually orthogonal division lines 13 are set on the front surface 11a.

[0131] Each of the multiple rectangular regions demarcated by multiple division lines 13 is provided with a device 15 such as an IC (Integrated Circuit). Each division line 13 and each device 15 is covered with a low-k film (not shown). However, the low-k film is not mandatory.

[0132] Furthermore, there are no restrictions on the material, shape, structure, size, etc. of the wafer. The workpiece 11 may have a substrate made of semiconductors other than silicon (GaAs, SiC, GaN, etc.), diamond, sapphire, glass, ceramics, resin, metal, etc. There are also no restrictions on the type, number, shape, structure, size, arrangement, etc. of the devices 15. The workpiece 11 does not have to have devices 15.

[0133] An annular metal frame 17 is positioned around the outer periphery of the workpiece 11, enclosing the workpiece 11. A dicing tape 19 with a diameter larger than the opening of the frame 17 is attached to one surface of the frame 17 and the back surface 11b of the workpiece 11.

[0134] In other words, the workpiece 11 is supported by the frame 17 via the dicing tape 19. The workpiece 11, the frame 17, and the dicing tape 19 constitute the workpiece unit 21.

[0135] When forming the processing marks 11d, the workpiece 11 is held by the chuck table 4 via the dicing tape 19 so that the surface 11a is exposed upwards (see Figure 9(A)). Note that in Figure 9(A), the frame 17 and dicing tape 19 are omitted as described above.

[0136] Then, after adjusting the position of the chuck table 4 so that the focal point of the laser beam L is located near the surface 11a in the Z-axis direction and on the extension of one planned division line 13 in the X-axis direction, the chuck table 4 is moved along the X-axis direction while the laser beam L is irradiated. The laser processing conditions are set as follows, for example.

[0137] Wavelength: 355nm Average output: 0.7W Repeat frequency: 200kHz Pulse width: 9 ps Machining feed rate: 600 mm / s

[0138] Figure 9(A) shows the process of performing ablation on the workpiece 11. After machining one division line 13 from one end in the X-axis direction to the other, the chuck table 4 is moved along the Y-axis direction by a predetermined index feed amount. Then, ablation is performed in the same manner on other division lines 13 adjacent to the machined division line 13 in the Y-axis direction.

[0139] After ablation processing is performed on all planned division lines 13 along the X-axis in this manner, the chuck table 4 is rotated by approximately 90 degrees using the θ table of the moving mechanism 6. Then, ablation processing is performed similarly on each of the unprocessed planned division lines 13.

[0140] Figure 9(B) is a cross-sectional view of a workpiece 11 including a processing mark 11d formed on the surface 11a by ablation. The processing mark 11d is a processing groove having a predetermined depth from the surface 11a (i.e., a depth that does not completely cut the workpiece 11), and comprises a processing groove 11d1 and molten deposit portions 11d2 located at both edges of the processing groove 11d1.

[0141] Note that the molten deposit area 11d2 shown in Figure 9(B) is explicitly shown for explanatory purposes, but it is not always formed uniformly along the longitudinal direction of the planned division line 13. When the processing mark 11d is viewed from above, there are also regions where the molten deposit area 11d2 is partially formed.

[0142] Furthermore, it is not essential to form machining marks 11d on all planned division lines 13; machining marks 11d may be formed on at least one planned division line 13. Alternatively, to confirm the machining performance, circular or elliptical machining marks 11d may be formed on a portion of one planned division line 13 or on a portion of a test piece.

[0143] In this embodiment, after forming a processing mark 11d on at least one planned division line 13 (i.e., after the processing step S10), the relative position between the measuring optical system 20 and the workpiece 11 is changed along the X-axis direction, and the processing mark 11d is imaged (imaging step S20).

[0144] In the imaging process S20, as described above, the measurement light 24a1 from the branched low-coherence light 24a is irradiated onto the workpiece 11 held on the holding surface 4a via the objective lens 30, and the interference light resulting from the interference of the reference light 24a2 and the measurement light 24a1 reflected from the surface 11a of the workpiece 11 is received by the camera 38 (light receiving process), and the relative position between the camera 38 and the workpiece 11 is changed along the X-axis (X-axis movement process).

[0145] For example, in the imaging process S20, first, the measurement light 24a1 is irradiated onto one end of the processing mark 11d in the X-axis direction (see Figure 10(A)). Figure 10(A) shows the start of imaging of one processing mark 11d.

[0146] With the measurement light 24a1 irradiated onto one end of the processed mark 11d, the chuck table 4 is moved in the +X direction at a predetermined speed (for example, 50 mm / s) while imaging approximately the entire processed mark 11d up to the other end in the X-axis direction. Figure 10(B) shows the point at which imaging of one processed mark 11d is completed.

[0147] In this embodiment, substantially the entire processing mark 11d is imaged with a single scan in the X-axis direction, but this is not limited to this, and substantially the entire processing mark 11d may be imaged with multiple scans in the X-axis direction.

[0148] For example, the spot diameter of the measurement light 24a1 on the surface 11a may be narrowed, and the entire area of ​​a single processing mark 11d may be imaged by scanning slightly shifted areas in the Y-axis direction. Alternatively, the entire area of ​​a single processing mark 11d may be imaged by changing the focal point position of the measurement light 24a1 in the Z-axis direction with each scan.

[0149] As described above, the measuring optical system 20 scans the workpiece 11 with the measuring light 24a1 by moving the camera 38, the objective lens 30, and the chuck table 4 relatively along the X-axis, with the optical axis 30a of the objective lens 30 tilted at a predetermined angle φ with respect to the X-axis.

[0150] Then, the three-dimensional shape of the workpiece 11 is measured by receiving the interference light of the measurement light 24a1 reflected from the workpiece 11 and the reference light 24a2 with the camera 38 (light receiving step), and by changing the relative position of the camera 38 and the workpiece 11 in the X-axis direction (X-axis direction movement step).

[0151] In contrast, conventionally, height information of the workpiece 11 was acquired in a band-shaped region along the X-axis by repeatedly moving a measuring optical system 20 having an objective lens 30 with an optical axis 30a arranged parallel to the Z-axis direction, and a chuck table 4 relatively in the Z-axis direction, and then shifting the measuring optical system 20 and the chuck table 4 relatively by a predetermined distance in the X-axis direction.

[0152] In this embodiment, this repetitive operation can be significantly simplified, thus shortening the measurement time when measuring the three-dimensional shape of the workpiece 11 compared to conventional methods. After the imaging step S20, the controller 40 determines whether or not all processing marks 11d formed on the workpiece 11 have been imaged (determination step S30).

[0153] If the entire machining mark 11d has not been imaged (NO in S30), the measuring optical system 20 and the workpiece 11 are moved a predetermined distance relative to each other along the Y-axis (index feed step S40). Then, the process returns to the imaging step S20, and other machining marks 11d adjacent to the already imaged machining marks 11d are imaged in the same manner.

[0154] The imaging process S20 and the index feeding process S40 are repeated until all processing marks 11d are captured. If all processing marks 11d have been captured (YES in S30), the process ends.

[0155] (Second Embodiment) Next, a second embodiment will be described with reference to Figures 11 to 12(C). The second embodiment relates to a method for manufacturing a device chip 23 in which a plurality of device chips 23 (see Figure 12(C)) are manufactured from a workpiece 11.

[0156] Figure 11 is a flowchart showing an example of a method for manufacturing the device chip 23. In the second embodiment, as in the processing step S10 of the first embodiment, processing grooves (i.e., the processing marks 11d described above) are formed on each planned division line 13 by ablation processing using a laser beam L (processing step S12).

[0157] After the machining process S12, the imaging process S20, judgment process S30, and index feeding process S40 in the first embodiment are repeated as appropriate (shape measurement process S50). This allows all machined grooves to be imaged and their three-dimensional shapes to be measured. Of course, in the shape measurement process S50, the measurement time for measuring the three-dimensional shape can be shortened compared to conventional vertical scanning low-coherence interferometry.

[0158] After imaging all the machined grooves and measuring the three-dimensional shape of each machined groove, the process proceeds to the workpiece division step S60. In the division step S60 of this embodiment, first, a modified region 11e with reduced mechanical strength is formed along each planned division line 13 between the bottom 11d3 of the machined groove 11d1 and the back surface 11b of the workpiece 11 (see the modified region formation step in the division step S60, Figure 12(A)).

[0159] Figure 12(A) shows the modification region formation process. The modification region formation process can use essentially the same processing apparatus 2 as the ablation process. Therefore, in Figure 12(A), components having the same function as processing apparatus 2 are denoted by the same reference numerals as the components shown in processing apparatus 2 in Figure 1.

[0160] However, the laser beam irradiation unit 8 shown in Figure 12(A) irradiates the workpiece 11 with a pulsed laser beam L from the condenser 10 that has a wavelength that can penetrate the workpiece 11 (for example, 1064 nm), rather than a wavelength that is absorbed by the workpiece 11.

[0161] In the modification region formation process, the chuck table 4 is moved along the X-axis while the focusing point of the laser beam L is positioned between the bottom 11d3 of the processing groove 11d1 and the back surface 11b of the workpiece 11 in the Z-axis direction. The laser processing conditions are set as follows, for example.

[0162] Wavelength: 1064nm Average output: 1.0W Repeat frequency: 30kHz Pulse width: 3ns Machining feed rate: 165 mm / s Number of passes: 2

[0163] In the modification region formation process, the depth position of the focusing point of the laser beam L in the Z-axis direction can be adjusted by utilizing the three-dimensional shape of the processed groove (i.e., the processed mark 11d) obtained in the shape measurement process S50.

[0164] For example, the position of the focusing point may vary for each division line 13 due to rough surface conditions at the bottom 11d3 of the processed groove 11d1. However, by fine-tuning the position of the focusing device 10 in the Z-axis direction, the possibility of the modified region 11e varying for each division line 13 due to the surface conditions at the bottom 11d3 can be reduced.

[0165] After forming a modified region 11e along each planned division line 13, the back surface 11b of the workpiece 11 is ground with the grinding device 50 (back surface grinding step in division step S60, see Figure 12(B)). The grinding device 50 has a disc-shaped chuck table 52.

[0166] The chuck table 52 has a holding surface 52a that holds the workpiece 11 by suction. The holding surface 52a has a conical shape in which the central part protrudes slightly (for example, by only 20 μm) compared to the outer circumference, but since the amount of protrusion is small, the holding surface 52a is shown as substantially flat in Figure 12(B).

[0167] A rotational drive source (not shown), such as a servo motor, is provided at the bottom of the chuck table 52. The chuck table 52 is rotatable around a rotation axis 52b located at the radial center of the chuck table 52.

[0168] The rotating shaft 52b is tilted such that a portion of the holding surface 52a is approximately parallel to the XY plane. However, since the tilt is slight, in Figure 12(B), the rotating shaft 52b is shown as approximately parallel to the Z-axis direction. A grinding unit 54 is provided above the chuck table 52.

[0169] The grinding unit 54 is configured to be movable along the Z-axis direction by a Z-axis direction movement mechanism (not shown). It has a cylindrical spindle 56 arranged along the Z-axis direction. A disc-shaped mount 58 is fixed to the lower end of the spindle 56.

[0170] An annular grinding wheel 60 is mounted on the lower surface of the mount 58. The grinding wheel 60 has an annular wheel base 60a. Multiple grinding wheels 60b are arranged on the lower surface of the wheel base 60a at approximately equal intervals along the circumferential direction of the wheel base 60a.

[0171] In the back grinding process using the grinding device 50, the workpiece 11 is first held in place by suction on the holding surface 52a. In this example, the surface 11a of the workpiece 11 is covered with a resin protective member 25, and then the workpiece 11 is held in place by suction on the holding surface 52a so that the back surface 11b is exposed.

[0172] Next, the chuck table 52 and spindle 56 are rotated at predetermined speeds, and the grinding unit 54 is fed downward at a predetermined speed. The area of ​​the back surface 11b that is arranged approximately parallel to the XY plane is ground by contacting the multiple grinding wheels 60b.

[0173] The grinding area on the back surface 11b is semi-circular, but as the chuck table 52 rotates around the rotation axis 52b, the entire back surface 11b is ground almost uniformly, and the workpiece 11 is thinned. In the back surface grinding process, the workpiece 11 is thinned, and the impact applied to the workpiece 11 is used to propagate cracks from the modified area 11e.

[0174] As the cracks reach the surface 11a and back surface 11b, the workpiece 11 is cut along each planned division line 13 so as to pass through the bottom 11d3 of the machining groove, and is divided into multiple device chips 23. Figure 12(B) shows the back surface grinding process, and Figure 12(C) is a perspective view of one of the multiple device chips 23.

[0175] In this embodiment, the division process S60 is performed by a modified region formation process and a back grinding process. However, after measuring the three-dimensional shape of each processed groove, instead of the modified region formation process and grinding process, the workpiece 11 may be cut at each planned division line 13 by blade dicing using a cutting device (not shown) to divide it into multiple device chips 23.

[0176] (Modified Version of the Second Embodiment) Next, a modified version of the second embodiment will be described with reference to Figures 13 and 14. Figure 13 is a flowchart showing a modified version of the manufacturing method of the device chip 23.

[0177] In this modified example, an ablation process is performed on the unprocessed workpiece 11 to form processing grooves 11f that completely cut the workpiece 11 at each planned division line 13, thereby dividing the workpiece 11 into multiple device chips 23 (processing step S14). Figure 14 is a cross-sectional view of the workpiece 11 cut by the ablation process.

[0178] After the machining process S14, the shape measurement process S50 is performed (i.e., the imaging process S20, judgment process S30, and index feeding process S40 in the first embodiment are repeated as appropriate). This captures images of all machined grooves 11f and measures their three-dimensional shape.

[0179] After measuring the three-dimensional shape of all machined grooves 11f, the controller 40 determines whether the three-dimensional shape of the machined grooves 11f obtained in the shape measurement step S50 falls outside a predetermined tolerance range (determination step S70). After the determination step S70, the flow ends. Even in this modified example, the measurement time for measuring the three-dimensional shape can be shortened compared to the conventional vertical scanning low-coherence interferometry.

[0180] Furthermore, in this modified version, if the three-dimensional shape of the machined groove 11f is outside a predetermined tolerance range, the controller 40 determines that the device chip 23 constituting the machined groove 11f that is outside the tolerance range is a defective product.

[0181] By inspecting the micro-shape after laser processing in this way, for example, the worker can choose to discard the defective device chip 23. This reduces the possibility of malfunctions occurring in products equipped with the device chip 23.

[0182] Furthermore, the structures, methods, etc., of the embodiments described above can be modified as appropriate without departing from the scope of the object of the present invention. In the embodiments and modifications described above, the case in which a three-dimensional shape such as a processing mark 11d formed by laser processing is measured by the measuring optical system 20 has been explained, but the three-dimensional shape of a region processed by mechanical processing such as cutting, grinding, polishing may also be measured by the measuring optical system 20.

[0183] Alternatively, for example, an event camera, also known as an EVS (Event-based Vision Sensor), can be used instead of the camera 38 described above. The event camera also has a solid-state image sensor, and the imaging surface of this solid-state image sensor is provided with a plurality of photoelectric conversion elements arranged regularly (for example, in a grid pattern), each capable of receiving interference light.

[0184] Each photoelectric conversion element includes, for example, a photodiode and a current-voltage conversion circuit, and generates a voltage signal corresponding to the brightness of the received light. The solid-state image sensor includes a brightness change detection circuit that is integrally provided with the region on which the multiple photoelectric conversion elements are located.

[0185] The brightness change detection circuit does not always output voltage signals from all photoelectric conversion elements to the controller 40. Instead, the brightness change detection circuit selectively outputs event data to the controller 40 in response to brightness changes in the photoelectric conversion elements.

[0186] The event data includes information that can identify (A) the coordinates (x,y) (i.e., position) of the photoelectric element where the brightness change occurred, and (B) information (i) indicating that there was a positive or negative brightness change in that photoelectric element, and further includes (C) the time (t) (i.e., timestamp) when the brightness change occurred.

[0187] Furthermore, whether or not there has been a change in brightness in the photoelectric conversion element is determined by a brightness change detection circuit, for example, by comparing the measured brightness with a threshold value. However, this threshold value is updated as appropriate for each photoelectric conversion element over time.

[0188] The event camera does not output voltage signals from all photoelectric elements to the controller 40 for each frame, but rather outputs event data for only the photoelectric elements that have experienced a change in brightness to the controller 40. Therefore, the amount of information in each frame can be reduced compared to that of a typical CMOS image sensor.

[0189] By using an event camera, the amount of image information output to the controller 40 can be significantly reduced compared to when interference light is received with a general CMOS image sensor, allowing for a higher frame rate. Therefore, there is an advantage in being able to measure 3D shapes at a faster speed. [Explanation of Symbols]

[0190] 2: Processing equipment 4: Chuck table (holding unit), 4a: Holding surface, 6: Moving mechanism (moving unit) 8: Laser beam irradiation unit (processing unit) 10: Focuser, 10a: Mirror, 10b: Focusing lens 11: Workpiece (object), 11a: Surface (surface to be measured), 11b: Back surface 11c: Measured area 11d: Machining marks (machined groove area), 11d1: Machined groove, 11d2: Molten deposit area, 11d3: Bottom 11e: Modified area, 11f: Machined groove 12: Laser Oscillator 13: Dividing line, 15: Device, 17: Frame, 19: Dicing tape 20: Measuring optical system, 22: Housing 21: Workpiece unit, 23: Device chip, 25: Protective member 24: White light source (light source), 24a: Low coherence light 24a1: Measurement light, 24a2: Reference light, 24b: Coherence distance 26: First beam splitter, 28: Second beam splitter (branching point) 30: Objective lens, 30a: Optical axis 32: Mirror, 32a: Reference plane, 34: Imaging lens 36: Solid-state image sensor, 36a: Photoelectric conversion element, 38: Camera (Photoelectric conversion device) 40: Controller, 40a: Processor, 40b: Memory 42: Shape measurement system, 44: Display device 50: Grinding equipment 52: Chuck table, 52a: Holding surface, 52b: Rotating shaft 54: Grinding unit, 56: Spindle, 58: Mount 60: Grinding wheel, 60a: Wheel base, 60b: Grinding wheel L: Laser beam L0,L A ,L B ,L N ,L REF :Optical path length P,P0,P B ,PC ,P D ,P M ,P N :pixel Z0: Reference height, Z B : Height of point B ΔL B : Optical path difference, φ: predetermined angle S10,S12,S14: Machining process S20: Imaging process, S30: Decision process, S40: Index feeding process S50: Shape measurement process, S60: Division process, S70: Judgment process

Claims

1. A shape measuring system for measuring the three-dimensional shape of an object, A holding unit for holding the object, A measuring optical system positioned away from the holding unit in a first direction, A moving unit that moves the measuring optical system and the holding unit relative to each other along a second direction intersecting the first direction, A controller for controlling the mobile unit, Equipped with, The measuring optical system is, A light source that emits low-coherence light, A branching unit that splits the light from the light source into a reference light that irradiates a reference surface which is a reflective surface, and a measurement light that irradiates the object, An objective lens that focuses the measurement light onto the object, A photoelectric conversion device having a plurality of photoelectric conversion elements that receive interference light resulting from the interference of the reference light reflected from the reference surface and the measurement light reflected from the object, It has, A shape measuring system characterized in that, when the measuring optical system and the holding unit are moved relative to each other along the second direction by the moving unit and the object is scanned with the measuring light, the optical axis of the objective lens is inclined at a predetermined angle with respect to the second direction.

2. The controller is, The shape measurement system according to claim 1, characterized in that by controlling the moving unit, the relative position of the measuring optical system and the object is changed, and the photoelectric converter is made to image the object, thereby receiving the interference light of the measurement light reflected from the surface of the object to be measured and the reference light at each relative position with a plurality of photoelectric converters, and calculating the three-dimensional shape based on the information of each relative position and the amount of light received by each photoelectric converter at each relative position.

3. A processing device for processing a workpiece, A holding unit for holding the workpiece, A processing unit for processing the workpiece, A measuring optical system positioned at a distance in a first direction from the holding unit, A moving unit that moves the measuring optical system and the holding unit relative to each other along a second direction intersecting the first direction, A controller for controlling the mobile unit, Equipped with, The measuring optical system is, A light source that emits low-coherence light, A branching unit that splits the light from the light source into a reference light that irradiates a reference surface which is a reflective surface, and a measuring light that irradiates the workpiece, An objective lens that focuses the measuring light onto the workpiece, A photoelectric conversion device having a plurality of photoelectric conversion elements that receive interference light resulting from the interference of the reference light reflected from the reference surface and the measurement light reflected from the workpiece, It has, A processing apparatus characterized in that, when the measuring optical system and the holding unit are moved relative to each other along the second direction by the moving unit and the workpiece is scanned with the measuring light, the optical axis of the objective lens is inclined at a predetermined angle with respect to the second direction.

4. A shape measurement method for measuring the three-dimensional shape of an object, The low-coherence light is split into a reference light irradiated onto a reference surface and a measurement light irradiated onto the object, and the measurement light is irradiated onto the object held by the holding unit via an objective lens. The interference light resulting from the interference between the reference light reflected from the reference surface and the measurement light reflected from the object is received by a photoelectric converter. The relative positions of the photoelectric conversion device and the object held by the holding unit, which are arranged apart from each other in a first direction, are changed along a second direction that intersects the first direction. Equipped with, The three-dimensional shape of the object is measured by receiving the interference light with the photoelectric converter and changing the relative position between the photoelectric converter and the object in the second direction. A shape measurement method characterized by changing the relative positions of the photoelectric conversion device and the object along the second direction, so that the optical axis of the objective lens is tilted at a predetermined angle with respect to the second direction, and the photoelectric conversion device, the objective lens, and the holding unit are moved relative to each other along the second direction to scan the object with the measuring light.

5. A processing method for processing a workpiece, The process involves processing the workpiece with a processing unit to form processing marks, After the formation of the processing mark, the measurement light, which is split into a reference light irradiated onto the reference surface and a measurement light irradiated onto the processing mark, is irradiated onto the processing mark of the workpiece held by the holding unit via an objective lens, and the interference light resulting from the interference between the reference light reflected from the reference surface and the measurement light reflected from the processing mark is received by a photoelectric converter. After the formation of the processing mark, the relative positions of the photoelectric conversion device and the processing mark on the workpiece held by the holding unit, which are positioned apart from each other in a first direction, are changed along a second direction intersecting the first direction. Equipped with, The three-dimensional shape of the processing mark is measured by receiving the interference light with the photoelectric converter and changing the relative position between the photoelectric converter and the processing mark in the second direction. A processing method characterized by changing the relative positions of the photoelectric conversion device and the processing mark along the first direction, so that the optical axis of the objective lens is tilted at a predetermined angle with respect to the second direction, and the photoelectric conversion device, the objective lens, and the holding unit are moved relative to each other along the second direction to scan the processing mark with the measuring light.

6. A method for manufacturing chips from a workpiece, The workpiece is processed by a processing unit to form a processing groove having a depth that does not completely cut through the workpiece, After the formation of the machined groove, the three-dimensional shape of the machined groove is measured, After measuring the three-dimensional shape of the machined groove, the workpiece is cut so as to pass through the bottom of the machined groove and divided into chips. Equipped with, Measuring the three-dimensional shape of the machined groove is The low-coherence light is split into a reference light irradiated onto a reference surface and a measurement light irradiated onto the machining groove, and the measurement light is irradiated onto the machining groove of the workpiece held by the holding unit via an objective lens. The interference light resulting from the interference between the reference light reflected from the reference surface and the measurement light reflected from the machining groove is received by a photoelectric converter. The relative position of the photoelectric converters, which are arranged apart from each other in a first direction, and the machining groove portion of the workpiece held by the holding unit, is changed along a second direction that intersects the first direction. It has, A method for manufacturing a chip, characterized in that the relative positions of the photoelectric conversion device and the processed groove portion of the workpiece held by the holding unit are changed along the second direction, so that the optical axis of the objective lens is tilted at a predetermined angle with respect to the second direction, and the photoelectric conversion device, the objective lens, and the holding unit are moved relative to each other along the second direction to scan the processed groove portion with the measuring light.

7. A method for manufacturing chips from a workpiece, The workpiece is processed by a processing unit to form a processing groove that completely cuts the workpiece and divides it into chips, After the formation of the machined groove, the three-dimensional shape of the machined groove is measured, Equipped with, Measuring the three-dimensional shape of the machined groove is The low-coherence light is split into a reference light irradiated onto a reference surface and a measurement light irradiated onto the machining groove, and the measurement light is irradiated onto the machining groove of the workpiece held by the holding unit via an objective lens. The interference light resulting from the interference between the reference light reflected from the reference surface and the measurement light reflected from the machining groove is received by a photoelectric converter. The relative positions of the photoelectric converter and the workpiece groove held by the holding unit, which are positioned apart from each other in a first direction, are changed along a second direction that intersects the first direction. It has, A method for manufacturing a chip, characterized in that the relative positions of the photoelectric conversion device and the processed groove portion of the workpiece held by the holding unit are changed along the second direction, so that the optical axis of the objective lens is tilted at a predetermined angle with respect to the second direction, and the photoelectric conversion device, the objective lens, and the holding unit are moved relative to each other along the second direction to scan the processed groove portion with the measuring light.

Images