JP6560969B2 - Wafer division method - Google Patents

Description

  The present invention relates to a dividing method for dividing a wafer into individual devices.

  In the manufacture of devices, a plurality of chip areas are defined by a plurality of division lines (streets) arranged in a lattice pattern on the surface of the wafer, and devices such as ICs and LSIs are formed in these chip areas. A metal such as a low dielectric constant (Low-k) film or a Test Element Group (TEG), which is a functional film forming a device, is formed on the surface of the wafer. When such a wafer is divided along the street with a cutting blade, for example, if the low dielectric constant film is removed, the film may be peeled off. Etching cannot be performed. Therefore, in these cases, the division process is performed after the functional film and the metal member on the street are removed.

  Laser processing is performed as a method for removing functional layers and metal members on the street, and laser irradiation is performed in a state where a protective film containing a resin or the like is formed on the surface of the wafer. Thereby, the device can be protected from debris and the like generated by laser processing.

JP 2006-140311 A

  However, when dividing the wafer with metal bump electrodes such as bumps formed on the surface of the silicon substrate, even if the wafer is covered with a protective film, a protective film with a sufficient thickness is formed on the surface of the bump electrode. There are cases where it is not possible. Therefore, debris made of silicon produced by laser processing reacts with oxygen in the air and becomes a residue such as silicon oxide (SiOx) that cannot be removed by washing with water, and adheres to the surface of the protruding electrode, making it difficult to remove it. . Since the residue deteriorates the electrical characteristics of the device (increases the electrical resistance value), there is a problem of deteriorating the characteristics of the device.

  The present invention has been made in view of the above circumstances, and an object thereof is to remove a residue attached to an electrode by laser processing by plasma etching so as not to impair device characteristics (particularly electrical characteristics). .

  The present invention relates to a wafer dividing method for dividing, along a street, a wafer in which a device having a plurality of electrodes is formed in each region partitioned by a plurality of grid-like streets on a functional layer laminated on the surface of a substrate. A holding step for holding the back side of the wafer, a protective film forming step for forming a protective film by supplying a water-soluble resin to the surface of the wafer, and a functional layer by irradiating laser light along the street A laser beam irradiation process for removing the substrate and exposing the substrate, an etching process for removing the residue adhering to the electrode by plasma etching the wafer surface after the laser beam irradiation process, and a dividing process for dividing the wafer along the street And including.

  Examples of the dividing step include a method using a cutting blade, plasma etching, or the like. That is, the substrate may be divided by cutting with a cutting blade, or the substrate may be divided by plasma etching.

  The etching step and the dividing step are preferably performed in a state where the protective film is formed on the surface of the wafer.

  The wafer dividing method of the present invention includes a holding step for holding the back side of the wafer, a protective film forming step for forming a protective film on the surface of the wafer, and removing the functional layer by irradiating laser light along the street. And a laser beam irradiation process for exposing the substrate, an etching process for plasma etching the wafer surface after the laser beam irradiation process, and a dividing process for dividing the wafer along the street. Even if residue adheres to the surface of the wafer, it can be removed by plasma etching on the wafer surface, dividing the wafer into high quality devices without compromising the electrical characteristics of the device can do.

  When the substrate is divided by cutting with a cutting blade in the dividing step, the wafer can be separated into high-quality devices. Further, when the substrate is divided by plasma etching in the dividing step, the wafer can be efficiently divided into high-quality devices without generating microcracks or the like.

  In the plasma etching division process and the etching process, the protective film is formed on the surface of the wafer so that the protective film functions as a mask during the plasma etching, thereby reducing damage to the device. The

It is a perspective view which shows the structure of a laser processing apparatus. It is a perspective view which shows the structure of a protective film formation means. It is a perspective view which shows the structure of a wafer. It is sectional drawing which shows a holding process and a protective film formation process. It is an expanded sectional view showing a laser beam irradiation process while showing a part of a wafer in the state where a protective film was formed. It is an expanded sectional view showing a part of wafer W in which a groove along a street is formed and a residue adheres to a bump. It is sectional drawing which shows the structure of an example of a plasma etching apparatus. It is an expanded sectional view which shows an etching process. It is an expanded sectional view which shows the state from which the residue was removed from the bump, and shows a part of wafer from which the protective film was removed from the surface. It is an expanded sectional view showing a division process (blade dicing). It is an expanded sectional view showing a division process (plasma etching). It is a partially expanded sectional view for demonstrating the Example (1st irradiation and 2nd irradiation) of a laser beam irradiation process. It is a partially expanded sectional view for demonstrating the Example (3rd irradiation) of a laser beam irradiation process. It is an enlarged photograph which shows the upper surface of the bump before an etching process and after an etching process.

  A laser processing apparatus 10 shown in FIG. 1 is an example of a laser processing apparatus having a function of coating a protective film on a wafer W that is a workpiece. The laser processing apparatus 10 has an apparatus base 11, and a cassette 12 capable of accommodating a plurality of wafers W is disposed on the front side (−Y axis direction side) thereof. On the upper surface 11 a of the apparatus base 11, a temporary placement region 13 in which the wafer W is temporarily placed, and a loading / unloading means for unloading the wafer W before laser processing from the cassette 12 and loading the wafer W after laser processing into the cassette 12. 14 and a chuck table 15 for holding the wafer W are disposed.

  The chuck table 15 has a holding surface 15a for sucking and holding the wafer W, and is connected to a suction source (not shown). The chuck table 15 can suck and hold the wafer W on the holding surface 15a by the suction action of the suction source. The chuck table 15 is movable in the ± X-axis direction and is also movable in the ± Y-axis direction.

  On the moving path on the −X axis direction side of the chuck table 15, there are an imaging means 18 for imaging an area of the wafer W to be laser processed (ablation processing) and a laser processing means 20 for performing laser processing on the wafer W. It is arranged. The laser processing means 20 includes an oscillating means 22 that oscillates laser light, a condenser 21 that condenses the laser light oscillated by the oscillating means 22 at a desired condensing position, and a frequency setting that adjusts the repetition frequency of the laser light. At least means 23 and adjusting means 24 for adjusting the output of the laser beam are provided.

  In the center of the upper surface 11a of the apparatus base 11, a protective film forming means 30 having a function of forming a protective film on the wafer W before laser processing and a function of cleaning and drying the wafer W after laser processing is disposed. . In the vicinity of the protective film forming means 30, a first transport means 16 for transporting the wafer W between the temporary placement region 13 and the protective film forming means 30, and a wafer on which a protective film is formed by the protective film forming means 30. A second transfer means 17 is provided for transferring W to the chuck table 15 and transferring the laser-processed wafer W to the protective film forming means 30.

As shown in FIG. 2, the protective film forming means 30 includes a spinner table 31 having a holding surface 31a for holding the wafer W in a rotatable manner. A suction source is connected to the spinner table 31, and the wafer W can be sucked and held by the holding surface 31a by the suction force of the suction source.
The formation of the protective film and the cleaning / drying of the protective film may be performed by separate apparatuses.

Below the spinner table 31, there is a rotating means 3 for rotating the spinner table 31.
5 and an elevating means 36 for elevating and lowering the spinner table 31 in the vertical direction. The rotating means 35 includes a rotating shaft 350 having a vertical axis and a motor 351 connected to the rotating shaft 350, and the motor 351 rotates the rotating shaft 350 to rotate the spinner table 31. Can do. The lifting / lowering means 36 includes an air cylinder 360 and a piston 361, and at least three lifting / lowering means 36 are disposed on the outer peripheral side of the motor 351. As the piston 361 moves up and down in the air cylinder 360, the spinner table 31 can be moved up and down.

  Around the spinner table 31, a cover portion 37 for preventing liquid resin, washing water, etc. from splashing around is disposed. Inside the cover portion 37 and along the periphery of the spinner table 31, a wafer is provided. A resin supply nozzle 32 that drops liquid resin onto W, a cleaning water supply nozzle 33 that supplies cleaning water to the wafer W, and an air supply nozzle 34 that injects air onto the wafer W are disposed. The resin supply nozzle 32, the cleaning water supply nozzle 33 and the air supply nozzle 34 are all configured to be able to turn horizontally with respect to the holding surface 31 a of the spinner table 31, and can move to the upper side of the spinner table 31. It is possible to retreat from the upper side of the spinner table 31.

  The cover portion 37 is formed with a tray portion 370 that receives the liquid resin and the washing water, and the tray portion 370 is formed with a discharge port 371. A drain hose 38 is connected to the discharge port 371, and liquid resin and washing water can be discharged out of the apparatus through the discharge port 371.

  Next, a method for dividing the wafer will be described. A wafer W shown in FIG. 3 is an example of a workpiece, and includes a substrate 1 made of silicon (Si) or silicon carbide (SiC). On the surface Wa of the substrate 1, a functional layer (Low-k film 2) and a passivation 3 (shown in FIG. 5) are stacked, and a plurality of regions partitioned by streets S formed in a lattice shape in the functional layer. Device D is formed in Each device D has a plurality of bumps 4 that are protruding electrodes, as shown in the partially enlarged view. The surface on the side opposite to the front surface Wa is a back surface Wb held by the chuck table 15 shown in FIG. The wafer W is formed integrally with the annular frame F via the protective tape T, and a plurality of wafers W are accommodated in the cassette 12.

(1) Holding Step The wafer W before laser processing is pulled out from the cassette 12 to the temporary placement region 13 and temporarily placed by the loading / unloading means 14 shown in FIG. Subsequently, the first transport unit 16 transports the wafer W temporarily placed in the temporary placement region 13 to the protective film forming unit 30. Specifically, as shown in FIG. 4, the back surface Wb side of the wafer W integrated with the annular frame F is placed on the holding surface 31 a of the spinner table 31, and the surface Wa of the wafer W is exposed upward. The wafer W is sucked and held by the holding surface 31a by a suction action of a suction source (not shown).

(2) Protective Film Forming Step After the holding step, as shown in FIG. 4, the water-soluble resin 320 is supplied from the resin supply nozzle 32 to the surface Wa of the wafer W sucked and held by the spinner table 31. Specifically, a predetermined amount of water-soluble resin 320 is dropped from the resin supply nozzle 32 toward the surface Wa of the wafer W while the spinner table 31 is rotated in the direction of arrow A by the rotating means 35 shown in FIG. By doing so, the water-soluble resin 320 is applied to the entire surface Wa of the wafer W. The water-soluble resin 320 is not limited to application by spin coating, and may be applied by supplying from a spray or slit nozzle, for example.

  After a predetermined amount of the water-soluble resin 320 is applied on the entire surface Wa of the wafer W, the protective film shown in FIG. 5 is obtained by, for example, rotating the spinner table 31 to dry and solidify the water-soluble resin 320. 321 is formed on the surface Wa of the wafer W. Although the protective film 321 also covers the surface of the bump 4, the thickness of the protective film 321 in this portion tends to be thin. That is, the thickness of the protective film 321 covering the surface of the bump 4 is about several hundreds of 200 nm to 1 μm when it is dried by spin coating. The water-soluble resin 320 may be dried by irradiation with light from a lamp (for example, mercury, tungsten, xenon pulse, etc.) or LED, or may be dried by baking with a hot plate. In the case of drying with a lamp, the thickness of the protective film 321 covering the surface of the bump 4 is about 1 μm to 10 μm (preferably 2 to 5 μm). In order to obtain the protective film 321 having a thickness of 10 μm or more, it takes a long time to dry, and when the thickness of 10 μm or more is obtained, at least the thickness of the protective film 321 is 20 μm or more on the street. In some cases, the irradiation profile of the laser beam may be lost, and laser processing may not be performed properly. When it is desired to form a thick protective film, the supply and drying of the water-soluble resin may be set as one set, and a plurality of sets may be repeated.

Here, the protective film 321 will be described in detail. The water-soluble resin 320 dropped on the surface Wa of the wafer W shown in FIG. 4 is formed by mixing one or both of a metal oxide and an organic compound having absorptivity with respect to laser light. . As the water-soluble resin 320, for example, a water-soluble resin mainly composed of polyvinyl alcohol (PVA), polyvinyl pyrrolidone (PVP), or the like is preferably used. As the metal oxide, for example, fine particles such as titanium dioxide (TiO ₂ ), Fe ₂ O ₃ , ZnO, TiO ₂ , CeO ₂ , CuO, and Cu ₂ O are preferably used. The shape of titanium dioxide may be acicular. Thereby, the laser processing described later can be promoted. Further, the etching resistance is improved during the etching described later. In addition to a metal oxide such as titanium dioxide, silica (SiO ₂ ) may be used. Silica is useful because it has a low coefficient of thermal expansion and high thermal diffusivity, and can suppress film peeling.

  Examples of organic compounds that absorb laser light include azo dyes, phthalocyanine dyes, quinone dyes, methylene dyes, naphthalene dyes, 4,4′-dicarboxybenzophenone, ferulic acid, triazine dyes, and benzotriazole dyes. It is preferable to use an ultraviolet absorber such as a system. After forming the protective film 321 on the surface Wa of the substrate 1, the presence or absence of application of the protective film 321 may be determined by irradiating the organic compound with absorptive light and detecting the fluorescence. Alternatively, whether or not the protective film 321 is applied may be determined by irradiating infrared light and detecting peaks derived from O—H, C—H, and C═O of the protective film 321. In this way, the wafer W on which the protective film 321 is formed on the surface Wa is conveyed to the chuck table 15 by the second conveying means 17 shown in FIG. 1, and is sucked and held by the holding surface 15a.

(3) Laser light irradiation process After performing the protective film forming process, the chuck table 15 shown in FIG. 1 moves in the −X-axis direction, and the imaging unit 18 detects a region (street S) to be laser processed. Then, the light collector 21 and the street S are aligned in the Y-axis direction. Subsequently, while the chuck table 15 is further moved in the −X-axis direction, as shown in FIG. 5, the high-intensity laser beam LS is irradiated along the street S toward the surface Wa of the substrate 1 to perform laser processing (ablation). Process). The laser processing may be performed by irradiating the substrate 1 with the high-intensity laser beam LS to the extent that the groove is formed, or to a depth at which the groove is formed in the Low-k film 2 so that the surface Wa of the substrate 1 is exposed. High intensity laser light LS may be irradiated.

In the laser light irradiation step, for example, the following high intensity laser light irradiation conditions are set. When laser processing is performed under these conditions, a high-quality semiconductor chip (device) can be obtained after the division step described later.
[High-intensity laser light irradiation conditions]
Wavelength: 550 nm or less Pulse width: 1 ps to 500 ns
Output: 2-100W
Repeat frequency: 10 kHz to 500 MHz
Spot diameter: 10-100 μm
Feeding speed: 100-3000mm / sec

  In the present embodiment, the high-intensity laser beam LS is irradiated a plurality of times along the street S to a depth that allows the grooves to enter the Low-k film 2, and the Low-k film 2 and the passivation as shown in FIG. 3 is removed, and the surface Wa of the substrate 1 is exposed to form a laser processing groove 5. At this time, debris is generated by irradiation with the high-intensity laser beam LS shown in FIG. 5 and reacts with oxygen in the air to form silicon oxide (SiOx) and adhere to the protective film 321. Due to the presence of the protective film 321, adhesion of debris to the surface Wa of the substrate 1 is suppressed. However, if a residue 6 such as silicon oxide (SiOx) is formed on a thin portion of the protective film 321 on the surface of the bump 4, the residue 6 cannot be easily removed by subsequent water washing.

(4) Etching Step After the laser beam irradiation step, isotropic plasma etching is performed on the surface Wa of the wafer W to remove the residue 6 attached to the bumps 4. In the etching step, for example, a microwave plasma type plasma etching apparatus 40 shown in FIG. 7 is used. The plasma etching apparatus 40 includes a chamber 41 having a processing space 42 in which plasma etching is performed. An electrostatic chuck 48 that holds the wafer W is accommodated in the chamber 41. A gate valve 47 that can be opened and closed is attached to the side of the chamber 41. By opening the gate valve 47, the wafer W can be carried into the chamber 41 or carried out of the chamber 41.

  A double-structured pipe is connected to the upper portion of the chamber 41. That is, the inner pipe is a dielectric pipe 43 that serves as an etching gas supply passage, and a cover pipe 44 is disposed surrounding the dielectric pipe 43. The dielectric tube 43 is composed of, for example, a quartz tube. A waveguide 45 is disposed on the side of the dual-structure pipe, and an oscillator 50 that oscillates microwaves is connected to the waveguide 45. The oscillator 50 is composed of, for example, a magnetron, and can oscillate a microwave having a predetermined frequency (for example, 2.45 GHz). The microwave oscillated from the oscillator 50 propagates through the waveguide 45. It is introduced into the dielectric tube 43. The cover tube 44 is made of metal and can prevent leakage of microwaves introduced into the dielectric tube 43 through the waveguide 45.

A gas supply pipe 54 (two in the illustrated example) for supplying an etching gas into the dielectric pipe 43 is connected to the upper end of the dielectric pipe 43. An etching gas source 51 is connected to each gas supply pipe 54 via a mass flow controller 53 and a valve 52. The etching gas source 51 is filled with an etching gas containing fluorine such as SF ₆ , C ₄ F ₄ , CF ₄ , C ₂ F ₆ , and C ₂ F ₄ . The etching gas introduced into the dielectric tube 43 is excited by microwaves in the plasma excitation unit 430 (airtight space in the dielectric tube 43), and plasma is generated. A dispersion plate 46 having a plurality of passage holes is disposed immediately below the dielectric tube 43. The dispersion plate 46 is arbitrarily disposed, but by providing the dispersion plate 46 in the chamber 41, the plasma can be made uniform.

  The electrostatic chuck (ESC) 48 has a base for holding the wafer W, and includes an electrode 480 in the base. The electrostatic chuck 48 is supported by a support portion 49. The electrostatic chuck 48 is connected to a high frequency power supply 55 for applying a bias high frequency voltage to the etching gas. The bias high-frequency voltage may be adjusted to a frequency and power that allow the etching species to be drawn into the substrate 1. Further, a DC power source 56 is connected to the electrode 480 of the electrostatic chuck 48, and by applying a DC voltage to the electrode 480, the wafer W is attracted onto the base of the electrostatic chuck 48 by electrostatic force such as coulomb. be able to. A controller 57 is connected to the plasma etching apparatus 40 including the chamber 41, and various mechanisms of the plasma etching apparatus 40 can be controlled. Further, a vacuum pump 58 is connected to the chamber 41, and etching by-products (etching by product) can be discharged by the vacuum pump 58. Note that the plasma etching apparatus used in the etching process is not limited to the microwave plasma method, and plasma etching apparatuses such as inductive coupling (ICP) type, capacitive coupling (CCP) type, and electron cyclotron resonance (ECR) type are used. May be.

  When plasma etching is performed on the wafer W using the plasma etching apparatus 40, the wafer W is carried into the chamber 41 and held by the electrostatic chuck 43 with the surface facing upward.

As the plasma etching conditions, for example, the conditions shown in [Table 1] below are used.
[Table 1]

  The microwave 50 is introduced into the dielectric tube 43 through the waveguide 45 by oscillating the microwave 500 of 2.45 GHz by the oscillator 50. At this time, a predetermined etching gas is introduced into the dielectric tube 43 from each etching gas source 51 at a predetermined pressure through the gas supply tube 54, and the etching gas is excited by the microwave 500 introduced into the plasma excitation unit 430 to generate plasma. Is generated. The plasma moves downward through the dielectric tube 43 and acts in contact with the wafer W in the processing space 42. Then, by applying a bias high-frequency voltage (13.56 MHz) to the wafer W from the high-frequency power supply 55, ions in the plasma are drawn into the wafer W and plasma etching is performed. During etching, the pressure is maintained at a predetermined pressure, and an etching byproduct (etching by product) is exhausted by the vacuum pump 58.

By performing plasma etching in this way, the residue 6 attached to the surface of each bump 4 shown in FIG. 8 is removed. Here, when metal particles (10 nm to 150 nm) such as titanium dioxide (TiO ₂ ) are mixed in the protective film 321 formed on the surface Wa of the wafer W, the resistance of the protective film 321 to plasma is improved. Therefore, it is possible to prevent the protective layer 321 from being excessively etched during the etching and exposing the functional layer stacked on the surface Wa of the wafer W. The protective film 321 formed in this way can have a desired etching shape and can be easily cleaned later.

Since plasma etching is performed in a state where the protective film 321 is formed on the surface Wa of the wafer W, the protective film 321 can be used as a mask. After the plasma etching is completed, the protective film 321 is removed from the surface Wa of the wafer W. For example, after supplying cleaning water (for example, pure water) to the surface Wa of the wafer W to dissolve the protective film 321, the surface Wa of the wafer W is dried. In this way, as shown in FIG. 9, the residue 6 and the protective film 321 are removed from the surface of the bump 4, and the wafer W with the Low-k film 2 and the passivation 3 exposed is left. To do. When the wafer W contains carbon, carbon contained in the debris generated during the laser light irradiation process can become a part of the residue 6. Therefore, an ashing process using O ₂ is performed as necessary after etching. May be. Plasma etching with a mask formed on the device surface reduces plasma damage to the device.

  The processing time of plasma etching is not particularly limited. For example, by performing plasma etching for 2 minutes to 4 minutes, the residue 6 attached to the surface of the bump 4 after the plasma etching becomes the total surface area of the bump 4. It is good to remove until it becomes 30% or less. When the residue process is performed after the residue 6 is removed as described above, a high-quality semiconductor chip (device) can be obtained. Moreover, you may remove the residue 6 before performing a division | segmentation process.

(5) Division process After removing the functional layer on the street S by irradiating the laser beam, forming the groove and exposing the substrate 1, the substrate 1 is further irradiated with the high-intensity laser beam one or more times. You may make it divide | segment. Such division is performed while the protective film 321 is formed on the surface Wa of the wafer W. Thereby, adhesion of debris (residue) to the surface Wa of the wafer W can be suppressed. After the substrate 1 is divided, the protective film 321 is washed, and then the residue is removed by plasma etching. The number of times of irradiation with the high-intensity laser beam is not particularly limited.

  After performing the etching process, the protective film 321 is cleaned, and the wafer W is divided into individual devices D by blade dicing with a cutting blade 50, for example, as shown in FIG. When performing the dividing step, if the thickness of the wafer W is the finished thickness, the cutting blade 50 is cut along the streets S in which the laser processing grooves 5 are formed to cut the substrate 1. Is fully cut (completely cut) to divide the wafer W into individual devices D. Although the blade dicing is performed after the protective film 321 is washed, the present invention is not limited to this.

  When performing the dividing step, if the thickness of the wafer W is not the finished thickness, the cutting blade 50 is first diced to form a processed groove having a depth corresponding to the finished thickness, and the surface Wa of the wafer W After the protective member is attached to the wafer W, the back surface Wb of the wafer W may be ground with, for example, a grinding wheel so that the processing groove is exposed, and the wafer W may be divided into individual devices D.

  The dividing step may be performed by plasma etching based on the following conditions in addition to the case where the dividing step is performed by the cutting blade 50. The plasma etching apparatus used in this step is not particularly limited. In this embodiment, an inductively coupled (ICP) type plasma source is used. However, the present invention is not limited to this, and for example, a capacitively coupled (CCP) type or microwave excitation type plasma source may be used. Note that whether the plasma etching is performed in the same chamber or a plurality of chambers is appropriately selected depending on the processing time and the type of processing gas.

[Table 2]

As shown in FIG. 11, the surface Wa of the wafer W is covered with, for example, a mask (a resist film 70 or a protective film used in laser processing). Further, plasma etching may be performed using the passivation 3 as a mask without coating the mask on the surface Wa. Based on condition A, an etching gas containing SF ₆ is supplied at a predetermined pressure for a predetermined time, and a high-frequency power of 3 kW is applied to a plasma generator (not shown) to turn the etching gas into plasma. High frequency power of 300 W is applied to the substrate 1 side, and excited etching species (ions) are drawn into the substrate. Subsequently, the deposition gas composed of C ₄ F _{8 is} supplied at a predetermined pressure for a predetermined time toward the street S where the laser processing groove 5 is formed under the condition A to the condition B, and to a plasma generation unit (not shown). A high-frequency power of 3 kW is applied, the gas is turned into plasma, and a carbon-based protective film is deposited on the side wall formed by etching. As described above, the etching based on the condition A and the deposition (side wall protection) based on the condition B are alternately repeated, and anisotropic plasma etching is performed along the street S of the wafer W, whereby the wafer W can be separated into individual devices. Divide into D (divide into pieces). In this example, the wafer W may be the finished thickness of the device before plasma etching. When the thickness is not the finished thickness, the back surface Wb of the wafer W may be processed to a finished thickness by grinding with a grinding wheel or the like. In the case where the mask is a resist, the resist may be removed by performing ashing with oxygen plasma after plasma etching.

  In addition, by processing the wafer W by plasma etching, at least a processed groove having a depth corresponding to the finished thickness is formed, and a protective member is attached to the front surface Wa of the wafer W. The wafer W may be divided (separated) into individual devices D by exposing the processed grooves by grinding.

  The dividing step by plasma etching may also be performed in a state where the protective film 321 described above is formed on the surface Wa of the wafer W, as in the etching step. That is, after carrying out the etching process, the protective film 321 can be used as a mask by performing plasma etching without removing the protective film 321 from the surface Wa of the wafer W, so that the plasma to the device can be used. Damage can be reduced.

  As described above, in the wafer dividing method according to the present invention, the etching process for performing isotropic plasma etching on the surface Wa of the wafer W is performed after the protective film forming process and the laser beam irradiation process. The residue 6 attached to the surface of the bump 4 during the laser beam irradiation process is removed by anisotropic plasma etching, and the dividing process can be performed, and the wafer W is divided into high-quality devices D. be able to.

  In consideration of the processing time of the etching process and the processing time of the dividing process by plasma etching, a plurality of chambers used for dividing may be used for one chamber used for removing residues. A processing apparatus may be configured by configuring these chambers in a cluster type. In the case of configuring in a cluster type, a transfer chamber having a transfer arm for transferring a wafer is arranged, and each chamber is connected to the transfer chamber. . A stand-alone chamber may be used.

  The wafer W to be processed may be subjected to the respective steps shown in the present embodiment while being supported on an annular frame larger than the wafer size via a protective tape, or may be applied to the back surface Wb of the wafer W. For example, you may perform each process shown to this embodiment in the state which affixed the rigid support plate (a silicon plate, glass, etc.).

  The wafer dividing method according to the present invention can be widely applied to semiconductor chip (device) manufacturing methods.

In the laser beam irradiation step described above, an experiment was performed using the following conditions, and whether or not the residue 6 adhered to the bump 4 was confirmed. The laser irradiation conditions in this example are as shown in [Table 3] below.
[Table 3]

  As shown in FIG. 12, laser processing is performed by irradiating a relatively low-intensity laser beam for preventing delamination (film peeling) in the first irradiation and the second irradiation. Two laser processing grooves 8a and 8b were formed along the extending direction. That is, the wafer W is indexed by, for example, +0.02 mm in the + Y-axis direction with respect to the center position of the street S (0 in the illustrated example), and the laser beam irradiation position is adjusted to provide a weak intensity (output is 1.). 5W) After forming the laser processing groove 8a by performing the first irradiation with the laser beam, the wafer W is indexed by, for example, -0.02 mm in the -Y-axis direction with respect to the center position of the street S. The laser processing groove 8b was formed by performing the second irradiation of irradiating the laser beam with a weak intensity (output: 1.5 W) in accordance with the irradiation position of the laser beam. In the first irradiation and the second irradiation, it was not possible to confirm a phenomenon in which the residue 6 or the like adhered to the bumps 4 to the extent that the stacked body 7 such as a Low-K film stacked on the surface of the substrate 1a was removed.

  Subsequently, as shown in FIG. 13, laser processing grooves 9 along the streets S were formed by performing third irradiation. Specifically, the laser light irradiation position is aligned with the center position of the street S, and the laser light condensing point position is adjusted slightly upward to widen the laser light irradiation range. By irradiating a laser beam having a higher intensity (5.9 W) than the two irradiations, the laser processing groove 9 is formed so as to overlap (at least partially overlap) at least a part of the laser processing grooves 8a and 8b. did. In the third irradiation, which is a high-intensity laser irradiation condition, a part of the substrate 1a was removed, and a phenomenon in which residues 6 and the like adhered to the bumps 4 was confirmed.

  In order to obtain a desired processed groove, and when a metal member such as TEG is provided on the street, a plurality of laser light irradiations along the street S are performed so that they can be sufficiently removed. You may irradiate twice. In the multiple irradiation method, irradiation is performed so that at least one irradiation region overlaps the first irradiation region and the subsequent irradiation regions. In multiple times of irradiation, a branched beam obtained by branching light from one laser light source may be used to simultaneously perform the first irradiation and the second irradiation, or may be branched into three or more and processed. Processing may be performed in parallel, perpendicularly, and obliquely to the traveling method. In order to mitigate the thermal effects of the branched beams, it is better to process them with a gap between the beams. The intensity spatial distribution may be a Gaussian type or a top hat type. Furthermore, during laser processing, an assist gas such as oxygen, nitrogen, argon, or helium may be sprayed around the laser processing groove to remove the melted / evaporated material, or the melted / evaporated material may be sucked by a suction nozzle or the like. It may be removed by doing.

  Many bumps 4 exist on the surface of the wafer W. The adhesion rate of the residue 6 on the bump 4 decreases as the distance from the street S increases and the laser processing conditions become weaker. For example, under the above laser processing conditions, in the case of the bump 4 existing at a position of 50 μm or less from the street S, the adhesion rate of the residue 6 is 50% or more (a state where 50 or more residues are adhered to the bump 100). In the case of the bump 4 existing at a position separated from the street S by 50 μm or more, the adhesion rate of the residue 6 was 5% or less. Therefore, residue removal by plasma etching is particularly effective for the bumps 4 that are 50 μm or less from the street S.

  Photographs 100 to 102 shown in FIG. 14 compare the amount of the residue 6 attached to the surface of the bump 4 before and after plasma etching. The photograph 100 shows a state in which the bump 4 before plasma etching is viewed from the upper surface side. A photograph 101 shows a state where the bump 4 is viewed from the upper surface side when the plasma etching is performed for 2 minutes. The photograph 102 shows a state in which the bump 4 is viewed from the upper surface side when the plasma etching is performed for 4 minutes. Conditions other than the plasma etching processing time are the same as in [Table 1] above. In addition, in order to make it easy to see the portion of the residue 6 attached to the bump 4, the portion of the residue 6 is shaded.

  As shown in Photo 100, it was confirmed that a large amount of residue 6 adhered to the upper surface of bump 4 before plasma etching. Further, as shown in Photo 101, when plasma etching is performed for 2 minutes, although a small amount of residue 6 is attached to the upper surface of bump 4, a larger amount of residue 6 is removed from bump 4 compared to Photo 100. It was confirmed that Furthermore, as shown in Photo 102, it was confirmed that when plasma etching was performed for 4 minutes, most of the residue 6 was removed from the bumps 4 as compared to Photo 100.

  Next, an electrical property test was performed using the wafers of photographs 100 to 102. In the wafer of the photograph 100 before plasma etching in which the residue 6 adhered to the entire surface area of the bump 4 by 30% or more, an error occurred in the electrical property test. This is presumably because the residue 6 on the surface of the bump 4 acts as an insulating film. On the other hand, in the wafers of Photo 101 and Photo 102 in which plasma etching was performed for 2 minutes and 4 minutes, the residue 6 was 30% or less with respect to the total surface area of the bumps 4, and the electrical property test was passed. Thus, by removing the residue 6 on the surface of the bump 4 by plasma etching, it was possible to obtain a high-quality chip that does not impair electrical characteristics.

DESCRIPTION OF SYMBOLS 1,1a: Board | substrate 2: Low-k film | membrane 3: Passivation 4: Bump 5: Laser processing groove | channel 6: Residue 7: Laminated body 8a, 8b: Laser processing groove | channel 9: Laser processing groove | channel 10: Laser processing apparatus 11: Apparatus base 11a: Upper surface 12: Cassette 13: Temporary placement area 14: Loading / unloading means 15: Chuck table 15a: Holding surface 16: First conveying means 17: Second conveying means 18: Imaging means 20: Laser processing means 21: Collection Optical device 22: Oscillating means 23: Frequency setting means 24: Adjustment means 30: Protective film forming means 31: Spinner table 31a: Holding surface 32: Resin supply nozzle 320: Water-soluble resin 321: Protective film 33: Washing water supply nozzle 34 : Air supply nozzle 35: Rotating means 350: Rotating shaft 351: Motor 36: Lifting means 360: Air cylinder 361: Pis 37: Cover part 370: Dish part 371: Discharge port 38: Drain hose 40: Plasma etching apparatus 41: Chamber 42: Processing space 43: Dielectric tube 430: Plasma excitation unit 44: Metal tube 45: Waveguide 46: Dispersion plate 47: Gate valve 48: Electrostatic chuck (ESC) 480: Electrode 49: Support unit 50: Oscillator 500: Microwave 51: Etching gas source 52: Valve 53: Mass flow controller 54: Gas supply pipe 55: High frequency power supply 56 : DC power source 57: Control unit 58: Vacuum pump 60: Cutting blade 70: Resist film

Claims (4)

A wafer dividing method for dividing a wafer in which a device having a plurality of electrodes is formed in each region partitioned by a plurality of grid-like streets on a functional layer laminated on a surface of a substrate, along the streets. ,
Holding step of holding the back side of the wafer;
A protective film forming step of forming a protective film by supplying a water-soluble resin to the surface of the wafer;
Laser light irradiation step of removing the functional layer by irradiating laser light along the streets to expose the substrate;
An etching step of removing residues adhering to the electrode by plasma etching the surface of the wafer after the laser light irradiation step;
A dividing step of dividing the wafer along the street;
Method of dividing a wafer including   The wafer dividing method according to claim 1, wherein in the dividing step, the substrate is divided by cutting with a cutting blade.   The wafer dividing method according to claim 1, wherein in the dividing step, the substrate is divided by plasma etching.   The wafer dividing method according to claim 1, wherein the etching step and the dividing step are performed in a state where the protective film is formed on a surface of the wafer.