Method for dividing processing target

Abstract

Disclosed is a method for dividing a processing target in which a pattern layer is disposed between a first wafer layer and a second wafer layer. The dividing method disclosed herein may comprise the steps of: forming a processing groove extending in the depth direction from one surface of the processing target to at least the pattern layer; forming a modified region inside the processing target by focusing a laser beam onto a position spaced apart from the other surface of the processing target; and applying an external force to the processing target and thereby forming a crack extending along the modified region.

Description

Method for dividing the object to be processed

[0001] The present invention relates to a method for dividing a processing object.

[0002] In recent years, technologies that require a large number of calculations at once, such as artificial intelligence, have become increasingly important, and the demand for more powerful and efficient computing systems is increasing accordingly.

[0003] To meet these requirements, a hybrid bonding method can be used, which directly connects chips without intermediate conductors such as wires or bumps. This hybrid bonding method is also known as direct bond interconnect (DBI).

[0004] A workpiece bonded by a hybrid bonding method may have a pattern layer including metal wiring, etc., arranged at the bonding interface. In this way, a workpiece bonded by a hybrid bonding method with a pattern layer arranged at the bonding interface may be divided into a plurality of pieces.

[0005] The present disclosure can provide a method for dividing a workpiece that can easily cut a workpiece bonded by a hybrid bonding method.

[0006] The present disclosure can provide a method for dividing a processing object capable of easily cutting a processing object having a pattern layer disposed between wafers.

[0007] A method for dividing a processing object according to one aspect of the present disclosure is a method for dividing a processing object having a pattern layer disposed between a first wafer and a second wafer, the method comprising: forming a processing groove extending in a depth direction from one surface of the processing object to at least the pattern layer; forming a modified region inside the processing object by focusing a laser beam at a location spaced apart from the other surface of the processing object; and forming a crack extending along the modified region by applying an external force to the processing object.

[0008] In one embodiment, the step of forming the processing groove may include irradiating a first laser beam to form a processing groove extending from the surface of the second wafer layer to at least the pattern layer.

[0009] The depth of the above-mentioned processing groove may be equal to or greater than the sum of the thickness of the second wafer layer and the thickness of the pattern layer.

[0010] The frequency of the first laser beam may be 1.5 MHz to 2.5 MHz.

[0011] The intensity profile of the first laser beam has a Gaussian distribution, and the processing groove formed by the first laser beam may have a shape in which the width becomes narrower as the depth increases from the surface of the second wafer layer.

[0012] The step of forming the above-mentioned modified region can form the modified region inside the first wafer layer by irradiating a second laser beam.

[0013] The step of forming the above-mentioned modified region may irradiate the second laser beam so that it is focused at a position spaced apart in the extension direction of the processing groove within the first wafer layer so that the modified region overlaps the processing groove along the extension direction of the processing groove.

[0014] The thickness of the second wafer layer may be smaller than the thickness of the first wafer layer in which the modified region is formed.

[0015] The thickness of the second wafer layer may be 150 μm or less.

[0016] The thickness of the first wafer layer may be 1000 ㎛ or less.

[0017] In the step of forming the crack, a crack extending from the modified region to the processing groove can be formed inside the first wafer layer.

[0018] The pattern layer may include a first pattern layer including a first conductor region and a first dielectric region, and a second pattern layer including a second conductor region bonded to the first conductor region and a second dielectric region bonded to the first dielectric region.

[0019] The first pattern layer and the second pattern layer can be bonded by a hybrid bonding method.

[0020] The first pattern layer may be disposed on one surface of the first wafer layer facing the second wafer layer, and the second pattern layer may be disposed on one surface of the second wafer layer facing the first wafer layer.

[0021] The thickness of the above pattern layer may exceed 30 μm.

[0022] A method for dividing a processing object according to an embodiment of the present disclosure can easily cut and divide the processing object into a plurality of pieces by performing a grooving process on one side of the processing object and performing a process of forming a modified region inside the processing object by irradiating a laser beam on the other side of the processing object.

[0023] Figure 1 is a schematic drawing showing an example of a processing target according to an embodiment.

[0024] Figure 2 is an enlarged view of a portion of the processing target of Figure 1.

[0025] Figure 3 is a drawing for explaining a state when a process of forming a modified area inside the processing target of Figure 1 is performed.

[0026] Figure 4 is a flowchart for explaining a method for dividing a processing object according to one embodiment.

[0027] FIG. 5 and FIG. 6 are drawings for explaining an example of a grooving process of a method for dividing a processing object according to one embodiment.

[0028] Figures 7 and 8 are drawings for explaining a process of forming a modified area inside a processing target after the grooving process has been completed.

[0029] FIG. 9 is a drawing showing a processing object separated by a method for dividing a processing object according to one embodiment.

[0030] FIG. 10a and FIG. 10b are schematic drawings showing other examples of a processing object according to an embodiment.

[0031] Hereinafter, embodiments of the present invention will be described in detail with reference to the attached drawings. In the drawings, the same reference numerals designate the same components, and the size and thickness of each component may be exaggerated for clarity of explanation.

[0032] Terms that include ordinal numbers, such as "first," "second," etc., may be used to describe various components, but the components are not limited by the terms. The terms are used solely to distinguish one component from another. For example, without departing from the scope of the present invention, the first component could be referred to as the second component, and similarly, the second component could also be referred to as the first component. The term "and / or" includes any combination of multiple related items or any one of multiple related items.

[0033]

[0034] Fig. 1 is a schematic drawing showing an example of a processing target (1) according to an embodiment, and Fig. 2 is an enlarged drawing showing area A, which is a part of the processing target (1) of Fig. 1. Fig. 3 is a drawing for explaining a state when a process of forming a modified area inside the processing target (1) of Fig. 1 is performed.

[0035] Referring to FIGS. 1 and 2, the processing object (1) may include a first wafer layer (10), a second wafer layer (20), and a pattern layer (30) disposed therebetween.

[0036] The first wafer layer (10) may be a silicon wafer. The first wafer layer (10) may include silicon. However, the material of the first wafer layer (10) is not necessarily limited thereto, and may be modified in various ways as long as it is used as a material for a semiconductor substrate. For example, the material of the first wafer layer (10) may include silicon carbide (SiC) or sapphire.

[0037] The thickness of the first wafer layer (10) may be greater than the thickness of the second wafer layer (20). The first wafer layer (10) may correspond to a substrate of a chip from which the object to be processed or the object to be processed (1) is separated, and the second wafer layer (20) may correspond to a die of the object to be processed (1) or the chip from which the object to be processed (1) is separated. The thickness of the first wafer layer (10) may be 1000 um or less. However, the relationship between the thickness of the first wafer layer (10) and the thickness of the second wafer layer (20) is not necessarily limited thereto, and the thickness of the first wafer layer (10) may be equal to or less than the thickness of the second wafer layer (20) as needed.

[0038] The second wafer layer (20) may be a silicon wafer. The second wafer layer (20) may include silicon. However, the material of the second wafer layer (20) is not necessarily limited thereto, and may be modified in various ways as long as it is used as a material for a semiconductor substrate. For example, the material of the second wafer layer (20) may include silicon carbide (SiC) or sapphire.

[0039] The material of the second wafer layer (20) may be the same as the material of the first wafer layer (10). However, the material of the first wafer layer (10) and the material of the second wafer layer (20) are not necessarily the same and may be different as needed.

[0040] The pattern layer (30) may include a first pattern layer (31) disposed on the first wafer layer (10) and a second pattern layer (32) disposed on the lower side of the second wafer layer (20). The first pattern layer (31) may be disposed between the first wafer layer (10) and the second pattern layer (32), and the second pattern layer (32) may be disposed between the first pattern layer (31) and the second wafer layer (20). The first pattern layer (31) may be disposed on one surface of the first wafer layer (10) facing the second wafer layer (20), and the second pattern layer (32) may be disposed on one surface of the second wafer layer (20) facing the first wafer layer (10).

[0041] For example, the first pattern layer (31) may include at least one first conductor region (311) and at least one first dielectric region (312). The first conductor region (311) may include a conductive metal material. For example, the first conductor region (311) may include copper (Cu), but is not necessarily limited thereto. The first dielectric region (312) may be disposed between the first conductor regions (311). The first dielectric region (312) may include a low-conductivity dielectric material. For example, the first dielectric region (312) may include silicon oxide, but is not necessarily limited thereto.

[0042] For example, the second pattern layer (32) may include at least one second conductor region (321) and at least one second dielectric region (322). The second conductor region (321) may include a conductive metal material. For example, the second conductor region (321) may include copper (Cu), but is not necessarily limited thereto. The second dielectric region (322) may be disposed between the second conductor regions (321). The second dielectric region (322) may include a low-conductivity dielectric material. For example, the second dielectric region (322) may include silicon oxide, but is not necessarily limited thereto.

[0043] The pattern layer (30) of the processing target (1) may have a structure in which a first pattern layer (31) and a second pattern layer (32) are bonded. For example, the processing target (1) may have a structure in which the first pattern layer (31) and the second pattern layer (32) are bonded by a hybrid bonding method.

[0044] For example, the first conductor region (311) and the second conductor region (321) may be arranged at positions corresponding to each other. The first dielectric region (312) and the second dielectric region (322) may be arranged at positions corresponding to each other. In this state, the first pattern layer (31) and the second pattern layer (32) may be bonded by at least one of surface treatment, pressure treatment, and heat treatment.

[0045] The first conductor region (311) and the second conductor region (321) are bonded, and the first dielectric region (312) and the second dielectric region (322) can be bonded. The bonding method of the first pattern layer (31) and the second pattern layer (32) is a technique known as hybrid bonding, so a detailed description thereof will be omitted. In the drawing, for the convenience of explanation, a bonding interface between the first pattern layer (31) and the second pattern layer (32) is illustrated, but it may not appear in an actual structure.

[0046] The total thickness of the pattern layer (30) in which the first pattern layer (31) and the second pattern layer (32) are bonded may be a predetermined thickness or more. The thickness of the pattern layer (30) may exceed 30 ㎛. The thickness of the pattern layer (30) may be the sum of the thickness of the first pattern layer (31) and the thickness of the second pattern layer (32). However, the thickness of the pattern layer (30) may be smaller than the thickness of the second wafer layer (20).

[0047] In this way, in order to divide the processing object (1) bonded by the hybrid bonding method into a plurality of chips, the use of a method of forming a modified area inside the processing object and then applying an external force to cut it, which is one of the laser cutting methods, can be considered.

[0048] Referring to FIG. 3, a laser beam may be focused at a position spaced apart from the other surface of the object to be processed (1) to form a modified region (110) inside the object to be processed (1). For example, a laser beam may be focused at a position spaced apart from the surface of the first wafer layer (10) of the object to be processed (1) to form a modified region (110) inside the first wafer layer (10). The object to be processed (1) may be pressed so that a crack (C) occurs in the modified region (110) of the first wafer layer (10). However, even if a crack (C) occurs inside the first wafer layer (10), the crack (C) may not extend to the extent of cutting the pattern layer (30). In particular, when the thickness of the pattern layer (30) becomes thick, such as a pattern layer (30) bonded by a hybrid bonding method, for example, when the thickness of the pattern layer (30) exceeds 30 um, the object to be processed (1) may not be cut only by a cutting method that applies an external force after forming a modified area inside the object to be processed.

[0049] In the method for dividing a processing object (1) according to an embodiment, a dividing method (or cutting method) that can implement good cutting quality can be provided even for a processing object (1) in which a thick pattern layer (30) is arranged between wafer layers.

[0050] Fig. 4 is a flowchart for explaining a method for dividing a processing object (1) according to one embodiment. Figs. 5 and 6 are drawings for explaining an example of a grooving process of a method for dividing a processing object (1) according to one embodiment. Figs. 7 and 8 are drawings for explaining a process for forming a modified region inside a processing object (1) after a grooving process has been completed. Fig. 9 is a drawing showing a processing object (1) separated by a method for dividing a processing object (1) according to one embodiment.

[0051] Referring to FIGS. 4 to 6, in a method for dividing a processing object (1) according to one embodiment, a processing groove (40) extending in the depth direction from one surface of the processing object (1) to at least the pattern layer (30) can be formed (S10). For example, a processing groove (40) extending from the surface (201) of the second wafer layer (20) of the processing object (1) to at least the pattern layer (30) can be formed. In other words, a processing groove (40) extending in the depth direction from the surface (201) of the second wafer layer (20) of the processing object (1) to penetrate the pattern layer (30) can be formed.

[0052] For example, a laser grooving process may be performed in which a first laser beam (L1) is irradiated toward a second wafer layer (20) of a processing target (1). The first laser beam (L1) may be sequentially irradiated in a depth direction from the surface (201) of the second wafer layer (20) to the pattern layer (30). In other words, the first laser beam (L1) may be sequentially irradiated in a depth direction from the surface (201) of the second wafer layer (20), so that at least a portion of the second wafer layer (20) and the pattern layer (30) may be sequentially removed.

[0053] The frequency of the first laser beam (L1) may be high in consideration of local removal of the pattern layer (30). For example, the frequency of the first laser beam (L1) may be 1.5 MHz to 2.5 MHz. In this way, by setting the frequency of the first laser beam high, precise ablation of the pattern layer (30) can be implemented. The frequency of the first laser beam (L1) may be higher than the frequency of the second laser beam (L2) irradiated onto the first wafer layer (10), which will be described later.

[0054] The second wafer layer (20) and the pattern layer (30) of the processing target (1) can be locally removed by the first laser beam (L1). In other words, the second wafer layer (20) and the pattern layer (30) can be locally removed by laser ablation.

[0055] The thickness of the second wafer layer (20) on which laser grooving is performed may be smaller than the thickness of the first wafer layer (10). For example, the thickness of the second wafer layer (20) may be 150 μm or less. For example, the thickness of the second wafer layer (20) may be 60 μm or less. In this way, by performing the laser grooving process on the second wafer layer (20) having a predetermined thickness or less, the problem of the powders that are decomposed during the processing by the first laser beam (L1) not being discharged to the outside but sticking to the inside of the processing groove (40) can be prevented or minimized.

[0056] Referring to FIG. 6, the processing groove (40) formed by the laser grooving process can penetrate the second wafer layer (20) and the pattern layer (30).

[0057] The depth of the machining groove (40) may be greater than the thickness of the second wafer layer (20). The depth of the machining groove (40) may be equal to or greater than the sum of the thicknesses of the second wafer layer (20) and the thicknesses of the pattern layer (30). For example, the end of the machining groove (40) may be located on the lower surface of the pattern layer (30). In other words, the machining groove (40) may extend from the upper surface of the second wafer layer (20) to the lower surface of the pattern layer (30). In another example, the end of the machining groove (40) may be located inside the first wafer layer (10). In other words, the machining groove (40) may extend from the upper surface of the second wafer layer (20) to the interior of the first wafer layer (10). The depth of the machining groove (40) may be 150 ㎛ or less. For example, the depth of the machining groove (40) may be 60 ㎛ or less.

[0058] The width of the machining groove (40) may be smaller than the depth of the machining groove (40). The width of the machining groove (40) may be smaller than half the depth of the machining groove (40). The width of the machining groove (40) may be smaller than 1 / 3 the depth of the machining groove (40).

[0059] The intensity profile of the first laser beam (L1) may have a Gaussian distribution. For example, the first laser beam (L1) may have an intensity distribution in the form of a Gaussian function in the cross-section, and may have a distribution in which the intensity is maximum in the center and gradually decreases toward the periphery. The width of the processing groove (40) formed in the pattern layer (30) by the first laser beam (L1) may be smaller than the width of the processing groove (40) formed in the second wafer layer (20). The cross-sectional shape of the processing groove (40) may be a 'V' shape. However, the cross-sectional shape of the processing groove (40) is not limited thereto and may vary. For example, the cross-sectional shape of the processing groove (40) may be a 'U' shape.

[0060] In the above-described embodiment, a laser grooving process of irradiating a first laser beam (L1) is exemplified as a grooving process for forming a processing groove (40). However, the process for forming a processing groove (40) is not necessarily limited thereto, and other grooving processes, for example, a process using a rotating blade, may be performed depending on the material, thickness, etc. of the processing target (1).

[0061] Referring to FIGS. 4, 7 and 8, in the method for dividing a processing object (1) according to an embodiment, a process of forming a modified region inside a processing object (1) in which a processing groove (40) is formed by a grooving process can be performed.

[0062] For example, the second laser beam (L2) can be irradiated so as to be focused on the inside of the processing target (1) (S20). For example, the second laser beam (L2) can be irradiated so as to be focused on a location spaced apart from the surface (101) of the first wafer layer (10).

[0063] In order to irradiate the second laser beam (L2), the second laser beam (L2) can be irradiated from the surface (101) side of the first wafer layer (10). The second laser beam (L2) can be focused on the inside of the first wafer layer (10) to form a modified region (110) on the inside of the first wafer layer (10). The second laser beam (L2) can be different from the first laser beam (L1). For example, the second laser beam (L2) can have a higher peak power than the first laser beam (L1). For example, the second laser beam (L2) can have a shorter pulse width than the first laser beam (L1).

[0064] The thickness of the second wafer layer (20) to which the second laser beam (L2) is irradiated may be greater than the thickness of the first wafer layer (10). In other words, when the thicknesses of the first wafer layer (10) and the second wafer layer (20) are different, a laser grooving process may be performed on the relatively thin second wafer layer (20), and a process of forming a modified region may be performed on the relatively thick first wafer layer (10).

[0065] The second laser beam (L2) can be irradiated so as to be focused at a position spaced apart from the extension direction of the processing groove (40) within the first wafer layer (10) so that the modified region (110) overlaps with the processing groove (40) along the extension direction of the processing groove (40). Accordingly, the modified region (110) can be formed at a position overlapping with the processing groove (40) along the extension direction of the processing groove (40) within the first wafer layer (10). In FIGS. 7 and 8, an example in which the number of the modified region (110) is one is shown, but the present invention is not limited thereto, and the number of the modified region (110) may be plural as needed.

[0066] Referring to FIGS. 4 and 8, an external force may be applied (S30) to form a crack (C) inside the first wafer layer (10) in which the modified area (110) is formed. For example, by applying an external force in the direction of the arrow to the surface (101) of the first wafer layer (10), a crack (C) may be formed from the modified area (110) inside the first wafer layer (10). The crack (C) may extend vertically from the modified area (110). The crack (C) may extend from the modified area (110) to the processing groove (40), and may extend from the modified area (110) to the surface (101) of the first wafer layer (10).

[0067] Referring to Fig. 9, as the crack (C) extends to the machining groove (40), the machining target (1) can be divided into a plurality of chips (2). Each of the plurality of chips (2) can be a structure having a pattern layer (30) disposed between a first wafer layer (10) and a second wafer layer (20).

[0068] Meanwhile, in the above-described embodiment, a structure bonded by hybrid bonding was exemplified as an example of a processing object (1) to which the method for dividing the processing object (1) of the present invention is applied. However, the processing object to which the method for dividing the processing object (1) according to the embodiment is applied may be modified in various ways as long as it is a structure in which a pattern layer (30) is arranged between a first wafer layer (10) and a second wafer layer (20). For example, as shown in FIGS. 10A and 10B, the processing objects (1A, 1B) to which the method for dividing the processing object (1) according to the embodiment is used may be a structure in which a thin pattern layer (30A) or a single pattern layer, for example, a first pattern layer (31), is arranged between the first wafer layer (10) and the second wafer layer (20).

[0069] In addition, in Fig. 7, for convenience, the second laser beam (L2) is illustrated as being irradiated from the lower portion of the processing target (1), but the irradiation direction of the second laser beam (L2) is not necessarily limited to this. For example, the processing target (1) may be turned over and the second laser beam (L2) may be irradiated from the upper portion, if necessary.

[0070]

[0071] Although embodiments of the present invention have been described above, these are merely exemplary, and those skilled in the art will understand that various modifications and equivalent other embodiments are possible.

Claims

1. A method for dividing a processing target having a pattern layer disposed between a first wafer layer and a second wafer layer, A step of forming a processing groove extending in the depth direction from one surface of the processing target to at least the pattern layer; A step of forming a modified area inside the processing object by focusing a laser beam at a location spaced from the other surface of the processing object; and A method for dividing a processing object, comprising the step of applying an external force to the processing object to form a crack extending along the modified area.

2. In paragraph 1, The step of forming the above processing groove is: A method for dividing a processing object by irradiating a first laser beam to form a processing groove extending from the surface of the second wafer layer to at least the pattern layer.

3. In paragraph 2, A method for dividing a processing object, wherein the depth of the processing groove is equal to or greater than the sum of the thickness of the second wafer layer and the thickness of the pattern layer.

4. In paragraph 2, A method for dividing a processing object, wherein the frequency of the first laser beam is 1.5 MHz to 2.5 MHz.

5. In paragraph 2, The intensity profile of the first laser beam has a Gaussian distribution, A method for dividing a processing object, wherein the processing groove formed by the first laser beam has a shape in which the width becomes narrower as the depth increases from the surface of the second wafer layer.

6. In paragraph 1, The step of forming the above modified area is: A method for dividing a processing object by irradiating a second laser beam to form a modified region inside the first wafer layer.

7. In paragraph 6, The step of forming the above modified area is: A method for dividing a processing object, wherein the second laser beam is irradiated so that the modified area overlaps the processing groove along the extension direction of the processing groove at a location spaced apart from the inside of the first wafer layer.

8. In paragraph 1, A method for dividing a processing object, wherein the thickness of the second wafer layer is smaller than the thickness of the first wafer layer on which the modified area is formed.

9. In paragraph 1, A method for dividing a processing target, wherein the thickness of the second wafer layer is 150 ㎛ or less.

10. In paragraph 1, A method for dividing a processing target, wherein the thickness of the first wafer layer is 1000 ㎛ or less.

11. In paragraph 1, In the step of forming the above crack, A method for dividing a processing object, wherein a crack is formed extending from the modified area to the processing groove within the first wafer layer.

12. In paragraph 1, The above pattern layer is, A first pattern layer including a first conductor region and a first dielectric region, A method for dividing a processing object, comprising a second pattern layer including a second conductive region bonded to the first conductive region and a second dielectric region bonded to the first dielectric region.

13. In paragraph 12, A method for dividing a processing object, wherein the first pattern layer and the second pattern layer are bonded by a hybrid bonding method.

14. In paragraph 12, The first pattern layer is arranged on one side of the first wafer layer facing the second wafer layer, A method for dividing a processing object, wherein the second pattern layer is arranged on a surface of the second wafer layer facing the first wafer layer.

15. In paragraph 1, A method for dividing a processing object, wherein the thickness of the above pattern layer exceeds 30 ㎛.

Images