Ultrasonic vibration device, cleaning method, substrate manufacturing method, and chip manufacturing method

Abstract

To provide a novel ultrasonic vibration application device that facilitates the propagation of ultrasonic vibrations into liquids. [Solution] An ultrasonic vibration device 2 for applying ultrasonic vibrations to a liquid, comprising: a main body 4 having an inlet 4h1 into which the liquid enters; an ultrasonic transducer 8 provided inside the main body and including a vibrating surface 8c1 that contacts the liquid; and a planar region 4i provided inside the main body and positioned opposite to and separated from the vibrating surface. 2A The main body includes an adjustment wall 4i2 defined by a vibration-applying channel 4j with respect to a planar region and a vibrating surface, and the main body includes an outflow channel 4i4, the outflow channel having one end 4i connected to the vibration-applying channel. 4A The other end 4i corresponds to the outlet through which the liquid, which has been subjected to ultrasonic vibrations in the vibration-applying channel by the ultrasonic vibrations of the ultrasonic transducer, flows out of the main body. 4B Including the above, and with λ being the wavelength of the liquid in the vibrating channel and N being a natural number, the distance X between the vibrating surface and the planar region is expressed as N·λ / 2 - 0.1·λ / 2 ≤ X ≤ N·λ / 2 + 0.1·λ / 2.

Description

[Technical Field]

[0001] The present invention relates to an ultrasonic vibration device for applying ultrasonic vibrations to a liquid, a cleaning method for cleaning an object to be cleaned using the ultrasonic vibration device, a method for manufacturing a substrate using the ultrasonic vibration device to produce a substrate having a thickness less than the thickness of the workpiece, and a method for manufacturing a chip by dividing a workpiece using an ultrasonic generator to produce multiple chips. [Background technology]

[0002] A cleaning nozzle is known that cleans an object by supplying cleaning water with ultrasonic vibrations to the object from the cleaning nozzle (see, for example, Patent Document 1). This cleaning nozzle has a rectangular box, and a slit-shaped nozzle is formed on the bottom surface of the box.

[0003] Furthermore, a long ultrasonic vibrating plate is provided inside the box, opposite the nozzle, so as to be aligned with the longitudinal direction of the box. The ultrasonic vibrating plate has a first electrode, a piezoelectric material layer, and a second electrode, each being a thin plate with its long side aligned with the longitudinal direction of the box.

[0004] The second electrode has a semi-cylindrical curved surface and is a thin, convex plate that protrudes from the bottom surface of the box to the top surface of the box. The second electrode is positioned to cover the nozzle. A piezoelectric material layer is fixed on top of the second electrode in a manner that conforms to the shape of the second electrode, and furthermore, the first electrode is fixed on top of the piezoelectric material layer in a manner that conforms to the shape of the piezoelectric material layer.

[0005] When the first electrode and the second electrode are connected to a high-frequency power supply, and power is supplied from the high-frequency power supply to the ultrasonic vibrator at the ultrasonic frequency, the ultrasonic vibrator vibrates at the ultrasonic frequency. Due to the shape of the ultrasonic vibrator, the ultrasonic vibrations transmitted from the ultrasonic vibrator to the cleaning water are focused toward the nozzle, making it easier for the ultrasonic vibrations to be transmitted to the cleaning water and thus increasing the cleaning power.

[0006] However, if air bubbles are present in the liquid reservoir (i.e., the vibration-impregnating channel) between the second electrode and the nozzle, the convex shape of the ultrasonic diaphragm causes the vibration-impregnating channel to become semi-cylindrical, making it easy for air bubbles to accumulate in the channel. These accumulated air bubbles can then hinder the propagation of ultrasonic vibrations, potentially reducing the sound pressure near the nozzle.

[0007] On the other hand, if the ultrasonic vibrating plate has a nearly flat shape without any convex shape, the accumulation of air bubbles in the vibration-impregnating channel can be suppressed compared to when the vibration-impregnating channel is semi-cylindrical, but the ultrasonic vibrations become less likely to propagate to the cleaning water. However, in order to improve cleaning power, it is desirable to efficiently propagate ultrasonic vibrations to the cleaning water. [Prior art documents] [Patent Documents]

[0008] [Patent Document 1] Japanese Patent Publication No. 2021-041353 [Overview of the project] [Problems that the invention aims to solve]

[0009] This invention has been made in view of the aforementioned problems, and aims to provide a novel ultrasonic vibration application device that facilitates the propagation of ultrasonic vibrations into a liquid without requiring the vibration application channel to be semi-cylindrical in shape. [Means for solving the problem]

[0010] According to one aspect of the present invention, there is provided an ultrasonic vibration applying device for applying ultrasonic vibration to a liquid, comprising: a main body having an inlet through which the liquid enters; an ultrasonic vibrator provided inside the main body and including a vibration surface that contacts the liquid; and an adjustment wall provided inside the main body and having a planar region disposed to face the vibration surface at a predetermined distance therefrom, the vibration applying flow path being defined by the planar region and the vibration surface. The main body includes an outflow path, the outflow path including one end connected to the vibration applying flow path and the other end corresponding to an outlet through which the liquid to which ultrasonic vibration has been applied in the vibration applying flow path flows out of the main body. When the wavelength of the liquid in the vibration applying flow path is λ and N is a natural number, there is provided an ultrasonic vibration applying device in which the distance X between the vibration surface and the planar region is represented by the following formula (1).

[0011] [Number]

[0012] Preferably, N is 1.

[0013] Preferably, the adjustment wall and the liquid have the same acoustic impedance.

[0014] According to another aspect of the present invention, there is provided a cleaning method for cleaning an object to be cleaned using the above-described ultrasonic vibration applying device, comprising: an ultrasonic vibration applying step of applying ultrasonic vibration to the liquid supplied to the vibration applying flow path of the main body; and a supply step of supplying the liquid to which ultrasonic vibration has been applied from the outlet of the main body to the object to be cleaned to clean the object to be cleaned.

[0015] According to yet another aspect of the present invention, a method for manufacturing a substrate having a thickness less than the thickness of a workpiece is provided, using the ultrasonic vibration application device described above, comprising: a separation point formation step of forming a separation point inside the workpiece; and a crack extension step of applying ultrasonic vibration to the liquid supplied to the vibration application channel of the main body after the separation point formation step, and supplying the liquid to which ultrasonic vibration has been applied to the workpiece from the outlet of the main body, thereby extending a crack from the separation point.

[0016] According to yet another aspect of the present invention, a method for manufacturing a chip is provided for manufacturing a chip by dividing a workpiece using the ultrasonic vibration application device described above, comprising: a division starting point formation step of forming a division starting point at each of a plurality of division planned lines set on one surface of the workpiece; and a division step of dividing the workpiece into a plurality of chips starting from the plurality of division starting points, after the division starting point formation step of applying ultrasonic vibration to the liquid supplied to the vibration application channel of the main body, and supplying the liquid to which ultrasonic vibration has been applied from the outlet of the main body to each of the plurality of division starting points on the workpiece. [Effects of the Invention]

[0017] An ultrasonic vibration application device according to one aspect of the present invention comprises an ultrasonic transducer including a vibrating surface in contact with a liquid, and an adjustment wall having a planar region positioned opposite to the vibrating surface at a predetermined distance, the planar region and the vibrating surface define a vibration application channel. Furthermore, when the wavelength in the liquid in the vibration application channel is λ and N is a natural number, the distance X between the vibrating surface and the planar region is expressed by (1) below.

[0018] As shown in (1), the ultrasonic transducer and adjustment wall are arranged in such a way that the vibration-applying channel does not need to be semi-cylindrical in shape, and the ultrasonic vibrations are more easily transmitted to the liquid compared to when the ultrasonic transducer and adjustment wall are not arranged as shown in (1).

[0019]

number

[0020] [Figure 1] This is a perspective view of the nozzle unit. [Figure 2] This is a partial cross-sectional side view of the nozzle unit. [Figure 3] Figure 3(A) is a cross-sectional view of the lower half of the nozzle unit, and Figure 3(B) shows the sound pressure distribution obtained from the simulation superimposed on the lower half of the nozzle unit. [Figure 4] This is a simulation result showing the dependence of acoustic output on (interval X / wavelength λ). [Figure 5] Figure 5(A) shows the simulation results when the adjustment wall is made of stainless steel, and Figure 5(B) shows the simulation results when the adjustment wall is made of fused silica. [Figure 6] This is a flowchart of the cleaning method. [Figure 7] This is a perspective view of the workpiece. [Figure 8] Figure 8(A) is a perspective view of the cutting process, and Figure 8(B) is a partial cross-sectional side view of the cutting process. [Figure 9] This is a partial cross-sectional side view showing the cleaning process. [Figure 10] This is a flowchart of the manufacturing method for the substrate. [Figure 11] Figure 11(A) is a partial cross-sectional side view showing the separation point formation process, and Figure 11(B) is a top view of the workpiece showing the separation point formation process. [Figure 12] This is a partial cross-sectional side view showing the crack propagation process. [Figure 13] Figure 13(A) is a partial cross-sectional side view showing the separation apparatus, and Figure 13(B) is a partial cross-sectional side view showing the separation process. [Figure 14] This is a flow chart of the chip manufacturing method. [Figure 15]Figure 15(A) is a partial cross-sectional side view showing the splitting point formation process, and Figure 15(B) is a perspective view of the workpiece unit after the splitting point formation process. [Figure 16] Figure 16(A) is a partial cross-sectional side view showing the splitting process, and Figure 16(B) is an enlarged perspective view of the chip. [Modes for carrying out the invention]

[0021] (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 perspective view of a nozzle unit (i.e., an ultrasonic vibration device) 2 that imparts ultrasonic vibration to a liquid 11, and Figure 2 is a partial cross-sectional side view of the nozzle unit 2.

[0022] In Figures 1 and 2, for the sake of explanation, the liquid 11 is indicated by an arrow. Also, the A-axis shown in Figures 1 and 2 is parallel to the height direction of the nozzle unit 2, and in one example, it is parallel to the vertical direction.

[0023] The nozzle unit 2 has a cylindrical body portion 4 made of a hard resin or the like. The longitudinal direction of the body portion 4 defines the height direction of the nozzle unit 2. The body portion 4 includes a disc-shaped top cover 4a.

[0024] A small cylindrical portion 4b, which is cylindrical and has a smaller diameter than the outer diameter of the top lid 4a, is provided in the center of the top lid 4a so as to protrude from the upper surface of the top lid 4a. The top lid 4a and the small cylindrical portion 4b are arranged concentrically. A cylindrical cavity 4c is formed in the center of both the top lid 4a and the small cylindrical portion 4b so as to penetrate the top lid 4a and the small cylindrical portion 4b.

[0025] Wirings 6a and 6b are inserted into the cavity 4c. Wirings 6a and 6b electrically connect the disc-shaped ultrasonic transducer 8 (see Figure 2) and the ultrasonic oscillator 10. In Figures 1 and 2, the ultrasonic oscillator 10 is shown as a functional block.

[0026] The ultrasonic oscillator 10 is located outside the cavity 4c and vibrates the ultrasonic transducer 8 at an ultrasonic vibration frequency (for example, 300 kHz to 3 MHz). A large cylindrical part 4d, having an outer diameter approximately the same as the outer diameter of the top cover 4a, is fixed to the lower surface of the top cover 4a.

[0027] The large cylindrical section 4d has a cylindrical upper section with the largest diameter, a cylindrical middle section with a slightly smaller diameter than the upper section, and an inverted truncated cone-shaped lower section. In Figure 2, for the sake of explanation, the boundaries of the upper, middle, and lower sections of the large cylindrical section 4d are shown with dashed lines. As shown in Figure 2, a cavity 4e is formed inside the large cylindrical section 4d, which is connected to the cavity 4c.

[0028] The cavity 4e includes a first cavity 4e1 with the largest diameter located at the top of the large cylindrical portion 4d, a second cavity 4e2 with a smaller diameter than the first cavity 4e1 located across the upper and middle portions of the large cylindrical portion 4d, and a third cavity 4e3 with a smaller diameter than the second cavity 4e2 located across the middle and lower portions of the large cylindrical portion 4d.

[0029] The first cavity 4e1, the second cavity 4e2, and the third cavity 4e3 are arranged concentrically. An ultrasonic transducer 8 is provided in the annular stepped portion 4f defined by the first cavity 4e1 and the second cavity 4e2. In other words, the ultrasonic transducer 8 is located inside the main body 4.

[0030] Multiple bolts 12 are arranged in a ring shape at approximately equal intervals around the outer circumference of the ultrasonic transducer 8. The shaft of each bolt 12 passes through the outer circumference of the ultrasonic transducer 8 and is screwed into the stepped portion 4f.

[0031] Furthermore, the ultrasonic transducer 8 is fixed to the large cylindrical portion 4d by the heads of each bolt 12 pressing against the outer circumference of the ultrasonic transducer 8 against the large cylindrical portion 4d. Alternatively, an annular spacer may be provided between the heads of each bolt 12 and the ultrasonic transducer 8, and the heads of each bolt 12 may press against the ultrasonic transducer 8 via this spacer.

[0032] The ultrasonic transducer 8 has a disc-shaped first electrode layer 8a, a piezoelectric material layer 8b (i.e., a layer composed of piezoelectric material), and a second electrode layer 8c. The first electrode layer 8a and the second electrode layer 8c are made of a metal having good conductivity.

[0033] The metal materials constituting the first electrode layer 8a and the second electrode layer 8c are, for example, copper (Cu) or silver (Ag), but are not limited to these. A nickel coating (i.e., nickel plating) may be provided on the surfaces of the first electrode layer 8a and the second electrode layer 8c.

[0034] Furthermore, the lower surface of the second electrode layer 8c that is in contact with the liquid 11 (i.e., the vibrating surface 8c1 described later) is covered with a metal such as stainless steel, which has relatively high corrosion resistance. Alternatively, a thin metal plate may be provided on the lower surface of the second electrode layer 8c, either in place of the coating layer or together with the coating layer, so as to cover the second electrode layer 8c.

[0035] The piezoelectric material layer 8b is formed of, for example, lead zirconate titanate (PZT). The upper surface of the piezoelectric material layer 8b is in contact with the first electrode layer 8a, and the lower surface of the piezoelectric material layer 8b is in contact with the second electrode layer 8c. The ultrasonic transducer 8 vibrates at the ultrasonic frequency due to the high-frequency power supplied from the ultrasonic oscillator 10 via the wiring 6a and 6b.

[0036] A cylindrical wall portion 4g is provided at the lower part of the second cavity portion 4e2. Outer from the wall portion 4g in the radial direction of the large cylindrical portion 4d, an inflow passage 4h is provided so as to reach the outer peripheral surface of the middle part of the large cylindrical portion 4d along the radial direction of the large cylindrical portion 4d.

[0037] The inflow passage 4h includes a cylindrical first region 4h1. One end of the first region 4h1 is located on the outer surface of the large cylindrical portion 4d. The opening of the first region 4h1 corresponds to an inlet 4h3 through which the liquid 11 enters the main body portion 4. The inflow passage 4h further includes a cylindrical second region 4h2, which has a smaller diameter than the first region 4h1 and is connected to the second cavity portion 4e2.

[0038] An inlet pipe section 14 is inserted into the first region 4h1, and the tip of the inlet pipe section 14 is fixed to the large cylindrical section 4d. A part of the nozzle member 4i is fixed to the wall section 4g. In this embodiment, the nozzle member 4i is part of the main body section 4.

[0039] The nozzle member 4i includes a cylindrical nozzle 4i1 having an outer diameter approximately the same as the inner diameter of the wall portion 4g. The upper end of the nozzle 4i1 forms a substantially flat annular surface located approximately at the same position as the upper end of the wall portion 4g. The lower end of the nozzle 4i1 protrudes below the lowest surface of the inverted frustoconical lower part of the large cylindrical portion 4d.

[0040] In the A-axis direction, an annular adjustment wall 4i2 is provided between the nozzle 4i1 and the ultrasonic transducer 8. In other words, the adjustment wall 4i2 is located in the second cavity 4e2 (i.e., inside the main body 4) and is positioned closer to the ultrasonic transducer 8 than the wall 4g.

[0041] The adjustment wall 4i2 is positioned opposite the vibration surface 8c1 of the ultrasonic transducer 8 at a predetermined distance (for example, the interval X shown in Figure 3(A)) in the A-axis direction, forming an annular planar region (i.e., a planar region) 4i 2A It holds.

[0042] In this embodiment, the outer diameter of the adjustment wall 4i2 is larger than the outer diameter of the wall portion 4g and smaller than the inner diameter of the second cavity portion 4e2. With this configuration, the liquid 11 flowing into the second cavity portion 4e2 via the inflow passage 4h is distributed almost uniformly from the entire circumference of the outer periphery of the adjustment wall 4i2, without being biased towards a part of the outer periphery of the adjustment wall 4i2 closest to the inlet 4h3, to the vibration surface 8c1 and the annular planar region 4i 2A It can lead you to the space between them.

[0043] Vibration surface 8c1 and annular plane region 4i 2A The substantially disc-shaped region defined by the above functions as a vibration-applying channel 4j to which ultrasonic vibrations are applied to the liquid 11 by the ultrasonic vibrations of the ultrasonic transducer 8. The annular planar region 4i is adjusted in distance from the vibration surface 8c1. 2AThe adjustment wall 4i2 has the function of amplifying the ultrasonic vibrations of the liquid 11 in the vibration-applying channel 4j.

[0044] By providing such an adjustment wall 4i2, the vibration-applying channel 4j does not need to be semi-cylindrical in shape, and the annular planar region 4i has an adjusted distance from the vibration surface 8c1. 2A Compared to the case where the adjustment wall 4i2 is not provided, ultrasonic vibrations from the ultrasonic transducer 8 are more easily transmitted to the liquid 11.

[0045] The lower surface of the adjustment wall 4i2 is separated by a predetermined distance from the upper surfaces of the nozzle 4i1 and the wall portion 4g, and the upper end of the nozzle 4i1 and the lower end of the adjustment wall 4i2 are connected in the A-axis direction by an annular connecting portion 4i3 having an outer diameter smaller than the outer diameter of the nozzle 4i1.

[0046] In this embodiment, the nozzle member 4i (i.e., the nozzle 4i1, the adjustment wall 4i2, and the connecting portion 4i3) is integrally formed from a predetermined material. However, it is not limited to integral formation, and the adjustment wall 4i2 may be formed separately from the nozzle 4i1 and the connecting portion 4i3.

[0047] The predetermined materials constituting the nozzle member 4i include, for example, SUS316L (i.e., low-carbon austenitic stainless steel), fused silica, and PVDF (i.e., polyvinylidene difluoride), but are not limited to these materials.

[0048] Furthermore, SUS316L is defined by Japanese Industrial Standards (JIS) and corresponds to European Standards (i.e., Euronorm) 1.4404, 1.4432, 1.4435, etc.

[0049] Furthermore, the nozzle 4i1, the adjustment wall 4i2, and the connecting portion 4i3 may be formed from different materials, or any two locations (for example, the nozzle 4i1 and the adjustment wall 4i2, or the adjustment wall 4i2 and the connecting portion 4i3) may be formed from the same material.

[0050] The nozzle 4i1, the adjustment wall 4i2, and the connection part 4i3 have the same inner diameter and are arranged concentrically. A cylindrical outflow passage 4i4 is formed in the central part of the nozzle member 4i so as to penetrate the nozzle 4i1, the adjustment wall 4i2, and the connection part 4i3.

[0051] The outflow passage 4i4 has an upper end (i.e., one end) 4i 4A and a lower end (i.e., the other end) 4i 4B and includes them. The upper end 4i 4A of the outflow passage 4i4 is connected to the vibration-imparting flow passage 4j through a flow passage formed in the connection part 4i3 and an opening provided in the adjustment wall 4 i2 . This opening provided in the adjustment wall 4i2 is located in the same plane as the annular planar region 4i 2A .

[0052] The lower end 4i 4B of the outflow passage 4i4 is located at the tip of the nozzle 4i1. The lower end 4i 4B of the outflow passage 4i4 corresponds to the outflow port from which the liquid 11 to which ultrasonic vibration is imparted in the vibration-imparting flow passage 4j flows out of the main body part 4.

[0053] That is, the liquid 11 enters the main body part 4 from the inlet 4h3 through the inflow pipe part 14, is subjected to ultrasonic vibration in the vibration-imparting flow passage 4j, and then passes through the outflow passage 4i4 and is jetted downward from the lower end 4i 4B of the outflow passage 4i4.

[0054] Next, referring to FIGS. 3(A), 3(B), and 4, the simulation results of the ultrasonic vibration imparted to the liquid 11 are shown. FIG. 3(A) is a cross-sectional view of the lower half of the nozzle unit 2 for clarifying the main dimensions.

[0055] The length in the A-axis direction from the vibration surface 8c1 defining the vibration-imparting flow passage 4j to the annular planar region 4i 2A is the interval X. The interval X is, for example, several millimeters. The inner diameter Y of the outflow passage 4i4 is, for example, about several millimeters. Also, from the vibration surface 8c1 to the lower end 4i 4BThe distance Z in the A-axis direction to this point is, for example, about several tens of millimeters.

[0056] Note that the interval X shown in Figure 3(A) is different from the X representing the X-axis. Similarly, the inner diameter Y is different from the Y representing the Y-axis, and the distance Z is different from the Z representing the Z-axis. The interval X, inner diameter Y, and distance Z are merely symbols for explanatory purposes.

[0057] The simulation was performed using software that numerically obtains approximate solutions to differential equations using the Finite Element Method (FEM). Specifically, Femtet, an analysis simulation software sold by Murata Software Co., Ltd., was used.

[0058] In the simulation, the ultrasonic vibration frequency f was set to 400 kHz, the wavelength λ of the ultrasonic vibration occurring in the liquid 11 of the vibration-applying channel 4j was set to 3.8 mm, the maximum value of the vibration velocity v of the vibrating surface 8c1 was set to 0.1 m / s and the minimum value to 0.0 m / s, the inner diameter Y was set to 6.0 mm, and the distance Z was set to 36.4 mm. The speed of sound in water was also set to 1500 m / s.

[0059] Then, the acoustic output P1(W) (see Figure 4) was calculated for each (spacing X / wavelength λ) while keeping the thickness of the adjustment wall 4i2 (i.e., the length in the A-axis direction) constant. In addition, the acoustic output P1(W) was calculated for each (spacing X / wavelength λ) while changing the spacing X (i.e., (spacing X / wavelength λ)) by bringing the ultrasonic transducer 8 closer to the adjustment wall 4i2.

[0060] Furthermore, the spatial distribution of sound pressure P2 (kPa) when the interval X is fixed (see Figures 3(B), 5(A), and 5(B)) was also calculated. Figure 3(B) shows the distribution of sound pressure P2 obtained from the simulation superimposed on the lower half of the nozzle unit 2 (particularly the vibration-applying channel 4j and the outflow channel 4i4).

[0061] In Figure 3(B), the contour lines of sound pressure are shown in grayscale. In Figure 3(B), the lighter-colored regions 11a, which are shown relatively brightly, represent regions with relatively high sound pressure. In the vibration-impacting channel 4j and the outflow channel 4i4, the lighter-colored regions 11a occur periodically in the A-axis direction.

[0062] In the outflow path 4i4, there is a region with relatively low sound pressure between two adjacent light-colored regions 11a, and in Figure 3(B), this region with relatively low sound pressure is shown as relatively dark. Incidentally, the dashed line drawn along the A-axis at the radial center of the outflow path 4i4 is a reference line and is unrelated to sound pressure, so it can be ignored.

[0063] As shown in Figure 3(B), in the annular region of the vibration-impregnating channel 4j, excluding the area directly above the outflow channel 4i4, the positions of the antinodes and nodes in the A-axis direction are substantially fixed. In the cylindrical region of the vibration-impregnating channel 4j located directly above the outflow channel 4i4, the sound field changes over time such that a traveling wave is formed in which the region with relatively high sound pressure propagates downward.

[0064] However, some sound waves propagating through the outflow path 4i4 enter the interior of the nozzle 4i1's wall thickness in the radial direction from the inner circumferential wall surface of the nozzle 4i1, and then return to the outflow path 4i4. These sound waves returning to the outflow path 4i4, along with the traveling waves propagating downwards through the outflow path 4i4, form a pseudo-standing wave in the outflow path 4i4. Therefore, the positions of the antinodes and nodes are substantially fixed in the outflow path 4i4 as well.

[0065] Figure 3(B) shows the results of a two-dimensional simulation. However, since the vibration-impregnating channel 4j is disc-shaped and the outflow channel 4i4 is cylindrical, it can be reasonably inferred that the three-dimensional sound pressure distribution is approximately the same as the result shown in Figure 3(B).

[0066] Figure 4 shows the simulation results illustrating the dependence of acoustic output P1 on (spacing X / wavelength λ). The horizontal axis represents the ratio of spacing X to wavelength λ (a dimensionless quantity), and the vertical axis represents the acoustic output P1(W). As shown in Figure 4, when (spacing X / wavelength λ) is near 0.50, near 1.00, near 1.50, and near 2.00, the acoustic output P1 becomes locally high.

[0067] In particular, when N is a natural number, the acoustic output P1 tends to be relatively high within the range where the interval X is expressed by (1) below. Among these, the acoustic output P1 of nozzle unit 2 is highest when N is 1. Note that the coefficient (N±0.1) in (1) is determined taking error into consideration.

[0068]

number

[0069] By the way, if the spacing X is too narrow and the flow rate of liquid 11 is insufficient when N=1, then N may be set to a natural number greater than or equal to 2 (i.e., N=2, 3, ...). This has the advantage of increasing the flow rate of liquid 11, although it slightly sacrifices the acoustic output P1 of the nozzle unit 2.

[0070] Figures 5(A) and 5(B) show the simulation results of calculating the sound pressure P2 (kPa) at a distance Z while keeping the (spacing X / wavelength λ) fixed. The distance Z is shown with the vibration surface 8c1 as the origin in the A-axis direction.

[0071] Figure 5(A) shows the simulation results when the adjustment wall 4i2 is made of stainless steel SUS316L, and Figure 5(B) shows the simulation results when the adjustment wall 4i2 is made of fused silica.

[0072] The acoustic impedance of SUS316L is approximately 45.2 × 10⁻¹⁰. 6 (Ns / m 3 ) and the acoustic impedance of fused silica is approximately 12.6 × 10⁻¹⁰. 6 (Ns / m 3) and the acoustic impedance of pure water, a typical liquid 11, is approximately 1.46 × 10⁻¹⁰. 6 (Ns / m 3 )

[0073] Generally, the greater the difference in acoustic impedance between media, the higher the reflectivity of ultrasound at the interface between the media, and the smaller the difference in acoustic impedance between the media, the lower the reflectivity of ultrasound at the interface (i.e., it is easier to transmit).

[0074] However, as can be seen from the comparison of Figure 5(A) and Figure 5(B), using an adjustment wall 4i2 with an acoustic impedance closer to the acoustic impedance of the liquid 11 allows for a higher maximum sound pressure P2 (i.e., the maximum sound pressure P2 in Figure 5(A) < the maximum sound pressure P2 in Figure 5(B)). In other words, it is preferable that the adjustment wall 4i2 and the liquid 11 have equivalent acoustic impedances.

[0075] In this specification, it is said that the adjustment wall 4i2 and the liquid 11 have equivalent acoustic impedances, which means that the acoustic impedance of the adjustment wall 4i2 is 10% or more and 1000% or less of the acoustic impedance of the liquid 11, and more preferably 50% or more and 200% or less.

[0076] When a material with a higher acoustic impedance than the liquid 11 (for example, SUS316L) is used for the adjustment wall 4i2, it is thought that the energy entering the adjustment wall 4i2 is more likely to accumulate in the adjustment wall 4i2 and less likely to return to the liquid 11.

[0077] In contrast, when a material with low acoustic impedance (for example, fused silica) is used for the adjustment wall 4i2, although the energy entering the adjustment wall 4i2 is stored in the adjustment wall 4i2, the energy entering the adjustment wall 4i2 is more likely to return to the liquid 11 compared to an adjustment wall 4i2 made of a material with high acoustic impedance.

[0078] Thus, the inventors of this application hypothesize that differences in materials manifest as differences in the energy stored in the adjustment wall 4i2, ultimately resulting in differences in sound pressure P2. However, it should be noted that the above explanation is merely a hypothesis, and in reality, differences in sound pressure P2 may be caused by other factors.

[0079] The acoustic impedance of PVDF is 2.9 × 10⁻⁶. 6 (Ns / m 3 Therefore, among SUS316L, fused silica, and PVDF, PVDF is most preferably used as the material for the adjustment wall 4i2, followed by fused silica. However, the material for the adjustment wall 4i2 is not limited to the materials described above.

[0080] In this embodiment, the ultrasonic transducer 8 and the adjustment wall 4i2 are arranged such that the above-mentioned interval X satisfies (1) above. Therefore, compared to the case where the ultrasonic transducer 8 and the adjustment wall 4i2 are not arranged such that the interval X satisfies (1) above, ultrasonic vibrations are more easily transmitted to the liquid 11.

[0081] Furthermore, in this embodiment, the vibrating surface 8c1 and the annular planar region 4i 2A However, since each is approximately flat and arranged approximately parallel to each other, the annular planar region 4i 2A Another advantage is that ultrasonic vibrations are more easily propagated throughout the entire structure.

[0082] Note that the vibration surface 8c1 and the annular planar region 4i 2A These do not necessarily have to be perfectly flat, nor do they have to be perfectly parallel to one another. Therefore, the vibrating surface 8c1 and the annular planar region 4i 2A In this context, the presence of irregularities, undulations, and non-parallel regions is tolerated to a certain extent.

[0083] Of course, the vibration surface 8c1 and the annular plane region 4i 2A It should be noted that even if condition (1) above is satisfied in some cases, the effect of making ultrasonic vibrations easier to propagate into the liquid 11 can still be obtained, albeit to a limited extent.

[0084] Furthermore, the opening of the adjustment wall 4i2 is not limited to the radial center of the adjustment wall 4i2. The position of the opening of the adjustment wall 4i2 can be changed as appropriate, as long as ultrasonic vibrations can be appropriately applied to the liquid 11 during the process of the liquid 11 flowing into and out of the main body 4.

[0085] (Second Embodiment) Next, a second embodiment will be described with reference to Figures 6 to 9. Figure 6 is a flowchart of a method for cleaning a workpiece 21, in which the workpiece (i.e., the object to be cleaned) 21 is cut and then cleaned using a cutting device 22 equipped with a cutting unit 26 on which the nozzle unit 2 described above is mounted.

[0086] First, the workpiece 21 will be described with reference to Figure 7. Figure 7 is a perspective view of the workpiece 21, etc. The workpiece 21 has, for example, a disc-shaped wafer made of silicon (Si), but the wafer may be made of a semiconductor material other than silicon, such as gallium nitride (GaN) or silicon carbide (SiC).

[0087] Multiple division lines 21a1 are set in a grid pattern on the surface 21a of the workpiece 21, and a device such as an IC is formed in each of the multiple rectangular regions 21a2 partitioned by the multiple division lines 21a1. A ring frame 25 made of metal is positioned on the radially outer side of the workpiece 21.

[0088] The ring frame 25 has an opening with a larger diameter than the workpiece 21. With the workpiece 21 positioned in this opening, the dicing tape 23 is attached to the back surface 21b of the workpiece 21 and one surface of the ring frame 25. This forms a workpiece unit 27 in which the workpiece 21 is supported by the ring frame 25 via the dicing tape 23.

[0089] Next, the cutting apparatus 22 will be described with reference to Figures 8(A) and 8(B). The X, Y, and Z axes shown in Figures 8(A) and 8(B) are orthogonal to each other. The X axis is parallel to the machining feed direction, and the Y axis is parallel to the indexing feed direction. The Z axis is parallel to the depth of cut feed direction and the vertical direction.

[0090] The cutting device 22 has a disc-shaped chuck table 24 (see Figure 8(B)). The chuck table 24 has a disc-shaped frame made of non-porous metal. A disc-shaped recess with a diameter smaller than the outer diameter of the frame is formed on the upper surface of the frame.

[0091] A disc-shaped porous plate (not shown) made of porous ceramics is fixed to this recess using an adhesive or the like. The upper surfaces of the frame and the porous plate are substantially flush, forming a holding surface 24a that is substantially parallel to the XY plane.

[0092] A suction source (not shown), such as a vacuum pump, is connected to the frame via a flow path (not shown) including a rotary joint. The negative pressure generated by the suction source is transmitted to the upper surface of the porous plate through the frame. This negative pressure causes the workpiece 21 to be held in place by the holding surface 24a.

[0093] When the workpiece 21 is held by suction on the holding surface 24a, the workpiece 21 is held by suction on the holding surface 24a via the dicing tape 23. The ring frame 25 is clamped by a clamping unit provided on the outer circumference of the chuck table 24.

[0094] A cutting unit 26 is provided above the chuck table 24. The cutting unit 26 has a spindle housing 28 whose longitudinal portion is arranged along the Y-axis. A portion of a cylindrical spindle (not shown) is rotatably housed in the spindle housing 28.

[0095] The spindle is equipped with a motor (not shown), and the power supplied to the motor allows the spindle to rotate at high speed. The tip of the spindle protrudes outside the spindle housing 28.

[0096] A cutting blade 30 is mounted at the tip of the spindle. The cutting blade 30 is, for example, a hub blade having an annular base and an annular cutting edge fixed to one surface of the base, but is not limited to this. The cutting blade 30 may also be a hubless type (i.e., a washer type) that does not have an annular base and is composed only of an annular cutting edge.

[0097] A blade cover 32 is mounted on the tip of the spindle housing 28 so as to cover the cutting blade 30. The blade cover 32 has a pair of cooler nozzles 34, a shower nozzle 36, and the nozzle unit 2 described above.

[0098] The pair of cooler nozzles 34 have a roughly L-shape. The pair of cooler nozzles 34 supply cutting fluid 42, such as pure water, which is not subjected to ultrasonic vibration, to the contact area between the cutting edge of the cutting blade 30 and the workpiece 21.

[0099] The shower nozzle 36 supplies cutting water 42, such as pure water without ultrasonic vibration, to the outer peripheral edge of the cutting edge of the cutting blade 30 from the radially outside of the cutting blade 30, similar to the pair of cooler nozzles 34.

[0100] Furthermore, the nozzle unit 2 supplies pure water, to which ultrasonic vibrations are applied, as cutting fluid 42 to the upper surface (e.g., surface 21a) of the workpiece 21 held by suction on the holding surface 24a. Note that the cutting fluid 42 is not shown in Figure 8(A), and the cutting fluid 42 supplied from the pair of cooler nozzles 34 is not shown in Figure 8(B).

[0101] Figure 8(A) is a perspective view of the cutting process S10, and Figure 8(B) is a partial cross-sectional side view of the cutting process S10. In the cutting process S10, the workpiece 21 is held by suction on the holding surface 24a, and then alignment is performed. This makes the planned division line 21a1 approximately parallel to the X-axis.

[0102] Then, the spindle and cutting blade 30 are rotated at high speed, and the lower end of the cutting blade 30 is positioned at a height between the back surface 21b and the holding surface 24a. In this state, cutting water 42 is supplied at a predetermined flow rate from the pair of cooler nozzles 34, the shower nozzle 36, and the nozzle unit 2, respectively, while the chuck table 24 is machined along the X axis.

[0103] This cuts one planned division line 21a1, forming a so-called full-cut groove. After cutting one planned division line 21a1, the machine is indexed and fed by a predetermined amount along the Y-axis.

[0104] Then, other division lines 21a1 adjacent to the division line 21a1 after cutting are cut in the same manner. After cutting all division lines 21a1 along the X axis in this manner, the chuck table 24 is rotated 90 degrees.

[0105] Next, all planned division lines 21a1 along the X-axis are cut in the same manner. This divides the workpiece 21 into device units. An example of cutting conditions is as follows:

[0106] Spindle rotation speed: 30,000 rpm X-axis movement speed: 50 mm / s Cutting fluid flow rate: 4.0 L / min

[0107] After the cutting process S10, a cleaning process S20 is performed. Figure 9 is a partial cross-sectional side view showing the cleaning process S20. In the cleaning process S20, cutting fluid 42, which is pure water to which ultrasonic vibrations have been applied, is supplied from the nozzle unit 2 as a cleaning solution 42a to clean the entire surface 21a of the workpiece 21.

[0108] In the cleaning process S20, first, the cutting unit 26 is raised to the extent that the lower end of the cutting blade 30 does not come into contact with the surface 21a. Then, the cleaning fluid 42a is sprayed from the nozzle unit 2, and the chuck table 24 is moved along the X axis.

[0109] After cleaning the workpiece 21 from one end to the other in the X-axis direction with the cleaning solution 42a, the cutting unit 26 is indexed and fed by a predetermined amount, and then, similarly, the workpiece 21 from one end to the other in the X-axis direction is cleaned with the cleaning solution 42a.

[0110] The nozzle unit 2 sprays cleaning fluid 42a into a predetermined circular area in the XY plane, so in the cleaning step S20, a strip-shaped area with a predetermined width in the Y-axis direction is cleaned with a single processing feed. Ultimately, the entire surface 21a is cleaned by multiple strip-shaped areas.

[0111] In the cleaning process S20, when the workpiece 21 and the nozzle unit 2 are moved relative to each other along the X-axis by feeding the chuck table 24 for machining, the cleaning fluid 42a described above is used.

[0112] In other words, the cleaning process S20 consists of an ultrasonic vibration application process S22 which applies ultrasonic vibration to a liquid such as pure water supplied to the vibration application channel 4j, and the lower end 4i of the outflow channel 4i4 in the nozzle 4i1. 4B The process includes a supply step S24 in which a cleaning solution 42a (i.e., a liquid to which ultrasonic vibrations are applied) is supplied to the workpiece 21 from (i.e., the outlet) to clean the workpiece 21. An example of cleaning conditions is as follows:

[0113] Washing solution flow rate: 1 L / min or more and 8 L / min or less (for example, 3 L / min) Ultrasonic frequency: 380kHz to 3MHz (for example, 400kHz) Power required to drive the ultrasonic transducer: 20W to 100W (e.g., 50W) Relative movement speed: 1 mm / s to 100 mm / s (for example, 30 mm / s)

[0114] In Figure 8(B), the liquid supplied from the nozzle unit 2 is cutting fluid 42, and in Figure 9, it is cleaning fluid 42a. Both are, for example, pure water to which ultrasonic vibrations have been applied. In this embodiment, the cutting process S10 and the cleaning process S20 are performed without stopping the injection of cutting fluid 42 or cleaning fluid 42a from the nozzle unit 2.

[0115] In this embodiment, the entire surface 21a of the workpiece 21 after cutting can be cleaned with the cleaning solution 42a. In particular, since ultrasonic vibrations are applied to the cleaning solution 42a using the nozzle unit 2, the cleaning effect can be improved compared to when cleaning with a liquid that is not subjected to ultrasonic vibrations by the nozzle unit 2.

[0116] In addition, in the cleaning step S20, the surfaces 21a other than the dividing line 21a1 may be cleaned with the cleaning solution 42a, in addition to the full-cut groove formed on the dividing line 21a1. Furthermore, in the cleaning step S20, the full-cut groove may be cleaned more carefully than the areas other than the full-cut groove by moving the nozzle unit 2 relatively multiple times along the full-cut groove.

[0117] In areas other than the full-cut grooves, the surface 21a may be cleaned with cleaning solution 42a by setting N to 2 or more in order to perform cleaning at a relatively high flow rate, or the full-cut grooves may be cleaned with cleaning solution 42a by setting N to 1 in order to relatively increase the cleaning power.

[0118] (Third Embodiment) Next, a third embodiment will be described with reference to Figures 10 to 13(B). Figure 10 is a flowchart of a method for manufacturing a substrate 33, in which a substrate 33 having a thickness less than the thickness of the workpiece 31 (see Figure 13(B)) is manufactured from a workpiece 31 using the nozzle unit 2 described above.

[0119] First, the laser processing apparatus 50 will be described with reference to Figure 11(A). In Figure 11(A), some of the components of the laser processing apparatus 50 are shown as functional blocks. The laser processing apparatus 50 has a disc-shaped chuck table 52 that holds the workpiece 31 by suction. The chuck table 52 is substantially the same as the chuck table 24 described above, so a redundant explanation will be omitted.

[0120] The workpiece 31 is an ingot of gallium nitride (GaN), silicon carbide (SiC), etc. However, the workpiece 31 is not limited to an ingot. The workpiece 31 may be a polycrystalline substrate thicker than the thickness of the substrate 33.

[0121] A laser beam irradiation unit 54 is provided above the chuck table 52. The laser beam irradiation unit 54 has a laser oscillator 56. The laser oscillator 56 has, for example, a crystal such as Nd:YAG as the laser medium.

[0122] By irradiating the crystal with excitation light from a light source such as a flash lamp or laser diode, the laser oscillator 56 emits a pulsed laser beam L having a wavelength (for example, 1064 nm) that penetrates the workpiece 31.

[0123] The laser beam L emitted from the laser oscillator 56 is powered by an output adjustment unit 58, which includes an attenuator, a spatial light phase modulator, etc., before proceeding to the irradiation head 60. Note that the arrangement of the laser oscillator 56 and the output adjustment unit 58 is shown in general terms, and it is not necessary for them to be arranged in a straight line along the X-axis.

[0124] The irradiation head 60 is equipped with a mirror 62, a focusing lens 64, and the like. The laser beam L, which is reflected by the mirror 62 and has its direction of travel changed, passes through the focusing lens 64 and is irradiated toward the holding surface 52a.

[0125] With the laser beam L focused at a predetermined depth position in the workpiece 31, the irradiation head 60 and the chuck table 52 are moved relative to each other along the X-axis, thereby forming a modified region 31c (i.e., a separation point) inside the workpiece 31, where the mechanical strength is reduced compared to the non-irradiated area of ​​the laser beam L (separation point formation step S30).

[0126] In the separation starting point formation step S30 of this embodiment, a plurality of linearly shaped modified regions 31c are formed at predetermined depth positions inside the workpiece 31. Figure 11(A) is a partial cross-sectional side view showing the separation starting point formation step S30.

[0127] In the separation point formation step S30, first, the workpiece 31 is held by the chuck table 52 by suction so that one surface 31a of the workpiece 31 is exposed upwards and the other surface 31b of the workpiece 31 faces the holding surface 52a.

[0128] Next, the focusing point P of the laser beam L is positioned at a predetermined location between one surface 31a and the other surface 31b in the thickness direction 31d of the workpiece 31, and the workpiece 31 and the focusing point P are moved relative to each other in the X-axis direction from one end to the other of the surface 31a. Then, the irradiation head 60 and the chuck table 52 are indexed and fed by a predetermined amount along the Y-axis.

[0129] Subsequently, the workpiece 31 and the focusing point P are similarly moved relative to each other in the X-axis direction from one end to the other of a surface 31a. In this way, a plurality of modified regions 31c are formed at predetermined positions in the thickness direction 31d.

[0130] Figure 11(B) is a top view of the workpiece 31 showing the separation point formation process. In this embodiment, as shown in Figure 11(B), the modified region 31c is formed along the movement trajectory of the focusing point P indicated by the dashed arrow. An example of the processing conditions in the separation point formation process S30 is shown below.

[0131] Wavelength: 1064nm Average output: 1.0W to 5.0W (for example, 3.0W) Repetition frequency: 100kHz to 50000kHz (for example, 300kHz) Machining feed rate: 20 mm / s or more and 5000 mm / s or less (for example, 1000 mm / s)

[0132] After the separation point formation step S30, the liquid 35 to which ultrasonic vibrations are applied is supplied to one surface 31a of the workpiece 21 using the nozzle unit 2 described above, thereby causing cracks to extend from each modified region 31c (crack extension step S40).

[0133] Figure 12 is a partial cross-sectional side view showing the crack propagation process S40. In the crack propagation process S40, a box-shaped drain receiver 70 made of metal such as stainless steel is used. A support table 72 for supporting the workpiece 31 is provided on the bottom plate of the drain receiver 70.

[0134] The mounting table 72 has positioning members such as protrusions (not shown) to prevent the workpiece 31 from moving on the mounting table 72. Alternatively, the mounting table 72 may hold the workpiece 31 by suction using negative pressure suction or electrostatic attraction instead of, or in conjunction with, the positioning members.

[0135] In this embodiment, since multiple modified regions 31c are formed in the thickness direction 31d at a position closer to one surface 31a than to the other surface 31b, the workpiece 31 is placed on the mounting table 72 in such a manner that one surface 31a is exposed upwards.

[0136] In the crack propagation process S40, the liquid 35 such as pure water, to which ultrasonic vibrations are applied in the vibration-applying channel 4j described above, is sent to the lower end 4i of the outflow channel 4i4. 4B (That is, it is supplied to the workpiece 31 from the outlet.)

[0137] More specifically, the nozzle unit 2 is moved linearly relative to the workpiece 31 such that the landing point of the liquid 35 supplied from the nozzle unit 2 traces a linear region of one surface 31a corresponding to each modification region 31c in the thickness direction 31d.

[0138] This applies ultrasonic vibration to each modified region 31c, causing cracks to extend from each modified region 31c. Note that crack extension from the modified region 31c includes the formation of new cracks from the modified region 31c and the further extension of existing cracks. An example of processing conditions in the crack extension process S40 is shown below.

[0139] Washing solution flow rate: 1 L / min or more and 8 L / min or less (for example, 3 L / min) Ultrasonic frequency: 380kHz to 3MHz (for example, 400kHz) Power required to drive the ultrasonic transducer: 20W to 100W (e.g., 50W) Relative movement speed: 1 mm / s to 100 mm / s (for example, 30 mm / s)

[0140] In this embodiment, crack propagation can be promoted by applying ultrasonic vibrations to the workpiece 31 via the liquid 35. In particular, since ultrasonic vibrations are applied to the liquid 35 using the nozzle unit 2, crack propagation can be performed more efficiently compared to using a liquid that is not subjected to ultrasonic vibrations by the nozzle unit 2.

[0141] After the crack propagation process S40, the workpiece 31 is separated into the substrate 33 and the remaining ingot 37 using a separation device 76 (see Figure 13(A)) starting from multiple modified regions 31c (see Figure 13(B)).

[0142] Figure 13(A) is a partial cross-sectional side view showing the separation device 76. The separation device 76 has a chuck table 78 which has approximately the same diameter as the chuck table 52 described above. The upper surface of the chuck table 78 functions as a holding surface 78a that sucks and holds the workpiece 31. A separation unit 80 is provided above the chuck table 52.

[0143] The separation unit 80 has a cylindrical movable part 82 whose longitudinal portion is arranged along the Z-axis direction. A Z-axis direction movement mechanism (not shown) is connected to the movable part 82, and the movable part 82 is movable along the Z-axis direction. The Z-axis direction movement mechanism is, for example, a ball screw type movement mechanism, but may be composed of other actuators.

[0144] A disc-shaped suction head 84 is provided at the bottom of the movable part 82. The suction head 84, like the chuck table 52, has a frame and a porous plate. The lower surfaces of the frame and the porous plate are substantially flush and perpendicular to the Z-axis, and function as a holding surface 84a.

[0145] Figure 13(B) is a partial cross-sectional side view showing the separation process S50. In the separation process S50, the other surface 31b is held by the holding surface 78a through suction, while one surface 31a is held by the holding surface 84a of the suction head 84 through suction.

[0146] Next, the Z-axis movement mechanism is activated to raise the suction head 84. This separates the substrate 33 from the workpiece 31, starting from multiple modified regions 31c. Similarly, the remaining ingot (i.e., workpiece) 37 can also be subjected to the separation starting point formation process S30, the crack propagation process S40, and the separation process S50 to produce more substrates 33 from the remaining ingot 37.

[0147] By the way, in the third embodiment, separation points were formed in the workpiece 31 by forming a plurality of modified regions 31c in the workpiece 31 in the separation point formation step S30, but instead, regions with reduced mechanical strength (i.e., separation points) may be formed within the workpiece 31 by ion implantation.

[0148] Furthermore, during crystal growth of the workpiece 31, which is an ingot or substrate, crystal defects may be introduced at a predetermined depth in the workpiece 31 to form regions within the workpiece 31 where the mechanical strength is reduced (i.e., separation points).

[0149] (Fourth Embodiment) Next, the fourth embodiment will be described with reference to Figures 14 to 16(B). Figure 14 is a flowchart of a method for manufacturing a chip 29, in which a workpiece 21 (see Figure 7) is divided using the nozzle unit 2 described above to produce multiple chips 29 (see Figure 16(B)).

[0150] Figure 15(A) is a partial cross-sectional side view showing the division starting point formation process S60. In the division starting point formation process S60, the laser processing apparatus 50 described above is used, but a nonlinear optical crystal is mounted on the laser oscillator 56 to convert the laser beam L into a harmonic L1 (for example, the fourth harmonic with a wavelength of 266 nm), and this harmonic L1 is used for laser processing.

[0151] During laser processing, the irradiation head 60 of the laser processing device 50 focuses harmonics L1 onto the surface 21a, and the irradiation head 60 and the workpiece 21 are moved relative to each other along the X-axis. This performs ablation on the workpiece 21, forming half-cut grooves (i.e., division starting points) 21a3 in each of the multiple planned division lines 21a1.

[0152] Figure 15(B) is a perspective view of the workpiece unit 27 after the division starting point formation process S60. An example of the processing conditions in the division starting point formation process S60 is shown below.

[0153] Wavelength: 266nm Average output: 1.0W to 5.0W (for example, 3.0W) Repetition frequency: 100kHz to 50000kHz (for example, 300kHz) Machining feed rate: 20 mm / s or more and 5000 mm / s or less (for example, 1000 mm / s)

[0154] Furthermore, the term "half-cut groove 21a3" does not simply mean a groove with a depth exactly half the thickness of the workpiece 21, but rather a groove that does not completely divide the workpiece 21 from the surface 21a to the back surface 21b. For example, when the thickness of the workpiece 21 is 775 μm, the half-cut groove 21a3 has a remaining area with a thickness of 50 μm.

[0155] By the way, in the division starting point formation step S60, instead of the laser processing device 50, the above-described cutting device 22 may be used to form the half-cut groove 21a3 in the workpiece 21. When the cutting device 22 is used in the division starting point formation step S60, cutting fluid 42 without ultrasonic vibration may be supplied from the nozzle unit 2, or cutting fluid 42 with ultrasonic vibration may be supplied from the nozzle unit 2.

[0156] After the splitting starting point formation step S60, the workpiece 21 is split into multiple chips 29 starting from multiple half-cut grooves 21a3 using the nozzle unit 2, the drain receiver 70 and the mounting table 72, similar to the crack propagation step S40 in the third embodiment (splitting step S70).

[0157] Figure 16(A) is a partial cross-sectional side view showing the splitting process S70. In the splitting process S70, ultrasonic vibrations are applied to the liquid 35 such as pure water supplied to the vibration-applying channel 4j, and the ultrasonically vibrated liquid 35 is sent to the lower end 4i of the outflow channel 4i4. 4B (That is, from the outlet) the material is sequentially supplied to each of the half-cut grooves 21a3 of the workpiece 31.

[0158] More specifically, ultrasonic vibrations are applied to each half-cut groove 21a3 by moving the nozzle unit 2 linearly relative to the workpiece 21 so that the landing point of the liquid 35 supplied from the nozzle unit 2 traces each half-cut groove 21a3.

[0159] As a result, cracks are formed from the bottom of each half-cut groove 21a3 to the back surface 21b, and the workpiece 21 is divided into individual chips 29. Figure 16(B) is an enlarged perspective view of the chip 29. An example of the processing conditions in the division process S70 is shown below.

[0160] Washing solution flow rate: 1 L / min or more and 8 L / min or less (for example, 3 L / min) Ultrasonic frequency: 380kHz to 3MHz (for example, 400kHz) Power required to drive the ultrasonic transducer: 20W to 100W (e.g., 50W) Relative movement speed: 1 mm / s to 100 mm / s (for example, 30 mm / s)

[0161] In this embodiment, the remaining portion at the bottom of the half-cut groove 21a3 can be cut by applying ultrasonic vibrations to the workpiece 21 via the liquid 35. In particular, ultrasonic vibrations can be efficiently applied to the liquid 35 using the nozzle unit 2.

[0162] Furthermore, after the splitting process S70, the dicing tape 23 may be expanded radially to widen the spacing between the chips 29, making it easier to pick up individual chips 29 with a picking device such as a die picker or chip picker.

[0163] Furthermore, after the splitting process S70, or after expanding the dicing tape 23, the nozzle unit 2 may be used to clean all the chips 29 (i.e., the entire workpiece 21) to remove any chips generated due to the cutting.

[0164] Furthermore, the structures, methods, etc., according to the embodiments described above can be modified as appropriate without departing from the scope of the object of the present invention. [Explanation of symbols]

[0165] 2: Nozzle unit (ultrasonic vibration device) 4: Main body 4a: Upper lid, 4b: Small cylindrical part, 4c: Cavity part, 4d: Large cylindrical part 4e: Cavity part, 4e1: First cavity part, 4e2: Second cavity part, 4e3: Third cavity part 4f: Stepped section, 4g: Wall section 4h: Inflow path, 4h1: 1st area, 4h2: 2nd area, 4h3: Inlet 4i: Nozzle component 4i1: Nozzle 4i2: Adjustment wall, 4i 2A :Annular plane area (plane area) 4i3: Connection part 4i4:Outflow channel, 4i4A :Top end (one end), 4i 4B : Lower end (other end) 4j: Vibration-applying channel 6a, 6b: Wiring 8: Ultrasonic transducer 8a: First electrode layer, 8b: Piezoelectric material layer, 8c: Second electrode layer, 8c1: Vibration surface 10: Ultrasonic oscillator 11: Liquid, 11a: Bright color area 12: Bolt, 14: Inlet pipe section 21: Workpiece (object to be cleaned), 21a: Front surface, 21b: Back surface 21a1: Planned division line, 21a2: Rectangular region 21a3: Half-cut groove (starting point of division) 22:Cutting device 23: Dicing tape, 25: Ring frame, 27: Workpiece unit 24: Chuck table, 24a: Holding surface 26: Cutting unit, 28: Spindle housing, 30: Cutting blade 29: Tip 31: Workpiece, 31a: One side, 31b: Other side, 31c: Modified area, 31d: Thickness direction 32: Blade cover 33: Substrate, 35: Liquid, 37: Ingot 34: Cooler nozzle, 36: Shower nozzle 42: Cutting fluid, 42a: Cleaning fluid 50: Laser processing equipment 52: Chuck table, 52a: Holding surface 54: Laser beam irradiation unit, 56: Laser oscillator, 58: Output adjustment unit 60: Irradiation head, 62: Mirror, 64: Focusing lens 70: Drain receiver, 72: Mounting platform 76: Separation device, 78: Chuck table, 78a: Holding surface 80: Separation unit, 82: Movable part, 84: Suction head, 84a: Holding surface L: Laser beam, L1: Harmonic, P: Focus point P1: Acoustic output, P2: Sound pressure S10: Cutting process, S20: Cleaning process, S22: Ultrasonic vibration application process, S24: Supply process S30: Separation point formation process, S40: Crack extension process S50: Separation process, S60: Division starting point formation process, S70: Division process X: Spacing, Y: Inner diameter, Z: Distance

Claims

1. An ultrasonic vibration device that applies ultrasonic vibrations to a liquid, A main body having an inlet into which the liquid enters, An ultrasonic transducer is provided inside the main body and includes a vibrating surface that comes into contact with the liquid, An adjustment wall is provided inside the main body and has a planar region that is positioned opposite to the vibrating surface at a predetermined distance, and the vibration-applying flow path is defined by the planar region and the vibrating surface, Equipped with, The main body has an outflow passage, The outflow channel includes one end connected to the vibration-applying channel and the other end corresponding to the outlet through which the liquid, which has been subjected to ultrasonic vibration in the vibration-applying channel by the ultrasonic vibration of the ultrasonic transducer, flows out of the main body. An ultrasonic vibration device characterized in that, when the wavelength of the liquid in the vibration-applying channel is λ and N is a natural number, the distance X between the vibration surface and the planar region is expressed by (1) below. [Math 1]

2. The ultrasonic vibration device according to claim 1, characterized in that N is 1.

3. The ultrasonic vibration device according to claim 1, characterized in that the adjustment wall and the liquid have equivalent acoustic impedances.

4. A cleaning method for cleaning an object to be cleaned using the ultrasonic vibration device described in claim 1, An ultrasonic vibration application step of applying ultrasonic vibration to the liquid supplied to the vibration application channel of the main body, A supply step in which the liquid to be cleaned is supplied to the object to be cleaned from the outlet of the main body to which ultrasonic vibrations are applied, A cleaning method characterized by comprising the following:

5. A method for manufacturing a substrate having a thickness less than the thickness of a workpiece, using an ultrasonic vibration device as described in claim 1, A separation point formation step in which separation points are formed inside the workpiece, After the separation point formation step, a crack extension step is performed in which ultrasonic vibrations are applied to the liquid supplied to the vibration-applying channel of the main body, and the ultrasonically vibrated liquid is supplied to the workpiece from the outlet of the main body, thereby extending the crack from the separation point. A method for manufacturing a substrate, characterized by comprising the following:

6. A method for manufacturing chips, which involves dividing a workpiece to produce a plurality of chips using the ultrasonic vibration device described in claim 1, A division starting point formation step in which a division starting point is formed in each of a plurality of division planned lines set on one surface of the workpiece, After the division starting point formation step, ultrasonic vibration is applied to the liquid supplied to the vibration-applying channel of the main body, and the ultrasonically vibrated liquid is supplied from the outlet of the main body to each of the multiple division starting points in the workpiece, thereby dividing the workpiece into the multiple chips starting from the multiple division starting points. A method for manufacturing chips, characterized by comprising the following features.

Images