Substrate processing method and substrate processing apparatus

Abstract

A substrate processing method for processing a bonded substrate in which a first substrate and a second substrate are bonded, the second substrate having a laser absorption layer formed thereon, the processing method including: irradiating the laser absorption layer with pulsed laser light to form a separation modification layer and accumulate stress inside the laser absorption layer; and releasing the accumulated stress in a chain reaction to separate the second substrate.

Description

Technical Field

[0001] The present disclosure relates to a substrate processing method and a substrate processing apparatus. Background Art

[0002] Patent Document 1 discloses a method for manufacturing a semiconductor device. This method includes the following steps: a heating step in which a CO2 laser is irradiated from the back surface of a semiconductor substrate to locally heat a debonding oxide film; and a transfer step in which debonding is caused in the debonding oxide film and / or at the interface between the debonding oxide film and the semiconductor substrate, thereby transferring the semiconductor element to a transfer target substrate.

[0003] Prior art literature

[0004] Patent Literature

[0005] Patent Document 1: Japanese Patent Application Laid-Open No. 2007-220749 Summary of the Invention

[0006] Problems to be solved by the invention

[0007] The technology according to the present disclosure appropriately peels the second substrate from the first substrate in a superimposed substrate in which the first substrate and the second substrate are bonded together.

[0008] Solutions for solving problems

[0009] One method disclosed herein is a substrate processing method for processing a superimposed substrate formed by bonding a first substrate and a second substrate, wherein a laser absorption layer is formed on the second substrate, and the processing method includes: irradiating the laser absorption layer with a laser in a pulsed manner to form a peeling modification layer, and accumulating stress inside the laser absorption layer; and releasing the accumulated stress in a chain reaction to peel off the second substrate.

[0010] Effects of the Invention

[0011] According to the present disclosure, in a superposed substrate in which a first substrate and a second substrate are bonded together, the second substrate can be appropriately peeled from the first substrate. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 FIG. 1 is a side view showing an example of overlapping wafers processed by a wafer processing system.

[0013] Figure 2 It is a plan view schematically showing the outline of the structure of the wafer processing system.

[0014] Figure 3 It is a side view schematically showing the structure of the interface laser irradiation device.

[0015] Figure 4 It is a plan view schematically showing the structure of the interface laser irradiation device.

[0016] Figure 5 It is an explanatory diagram showing a state of forming a release-modifying layer according to this embodiment.

[0017] Figure 6 It is a plan view showing a formation example of the release-modifying layer according to this embodiment.

[0018] Figure 7 It is an explanatory diagram showing the flow of wafer processing according to this embodiment.

[0019] Figure 8 It is a plan view showing another formation example of the release-modifying layer according to this embodiment.

[0020] Figure 9 It is an explanatory diagram showing a state of peeling of the second wafer according to the present embodiment.

[0021] Figure 10 It is a plan view showing another formation example of the release-modifying layer according to this embodiment.

[0022] Figure 11 It is an explanatory diagram schematically showing the outline of the structure of a laser irradiation unit according to another embodiment.

[0023] Figure 12 This is an explanatory diagram showing how the frequency of laser light is changed by an acousto-optic modulator in another embodiment.

[0024] Figure 13 This is an explanatory diagram showing how the frequency of laser light is changed by an acousto-optic modulator in another embodiment.

[0025] Figure 14 It is an explanatory diagram schematically showing the outline of the structure of a laser irradiation unit according to another embodiment.

[0026] Figure 15 It is an explanatory diagram schematically showing the outline of the structure of a laser irradiation unit according to another embodiment.

[0027] Figure 16 This is an explanatory diagram showing a state in which a reforming layer formed in this embodiment is peeled off.

[0028] Figure 17 It is a plan view showing another formation example of the release-modifying layer according to this embodiment.

[0029] Figure 18 It is an explanatory diagram showing another example of separation of the second wafer according to the present embodiment.

[0030] Figure 19 It is an explanatory diagram showing a state of pressing the second wafer.

[0031] Figure 20 It is an explanatory diagram showing a state of pressing the second wafer.

[0032] Figure 21 This is a side view schematically showing the structure of stacked wafers in another embodiment.

[0033] Figure 22 It is an explanatory diagram showing the flow of edge trimming processing according to this embodiment. DETAILED DESCRIPTION

[0034] In recent years, laser lift-off (LAL) has been used in LED manufacturing processes to remove GaN (gallium nitride) compound crystal layers (material layers) from sapphire substrates using lasers. In this LLA process, sapphire substrates are transparent to short-wavelength lasers (such as UV light), allowing the use of short-wavelength lasers with high absorption rates in the absorption layer, and a wide range of laser options are available.

[0035] On the other hand, in the semiconductor device manufacturing process, a device layer formed on the surface of one substrate (a semiconductor, such as a silicon substrate) is transferred to another substrate. Generally speaking, silicon substrates are transparent to lasers in the NIR (near-infrared) region, but the absorption layer is also transparent to NIR lasers, which can damage the device layer. Therefore, in the semiconductor device manufacturing process, laser lift-off is performed using lasers in the FIR (far-infrared) region.

[0036] Generally, a laser with an FIR wavelength can be used, for example, a CO2 laser. In the method described in Patent Document 1, a CO2 laser is irradiated onto a release oxide film serving as an absorption layer to cause release at the interface between the release oxide film and the substrate.

[0037] The inventors of the present invention, after careful research, discovered that simply irradiating the absorption layer with a laser (CO2 laser) sometimes prevents separation between the substrate and the release oxide film (device layer), effectively preventing proper transfer. Specifically, they discovered that the primary cause of separation is not the amount of laser energy, but rather the peak power of the laser (the maximum intensity of the irradiated laser). For example, the peak power can be increased by reducing the laser frequency.

[0038] As described above, in order to produce peeling between the substrate and the absorption layer (device layer) by irradiating a laser, as in the method described in Patent Document 1, it is necessary to increase the peak power by, for example, reducing the frequency of the laser irradiating the absorption layer. However, when the frequency of the laser is reduced in this way, for example, the time required to peel the substrate and the absorption layer over the entire surface increases, and the productivity related to the transfer of the device layer decreases. Moreover, in the method of Patent Document 1, the frequency of the laser is not considered at all, and there is no suggestion related to it. Therefore, there is room for improvement in the conventional device layer transfer method.

[0039] The technology disclosed herein appropriately separates a second substrate from a first substrate in a superimposed substrate formed by bonding a first substrate and a second substrate. The following describes a wafer processing system as a substrate processing apparatus and a wafer processing method as a substrate processing method according to this embodiment with reference to the accompanying drawings. Elements having substantially the same functional structure are denoted by the same reference numerals in this specification and the accompanying drawings to omit repeated descriptions.

[0040] like Figure 1 As shown, the superposed wafer T, which serves as a superposed substrate and is processed in the wafer processing according to this embodiment, is formed by bonding a first wafer W1 serving as a first substrate to a second wafer W2 serving as a second substrate. Hereinafter, the surface of the first wafer W1 that is bonded to the second wafer W2 is referred to as the front surface W1a, and the surface opposite to the front surface W1a is referred to as the back surface W1b. Similarly, the surface of the second wafer W2 that is bonded to the first wafer W1 is referred to as the front surface W2a, and the surface opposite to the front surface W2a is referred to as the back surface W2b.

[0041] The first wafer W1 is, for example, a semiconductor wafer such as a silicon substrate. A device layer D1 including a plurality of devices is formed on the surface W1a of the first wafer W1. A surface film F1 is also formed on the device layer D1, through which the device layer D1 is bonded to the second wafer W2. Examples of the surface film F1 include an oxide film (SiO2 film, TEOS film), a SiC film, a SiCN film, or an adhesive. In some cases, the device layer D1 and the surface film F1 are not formed on the surface W1a.

[0042] The second wafer W2 is also a semiconductor wafer such as a silicon substrate. On the surface W2a of the second wafer W2, a laser absorption layer P, a device layer D2, and a surface film F2 are stacked in the order of the laser absorption layer P, the device layer D2, and the surface film F2 from the surface W2a side, and are bonded to the first wafer W1 via the surface film F2. The device layer D2 and the surface film F2 are respectively the same as the device layer D1 and the surface film F1 of the first wafer W1. As the laser absorption layer P, there are listed, for example, oxide films (SiO2 films, TEOS films) that can absorb laser light (for example, CO2 lasers) as described later. In addition, sometimes the laser absorption layer P, the device layer D2, and the surface film F2 are not formed on the surface W2a. In this case, the laser absorption layer P is formed on the surface W1a of the first wafer W1 on which the device layer D1 and the surface film F1 are formed, and the device layer D1 is transferred to the second wafer W2 side.

[0043] The peripheral portion We of the second wafer W2 is chamfered, and the thickness of the cross section of the peripheral portion We decreases as it goes toward its front end. In the manufacturing process of semiconductor devices, the back side of the second wafer W2 formed in this way is sometimes removed to thin it. During this thinning process, the peripheral portion We may become a sharp shape (a so-called blade shape). As a result, fragments are generated in the peripheral portion We of the second wafer W2, and the second wafer W2 may be damaged. Therefore, the edge trimming described later is sometimes performed in advance to remove the peripheral portion We of the second wafer W2 before the thinning process. The peripheral portion We is the portion that is removed during this edge trimming, for example, a radial range of 0.5 mm to 3 mm from the outer end of the second wafer W2.

[0044] In the wafer processing system 1 described later in this embodiment, the aforementioned laser peeling process is performed as wafer processing, that is, the transfer process of transferring the device layer D2 to the side of the first wafer W1, or the aforementioned edge trimming process is performed as wafer processing, that is, the removal process of removing the peripheral portion We of the second wafer W2.

[0045] like Figure 2 As shown, the wafer processing system 1 has a structure in which the loading / unloading block G1, the transfer block G2, and the processing block G3 are connected as a whole. The loading / unloading block G1, the transfer block G2, and the processing block G3 are arranged in the order shown from the negative side in the X-axis direction.

[0046] The loading / unloading block G1, for example, loads and unloads cassettes Ct, Cw1, and Cw2, each capable of accommodating a plurality of overlapping wafers T, a plurality of first wafers W1, and a plurality of second wafers W2, to and from the outside. A cassette mounting table 10 is provided in the loading / unloading block G1. In the illustrated example, a plurality of, for example, three, cassettes Ct, Cw1, and Cw2 are freely mounted in a row along the Y-axis on the cassette mounting table 10. The number of cassettes Ct, Cw1, and Cw2 mounted on the cassette mounting table 10 is not limited to this embodiment and can be arbitrarily determined.

[0047] A wafer conveying device 20 is provided adjacent to the box loading platform 10 on the X-axis positive direction side of the conveying block G2. The wafer conveying device 20 is configured to be movable on a conveying path 21 extending in the Y-axis direction. In addition, the wafer conveying device 20 has, for example, two conveying arms 22, 22 for holding and conveying the overlapping wafers T, the first wafer W1, and the second wafer W2. Each conveying arm 22 is configured to be movable in the horizontal direction and the vertical direction and to be movable around the horizontal axis and the vertical axis. In addition, the structure of the conveying arm 22 is not limited to this embodiment, and any structure can be adopted. Moreover, the wafer conveying device 20 is configured to be able to convey the overlapping wafers T, the first wafer W1, and the second wafer W2 to the boxes Ct, Cw1, Cw2 of the box loading platform 10 and the conveying device 30 described later.

[0048] A transfer device 30 for transferring and superimposing the wafer T, the first wafer W1 , and the second wafer W2 is provided adjacent to the wafer transfer device 20 in the transfer block G2 on the X-axis positive side of the wafer transfer device 20 .

[0049] The processing block G3 includes a wafer transfer device 40 , a peripheral edge removal device 50 , a cleaning device 60 , an internal laser irradiation device 70 , and an interface laser irradiation device 80 .

[0050] The wafer conveying device 40 is configured to be movable on a conveying path 41 extending in the X-axis direction. In addition, the wafer conveying device 40 has, for example, two conveying arms 42, 42 for holding and conveying the overlapping wafer T, the first wafer W1, and the second wafer W2. Each conveying arm 42 is configured to be movable in the horizontal direction and the vertical direction and to be movable around the horizontal axis and the vertical axis. In addition, the structure of the conveying arm 42 is not limited to this embodiment, and any structure can be adopted. Moreover, the wafer conveying device 40 is configured to be able to convey the overlapping wafer T, the first wafer W1, and the second wafer W2 to the conveying device 30, the peripheral edge removal device 50, the cleaning device 60, the internal laser irradiation device 70, and the interface laser irradiation device 80.

[0051] The peripheral edge removal device 50 is arranged on the Y-axis positive direction side of the wafer conveying device 40, and is used to remove the peripheral edge portion We of the second wafer W2, that is, to perform edge trimming processing. The cleaning device 60 is arranged on the Y-axis negative direction side of the wafer conveying device 40, and is used to clean the overlapping wafer T after peeling or removing the peripheral edge portion We. The internal laser irradiation device 70 as the second laser irradiation unit is arranged on the Y-axis positive direction side of the wafer conveying device 40, and is used to irradiate the interior of the second wafer W2 with laser (internal laser, such as YAG laser) to form the peripheral edge modification layer M2 described later as the base point for peeling the peripheral edge portion We. The interface laser irradiation device 80 is arranged on the Y-axis negative direction side of the wafer conveying device 40, and is used to irradiate the laser absorption layer P formed on the surface W2a of the second wafer W2 with laser (interface laser, such as CO2 laser). In addition, the structure of the interface laser irradiation device 80 will be described later.

[0052] The above wafer processing system 1 is provided with a control device 90 as a control unit. The control device 90 is, for example, a computer, and has a program storage unit (not shown). The program storage unit stores a program for controlling the processing of the overlapping wafers T in the wafer processing system 1. In addition, the program storage unit also stores a program for controlling the operation of the drive system of the various processing devices, conveying devices, etc. mentioned above to realize the wafer processing described later in the wafer processing system 1. In addition, the above program can be recorded in a storage medium H that can be read by a computer, and installed in the control device 90 from the storage medium H.

[0053] The wafer processing system 1 is configured as described above. The wafer processing system 1 can perform the laser lift-off process described above on the overlapped wafers T, i.e., the transfer process of the device layer D2 to the first wafer W1, and the edge trimming process described above on the second wafer W2. Furthermore, if the wafer processing system 1 does not perform the edge trimming process on the second wafer W2, for example, the peripheral edge removal device 50 and the internal laser irradiation device 70 can be omitted.

[0054] Next, the above-mentioned interface laser irradiation device 80 will be described.

[0055] like Figure 3 and Figure 4 As shown, the interface laser irradiation device 80 includes a suction cup 100 that holds the overlapped wafer T by its upper surface. The suction cup 100 suction-holds a portion or the entire back surface W1b of the first wafer W1. The suction cup 100 is provided with lift pins (not shown) for transferring the overlapped wafer T to and from the transfer arm 42. The lift pins are configured to pass through through holes (not shown) formed through the suction cup 100 and are movable upward and downward. The lift pins support the overlapped wafer T from below and allow it to be raised and lowered.

[0056] The chuck 100 is supported on a slide table 102 via an air bearing 101. A rotation mechanism 103 is provided on the lower surface side of the slide table 102. The rotation mechanism 103 is provided, for example, with a motor as a drive source. The chuck 100 is configured to be rotatable about a θ axis (a vertical axis) via the air bearing 101 by the rotation mechanism 103. The slide table 102 is configured to be freely movable along a guide rail 105 provided on a base 106 and extending in the Y axis direction, by a moving mechanism 104 provided on the lower surface side thereof. The drive source of the moving mechanism 104 is not particularly limited, and a linear motor is used, for example.

[0057] A laser irradiation unit 110 is provided above the chuck 100. The laser irradiation unit 110 has a laser head 111, an optical system 112, and a lens 113. The laser head 111 oscillates pulsed laser light. The optical system 112 controls the intensity and position of the laser light, or attenuates the laser light to adjust the output. The lens 113 is a cylindrical member for irradiating the laser light to the bonded wafer T held on the chuck 100. In the present embodiment, the laser light is pulsed CO2 laser light, and the laser light emitted from the laser irradiation unit 110 passes through the second wafer W2 and is irradiated to the laser absorption layer P. The wavelength of the CO2 laser light is, for example, 8.9 μm to 11 μm. The lens 113 is configured to be freely raised and lowered by a raising and lowering mechanism (not shown).

[0058] Further, a transport pad 120 having a suction surface for suction-holding the back surface W2b of the second wafer W2 on the lower surface thereof is provided above the chuck 100. The transport pad 120 is configured to be freely raised and lowered by a raising and lowering mechanism (not shown). The transport pad 120 transports the second wafer W2 between the chuck 100 and the transport arm 42. Specifically, after the chuck 100 is moved to below the transport pad 120 (a handover position at which the transport arm 42 is to be handed over), the transport pad 120 is lowered to suction-hold the back surface W2b of the second wafer W2, and then the transport pad 120 is raised again to peel the second wafer W2 from the first wafer W1. The peeled second wafer W2 is handed over from the transport pad 120 to the transport arm 42, and is carried out from the interface laser irradiation device 80. The transport pad 120 can be configured to flip the front and back surfaces of the wafer by a flipping mechanism (not shown).

[0059] Next, wafer processing using the wafer processing system 1 configured as described above will be described. In the following description, a case in which laser peeling processing is performed in the wafer processing system 1, that is, a case in which the device layer D2 of the second wafer W2 is transferred to the first wafer W1, will be described. In the present embodiment, the first wafer W1 and the second wafer W2 are bonded in a bonding device (not shown) outside the wafer processing system 1 to form the bonded wafer T.

[0060] First, a cassette Ct containing a plurality of overlapped wafers T is placed on the cassette stage 10 of the loading / unloading block G1. Next, the overlapped wafers T in the cassette Ct are removed by the wafer transfer device 20. The overlapped wafers T removed from the cassette Ct are transferred to the wafer transfer device 40 via the conveyor device 30 and then transported to the interface laser irradiation device 80. In the interface laser irradiation device 80, the second wafer W2 is separated from the first wafer W1 (laser trimming).

[0061] Specifically, the transfer arm 42 and the superposed wafer T held by suction on the chuck 100 via lift pins are first moved to a processing position by the moving mechanism 104. The processing position is a position where the laser irradiation unit 110 can irradiate the superposed wafer T (laser absorption layer P) with laser light.

[0062] Then, if Figure 5 and Figure 6 As shown, the laser L (CO2 laser) is irradiated in a pulsed manner from the laser irradiation unit 110 toward the back surface W2b of the second wafer W2. At this time, the laser L passes through the second wafer W2 from the back surface W2b side of the second wafer W2 and is absorbed by the laser absorption layer P. Moreover, stress is generated inside the laser absorption layer P that has absorbed the laser L. Hereinafter, the stress accumulation layer formed by laser irradiation in this way, which serves as the base point for peeling of the second wafer W2 (the base point for transferring the device layer D2), is sometimes referred to as a "peeling modification layer M1". In addition, by forming the peeling modification layer M1, almost all the energy of the laser L irradiated on the laser absorption layer P is absorbed and does not reach the device layer D2. Therefore, damage to the device layer D2 can be suppressed.

[0063] Here, the output of the laser light L irradiated on the laser absorption layer P is controlled so as not to cause separation between the second wafer W2 and the laser absorption layer P due to irradiation with the laser light L. In other words, for example, by increasing the frequency of the laser light L to reduce the peak power, the separation modification layer M1 is formed by irradiation with the laser light L in a manner that does not cause separation between the second wafer W2 and the laser absorption layer P.

[0064] In this manner, by eliminating the space for releasing the generated stress in a manner that prevents separation between the second wafer W2 and the laser absorption layer P due to irradiation with laser light L, the generated stress is accumulated within the laser absorption layer P, thereby forming a peeling modification layer M1. More specifically, for example, by gasifying the laser absorption layer P through irradiation with laser light, the space for releasing the generated gas is eliminated as described above, thereby accumulating compressive stress in the peeling modification layer M1. Furthermore, for example, by absorbing laser light, the laser absorption layer P generates heat, and shear stress is accumulated in the peeling modification layer M1 due to the difference in thermal expansion coefficient between the laser absorption layer P and the second wafer W2 or the device layer D2. Furthermore, by accumulating the stress generated by laser irradiation in a manner that prevents separation between the second wafer W2 and the laser absorption layer P, the bonding strength between the laser absorption layer P and the second wafer W2 at the location where the peeling modification layer M1 is formed is reduced.

[0065] When the laser absorption layer P is irradiated with the laser light L, the chuck 100 (overlapping wafer T) is rotated by the rotation mechanism 103, and the chuck 100 is moved in the Y-axis direction by the movement mechanism 104. Thus, the laser light L is irradiated from the radial inside toward the outside to the laser absorption layer P, resulting in the laser light L being irradiated in a spiral shape from the inside toward the outside. Figure 6 The black arrow shown indicates the rotation direction of the suction cup 100 .

[0066] Here, the interval between adjacent peeling and modifying layers M1, in other words, the pulse interval (frequency) of the laser light L is controlled to be an interval at which the adjacent peeling and modifying layers M1 will not peel off due to the impact generated when the peeling and modifying layers M1 are formed. Specifically, for example, Figure 7 As shown in (a), it is preferable that adjacent release-modifying layers M1 are formed so as not to overlap each other in a plan view. In addition, in this case, it is preferable that adjacent release-modifying layers M1 are formed so as to be close to each other.

[0067] In addition, if Figure 8 As shown, the laser light L may be irradiated concentrically and annularly in the laser absorption layer P. However, in this case, since the rotation of the chuck 100 and the movement of the chuck 100 in the Y direction are performed alternately, the irradiation time can be shortened by irradiating the laser light L in a spiral shape as described above, thereby improving productivity.

[0068] In this embodiment, the chuck 100 is rotated when irradiating the laser light L onto the laser absorption layer P. However, the lens 113 may be moved to rotate relative to the chuck 100. Furthermore, the chuck 100 is moved in the Y-axis direction, but the lens 113 may also be moved in the Y-axis direction. Furthermore, the direction in which the release-modifying layer M1 is formed is not limited to from the radially inner side to the outer side of the laser absorption layer P; it may also be formed from the radially outer side to the inner side.

[0069] When a plurality of peeling reforming layers M1 are continuously formed in this manner, as shown in FIG. Figure 7 As shown in (a), regions (hereinafter referred to as "non-peeled regions R1") in which the peeling modified layers M1 are formed so as to prevent peeling between the second wafer W2 and the laser absorption layer P are not formed are formed sequentially from the radially inner side toward the outer side of the laser absorption layer P. As described above, stress generated during the formation of each peeling modified layer M1 accumulates in the non-peeled regions R1.

[0070] When the unpeeled region R1 is continuously formed, as shown in FIG. Figure 7 As shown in (b), the formation position of the peeling modified layer M1 reaches near the end of the second wafer W2, in other words, reaches the boundary Ad between the bonding area Ac where the first wafer W1 and the second wafer W2 are bonded and the unbonded area Ae radially outside the bonding area Ac. In addition, here, the boundary Ad can be, for example, a bonding end formed by bonding the first wafer W1 and the second wafer W2, or can be intentionally formed by removing the bonding interface between the first wafer W1 and the second wafer W2. That is, the unbonded area Ae is an area radially outside the boundary Ad, and can be, for example, an area where the bonding strength between the first wafer W1 and the second wafer W2 is intentionally eliminated by removing the bonding interface, or can be, for example, only an area radially outside the bonding area Ac where the first wafer W1 and the second wafer W2 are actually bonded.

[0071] When the formation position of the peeling modified layer M1 reaches the boundary Ad, the stress accumulated as the peeling modified layer M1 is released to the formation space of the unbonded region Ae, that is, outside the superimposed wafer T. When the accumulated stress is released, Figure 7 As shown in (b), at the formation position of the peeling modification layer M1 formed near the boundary Ad, a force acts in the thickness direction of the laser absorption layer P, that is, the peeling direction of the laser absorption layer P and the second wafer W2, resulting in peeling of the laser absorption layer P and the second wafer W2.

[0072] Next, when the laser absorption layer P and the second wafer W2 are peeled off near the boundary Ad, due to the influence of the force acting on the laser absorption layer P in the thickness direction due to this peeling, the peeling between the laser absorption layer P and the second wafer W2 progresses radially inward of the laser absorption layer P. Furthermore, the peeling progressing radially inward reaches the adjacent peeling reformed layer M1. In other words, the peeling between the laser absorption layer P and the second wafer W2 occurs at the position where the adjacent peeling reformed layer M1 is formed.

[0073] When delamination occurs at the location where the adjacent delamination reforming layer M1 is formed, the stress accumulated in the delamination reforming layer M1 is released. Consequently, a force acts in the thickness direction of the laser absorption layer P at the location where the delamination reforming layer M1 is formed, and the delamination between the laser absorption layer P and the second wafer W2 progresses further radially inward.

[0074] Furthermore, by repeating the peeling of the laser absorption layer P and the second wafer W2, the stress release, and the peeling progressing toward the radial inner side in this chain, as shown in FIG. Figure 7 As shown in (c), the peeling regions R2 are sequentially formed from the radially outer side toward the inner side of the laser absorption layer P. Furthermore, by peeling the entire surface of the second wafer W2 from the laser absorption layer P (first wafer W1), the device layer D2 of the second wafer W2 is transferred to the first wafer W1 side.

[0075] According to this embodiment, during the formation of the unpeeled region R1, that is, the continuous formation of the peeling modified layer M1, the peak power (frequency) of the laser L is controlled so that the laser absorption layer P and the second wafer W2 do not peel off. Moreover, the laser absorption layer P and the second wafer W2 form a peeling modified layer M1 (hereinafter referred to as the "starting point modified layer M1s") as the starting point of peeling near the boundary Ad, so that the peeling is naturally carried out with the peeling modified layer M1 of the unpeeled region R1 as the base point. As a result, there is no need to reduce the frequency of the laser L during the peeling of the laser absorption layer P and the second wafer W2, so that the time spent on the transfer process of the device layer D2 can be shortened, that is, the decline in productivity can be suppressed. In addition, as described above, there is no need to increase the peak power of the laser L, so the energy efficiency spent on the transfer process of the device layer D2 can be improved.

[0076] When the entire surface of the second wafer W2 is peeled off from the laser absorption layer P, the suction cup 100 is then moved to the handover position by the moving mechanism 104. Figure 9 As shown in (a) of FIG. 1 , the back side W2b of the second wafer W2 is adsorbed and held by the transfer pad 120, and then, as shown in FIG. Figure 9 As shown in (b), the transfer pad 120 is raised to peel the second wafer W2 from the laser absorption layer P (first wafer W1). At this time, as described above, peeling occurs at the interface between the laser absorption layer P and the second wafer W2, so the second wafer W2 can be peeled from the laser absorption layer P without applying a large load.

[0077] The peeled second wafer W2 is transferred from the transfer pad 120 to the transfer arm 42 of the wafer transfer device 40 and transferred to the cassette Cw2 on the cassette stage 10. Furthermore, before the second wafer W2 unloaded from the interface laser irradiation device 80 is transferred to the cassette Cw2, the surface W2a of the second wafer W2 may be cleaned in the cleaning device 60.

[0078] Meanwhile, the first wafer W1 held by the suction cup 100 is transferred to the transfer arm 42 of the wafer transfer device 40 via the lift pins and transferred to the cleaning device 60. The surface of the laser absorption layer P, which serves as the peeling surface, is scrubbed in the cleaning device 60. Furthermore, the back surface W1b of the first wafer W1 can be cleaned together with the surface of the laser absorption layer P in the cleaning device 60.

[0079] After that, the first wafer W1 subjected to all the processes related to transferring the device layer D2 onto the first wafer W1 is transported by the wafer transport device 20 to the cassette Cw1 on the cassette stage 10 via the conveyor device 30. This completes a series of wafer processes in the wafer processing system 1.

[0080] According to the above embodiment, the output of the laser light L emitted by the interface laser irradiation device 80 is controlled to a peak power that prevents delamination between the laser absorption layer P and the second wafer W2. In other words, there is no need to reduce the frequency of the laser light L during delamination of the laser absorption layer P and the second wafer W2, thereby suppressing a decrease in the productivity of transferring the device layer D2 to the first wafer W1. Furthermore, even when the peak power of the laser light L is reduced in this manner, stress accumulated by the formation of the delamination-modifying layer M1 can be released, allowing for appropriate delamination of the laser absorption layer P and the second wafer W2.

[0081] In the above embodiment, the peeling reforming layer M1 is formed near the boundary Ad as the starting point reforming layer M1s to release stress and start the chain peeling of the laser absorption layer P and the second wafer W2. However, the peeling start method is not limited to this.

[0082] Specifically, for example, a starting point modification layer M1s serving as a base point for peeling can be formed outside the non-peeled region R1 formed on the laser absorption layer P, thereby starting the chain peeling of the laser absorption layer P and the second wafer W2. At this time, the starting point modification layer M1s is formed by irradiating the laser L with a high peak power (low frequency) that causes peeling of the laser absorption layer P and the second wafer W2. In this way, peeling is caused by irradiating the laser L, thereby releasing the compressive stress, and then the chain peeling of the laser absorption layer P and the second wafer W2 is performed. Moreover, even in the case where the frequency is lowered to increase the peak power when forming the starting point modification layer M1s, the non-peeled region R1 can be formed by the same method as the above-mentioned embodiment, thereby suppressing the decrease in the productivity of transferring the device layer D2 to the first wafer W1.

[0083] Furthermore, in this case, the starting point modified layer M1s can be formed before forming the unpeeled region R1. Specifically, by forming the starting point modified layer M1s, the laser absorption layer P and the second wafer W2 are pre-peeled. Consequently, the compressive stress in the starting point modified layer M1s can be released by passing from the position where the peeled modified layer M1, which is the unpeeled region R1, to the position where the starting point modified layer M1s is formed. This allows the chain peeling process to begin.

[0084] Here, in order to uniformly perform the peeling of the laser absorption layer P and the second wafer W2 within the surface, it is preferable to keep the interval of the irradiated laser L, that is, the interval of the pulses, fixed. However, when the suction cup 100 (overlapping wafer T) is rotated during the irradiation of the laser L as described above, the relative rotation speed of the suction cup 100 with respect to the laser irradiation unit 110 (lens 113) is greater on the radial inner side than on the radial outer side. That is, even when the rotation speed of the suction cup 100 is fixed, if the irradiation position of the laser L is radially inner, the interval of the laser L becomes smaller, and sometimes the laser L overlaps in the center of the laser absorption layer P. Moreover, when the laser L overlaps like this, there is a risk that the peeling of the laser absorption layer P and the second wafer W2 cannot be properly performed in the center, or that leakage of the laser L occurs, which affects the device layer D2.

[0085] Therefore, in this embodiment, when the second wafer W2 and the laser absorption layer P are peeled off, Figure 10 As shown, the formation of the peeling modification layer M1 can be omitted in the center portion Pc of the laser absorption layer P, within the range formed by the natural extension of the delamination region R2 due to stress release. Even when the peeling modification layer M1 is not formed in the center portion of the laser absorption layer P, the peeling can be extended by the action of delamination (stress release) that develops from the radially outer side, thereby peeling the laser absorption layer P from the second wafer W2 in the center portion.

[0086] Furthermore, in this embodiment, the frequency of laser light L is reduced radially inward, where the relative rotational speed of the chuck 100 relative to the laser irradiation unit 110 (lens 113) increases, and the frequency of laser light L is increased radially outward, thereby controlling the relative irradiation interval of laser light L onto the laser absorption layer P to be substantially constant. However, when the frequency is varied in this manner, the pulse waveform of laser light L also changes when the frequency of laser light L is changed in the laser oscillator of the laser head 111. Consequently, complex adjustments are required that take into account the output and pulse waveform of laser light L, making process control of the laser processing difficult.

[0087] Therefore, in this embodiment, an acousto-optic modulator is used to control the frequency of the laser light L. As described above, the laser irradiation unit 110 includes the laser head 111 , the optical system 112 , and the lens 113 .

[0088] As Figure 11 shown, the laser head 111 has a laser oscillator 130 that oscillates a pulsed laser. The frequency of the laser oscillated from the laser oscillator 130 is the highest frequency that can be controlled by a later-described acousto-optic modulator 131. In addition, the laser head 111 can have a device other than the laser oscillator 130, such as an amplifier or the like.

[0089] The optical system 112 has the acousto-optic modulator (AOM) 131 that turns the laser from the laser oscillator 130 in a different direction as an optical element, and an attenuator 132 that attenuates the laser from the laser oscillator 130 to adjust the output of the laser as an attenuator. The acousto-optic modulator 131 and the attenuator 132 are provided in this order from the laser oscillator 130 side.

[0090] The acousto-optic modulator 131 is an optical modulator that electrically controls the intensity and position of the laser at high speed. As Figure 12 shown, when the laser LI from the laser oscillator 130 is incident on the acousto-optic modulator 131, a voltage is applied to change the refractive index of the laser LI, thereby turning the laser LI in a different direction. Specifically, by adjusting the voltage, the angle of change of the laser LI can be controlled. In the present embodiment, for example, the laser LI is turned in two different directions, one direction of the laser L2 is irradiated on the laser absorption layer P, and the other direction of the laser L3 is not irradiated on the laser absorption layer P. By controlling the turning of the lasers L2, L3, the frequency of the laser L2 irradiated on the laser absorption layer P can be adjusted.

[0091] In this case, by using the acousto-optic modulator 131 to interval-eliminate the pulses of the laser LI, the frequency of the laser L2 irradiated on the laser absorption layer P can be adjusted. For example, at a certain timing, if the turning rate of the laser L2 and the laser L3 with respect to the laser LI is 100:0, the laser LI directly becomes the laser L2 that is irradiated on the laser absorption layer P. On the other hand, at another timing, if the turning rate of the laser L2 and the laser L3 with respect to the laser LI is 0:100, the laser L2 is 0 (zero), and the laser L2 is not irradiated on the laser absorption layer P. In this case, the frequency of the laser L2 shown in (b) after being turned by the acousto-optic modulator 131 can be adjusted with respect to the frequency of the laser LI from the laser oscillator 130 shown in (a). Figure 13 Figure 13 In this case, by using the acousto-optic modulator 131 to interval-eliminate the pulses of the laser LI, the frequency of the laser L2 irradiated on the laser absorption layer P can be adjusted. For example, at a certain timing, if the turning rate of the laser L2 and the laser L3 with respect to the laser LI is 100:0, the laser LI directly becomes the laser L2 that is irradiated on the laser absorption layer P. On the other hand, at another timing, if the turning rate of the laser L2 and the laser L3 with respect to the laser LI is 0:100, the laser L2 is 0 (zero), and the laser L2 is not irradiated on the laser absorption layer P. In this case, the frequency of the laser L2 shown in (b) after being turned by the acousto-optic modulator 131 can be adjusted with respect to the frequency of the laser LI from the laser oscillator 130 shown in (a). Figure 13 The horizontal axis of (c) indicates time, and the vertical axis indicates the intensity of the laser L2. That is,​ Figure 13 The density in the graph indicates the frequency of the laser L2.

[0092] Also, in this case, the frequency of the laser L1 oscillated from the laser oscillator 130 is not changed, and thus the pulse waveform of the laser L1 is not changed, and the pulse waveform of the laser L2 can be made the same as that of the laser L1. Thus, the frequency of the laser L2 can be easily adjusted, and the conventional complicated adjustment as described above is not needed, and the process control of the laser processing can be easily performed.

[0093] Further, in the present embodiment, the acousto-optic modulator 131 is used as the optical element, but is not limited thereto. For example, an electro-optic modulator (EOM) can be used as the optical element. In addition, an optical deflector such as an acousto-optic deflector (AOD) or an electro-optic deflector (EOD) can be used.

[0094] Next, the control method of the laser L2 when the laser L2 is irradiated to the laser absorption layer P from the laser irradiation section 110 will be described. As described above, in the case where the irradiation position of the laser L2 is on the outer side in the radial direction of the laser absorption layer P, the frequency is increased, and in the case where the irradiation position of the laser L2 is on the inner side, the frequency is decreased.

[0095] Next, a specific example will be described. Further, the values in the specific example are one example, and the present disclosure is not limited to the values. For example, the energy required for peeling on the outer side and the inner side in the radial direction of the laser absorption layer P is set to 400 μJ. The required frequency of the laser L2 on the outer side in the radial direction of the laser absorption layer P is set to 100 kHz, and the required frequency of the laser on the inner side is set to 50 kHz. The frequency of the laser L1 from the laser oscillator 130 is set to 100 kHz, and the output is set to 40 W.

[0096] In this case, on the outer side in the radial direction of the laser absorption layer P, the pulses of the laser L1 from the laser oscillator 130 are not interval-eliminated in the acousto-optic modulator 131. Thus, the frequency of the laser L2 irradiated to the laser absorption layer P can be the same as that of the laser L1, that is, 100 kHz. In addition, the output of the laser L2 is also the same as that of the laser L1, that is, 40 W. Also, the energy of the laser L2 is 400 μJ (= 40 W / 100 kHz), and peeling can be appropriately performed.

[0097] Meanwhile, on the radially inner side of the laser absorption layer P, the acousto-optic modulator 131 thins out half of the pulses of laser light L1 from the laser oscillator 130. This allows the frequency of laser light L2 irradiating the laser absorption layer P to be set to 50 kHz, half the frequency of laser light L1. Furthermore, this thinning out of laser light L1 reduces the output of laser light L2 to 20 W, half the output of laser light L1. Furthermore, the energy of laser light L2 is 400 μJ (= 20 W / 50 kHz), enabling appropriate ablation.

[0098] In this manner, the rotational speed of chuck 100 is controlled based on the frequency and irradiation position of laser light L2 to maintain a constant pulse interval. Furthermore, the maximum rotational speed of chuck 100 is maintained at the center of laser absorption layer P, and acousto-optic modulator 131 adjusts the frequency of laser light L2 based on this maximum rotational speed. This allows laser processing to be performed while maintaining the maximum possible rotational speed of chuck 100 and the high frequency of laser light L2, thereby achieving high-productivity laser processing.

[0099] Furthermore, in this case, the frequency of laser light L1 from laser oscillator 130 remains unchanged, so the pulse waveform of laser light L1 remains unchanged, and the pulse waveform of laser light L2 can be made identical to that of laser light L1. This makes it easy to adjust the frequency of laser light L2, enabling continuous and smooth processing. As a result, process control of the laser processing becomes easier, enabling a stable process.

[0100] In this embodiment, the output of laser light L1 from laser oscillator 130 is 40 W, so there is no need to adjust the output relative to the energy of 400 μJ required for ablation. For example, if the output of laser light L1 is 50 W, the output of laser light L1 can be adjusted by attenuating it by 20% using attenuator 132.

[0101] In the laser irradiation unit 110 of the above embodiment, the acousto-optic modulator 131 is provided on the upstream side of the attenuator 132 inside the optical system 112, but the installation position is not limited to this. Figure 14 As shown, the acousto-optic modulator 131 may also be provided on the downstream side of the attenuator 132 within the optical system 112. Alternatively, for example, Figure 15 As shown, the AOM 131 is provided inside the laser head 111 on the downstream side of the laser oscillator 130. Alternatively, the AOM 131 may be provided at two or more of the above-mentioned locations.

[0102] Furthermore, in the laser irradiation section 110, after the frequency and output of the laser light L2 are adjusted by the acousto-optic modulator 131, the output can be fine-tuned by the attenuator 132. Here, the output of the laser light L1 oscillated from the laser oscillator 130 sometimes deviates due to individual differences of the laser oscillator 130. Such output deviations can be adjusted in the attenuator 132. In addition, while the output of the laser light L1 from the laser oscillator 130 is monitored at all times, the attenuator 132 can be feedback-controlled to adjust the output. Moreover, from the viewpoint of fine-tuning the output of the laser light L2 by the attenuator 132 as described above, it is preferable that the acousto-optic modulator 131 be as Figure 11 As shown, it is provided on the upstream side of the attenuator 132.

[0103] In the laser irradiation unit 110 of the above embodiment, the attenuator 132 can be omitted. For example, the output of laser light L2 can be adjusted using the acousto-optic modulator 131 instead of the attenuator 132. For example, if the output of laser light L1 is 50W and the output of laser light L2 required for ablation is 40W, the output of laser light L2 can be adjusted to 40W by setting the deflection ratio of laser light L2 and laser light L3 relative to laser light L1 to 80:20 in the acousto-optic modulator 131.

[0104] Furthermore, in the above embodiment, a peeling modification layer M1 is formed on the laser absorption layer P to reduce the bonding strength between the second wafer W2 and the laser absorption layer P, and the second wafer W2 and the laser absorption layer P are peeled off based on the peeling modification layer M1. However, for example, Figure 16 As shown in (a), when the laser is not irradiated in the surface of the laser absorption layer P and a region where the bonding strength is not reduced (hereinafter referred to as "unformed region R3") is formed, the second wafer W2 and the laser absorption layer P may not be properly peeled off. Specifically, for example, Figure 16 As shown in FIG. 5 (b), in the non-formed region R3 where the bonding strength is not reduced, a portion of the second wafer W2 (silicon piece) may remain on the surface of the laser absorption layer P after the peeling.

[0105] Therefore, in this embodiment, it is desirable to form the peeling reforming layer M1 (non-peeling region R1) to reduce the area of ​​the non-formed region R3 within the surface of the laser absorption layer P. Specifically, for example, Figure 17 By controlling the formation position of the peeling modification layer M1 as shown in (a) of FIG. 1 and increasing the number of other peeling modification layers M1 adjacent to one peeling modification layer M1, the unformed region R3 can be reduced. Figure 17As shown in (b), the laser irradiation pattern for the laser absorption layer P is controlled to reduce the unformed region R3. Specifically, the laser irradiation pattern may be, for example, a quadrilateral. Furthermore, by reducing the area of ​​the unformed region R3, the area within the surface of the laser absorption layer P where the bonding strength with the second wafer W2 is reduced increases. Consequently, the laser absorption layer P and the second wafer W2 can be debonded appropriately.

[0106] In addition, in the above embodiment, the laser absorption layer P, the device layer D2 and the surface film F2 are stacked in the order of the laser absorption layer P, the device layer D2 and the surface film F2 on the surface W2a of the second wafer W2, but it can also be as follows Figure 18 As shown in (a), a peeling-promoting layer P2 is further formed between the second wafer W2 and the laser absorption layer P. As the peeling-promoting layer P2, a material is selected that is transparent to laser light (CO2 laser) and has a lower adhesion to the second wafer W2 (silicon) than to the laser absorption layer P (SiO2), such as silicon nitride (SiN).

[0107] like Figure 18 As shown in (b), when transferring the device layer D2 from the superimposed wafer T on which the peeling-facilitating layer P2 is formed, first, pulsed laser light L (CO2 laser) is irradiated toward the back surface W2b of the second wafer W2. At this time, the laser light L passes through the second wafer W2 and the peeling-facilitating layer P2 from the back surface W2b of the second wafer W2 and is absorbed by the laser absorption layer P. Furthermore, a peeling-modifying layer M1 is formed within the laser absorption layer P that has absorbed the laser light L.

[0108] Here, the stress generated by irradiation with the laser L usually remains at the irradiation position of the laser L (inside the laser absorption layer) as shown in the above embodiment, and forms the peeling modification layer M1. However, when the peeling promotion layer P2 is formed as in this embodiment, the adhesion between the peeling promotion layer P2 and the second wafer W2 is weaker than the adhesion between the peeling promotion layer P2 and the laser absorption layer P. Therefore, as shown in FIG. Figure 18 As shown in (c), the stress generated passes through the peeling-promoting layer P2 and accumulates at the interface between the peeling-promoting layer P2 and the second wafer W2. In other words, the stress generated by irradiating the laser L moves to the interface between the peeling-promoting layer P2 and the second wafer W2, where it can be more stably retained, and accumulates at this interface, thereby reducing the bonding strength between the peeling-promoting layer P2 and the second wafer W2.

[0109] Moreover, since the bonding strength between the peeling promotion layer P2 and the second wafer W2 is reduced as described above, the peeling promotion layer P2 and the second wafer W2 can be properly peeled off later. In addition, at this time, since the adhesion between the peeling promotion layer P2 and the second wafer W2 is low, Figure 16As shown, it is possible to appropriately inhibit a portion of the second wafer W2 from remaining on the surface of the separation-promoting layer P2 after separation. In addition, in this embodiment, the laser-absorbing layer P absorbs the laser light L, and thus it is possible to more appropriately inhibit damage from remaining on the surface W2a of the second wafer W2 and the surface of the separation-promoting layer P2 after separation.

[0110] Further, in the case where separation is appropriately performed at the interface between the separation-promoting layer P2 and the second wafer W2 as described above, gas generated by irradiation of laser light needs to pass through the separation-promoting layer P2. However, in the case where the separation-promoting layer P2 has a large film thickness, the generated gas cannot appropriately pass through the separation-promoting layer P2, and separation can occur at the interface between the separation-promoting layer P2 and the laser-absorbing layer P. Therefore, in order to appropriately perform separation at the interface between the separation-promoting layer P2 and the second wafer W2, it is preferable that the film thickness of the separation-promoting layer P2 be thin relative to the film thickness of the laser-absorbing layer P, and specifically, for example, it is preferable that the film thickness of the separation-promoting layer P2 be about one-tenth of the film thickness of the laser-absorbing layer P. By reducing the film thickness of the separation-promoting layer P2 as described above, it is possible to cause the generated gas to appropriately pass through the separation-promoting layer P2, and thus it is possible to separate the second wafer W2 from the separation-promoting layer P2.

[0111] However, even in the case where the separation-promoting layer P2 has a large film thickness and separation occurs at the interface between the separation-promoting layer P2 and the laser-absorbing layer P, the second wafer W2 is separated from the laser-absorbing layer P via the separation-promoting layer P2, and thus, as shown in FIG. 6, a portion of the second wafer W2 does not remain on the surface of the laser-absorbing layer P after separation. Figure 16 As shown, a portion of the second wafer W2 does not remain on the surface of the laser-absorbing layer P after separation. That is, it is thereby possible to protect the surface W2a of the second wafer W2 and inhibit roughening of the separation surface.

[0112] Further, in the above example, a material having low adhesion to the second wafer W2 (silicon) is used as the separation-promoting layer P2, but the material used for the separation-promoting layer P2 is not limited thereto, and for example, a material having a different coefficient of thermal expansion from the coefficient of thermal expansion of the second wafer W2 (silicon) can be used. In this case, since the amount of deformation caused by heat generated by irradiation of laser light L differs between the second wafer W2 and the separation-promoting layer P2, a shear force is generated at the interface between the second wafer W2 and the separation-promoting layer P2, and it is possible to separate the second wafer W2 from the separation-promoting layer P2.

[0113] Furthermore, in the above embodiment, the release of the compressive stress generated by laser irradiation and accumulated as the peeling modification layer M1 allows the peeling of the second wafer W2 and the peeling promotion layer P2 to progress. However, this stress may cause warping of the superposed wafer T. If warping occurs in the superposed wafer T, proper wafer processing may not be possible. Therefore, to suppress this warping of the superposed wafer T, the superposed wafer T may be pressed from above when the laser absorption layer P is irradiated with laser light L.

[0114] For example, when the overlapped wafer T is warped in a manner such that it is deformed into an upward convex shape, as shown in FIG. Figure 19 As shown, the center of the overlapping wafer T can be pressed by the pressing member 200. Specifically, when the second wafer W2 is peeled off, first, the pressing range of the pressing member 200, that is, the center of the laser absorption layer P, is preliminarily subjected to laser irradiation processing, that is, the unpeeled area R1 is preliminarily formed. When the unpeeled area R1 is formed, the unpeeled area R1 is then pressed by the pressing member 200. Moreover, thereafter, in a state where the unpeeled area R1 is pressed by the pressing member 200, when the formation position of the unpeeled area R1 reaches the outer peripheral end of the laser absorption layer P, the chain peeling of the second wafer W2 is started. At this time, since the center of the overlapping wafer T is pressed by the pressing member 200, the warping of the overlapping wafer T can be suppressed.

[0115] In addition, the unpeeled region R1 can also be formed from the radial outside toward the inside. That is, first, the unpeeled region R1 is formed from the outer periphery toward the center of the laser absorption layer P. At this time, the starting position of the formation of the unpeeled region R1, that is, the outer peripheral end, is determined to be a position slightly radially inward from the outer peripheral end of the laser absorption layer P, and stress is not released. When the unpeeled region R1 is formed, the unpeeled region R1 is then pressed by the pressing member 200. Moreover, thereafter, in a state where the unpeeled region R1 is pressed by the pressing member 200, the formation position of the unpeeled region R1 reaches the outer peripheral end of the laser absorption layer P. Thereafter, the chain peeling is started by forming the starting point modified layer M1s on the radial outside of the second wafer W2. At this time, the center of the overlapping wafer T is pressed by the pressing member 200, so that warping of the overlapping wafer T can be suppressed.

[0116] Furthermore, since the superposed wafer T is rotated when the laser light L is irradiated, it is desirable that the end portion of the pressing member 200 be configured to rotate together with the superposed wafer T.

[0117] In addition, for example, when the overlapped wafer T is warped in a manner such that it is deformed into a convex shape, as shown in FIG. Figure 20As shown, the peripheral edge We of the overlapping wafer T can be pressed by the pressing member 200. Specifically, when the second wafer W2 is peeled off, first, the pressing range of the pressing member 200, that is, the outer periphery of the laser absorption layer P, is preliminarily subjected to laser irradiation processing, that is, the peeling area R2 is preliminarily formed. When the peeling area R2 is formed, the peeling area R2 is then pressed by the pressing member 200. Moreover, thereafter, in a state where the peeling area R2 is pressed by the pressing member 200, the unpeeled area R1 begins to be formed from the radial inside to the outside in the center of the laser absorption layer P. Moreover, when the formation area of ​​the unpeeled area R1 reaches the peeling area R2, the chain peeling of the second wafer W2 is started. At this time, since the outer periphery of the overlapping wafer T is pressed by the pressing member 200, the warping of the overlapping wafer T can be suppressed.

[0118] Furthermore, in the overlapped wafer T processed by the above embodiment, as shown in FIG. Figure 21 As shown, a reflective film R may be provided between the laser absorption layer P and the device layer D2. Specifically, the reflective film R is formed on the surface of the laser absorption layer P opposite to the incident surface of the laser light L. A material having a high reflectivity and a high melting point for the laser light L, such as a metal film, is used as the reflective film R. Furthermore, the device layer D2 is a functional layer, distinct from the reflective film R.

[0119] In this case, the laser light L emitted from the laser irradiation unit 110 passes through the second wafer W2 and is almost completely absorbed by the laser absorption layer P. However, any unabsorbed laser light L is reflected by the reflective film R. As a result, the laser light L does not reach the device layer D2, and damage to the device layer D2 can be reliably suppressed.

[0120] Furthermore, the laser light L reflected by the reflective film R is absorbed by the laser absorption layer P. Therefore, the peeling efficiency of the second wafer W2 can be improved.

[0121] Furthermore, in the above embodiment, a case where the laser lift-off process of the overlapped wafer T, i.e., the transfer process of the device layer D2 on the first wafer W1, is performed in the wafer processing system 1 has been described. However, as described above, an edge trimming process of the second wafer W2 can be performed in the wafer processing system 1. Next, a case where the edge trimming of the second wafer W2 is performed in the wafer processing system 1 will be described.

[0122] First, the wafer transfer device 20 takes out the overlapped wafer T from the cassette Ct placed on the cassette stage 10 in the loading / unloading block G1 , transfers it to the wafer transfer device 40 via the conveyor device 30 , and then transfers it to the internal laser irradiation device 70 .

[0123] In the internal laser irradiation device 70, as Figure 22As shown in (a) of FIG. 1 , laser L2 (YAG laser) is irradiated into the interior of the second wafer W2 to form a peripheral edge modification layer M2, which serves as a base point for removing the peripheral portion We in the edge trimming described later. A crack C2 extends from the peripheral edge modification layer M2 along the thickness direction of the second wafer W2. The upper end and lower end of the crack C2 reach, for example, the back surface W2b and the surface W2a of the second wafer W2, respectively. Next, the wafer transfer device 40 transfers the overlapping wafer T with the peripheral edge modification layer M2 formed inside the second wafer W2 to the interface laser irradiation device 80.

[0124] In the interface laser irradiation device 80, in the overlapped wafer T, the bonding strength between the laser absorption layer P and the second wafer W2 at the peripheral edge portion We of the second wafer W2 to be removed is reduced. Specifically, Figure 22 As shown in (b), laser L (CO2 laser) is irradiated onto the laser absorption layer P to form a peeling modified layer M1 (non-peeling region R1) at a position radially outward of the peripheral modified layer M2 formed by the internal laser irradiation device 70.

[0125] Furthermore, when forming the release-modified layer M1 (non-released region R1), the chuck 100 (overlapping wafer T) is rotated by the rotation mechanism 103, and the chuck 100 is moved in the Y-axis direction by the movement mechanism 104. Thus, the laser light L is irradiated onto the laser absorption layer P from the radial inside toward the outside, resulting in the laser light L being irradiated in a spiral shape from the inside toward the outside.

[0126] When the peeling modification layer M1 is continuously formed and the formation position of the peeling modification layer M1 reaches the end portion of the second wafer W2, that is, the boundary Ad, as shown in FIG. Figure 22 As shown in (c), the laser absorption layer P begins to undergo chain delamination from the radially outer side toward the inner side. In this embodiment, the delamination modified layer M1 is formed only radially outward of the peripheral modified layer M2 (crack C2). Therefore, delamination of the laser absorption layer P and the second wafer W2 progresses only in the peripheral portion We, that is, radially outward of the peripheral modified layer M2.

[0127] Next, the wafer transfer device 40 transfers the superimposed wafer T, in which the laser absorption layer P at the peripheral edge portion We is separated from the second wafer W2 , to the peripheral edge removal device 50 .

[0128] In the peripheral edge removal device 50, in the overlapped wafer T, as shown in FIG. Figure 22As shown in (d), the peripheral portion We of the second wafer W2 is removed (edge ​​trimming) based on the peripheral modified layer M2 and the crack C2. Furthermore, the edge trimming method of the peripheral edge removal device 50 can be arbitrarily selected. In this case, when removing the peripheral portion We, the bonding strength between the second wafer W2 and the laser absorption layer P is reduced by forming the peeling modified layer M1, thereby facilitating the removal of the peripheral portion We.

[0129] Next, the wafer transfer device 40 transfers the overlapped wafer T, from which the peripheral edge portion We of the second wafer W2 has been removed, to the cleaning device 60. In the cleaning device 60, the overlapped wafer T is scrubbed. Afterwards, the wafer transfer device 40 unloads the fully processed overlapped wafer T from the cleaning device 60 and transfers it to the cassette Ct on the cassette stage 10 via the wafer transfer device 20 and the conveyor device 30. This completes a series of wafer processing operations in the wafer processing system 1.

[0130] As described above, according to the technology of the present disclosure, the interface laser irradiation device 80 can reduce the bonding strength between the second wafer W2 and the laser absorption layer P at the peripheral portion We, thereby enabling the peripheral portion We to be removed appropriately, i.e., edge trimming, in the peripheral edge removal device 50 .

[0131] In addition, the processing order of the overlapping wafer T by the internal laser irradiation device 70 and the interface laser irradiation device 80 is not limited to the above-mentioned embodiment. After the peripheral portion We is peeled off by the interface laser irradiation device 80, the peripheral modification layer M2 can be formed by the internal laser irradiation device 70.

[0132] The embodiments disclosed herein are to be considered in all respects as illustrative and non-restrictive, and the embodiments described above may be omitted, replaced, or modified in various ways without departing from the scope of the appended claims and the spirit thereof.

[0133] Description of Reference Numerals

[0134] D2: device layer; L: laser; M1: lift-off modification layer; M1s: starting point modification layer; P: laser absorption layer; T: overlapped wafer; W1: first wafer; W2: second wafer; W2a: surface; W2b: backside.

Claims

1. A substrate processing method for processing a superposed substrate formed by bonding a first substrate and a second substrate, wherein a laser absorption layer is formed on the second substrate. The substrate processing method comprises: irradiating the laser absorption layer with a laser beam in a pulsed manner to form a release modification layer, and accumulating stress in the laser absorption layer, wherein the laser beam is irradiated in a pulsed manner while forming a plurality of release modification layers so that the release modification layers do not overlap each other when viewed from above; and The accumulated stress is released in a chain reaction to peel off the second substrate.

2. The substrate processing method according to claim 1, wherein: When the release-modifying layer is formed, the laser absorption layer is prevented from being separated from the second substrate.

3. The substrate processing method according to claim 1 or 2, wherein: A starting point modified layer is formed as a starting point for the chain release of the stress.

4. The substrate processing method according to claim 3, wherein: The starting point modification layer is formed at an end portion of a bonding region where the first substrate and the second substrate are bonded and inside the laser absorption layer. When forming the starting point modification layer, the stress generated by forming the starting point modification layer is released to the non-bonded area in the radial direction outside the bonding area, thereby causing the laser absorption layer to peel off from the second substrate. The stress accumulated by forming the peeling reforming layer is released toward the peeling region generated by forming the starting point reforming layer, thereby starting a chain release of the stress.

5. The substrate processing method according to claim 1 or 2, wherein: When forming the starting point modification layer, the laser absorption layer and the second substrate are separated by irradiating the laser beam. The stress accumulated by forming the peeling reforming layer is released toward the peeling region generated by forming the starting point reforming layer, thereby starting a chain release of the stress.

6. The substrate processing method according to claim 4, wherein: The starting point reforming layer is formed radially outward of the release reforming layer.

7. The substrate processing method according to claim 1 or 2, wherein: The method further includes forming a peripheral edge modification layer along a boundary between a peripheral edge portion of the second substrate to be removed and a central portion of the second substrate, The release reforming layer is formed radially outward of the peripheral edge reforming layer.

8. The substrate processing method according to claim 1 or 2, wherein: The release reforming layer is not formed in the center portion of the laser absorption layer.

9. The substrate processing method according to claim 1 or 2, characterized in that: By controlling the formation position of the release reforming layer and / or the irradiation shape of the laser when forming the release reforming layer, the formation area of ​​the release reforming layer relative to the laser absorption layer is increased.

10. The substrate processing method according to claim 1 or 2, wherein: A peeling-promoting layer for promoting peeling of the second substrate is further formed between the second substrate and the laser absorption layer.

11. The substrate processing method according to claim 1 or 2, characterized in that: When the laser absorption layer is irradiated with the laser beam, pulsed laser beams are oscillated from a laser oscillator toward an optical element, and the frequency of the laser beams is adjusted in the optical element.

12. The substrate processing method according to claim 11, wherein: The frequency of the laser light from the laser oscillator is the highest frequency controllable by the optical element.

13. The substrate processing method according to claim 11, wherein: When the laser light is irradiated onto the laser light absorbing layer, the laser light from the laser oscillator is attenuated in an attenuator.

14. A substrate processing apparatus for processing a superposed substrate formed by bonding a first substrate and a second substrate, wherein a laser absorption layer is formed on the second substrate. The substrate processing device comprises: a laser irradiation unit configured to irradiate the laser absorption layer of the second substrate with a laser beam in a pulsed manner; and a control unit that controls the operation of the laser irradiation unit, in, The control unit controls the operation of the laser irradiation unit so as to form a peeling modification layer by irradiating the laser and accumulate stress inside the laser absorption layer, and then peel off the second substrate by a chain release of the accumulated stress, wherein the laser is irradiated in a pulsed manner while forming a plurality of peeling modification layers so that the peeling modification layers do not overlap with each other when viewed from above.

15. The substrate processing apparatus according to claim 14, wherein: The control unit controls the output of the laser so as to prevent separation between the laser absorption layer and the second substrate when forming the release reforming layer.

16. The substrate processing apparatus according to claim 14 or 15, wherein: The control unit controls the operation of the laser irradiation unit to form a starting point modified layer serving as a starting point for the chain release of the stress.

17. The substrate processing apparatus according to claim 16, wherein: The control unit controls the operation of the laser irradiation unit to form the starting point modification layer at an end portion of a bonding region where the first substrate and the second substrate are bonded and inside the laser absorption layer. When forming the starting point modification layer, the stress generated by forming the starting point modification layer is released to the non-bonded area in the radial direction outside the bonding area, thereby causing the laser absorption layer and the second substrate to peel off. The stress accumulated by forming the peeling reforming layer is released toward the peeling region generated by forming the starting point reforming layer, thereby starting a chain release of the stress.

18. The substrate processing apparatus according to claim 16, wherein: When forming the starting point modification layer, the control unit controls the output of the laser so as to cause separation of the laser absorption layer and the second substrate by irradiating the laser. Furthermore, the control unit controls the operation of the laser irradiation unit so that the stress accumulated by forming the peeling reforming layer is released to the peeling region generated by forming the starting point reforming layer, thereby starting the chain release of the stress.

19. The substrate processing apparatus according to claim 17 or 18, wherein: The control unit controls the operation of the laser irradiation unit so that the starting point reforming layer is formed radially outward of the release reforming layer.

20. The substrate processing apparatus according to claim 14 or 15, wherein: The invention further comprises a second laser irradiation unit for forming a peripheral edge modification layer along a boundary between a peripheral edge portion of the second substrate to be removed and a central portion of the second substrate. The control unit controls the operation of the laser irradiation unit so as to form the release reforming layer at a position radially outward of the peripheral edge reforming layer.

21. The substrate processing apparatus according to claim 14 or 15, wherein: The control unit controls the operation of the laser irradiation unit so as not to form the release reforming layer in the center portion of the laser absorption layer.

22. The substrate processing apparatus according to claim 14 or 15, wherein: The control unit controls a formation position of the release reforming layer and / or an irradiation shape of the laser light when forming the release reforming layer so as to increase a formation area of ​​the release reforming layer relative to the laser absorption layer.

23. The substrate processing apparatus according to claim 14 or 15, wherein: A peeling-promoting layer for promoting peeling of the second substrate is further formed between the second substrate and the laser absorption layer.

24. The substrate processing apparatus according to claim 14 or 15, wherein: It also includes a control unit that controls the laser irradiation unit. The laser irradiation unit has: a laser oscillator that oscillates pulsed laser light; and an optical element that redirects the laser light from the laser oscillator in different directions, The control unit controls the optical element to adjust the frequency of the laser beam irradiated onto the laser absorption layer.

25. The substrate processing apparatus according to claim 24, wherein: The frequency of the laser light from the laser oscillator is the highest frequency controllable by the optical element.

26. The substrate processing apparatus according to claim 24, wherein: The laser irradiation unit includes an attenuator for attenuating the laser light from the laser oscillator.

Images