JP7828160B2 - Device chip manufacturing method - Google Patents

Description

The present invention relates to a method for manufacturing device chips by dividing a plate-shaped workpiece having a device mounted on its surface into device chips.

Device chips, which include devices such as electronic circuits, are essential components of electronic devices such as mobile phones and personal computers. Device chips are obtained, for example, by dividing the surface of a wafer made of a semiconductor such as silicon (Si) into multiple regions along planned division lines called streets, forming devices in each region, and then dividing the wafer along these planned division lines.

When dividing a wafer into device chips, a cutting device with a circular tool called a cutting blade attached to a spindle is typically used. The cutting blade is rotated at high speed and cuts into the wafer from the front side along the intended division line while supplying a liquid such as pure water, cutting the wafer and dividing it into multiple device chips.

Wafer can also be divided into device chips using a laser processing machine equipped with a laser oscillator that can generate a laser beam with a wavelength that is absorbed by the wafer. In this case, the laser beam generated by the laser oscillator is irradiated from the front side of the wafer along the intended division line, ablating the wafer and dividing it into multiple device chips.

In order to secure the device chips obtained by the above-mentioned method to another device chip or a substrate, an adhesive resin layer called a non-conductive film (NCF) or die attach film (DAF) may be provided on the surface of each device chip (see, for example, Patent Document 1).

In this case, for example, before dividing the wafer into multiple device chips, a resin layer large enough to cover the entire surface of the wafer is provided on the front side of the wafer. The resin layer is then divided along with the wafer, yielding multiple device chips with an adhesive resin layer on the front side.

JP 2016-92188 A

The adhesive resin layer described above is provided on the wafer in an incompletely cured state (uncured or semi-cured state) so that it deforms appropriately when pressure is applied when fixing the device chip to the target. However, when the wafer is processed and divided into device chips using conventional methods, the heat generated during this processing hardens the resin layer, making it impossible to properly fix the device chip to the target.

Therefore, an object of the present invention is to provide a method for manufacturing device chips that can suppress hardening of the resin layer compared to conventional methods.

According to one aspect of the present invention, there is provided a method for manufacturing device chips, in which a plate-shaped workpiece having a device provided in an area on the surface side partitioned by planned dividing lines is divided along the planned dividing lines to manufacture device chips including the device, the method comprising: a resin layer forming step of forming a resin layer containing a heat-hardening resin in an uncured or semi-cured state on the surface side of the workpiece; and a workpiece cutting step, after the resin layer forming step, of manufacturing device chips by cutting the workpiece along the planned dividing lines from the back side of the workpiece on which the resin layer is provided on the surface side, in which the device chips are manufactured; and in the workpiece cutting step, a laser beam of a wavelength absorbed by the workpiece is shaped so that the irradiated area is linear or rectangular, and then irradiated onto the workpiece along the planned dividing lines . According to another aspect of the present invention, there is provided a method for manufacturing device chips by dividing a plate-shaped workpiece, in which devices are provided in areas on the surface side defined by planned dividing lines, along the planned dividing lines to manufacture device chips including the devices, the method comprising: a resin layer forming step of forming a resin layer on the surface side of the workpiece, the resin layer including a heat-hardening resin in an uncured or semi-cured state; and a workpiece cutting step, after the resin layer forming step, of cutting the workpiece along the planned dividing lines from the back side of the workpiece on which the resin layer is provided on the surface side to manufacture the device chips, wherein in the workpiece cutting step, a laser beam of a wavelength absorbed by the workpiece is irradiated onto the workpiece along the planned dividing lines so that a plurality of grooves parallel to each other are formed along each of the planned dividing lines.

Preferably, the method further includes a resin layer dividing step after the workpiece cutting step, in which an external force is applied to the resin layer to divide the resin layer to match the device chips. For example, the method further includes a tape applying step after the resin layer forming step and before the resin layer dividing step, in which an expandable tape is applied to the front side of the workpiece, and in which the resin layer dividing step divides the resin layer by expanding the tape. Alternatively, the method further includes a tape applying step after the workpiece cutting step and before the resin layer dividing step, in which an expandable tape is applied to the back side of the workpiece, and in which the resin layer dividing step divides the resin layer by expanding the tape.

Preferably, the method further includes a protective film formation step of forming a protective film on the back side of the workpiece before the workpiece cutting step.

Also preferably, in the workpiece cutting step, the workpiece is cut by irradiating the workpiece along the intended division line with a laser beam having a wavelength absorbed by the workpiece. Also preferably, after the workpiece cutting step, the method further includes a plasma etching step in which etching gas in a plasma state is supplied from the back side of the workpiece to remove processing strain or debris remaining on the device chip.

Furthermore, preferably, the workpiece cutting step includes a groove forming step in which a laser beam having a wavelength absorbed by the workpiece is irradiated onto the workpiece along the intended dividing line to form a groove opening on the back surface of the workpiece, and a plasma etching step in which, after the groove forming step, an etching gas in a plasma state is supplied from the back surface of the workpiece to remove the portion of the workpiece between the front surface and the bottom of the groove, thereby cutting the workpiece.

In a device chip manufacturing method according to one aspect of the present invention, a resin layer containing uncured or semi-cured resin is formed on the front surface of the workpiece, and then the workpiece is cut along the intended division lines from the back surface of the workpiece. This means that heat is less likely to be transferred to the resin layer on the front surface than when the workpiece is cut from the front surface. Therefore, according to a device chip manufacturing method according to one aspect of the present invention, curing of the resin layer can be suppressed compared to conventional methods.

FIG. 1 is a perspective view showing a workpiece. FIG. 2 is a perspective view showing a workpiece to which an expandable tape is attached. FIG. 3 is a cross-sectional view showing how the raw material for the protective film is applied to the back surface of the workpiece. FIG. 4 is a cross-sectional view showing a workpiece having a protective film formed on the back surface side thereof. FIG. 5 is a cross-sectional view showing how a laser beam is irradiated onto a workpiece. FIG. 6 is a cross-sectional view showing the workpiece after it has been cut. FIG. 7 is a cross-sectional view showing a schematic view of an etching gas in a plasma state being supplied to a workpiece. FIG. 8 is a cross-sectional view showing the state in which the tape is expanded. FIG. 9 is a cross-sectional view showing a state in which the resin layer is divided. FIG. 10 is a cross-sectional view of the workpiece after the grooves have been formed. FIG. 11 is a cross-sectional view schematically showing a plasma processing apparatus. FIG. 12 is a cross-sectional view showing the workpiece after the mask layer has been formed.

Embodiments of the present invention will be described below with reference to the accompanying drawings.

(First embodiment)
1 is a perspective view showing a workpiece 11 used in a device chip manufacturing method according to this embodiment. The workpiece 11 is, for example, a disk-shaped wafer made of a semiconductor material such as silicon. The workpiece 11 has a circular front surface 11a and a circular back surface 11b opposite the front surface 11a. The front surface 11a side of the workpiece 11 is partitioned into a plurality of small regions by a plurality of intersecting streets (planned division lines) 13, and a device 15 including an integrated circuit (IC) or the like is formed in each small region.

In this embodiment, the workpiece 11 is a disk-shaped wafer made of a semiconductor material such as silicon, but there are no restrictions on the material, shape, structure, size, etc. of the workpiece 11. For example, a substrate made of other semiconductors, ceramics, resin, metal, or other materials can also be used as the workpiece 11. Similarly, there are no restrictions on the type, quantity, shape, structure, size, arrangement, etc. of the devices 15.

In the device chip manufacturing method according to this embodiment, first, a resin layer 21 is formed on the surface 11a side of the workpiece 11 (resin layer formation step). The resin layer 21 is typically a non-conductive film (NCF) or die attach film (DAF), and contains uncured or semi-cured resin. This resin layer 21 has a predetermined adhesive strength and is used as an underfill material when mounting the device chips obtained by dividing the workpiece 11 onto another substrate, etc.

As shown in FIG. 1, the resin layer 21 is a disk-shaped film having roughly the same diameter as the workpiece 11, and is attached to the workpiece 11 so as to cover roughly the entire surface 11a. However, there are no limitations on the type of resin layer 21. For example, the resin layer 21 may be formed by applying a non-conductive paste (NCP) or the like to the surface 11a of the workpiece 11.

Furthermore, there are no limitations on the materials that make up the resin layer 21. Resin layer 21 may be made of a resin whose main component is, for example, an epoxy resin, an acrylic resin, a urethane resin, a silicone resin, or a polyimide resin. Furthermore, oxidizers, fillers, etc. may also be added to resin layer 21.

After the resin layer 21 containing uncured or semi-cured resin is formed on the surface 11a of the workpiece 11, an expandable tape is applied to the surface 11a of the workpiece 11 (tape application step). Figure 2 is a perspective view showing the workpiece 11 to which the expandable tape 23 has been applied.

The tape 23 includes, for example, a circular base film (substrate) with a diameter larger than the workpiece 11, and an adhesive layer (glue layer) provided on the base film. The base film is typically made of a resin such as polyolefin or polyvinyl chloride, and has expandable properties.

The adhesive layer is typically made of a resin whose main component is an epoxy resin, an acrylic resin, a urethane resin, a silicone resin, a polyimide resin, or a rubber resin, and has adhesive properties to the workpiece 11. The adhesive layer may also be made of an ultraviolet-curable resin that hardens when exposed to ultraviolet light.

As shown in FIG. 2, the tape 23 is attached to, for example, both the workpiece 11 and an annular frame 25 that is arranged to surround the workpiece 11. The frame 25 is annular and made of a metal such as stainless steel or aluminum, and has a circular opening 25a in the center that is larger in diameter than the workpiece 11.

With the workpiece 11 placed inside the opening 25a of the frame 25, tape 23 is attached to the surface 11a of the workpiece 11 (resin layer 21) and to the frame 25. This allows the surface 11a of the workpiece 11 to be supported by the frame 25 via the resin layer 21 and tape 23, making the workpiece 11 easier to handle.

After the tape 23 is attached to the front surface 11a of the workpiece 11, a protective film is formed on the back surface 11b of the workpiece 11 (protective film formation step). Figure 3 is a cross-sectional view showing how the protective film raw material 27 is applied to the back surface 11b of the workpiece 11. In this embodiment, the protective film raw material 27 is applied to the workpiece 11 using, for example, the spin coater 2 shown in Figure 3.

The spin coater 2 is equipped with a spinner table 4 configured to hold the workpiece 11. The top surface (holding surface) 4a of the spinner table 4 is generally flat and comes into contact with, for example, tape 23 attached to the resin layer 21. This top surface 4a is connected to a suction source (not shown) such as an ejector via a flow path 4b provided inside the spinner table 4 and a valve (not shown).

A rotary drive source (not shown), such as a motor, is connected to the bottom of the spinner table 4, rotating the spinner table 4 around a rotation axis that passes through the center of the top surface 4a and is generally parallel to the vertical direction (up-down direction). Additionally, multiple clamps 6 that can grip and secure the frame 25 are provided around the periphery of the spinner table 4. A nozzle 8 that can drip protective film raw material 27 is located above the center of the top surface 4a.

When forming a protective film on the back surface 11b of the workpiece 11, the workpiece 11 is first placed on the spinner table 4 so that the tape 23 attached to the resin layer 21 is in contact with the top surface 4a, i.e., with the back surface 11b facing upward. In this state, when negative pressure (suction force) generated by the suction source is applied to the top surface 4a, the tape 23 is attracted to the top surface 4a, and the workpiece 11 is held to the spinner table 4 via the tape 23. The frame 25 is fixed in place by multiple clamps 6.

Next, protective film raw material 27 is dropped from nozzle 8 toward the center of the back surface 11b of workpiece 11, and spinner table 4 is rotated. This causes protective film raw material 27 to spread from the center of back surface 11b outward, coating the entire back surface 11b of workpiece 11. Examples of protective film raw material 27 that can be used include water-soluble resins such as polyvinyl alcohol (PVA), polyethylene glycol (PEG), polyethylene oxide (PEO), and polyvinylpyrrolidone (PVP).

Then, the protective film raw material 27 applied to the workpiece 11 is dried or otherwise processed to form a water-soluble protective film that covers the entire back surface 11b of the workpiece 11. Figure 4 is a cross-sectional view showing the workpiece 11 with a protective film 29 formed on the back surface 11b. There are no specific limitations on the method for forming this protective film 29. For example, a tape made of resin or the like can be attached to the back surface 11b of the workpiece 11 and used as the protective film 29.

After forming a protective film 29 on the back surface 11b of the workpiece 11, the workpiece 11 is cut from the back surface 11b along the streets 13 to produce multiple device chips, each having a device 15 (workpiece cutting step). In this embodiment, the workpiece 11 is cut by irradiating the workpiece 11 along the streets 13 with a laser beam of a wavelength that is absorbed by the workpiece 11.

Figure 5 is a cross-sectional view showing how a laser beam 31 is irradiated onto a workpiece 11. In this embodiment, for example, the laser beam 31 is irradiated onto the workpiece 11 using the laser processing device 12 shown in Figure 5. Note that the X-axis direction (processing feed direction), Y-axis direction (indexing feed direction), and Z-axis direction (vertical direction) used in the following explanation are mutually perpendicular directions.

As shown in FIG. 5, the laser processing device 12 is equipped with a chuck table 14 configured to hold the workpiece 11. The upper surface (holding surface) 14a of the chuck table 14 is configured to be generally flat and parallel to the X-axis and Y-axis directions, and comes into contact with, for example, tape 23 attached to the resin layer 21. This upper surface 14a is connected to a suction source (not shown) such as an ejector via a flow path 14b provided inside the chuck table 14 and a valve (not shown).

A rotary drive source (not shown), such as a motor, is connected to the lower part of the chuck table 14. This drive source rotates the chuck table 14 around a rotation axis that passes through the center of the upper surface 14a and is generally parallel to the Z-axis direction. Additionally, multiple clamps 16 that can grip and secure the frame 25 are provided around the periphery of the chuck table 14. The chuck table 14, rotary drive source, and clamps 16 are supported by a ball-screw type movement mechanism (not shown), which moves them along the X-axis and Y-axis directions.

Above the chuck table 14 is a laser processing head 18 that directs a laser beam 31 generated by a laser oscillator (not shown) to the workpiece 11 held by the chuck table 14. The laser oscillator is equipped with a laser medium such as Nd:YAG that is suitable for laser oscillation, and generates a pulsed laser beam 31 with a wavelength that is absorbed by the workpiece 11 at a predetermined repetition frequency.

The laser processing head 18 is equipped with an optical system, such as mirrors and lenses, that guides the pulsed laser beam 31 emitted from the laser oscillator to the workpiece 11, and focuses the laser beam 31, for example, at a predetermined position above the chuck table 14. The laser beam 31 irradiated from the laser processing head 18 onto the workpiece 11 causes the workpiece 11 to be laser ablated.

When cutting the workpiece 11 with the laser beam 31, the workpiece 11 is placed on the chuck table 14 so that the tape 23 attached to the resin layer 21 is in contact with the top surface 14a, i.e., with the back surface 11b facing upward. In this state, when negative pressure (suction force) generated by the suction source is applied to the top surface 14a, the tape 23 is sucked to the top surface 14a, and the workpiece 11 is held to the chuck table 14 via the tape 23. The frame 25 is fixed in place by multiple clamps 16.

Next, the orientation of the chuck table 14 around its rotation axis is adjusted so that the length of the street 13 to be processed is aligned with the X-axis direction. Then, the position of the chuck table 14 in the Y-axis direction is adjusted so that the laser processing head 18 is positioned above the extension of the street 13 (above the line passing through the center of the street 13 in the width direction). Furthermore, the optical system of the laser processing head 18 is adjusted so that the laser beam 31 is focused at a position in the Z-axis direction suitable for processing the workpiece 11.

Then, while irradiating the laser beam 31 from the laser processing head 18, the chuck table 14 is moved along the X-axis direction at a predetermined speed (processing feed rate). In other words, the workpiece 11 held on the chuck table 14 and the focal point of the laser beam 31 are moved relatively along the X-axis direction. As a result, the laser beam 31 is irradiated onto the workpiece 11 from the back surface 11b side along the street 13.

The conditions for irradiating the laser beam 31 are adjusted within a range that allows for processing of the workpiece 11 (laser ablation processing). For example, the wavelength of the laser beam 31 is set to 266 nm to 1064 nm, typically 355 nm, the repetition frequency is set to 10 kHz to 1000 kHz, typically 200 kHz, the average output is set to 1 W to 20 W, typically 2 W, and the processing feed rate is set to 10 mm/s to 1000 mm/s, typically 400 mm/s. However, there are no restrictions on the specific conditions for irradiating the laser beam 31.

When the laser beam 31 is irradiated onto the workpiece 11 along the street 13, the portion of the workpiece 11 irradiated with the laser beam 31 is removed by laser ablation, and a groove 11c is formed along the street 13 on the back surface 11b side of the workpiece 11. After the groove 11c is formed on the target street 13, grooves 11c are formed on all streets 13 using the same procedure.

After grooves 11c have been formed on all streets 13, the laser beam 31 is repeatedly applied to deepen each groove 11c. Finally, the workpiece 11 is cut, yielding multiple device chips. The number of times (number of passes) the laser beam 31 is applied to each street 13 is approximately 5 to 20 times (5 to 20 passes) when the thickness of the workpiece 11 is approximately 20 μm to 100 μm.

However, there is no limit to the number of times (number of passes) that the laser beam 31 is irradiated onto each street 13. If the workpiece 11 is sufficiently thin or if the average output of the laser beam 31 is sufficiently high, the workpiece 11 may be cut with a single irradiation. Figure 6 is a cross-sectional view showing the workpiece 11 after it has been divided into device chips 33.

When the workpiece 11 is processed by laser ablation as described above, molten material from the workpiece 11 may scatter around and adhere to the back surface 11b as debris. However, in this embodiment, a protective film 29 is formed on the back surface 11b side of the workpiece 11, making it difficult for debris to adhere to the back surface 11b of the workpiece 11, preventing contamination of the workpiece 11 and device chip 33.

After cutting the workpiece 11 to produce multiple device chips 33, etching gas in a plasma state is supplied from the back surface 11b side of the workpiece 11 to remove any processing damage or debris remaining on each device chip 33 (plasma etching step). Figure 7 is a cross-sectional view that schematically shows how etching gas in a plasma state is supplied to the workpiece 11. In this embodiment, for example, the plasma processing apparatus 22 shown in Figure 7 is used to supply etching gas in a plasma state to the workpiece 11 from the back surface 11b side.

The plasma processing apparatus 22 includes a chamber 24. Inside the chamber 24, a processing space is provided in which plasma processing is performed on the workpiece 11. An opening 24b is formed in the sidewall 24a of the chamber 24, through which the workpiece 11 and frame 25 pass when the workpiece 11 is loaded and unloaded.

A cover 26 that closes the opening 24b is disposed on the outside of the side wall 24a. An opening/closing mechanism 28, such as an air cylinder, is connected to the cover 26. The opening/closing mechanism 28 moves the cover 26 downward to expose the opening 24b, allowing the workpiece 11 to be loaded into and unloaded from the processing space. The processing space is sealed by moving the cover 26 upward with the opening/closing mechanism 28 to close the opening 24b.

A decompression unit 32, such as a vacuum pump, is connected to the bottom wall 24c of the chamber 24 via piping 30. Therefore, for example, when the opening 24b is closed with the cover 26 and the decompression unit 32 is operated with the processing space sealed, the processing space of the chamber 24 is evacuated and the pressure in this processing space is reduced.

A table base 34 is provided inside the chamber 24. The table base 34 includes a disk-shaped holding portion 36 and a cylindrical support portion 38 that supports the holding portion 36 from below. The width (diameter) of the support portion 38 is, for example, smaller than the width (diameter) of the holding portion 36, and the upper end of the support portion 38 is connected to the lower end of the holding portion 36.

A chuck table 40 capable of holding the workpiece 11 is disposed on the upper surface of the holding portion 36. The chuck table 40 comprises a disk-shaped insulating portion 42 made of an insulating material and multiple electrodes 44 embedded inside the insulating portion 42. Each of the multiple electrodes 44 is connected to a DC power supply 46 capable of applying a predetermined DC voltage (e.g., a high DC voltage of about 5 kV) to the electrode 44.

The insulating portion 42 of the chuck table 40 is provided with multiple suction paths 42a that open to the upper surface of the insulating portion 42 (i.e., the upper surface of the chuck table 40). The suction paths 42a are connected to the suction pump 48 via suction paths 34a formed inside the table base 34, etc.

For example, when a workpiece 11 or the like is placed on the chuck table 40 and the suction pump 48 is operated, the workpiece 11 or the like is sucked onto the upper surface of the chuck table 40 by the suction force of the suction pump 48. Furthermore, when a DC voltage is applied to the electrode 44 by the DC power supply 46 to create a potential difference between the electrodes 44, the workpiece 11 or the like is attracted to the chuck table 40 by the electrical force acting between the electrode 44 and the workpiece 11. Therefore, even if the pressure inside the chamber 24 is reduced, the workpiece 11 can be held by the chuck table 40.

A flow path 34b is formed inside the table base 34. Both ends of the flow path 34b are connected to a circulation unit 50 that circulates a refrigerant such as water. When the circulation unit 50 is activated, the refrigerant flows from one end of the flow path 34b to the other, cooling the table base 34.

A gas supply unit 52 that supplies etching gas is connected to the top of the chamber 24. The gas supply unit 52 is configured to convert the etching gas into plasma outside the chamber 24 and supply the plasma-state etching gas to the processing space of the chamber 24. Specifically, the gas supply unit 52 is equipped with a supply pipe 54 through which the etching gas supplied to the chamber 24 flows.

One end (downstream side) of the supply pipe 54 is connected to the internal processing space via the upper wall 24d of the chamber 24. The other end (upstream side) of the supply pipe 54 is connected to a gas supply source 62a via a valve 56a, a flow controller 58a, and a valve 60a, to a gas supply source 62b via a valve 56b, a flow controller 58b, and a valve 60b, and to a gas supply source 62c via a valve 56c, a flow controller 58c, and a valve 60c.

When predetermined gases are supplied from the gas supply sources 62a, 62b, and 62c at predetermined flow rates, these gases are mixed in the supply pipe 54 to form an etching gas used for etching. For example, the gas supply source 62a supplies a fluorine-based gas such as _SF6 , the gas supply source 62b supplies oxygen gas ( _O2 gas), and the gas supply source 62c supplies an inert gas such as He. However, the components and flow rate ratios of the gases supplied from the gas supply sources 62a, 62b, and 62c can be changed as desired depending on the material of the object to be processed, the required processing quality, and the like.

The gas supply unit 52 includes an electrode 64 that applies a high-frequency voltage to the etching gas in the supply pipe 54. The electrode 64 is disposed so as to surround the midstream portion of the supply pipe 54 and is connected to a high-frequency power supply 66. The high-frequency power supply 66 applies a high-frequency voltage to the electrode 64, for example, with a Vpp (Voltage peak to peak) of 0.5 kV to 5 kV and a frequency of 450 kHz to 2.45 GHz.

When a high-frequency voltage is applied to the etching gas flowing through the supply pipe 54 using the electrode 64 and high-frequency power supply 66, some of the molecules in the etching gas are converted into ions and radicals. The plasma-state etching gas containing these ions and radicals is then supplied to the processing space inside the chamber 24 from the supply port 54a, which opens at the downstream end of the supply pipe 54. In this way, the etching gas that has been converted into plasma outside the chamber 24 is supplied to the processing space inside the chamber 24.

A dispersion member 68 that disperses (diffuses) the plasma-state etching gas is attached to the inner surface of the upper wall 24d of the chamber 24, covering the supply port 54a. The plasma-state etching gas that flows into the chamber 24 from the supply pipe 54 is dispersed above the chuck table 40 by this dispersion member 68.

A pipe 70 is connected to the sidewall 24a of the chamber 24, and a gas supply source (not shown) that supplies an inert gas is connected to the pipe 70. When the inert gas is supplied from this gas supply source to the chamber 24 through the pipe 70, the processing space of the chamber 24 is filled with the inert gas (inner gas). The pipe 70 may also be connected to the gas supply source 62c via a valve (not shown), a flow rate controller (not shown), or the like. In this case, the inert gas is supplied from the gas supply source 62c to the chamber 24 through the pipe 70.

The etching gas supplied from the gas supply unit 52 and converted into plasma within the supply pipe 54 is dispersed (diffused) by the dispersion member 68 located below the supply port 54a and supplied from above to the entire workpiece 11 held by the chuck table 40. As a result, the plasma-state etching gas acts on the workpiece 11, and the workpiece 11 is processed by this etching gas (plasma etching).

When removing processing strain or debris remaining on the workpiece 11 (device chip 33), the workpiece 11 is first loaded into the processing space of the chamber 24 through the opening 24b and placed on the chuck table 40. Here, the workpiece 11 is placed on the chuck table 40 so that the tape 23 attached to the resin layer 21 is in contact with the upper surface of the chuck table 40, i.e., with the back surface 11b facing upward.

Next, the suction pump 48 is activated. As a result, the workpiece 11, etc. is sucked onto the upper surface of the chuck table 40 by the suction force of the suction pump 48. Furthermore, a DC voltage is applied to the electrode 44 by the DC power supply 46. As a result, the workpiece 11, etc. is attracted to the chuck table 40 by the electrical force acting between the electrode 44 and the workpiece 11.

After the workpiece 11 is held on the chuck table 40, plasma-state etching gas is supplied to the workpiece 11 from the back surface 11b side. Specifically, the opening/closing mechanism 28 first moves the cover 26 upward to close the opening 24b. This seals the processing space of the chamber 24.

The decompression unit 32 is also operated to reduce the pressure in the processing space of the chamber 24. An appropriate amount of inert gas may be supplied to the processing space of the chamber 24 through the piping 70. In this state, etching gas is flowed through the supply pipe 54, and a high-frequency voltage is applied to the etching gas using the electrode 64 and high-frequency power supply 66.

As a result, etching gas in a plasma state containing ions and radicals is supplied from the supply port 54a to the workpiece 11 below. Since a protective film 29 is provided on the back surface 11b of the workpiece 11, this protective film 29 acts as a mask layer, and the etching gas in a plasma state acts almost exclusively on the back surface 11b of the workpiece 11, but mainly on the side surface of the device chip 33 (the area that was previously the groove 11c).

When the etching gas in a plasma state acts on the side of the device chip 33, processing strain generated on the side of the device chip 33 and its surroundings by, for example, irradiation with the laser beam 31 is removed. Furthermore, debris (foreign matter) attached to the side of the device chip 33 and its surroundings is removed by, for example, irradiation with the laser beam 31. This prevents a decrease in the flexural strength and quality of the device chip 33.

In this embodiment, the etching gas is converted into plasma outside the chamber 24, so the proportion of ions in the etching gas that reach the workpiece 11 is lower than when the etching gas is converted into plasma inside the chamber 24. As a result, deformation of the device chip 33 associated with etching of the back surface 11b, which is prone to progress due to ions, can be suppressed, while processing strain or debris can be removed from the entire side surface of the device chip 33.

However, if the workpiece 11 is cut under conditions that are unlikely to produce processing distortion or debris, if debris is reliably removed in a subsequent cleaning process, or if remaining processing distortion or debris does not impair the operation of the device 15 or the quality of the device chip 33, the supply of the plasma-state etching gas described above may be omitted.

After the etching gas in a plasma state removes any processing strain or debris remaining on each device chip 33, an external force is applied to the resin layer 21 to divide the resin layer 21 to fit the device chips 33 (resin layer division step). In this embodiment, the tape 23 is expanded to apply an external force to the resin layer 21, dividing the resin layer 21. Note that it is advisable to remove any protective film 29 remaining on the workpiece 11 by cleaning or other methods before applying an external force to the resin layer 21.

Figure 8 is a cross-sectional view showing the tape 23 being expanded, and Figure 9 is a cross-sectional view showing the resin layer 21 in a divided state. In this embodiment, the tape 23 is expanded using, for example, the expansion device 72 shown in Figures 8 and 9. As shown in Figures 8 and 9, the expansion device 72 has a cylindrical drum 74 with a circular opening at its upper end that is larger than the diameter of the workpiece 11.

A number of rollers 76 are arranged around the upper end of the drum 74 in the circumferential direction. A number of columnar support members 78 are also arranged on the outside of the drum 74. An air cylinder (not shown) is connected to the lower end of each support member 78, which moves (raises and lowers) the support member 78 vertically.

The upper end of each support member 78 is fixed to the underside of an annular table 80 with a circular opening in the center. The diameter of the opening in the table 80 is larger than the diameter (outer diameter) of the drum 74, and the upper part of the drum 74 is inserted into the opening in the table 80. Above the table 80, for example, an annular fixing member 82 is arranged to sandwich and fix the frame 25 between the table 80 and the top surface of the table 80.

When expanding the tape 23 to divide the resin layer 21, first, the support member 78 is moved using an air cylinder (not shown) so that the upper end of the roller 76 and the upper surface of the table 80 are positioned at approximately the same height. Then, the frame 25 is placed on the upper surface of the table 80 and fixed to the table 80 with the annular fixing member 82 (Figure 8). At this time, the workpiece 11 is positioned so that it overlaps with the opening at the upper end of the drum 74.

Next, the support member 78 is lowered by an air cylinder (not shown), and the table 80 is pulled down. A portion of the tape 23 is supported by the drum 74 and roller 76, and its height is maintained. Therefore, when the frame 25 is pulled down along with the table 80, the tape 23 is pulled radially outward by the frame 25, and is expanded radially.

When the tape 23 expands, an external force acting radially outward is also applied to the resin layer 21 to which the tape 23 is attached. As a result, the resin layer 21 is broken in the gaps between adjacent device chips 33. In other words, the resin layer 21 is divided into small pieces 21a to fit the device chips 33, and these small pieces 21a of the resin layer 21 are provided on the surface 11a side of the device chip 33 (Figure 9). The device chip 33, together with the small pieces 21a, is then picked up from the tape 23 and mounted on a substrate or the like.

As described above, in the device chip manufacturing method according to this embodiment, a resin layer 21 containing uncured or semi-cured resin is formed on the front surface 11a of the workpiece 11, and then the workpiece 11 is cut along the streets (planned division lines) 13 from the back surface 11b side of the workpiece 11. Therefore, heat is less likely to be transferred to the resin layer 21 provided on the front surface 11a side than when the workpiece 11 is cut from the front surface 11a side. Therefore, curing of the resin layer 21 can be suppressed compared to conventional methods.

In this embodiment, the same tape 23 is used without being replaced, but the tape 23 may be replaced as necessary. For example, if the tape 23 comes into contact with plasma etching gas, the tape 23 may be deteriorated. Therefore, the tape 23 may be replaced after the plasma etching gas is supplied.

When the tape is replaced, it is sufficient that the replaced tape has expandability. In other words, the application of the expandable tape 23 to the surface 11a of the workpiece 11 can be performed at any time after the resin layer 21 is formed on the workpiece 11 and before the resin layer 21 is divided.

Furthermore, when the tape is replaced, the expandable tape 23 does not necessarily have to be attached to the front surface 11a of the workpiece 11. For example, after the workpiece 11 has been cut into multiple device chips 33, the expandable tape 23 can be attached to the back surface 11b of the workpiece 11 at any time before the resin layer 21 is divided. In this case, the tape 23 can be expanded in a similar manner, and the resin layer 21 can be divided.

Furthermore, in this embodiment, the laser beam 31 is irradiated sequentially onto multiple streets 13 to form grooves 11c (first pass), and then the workpiece 11 is cut by irradiating the laser beam 31 to deepen each groove 11c (second pass and beyond). However, it is also possible to cut the workpiece 11 along one street 13 and then along another street 13.

The laser beam 31 irradiated onto the workpiece 11 may be shaped so that the shape of the irradiated area (beam profile) is linear or rectangular. In this case, for example, a wide groove 11c is formed by aligning the longitudinal direction of the irradiated area with the width direction of the street 13. It is also possible to form multiple grooves 11c parallel to each other on each street 13.

Furthermore, in this embodiment, a protective film 29 is used as a mask layer when the plasma etching gas acts on the workpiece 11. However, instead of this protective film 29, a mask layer made of a photosensitive resin formed by photolithography or the like can also be used. If the effect of the etching gas on the workpiece 11 is small, such as when the time during which the plasma etching gas acts on the workpiece 11 is sufficiently short, the plasma etching gas may be acted on the workpiece 11 without using a mask layer.

Second Embodiment
In the device chip manufacturing method according to this embodiment, the workpiece 11 is cut by combining irradiation of a laser beam 31 and supply of an etching gas in a plasma state. Before cutting the workpiece 11, a resin layer 21 is formed on the front surface 11a of the workpiece 11 (resin layer forming step), as in the first embodiment described above. Tape 23 is then applied to the front surface 11a of the workpiece 11 (resin layer 21) (tape application step), and a protective film 29 is then formed on the back surface 11b of the workpiece 11 (protective film forming step).

For example, after forming a protective film 29 on the back surface 11b of the workpiece 11, the workpiece 11 is cut from the back surface 11b along the streets 13 to produce multiple device chips 33, each having a device 15 (workpiece cutting step). More specifically, a laser beam 31 with a wavelength absorbed by the workpiece 11 is first irradiated onto the workpiece 11 along the streets 13 to form grooves 11c opening into the back surface 11b of the workpiece 11 (groove forming step).

In this embodiment, the laser beam 31 is irradiated onto the workpiece 11 using the laser processing device 12 described above (Figure 5). Specific procedures are the same as those for irradiating the workpiece 11 with the laser beam 31 in the first embodiment described above. However, the average output of the laser beam 31 and the number of times (number of passes) the laser beam 31 is irradiated onto each street 13 are adjusted within a range that does not cut the workpiece 11. Figure 10 is a cross-sectional view showing the workpiece 11 after the groove 11c has been formed.

After forming a groove 11c that opens into the back surface 11b of the workpiece 11, etching gas in a plasma state is supplied from the back surface 11b side to remove the portion between the front surface 11a of the workpiece 11 and the bottom of the groove 11c, thereby cutting the workpiece 11 (plasma etching step).

Figure 11 is a cross-sectional view schematically showing a plasma processing apparatus 92 that is different from the plasma processing apparatus 22 of the first embodiment described above. In this embodiment, for example, using the plasma processing apparatus 92 shown in Figure 11, an etching gas in a plasma state is supplied to the workpiece 11 from the rear surface 11b side. However, the plasma processing apparatus 22 of the first embodiment may also be used.

As shown in FIG. 11, the plasma processing apparatus 92 includes a chamber 94 with a processing space provided inside. An opening 94a large enough to allow the workpiece 11 and frame 25 to pass through is formed in the side wall of the chamber 94. A cover 96 large enough to cover the opening 94a is provided outside the opening 94a.

An opening/closing mechanism (not shown) is connected to the cover 96, and this opening/closing mechanism moves the cover 96. For example, by moving the cover 96 downward to expose the opening 94a, the workpiece 11 can be loaded into or unloaded from the processing space inside the chamber 94 through this opening 94a.

An exhaust port 94b is formed in the bottom wall of the chamber 94. This exhaust port 94b is connected to an exhaust unit 98 such as a vacuum pump. A lower electrode 100 is disposed within the space of the chamber 94. The lower electrode 100 is formed in a disk shape using a conductive material and is connected to a high-frequency power source 102 outside the chamber 94.

A chuck table 104 is disposed on the upper surface of the lower electrode 100. The chuck table 104 has a structure in which electrodes 106a and 106b are embedded in a plate-shaped insulating portion, and the workpiece 11 is attracted to the electrodes 106a and 106b by the electrical force acting between the electrodes 106a and 106b and the workpiece 11.

For example, electrode 106a is configured so that the positive pole of DC power supply 108a can be connected, and electrode 106b is configured so that the negative pole of DC power supply 108b can be connected. Note that DC power supplies 108a and 108b may be the same DC power supply. Furthermore, the top surface of chuck table 104 may have an open end of a suction path that transmits the suction force of a suction pump or the like.

An upper electrode 110, formed in a disk shape using a conductive material, is attached to the upper wall of the chamber 94 via an insulating member 112. Multiple gas outlet holes 110a are formed on the lower surface of the upper electrode 110. The gas outlet holes 110a are connected to a gas supply source 114 via gas supply holes 110b and other holes provided on the upper surface of the upper electrode 110. This allows etching gas to be supplied from the gas supply source 114 to the processing space of the chamber 94. This upper electrode 110 is also connected to a high-frequency power source 116 outside the chamber 94.

When supplying plasma etching gas to the back surface 11b of the workpiece 11, the workpiece 11 is carried into the processing space of the chamber 94 through the opening 94a and placed on the chuck table 104. Here, the workpiece 11 is placed on the chuck table 104 so that the tape 23 attached to the resin layer 21 is in contact with the upper surface of the chuck table 104, i.e., so that the back surface 11b faces upward.

Next, DC voltage is applied to electrodes 106a and 106b by DC power supplies 108a and 108b. As a result, the workpiece 11 and other objects are attracted to the chuck table 104 by the electrical force acting between electrodes 106a and 106b and the workpiece 11.

After the workpiece 11 is adsorbed onto the chuck table 104, plasma-state etching gas is supplied to the workpiece 11 from the back surface 11b side of the workpiece 11. Specifically, first, the cover 96 is moved using the opening/closing mechanism to close the opening 94a. This seals the processing space of the chamber 94. The exhaust unit 98 is then operated to reduce the pressure in the processing space of the chamber 94. An appropriate amount of inert gas may also be supplied to the processing space.

In this state, when etching gas is supplied at a predetermined flow rate from the gas supply source 114 and appropriate high-frequency power is supplied to the lower electrode 100 and upper electrode 110 from the high-frequency power supply 102 and high-frequency power supply 116, some of the molecules of the etching gas present between the lower electrode 100 and upper electrode 110 are converted into ions and radicals.

As a result, a plasma-state etching gas containing ions and radicals is supplied to the back surface 11b of the workpiece 11 held by the chuck table 104. Since a protective film 29 is provided on the back surface 11b of the workpiece 11, this protective film 29 acts as a mask layer, and the plasma-state etching gas acts mainly on the grooves 11c, with little effect on the back surface 11b of the workpiece 11.

If the portion between the surface 11a of the workpiece 11 and the bottom of the groove 11c is relatively thick, it is advisable to repeat the three steps of film formation, partial film removal, and groove processing in order to properly remove this portion. For example, if the workpiece 11 is a wafer made of silicon, the process is as follows:

In the film formation step, for example, while maintaining the pressure in the internal space of the chamber 94, _C4F8 is supplied from the gas supply source ₁₁₄ at a predetermined flow rate, and a predetermined high-frequency power is supplied to the lower electrode 100 and the upper electrode 110. This causes a fluorine-based material to be deposited inside the groove 11c, forming a thin film that covers the inner surface of the groove 11c. This film made of a fluorine-based material has a predetermined resistance to ions and radicals generated using _SF6 as a raw material.

In the partial film removal step, for example, while maintaining a constant pressure in the internal space of the chamber 94, _SF6 is supplied from the gas supply source 114 at a predetermined flow rate, and predetermined high-frequency power is supplied to the lower electrode 100 and the upper electrode 110. This makes it possible to generate ions and radicals using _SF6 as a raw material. Note that in this partial film removal step, the power supplied to the lower electrode 100 is made higher than that in the subsequent groove processing step.

Increasing the power supplied to the lower electrode 100 increases the anisotropy of the etching. Specifically, the portion of the film covering the groove 11c on the lower electrode 100 side (i.e., the bottom side of the groove 11c) is preferentially processed. In other words, the ions and radicals generated using _SF6 as a raw material can remove only the portion of the film covering the bottom of the groove 11c.

In the groove processing step, for example, while maintaining the pressure in the processing space of the chamber 94, _SF6 is supplied from the gas supply source 114 at a predetermined flow rate, and predetermined high-frequency power is supplied to the lower electrode 100 and the upper electrode 110. This generates ions and radicals using _SF6 as a raw material, and allows processing of the bottom of the groove 11c that is not covered with a film.

By repeating the three steps of film formation, partial film removal, and groove processing described above, the grooves 11c are gradually deepened, and ultimately the workpiece 11 can be cut along the streets 13. After cutting the workpiece 11 to obtain multiple device chips 33, an external force is applied to the resin layer 21, as in the first embodiment described above, and the resin layer 21 is divided to fit the device chips 33 (resin layer division step).

In this embodiment, a laser beam 31 is irradiated from the back surface 11b side to form a groove 11c, and then a plasma etching gas is supplied from the back surface 11b side to cut the workpiece 11. Therefore, heat is less likely to be transferred to the resin layer 21 on the front surface 11a side than when the workpiece 11 is cut using only the laser beam 31. This makes it possible to more appropriately suppress hardening of the resin layer 21. Note that the methods according to the first embodiment and its variations described above can be combined in any way with the method of this embodiment.

(Third embodiment)
In the device chip manufacturing method according to this embodiment, the workpiece 11 is cut by supplying a plasma etching gas from the back surface 11b of the workpiece 11. Before cutting the workpiece 11, a resin layer 21 is formed on the front surface 11a of the workpiece 11 (resin layer forming step), as in the first embodiment described above. Tape 23 is then applied to the front surface 11a of the workpiece 11 (resin layer 21) (tape application step), and a protective film 29 is then formed on the back surface 11b of the workpiece 11 (protective film forming step).

After forming the protective film 29 on the back surface 11b of the workpiece 11, the protective film 29 is processed to form a mask layer that covers the area corresponding to the device 15 on the back surface 11b of the workpiece 11 (mask layer formation step). Figure 12 is a cross-sectional view showing the workpiece 11 after the mask layer 35 has been formed.

The mask layer 35 in this embodiment is formed, for example, using the laser processing device 12 described above, using a procedure similar to the procedure used to form the grooves 11c in the first and second embodiments. In other words, the mask layer 35 is formed by processing the protective film 29 with a laser beam 31.

Specific procedures are the same as those used in the first embodiment described above when irradiating the workpiece 11 with the laser beam 31. However, the average output of the laser beam 31 and the number of times (number of passes) the laser beam 31 is irradiated are adjusted so that the workpiece 11 is hardly processed. Note that the back surface 11b of the workpiece 11 may be slightly processed.

This allows the protective film 29 to be cut along the streets 13, forming a mask layer 35 that covers the area corresponding to the device 15 on the back surface 11b of the workpiece 11. In this embodiment, the protective film 29 is processed into the mask layer 35, but the mask layer 35 may also be formed by processing a photosensitive resin using photolithography or the like.

After forming the mask layer 35 on the back surface 11b of the workpiece 11, plasma etching gas is supplied from the back surface 11b side to remove the portion of the workpiece 11 exposed from the mask layer 35, thereby cutting the workpiece 11 (workpiece cutting step). The equipment and procedures used are the same as those used in the second embodiment described above when plasma etching gas is supplied to the workpiece 11.

After cutting the workpiece 11 to obtain multiple device chips 33, an external force is applied to the resin layer 21, as in the first embodiment described above, and the resin layer 21 is divided to fit the device chips 33 (resin layer division step).

In this embodiment, a mask layer 35 is formed on the back surface 11b of the workpiece 11 to cover the area corresponding to the device 15. Then, plasma etching gas is supplied from the back surface 11b of the workpiece 11 on which the mask layer 35 is formed to cut the workpiece 11. Therefore, heat is less likely to be transferred to the resin layer 21 on the front surface 11a than when the workpiece 11 is cut using a laser beam 31. Therefore, hardening of the resin layer 21 can be more appropriately suppressed. The methods according to the first and second embodiments and their variations described above can be combined in any way with the method of this embodiment.

The present invention is not limited to the above-described embodiments and variations, and can be implemented in various ways. For example, in the above-described embodiments, the workpiece 11 is cut from the back surface 11b side using a laser beam or plasma etching gas, but the workpiece 11 may also be cut from the back surface 11b side using other methods. Specifically, the workpiece 11 can be cut from the back surface 11b side using an annular cutting blade in which abrasive grains are dispersed in a binder such as resin.

In addition, the structures, methods, etc. of the above-described embodiments and modifications may be implemented without departing from the scope of the present invention.

11: Workpiece 11a: Surface 11b: Back surface 11c: Groove 13: Street (planned division line)
15: Device 21: Resin layer 21a: Small piece 23: Tape 25: Frame 25a: Opening 27: Raw material 29: Protective film 31: Laser beam 33: Device chip 35: Mask layer 2: Spin coater 4: Spinner table 4a: Upper surface (holding surface)
4b: Flow path 6: Clamp 8: Nozzle 12: Laser processing device 14: Chuck table 14a: Upper surface (holding surface)
14b: Flow path 16: Clamp 18: Laser processing head 22: Plasma processing apparatus 24: Chamber 24a: Side wall 24b: Opening 24c: Bottom wall 24d: Top wall 26: Cover 28: Opening/closing mechanism 30: Piping 32: Decompression unit 34: Table base 34a: Suction path 34b: Flow path 36: Holding part 38: Support part 40: Chuck table 42: Insulation part 42a: Suction path 44: Electrode 46: DC power supply 48: Suction pump 50: Circulation unit 52: Gas supply unit 54: Supply pipe 54a: Supply port 56a: Valve 56b: Valve 56c: Valve 58a: Flow rate controller 58b: Flow rate controller 58c: Flow rate controller 60a: Valve 60b: Valve 60c: Valve 62a: Gas supply source 62b: Gas supply source 62c: Gas supply source 64: Electrode 66: High frequency power supply 68: Dispersion member 70: Pipe 72: Expansion device 74: Drum 76: Roller 78: Support member 80: Table 82: Fixing member 92: Plasma processing device 94: Chamber 94a: Opening 94b: Exhaust port 96: Cover 98: Exhaust unit 100: Lower electrode 102: High frequency power supply 104: Chuck table 106a: Electrode 106b: Electrode 108a: DC power supply 108b: DC power supply 110: Upper electrode 110a: Gas ejection hole 110b: Gas supply hole 112 : Insulating member 114 : Gas supply source 116 : High frequency power source

Claims (9)

A method for manufacturing device chips by dividing a plate-shaped workpiece, along planned dividing lines, in which devices are provided in areas on a front surface side defined by the planned dividing lines, to manufacture device chips including the devices, the method comprising:
a resin layer forming step of forming a resin layer containing an uncured or semi-cured heat-curable resin on the surface side of the workpiece;
a workpiece cutting step, after the resin layer forming step, of manufacturing the device chips by cutting the workpiece along the planned dividing lines from the back surface side of the workpiece on which the resin layer is provided on the front surface side ,
In the workpiece cutting step, a laser beam of a wavelength that is absorbed by the workpiece is shaped so that the irradiated area is linear or rectangular, and then irradiated onto the workpiece along the intended division line. This is a method for manufacturing device chips. A method for manufacturing device chips by dividing a plate-shaped workpiece, along planned dividing lines, in which devices are provided in areas on a front surface side defined by the planned dividing lines, to manufacture device chips including the devices, the method comprising:
a resin layer forming step of forming a resin layer containing an uncured or semi-cured heat-curable resin on the surface side of the workpiece;
a workpiece cutting step, after the resin layer forming step, of manufacturing the device chips by cutting the workpiece along the planned dividing lines from the back surface side of the workpiece on which the resin layer is provided on the front surface side ,
In the workpiece cutting step, a laser beam of a wavelength absorbed by the workpiece is irradiated onto the workpiece along the intended dividing line so that multiple grooves parallel to each other are formed along one of the intended dividing lines . 3. The method for manufacturing device chips according to claim 1, further comprising, after the workpiece cutting step, a resin layer dividing step of dividing the resin layer into pieces corresponding to the device chips by applying an external force to the resin layer. After the resin layer forming step and before the resin layer dividing step, a tape applying step of applying an expandable tape to the surface side of the workpiece is further included,
4. The method for manufacturing a device chip according to claim 3 , wherein in the resin layer dividing step, the resin layer is divided by applying an external force to the resin layer by expanding the tape. After the workpiece cutting step and before the resin layer dividing step, a tape applying step of applying an expandable tape to the back surface side of the workpiece is further included,
4. The method for manufacturing a device chip according to claim 3 , wherein in the resin layer dividing step, the resin layer is divided by applying an external force to the resin layer by expanding the tape. 6. The method for manufacturing a device chip according to claim 1 , further comprising a protective film forming step of forming a protective film on the back surface side of the workpiece before the workpiece cutting step. 7. A method for manufacturing a device chip according to claim 1 , wherein the workpiece cutting step cuts the workpiece by irradiating the workpiece with a laser beam having a wavelength that is absorbed by the workpiece along the intended dividing line. 8. The method for manufacturing a device chip according to claim 7 , further comprising a plasma etching step of removing processing damage or debris remaining in the device chip by supplying an etching gas in a plasma state from the back side of the workpiece after the workpiece cutting step. The workpiece cutting step includes:
a groove forming step of forming grooves that open on the back surface of the workpiece by irradiating the workpiece with a laser beam having a wavelength that is absorbed by the workpiece along the intended dividing line;
7. The method for manufacturing a device chip according to claim 1, further comprising, after the groove forming step, a plasma etching step of supplying an etching gas in a plasma state from the back side of the workpiece to remove a portion between the front surface of the workpiece and the bottom of the groove, thereby cutting the workpiece.