Manufacturing method of stacked device
By forming a cutting part on the retaining sheet, batch stacking of chip areas is achieved, solving the problems of cumbersome processes and insufficient bonding strength in the prior art, and significantly simplifying the manufacturing process of stacking devices.
Patent Information
- Application Number
- CN202380077856.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2022-11-11
- Filing Date
- 2023-09-29
- Publication Date
- 2025-06-20
AI Technical Summary
The existing COW and WOW methods are cumbersome in chip stacking manufacturing, especially in the thinning and heat treatment process, which leads to complex manufacturing processes.
Using a new stacking device manufacturing method, the process flow is simplified by forming cut portions on the first and second holding sheets so as to stack multiple chip areas in one go in the bonding step.
The manufacturing process of stacking devices is significantly simplified, the number of stacking operations is reduced, and the insufficient bonding strength and shear stress problems during heat treatment and grinding are avoided.
Smart Images
Figure CN120188262A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to a manufacturing technology for stacked devices formed by stacking multiple chips. Background Art
[0002] In the field of semiconductor technology, as stacking technologies for three-dimensionally stacking chips such as LSIs (Large Scale Integration), various methods such as the COW (Chip on Wafer) method and the WOW (Wafer on Wafer) method have been proposed.
[0003] In the COW method (for example, refer to Patent Documents 1 and 3), first, a semiconductor wafer is subjected to dicing processing to singulate (monolithicize) the semiconductor wafer into multiple chips. Then, the multiple chips are sequentially stacked on a base wafer. After the chips are stacked to the required number of stacks, the base wafer is subjected to dicing processing to singulate the base wafer into multiple bases (the parts that serve as the basis for the stacked device), thereby completing a stacked device in which multiple chips are stacked on the base.
[0004] In the WOW method (for example, refer to Patent Documents 2 and 3), first, semiconductor wafers are stacked on a base wafer. After the semiconductor wafers are stacked to the required number of stacks, the wafer stack is subjected to dicing processing to singulate the wafer stack into multiple stacked devices.
[0005] Prior Art Documents
[0006] Patent Documents
[0007] Patent Document 1: Japanese Patent Publication No. 5853754
[0008] Patent Document 2: Japanese Patent Publication No. 6391999
[0009] Patent Document 3: Japanese Patent Publication No. 6485897 Summary of the Invention
[0010] Technical Problem to be Solved by the Invention
[0011] However, in the existing COW method and WOW method, there are problems of complicated manufacturing processes as described below.
[0012] Assuming a case where there are N regions (base regions) on the base wafer that serve as the bases for the stacked devices, in the COW method, it is necessary to stack chips on each of the N base regions one by one. Therefore, when stacking the first layer of chips on the base wafer, N stacking operations are required. When it is necessary to stack up to the Mth layer of chips on the base wafer, N×M stacking operations are required.
[0013] In addition, in the WOW method, when it is necessary to gradually stack thinned semiconductor wafers on a base wafer, if one wants to stack the thinned semiconductor wafer on the base wafer, the operation is technically difficult to achieve. Therefore, in the prior art, the semiconductor wafer before thinning is bonded (stacked) to the base wafer, and then the semiconductor wafer is thinned by grinding. However, during grinding, since shear stress is generated at the bonding surface between the wafers, if the bonding strength is weak, the bonding (stacking) between the wafers may be damaged. For this reason, heat treatment of the entire stack is required before grinding to increase the bonding strength. Therefore, when stacking thinned semiconductor wafers on the base wafer to the required number of stacks, for each layer of semiconductor wafer stacked, heat treatment for increasing the bonding strength is required before grinding.
[0014] For this reason, an object of the present invention is to significantly simplify the manufacturing process in the manufacturing technology of a stacked device formed by stacking multiple chips.
[0015] Technical means for solving the problem
[0016] The manufacturing method of the stacked device of the present invention performs a first forming step, a second forming step, and a bonding step. In the first forming step, in a state where a first bonding object provided with a plurality of chip regions is held by a first holding sheet, a cutting portion for singulating each of the plurality of chip regions is formed in the first bonding object. In the second forming step, in a state where a second bonding object provided with a plurality of chip regions is held by a second holding sheet, a cutting portion for singulating each of the plurality of chip regions is formed in the second bonding object. In the bonding step, the first bonding object and the second bonding object are bonded in such a manner that their chip regions are stacked in a predetermined positional relationship by moving the first holding sheet and the second holding sheet relative to each other.
[0017] By adopting the above manufacturing method, all the chip regions provided in the second bonding object can be batch-bonded (stacked) to all the chip regions provided in the first bonding object through one bonding. Therefore, compared with the case of stacking the chip regions singulated from the second bonding object one by one on the first bonding object, the manufacturing process of the stacked device can be significantly simplified. Specifically, when each of the first bonding object and the second bonding object is provided with N chip regions, when stacking the chip regions singulated from the second bonding object one by one on the N chip regions of the first bonding object, N stacking operations are required, while only one bonding is required by adopting the above manufacturing method, so the manufacturing process of the stacked device can be significantly simplified.
[0018] Advantages of the invention
[0019] According to the present invention, it is possible to significantly simplify the manufacturing process in the manufacturing technology of a stacked device formed by stacking a plurality of chips. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] Figure 1 FIG. is a perspective view conceptually showing a stacked device obtained by the manufacturing method of the embodiment.
[0021] Figure 2 FIG. is a plan view showing an example of a chip region provided in a chip base material (bonding object) indicated by a one-dot chain line.
[0022] Figure 3 FIG. is a conceptual diagram showing the first forming step.
[0023] Figure 4 FIG. is Figure 3 an enlarged perspective view of a Ya region portion shown in the upper left of FIG.
[0024] Figure 5 (A) of FIG. and Figure 5 (B) of FIG. are conceptual diagrams showing an enlarged longitudinal section at a Za-Za line shown in the lower left of FIG. in the thickness direction, Figure 3 (A) of FIG. shows a case where a groove is formed as a cutting portion, Figure 5 (B) of FIG. shows a case where a cavity is formed as a cutting portion. Figure 5
[0025] Figure 6 FIG. is a conceptual diagram showing the second forming step.
[0026] Figure 7 Figure 6 FIG. is an enlarged perspective view of a Yb region portion shown in the upper left of FIG.
[0027] Figure 8 Figure 8 (A) of FIG. is a conceptual diagram showing a bonding step, Figure 8 (B) of FIG. is a conceptual diagram showing an enlarged longitudinal section at a Zb-Zb line shown in (A) of FIG. in the thickness direction. (A) of FIG. is a conceptual diagram showing a repeated bonding step,
[0028] (B) of FIG. is a conceptual diagram showing an enlarged longitudinal section at a Zc-Zc line shown in (A) of FIG. in the thickness direction. Figure 9 (A) of FIG. is a conceptual diagram showing a repeated bonding step, Figure 9 (B) of FIG. is a conceptual diagram showing an enlarged longitudinal section at a Zc-Zc line shown in (A) of FIG. in the thickness direction. Figure 9 (A) of FIG. is a conceptual diagram showing a repeated bonding step,
[0029] Figure 10 FIG. is a conceptual diagram showing a transfer step.
[0030] Figure 11 FIG. is a conceptual diagram showing a stretching step.
[0031] Figure 12 This is a conceptual diagram showing the second process.
[0032] Figure 13 This is a conceptual diagram showing, in an enlarged view in the thickness direction, a longitudinal cross-section of an example of a stack of chip substrates obtained through the first bonding step in the second modification. Detailed Description of the Embodiment
[0033] [1] Embodiment
[0034] [1-1] Structure of the Stacked Device
[0035] Figure 1 This is a perspective view conceptually showing the stacked device 10 obtained through the manufacturing method of the embodiment. As Figure 1 shown, the stacked device 10 has a substrate 11 and a chip stack 12. Here, the substrate 11 is the part that serves as the basis of the stacked device 10, and external terminals, wirings, etc. are formed on the substrate 11. In addition, the chip stack 12 is the part composed of a plurality of chips Mc three-dimensionally stacked on the substrate 11. In Figure 1 the example, a case where the chips Mc are three-dimensionally stacked from the first layer to the seventh layer on the substrate 11 is shown. The chip Mc is not particularly limited and is, for example, a semiconductor chip such as an LSI.
[0036] In addition, the substrate 11 is not limited to providing terminals and wirings, and circuits the same as or different from those of the chips Mc may also be formed. Further, among the chips Mc constituting the chip stack 12, chips having the same circuit or chips having different circuits may be included.
[0037] In the present embodiment, all the chips Mc have the same rectangular (including square) shape and the same size. Regarding the substrate 11, its shape is rectangular (including square) and its size is larger than that of the chips Mc. In addition, the shape and size are not limited thereto and can be appropriately changed. The shapes of the substrate 11 and the chips Mc can be changed to, for example, circular. Further, as shown in the second modification described later (refer to Figure 13 ), chips Mc having different sizes may also be stacked on the substrate 11.
[0038] [1-2] Manufacturing Method of the Stacked Device
[0039] In the manufacturing method of the present embodiment, as manufacturing processes, a first process for manufacturing the chip stack 12 in the stacked device 10 and a second process for completing the stacked device 10 using the chip stack 12 obtained through the first process are executed. In addition, these processes can be implemented using various known devices. Further, the manufacturing method described below is also applicable to the case of manufacturing the stacked device 10 in which chips Mc are stacked with a stacking number (two or more layers) not limited to seven layers (refer to Figure 1 ).
[0040] [1-2-1] First process
[0041] In the first process, two or more chip base materials Gc (refer to Figure 2 ) are used, and a chip stack 12 is manufactured by joining (stacking) them. In each chip base material, a plurality of regions (chip regions Rc) that become chips Mc are provided. Figure 2 FIG. is a plan view showing an example of the chip region Rc provided in the chip base material Gc, indicated by a one-dot chain line. The chip base material Gc is not particularly limited and is, for example, a semiconductor wafer.
[0042] In addition, in the present embodiment, in order to stack the chip regions Rc with a predetermined positional relationship when joining the chip base materials Gc, positioning marks Pt for determining the positional relationship with each other are provided in these chip base materials Gc. Here, the predetermined positional relationship is a positional relationship in which the circuits each having the stacked chip regions Rc can function properly in the stacked device 10 (a positional relationship considering the connection positions of the circuits with each other, etc.).
[0043] In the first process of the present embodiment, a thinning step, a first forming step, a second forming step, a joining step, a transfer step, and a stretching step are performed. Each step will be specifically described below. In addition, hereinafter, the two chip base materials Gc that are the objects of joining in the first joining (stacking) will be referred to as "first joining object Gc1" and "second joining object Gc2", respectively.
[0044] <Thinning step>
[0045] In the thinning step, the first joining object Gc1 and the second joining object Gc2 are thinned by performing a grinding process on them. As an example, the first joining object Gc1 and the second joining object Gc2 are thinned to a thickness of 150 μm or less. In the present embodiment, the thinning of the first joining object Gc1 is performed before the first forming step described below. In addition, the thinning of the second joining object Gc2 is performed before the second forming step described later.
[0046] Here, if thinning of at least any one of the first bonding object Gc1 and the second bonding object Gc2 is performed after bonding (stacking) as in the prior art, shear stress will be generated at their bonding surface during grinding. Therefore, when the bonding strength is weak, the stack may be damaged. For this reason, in the prior art method, in order to perform thinning, it is necessary to perform heat treatment on the entire stacked body before grinding to improve the bonding strength. Thus, in the case of bonding (stacking) three or more chip substrates Gc, it is necessary to perform bonding (stacking) two or more times, and heat treatment for improving the bonding strength needs to be performed before each bonding. On the other hand, by performing the above thinning step before bonding (stacking) as in the present embodiment, heat treatment does not need to be performed every time of bonding, and the manufacturing process is simplified.
[0047] <First forming step>
[0048] Figure 3 is a conceptual diagram showing the first forming step. In the first forming step, the first bonding object Gc1 thinned by the thinning step is held by the first holding sheet 21 ( Figure 3 the upper left and right views). Here, the first holding sheet 21 is a sheet mainly composed of at least a resin (such as polyolefin, silicone, etc.) or rubber (such as fluororubber, etc.) having an adsorption force, and can hold the first bonding object Gc1 by the adsorption force of the main component. In the present embodiment, the first holding sheet 21 also has the following structure.
[0049] Figure 4 is in the first holding sheet 21 Figure 3 an enlarged three-dimensional view of the Ya region portion shown in the upper left view. As shown in this figure, in the first holding sheet 21, a plurality of convex portions 21b are provided on the holding surface 21a for holding the first bonding object Gc1. In addition, the convex portions 21b are formed of the same component as the above main component constituting the first holding sheet 21 and are integrally formed with the first holding sheet 21. Thus, by the first bonding object Gc1 being in surface contact with the plurality of convex portions 21b, the first bonding object Gc1 is held on the first holding sheet 21 by the adsorption force of the convex portions 21b.
[0050] On the other hand, when the first bonding object Gc1 is held on the holding surface 21a, only the portion of the holding surface 21a where the convex portions 21b are provided is in surface contact with the first bonding object Gc1. Figure 4 In the example of, the convex portions 21b are formed such that the contact surface 21c with the first bonding object Gc1 is rectangular (including square). In addition, the shape of the contact surface 21c of the convex portions 21b may be appropriately changed to various shapes including polygons, circles, etc. as long as it can hold the first bonding object Gc1, and is not limited to a rectangle.
[0051] By using such a first holding piece 21, the contact area with the first joining object Gc1 can be adjusted by adjusting the number of convex portions 21b and the area of the contact surface 21c. As a result, the holding force of the first holding piece 21 on the first joining object Gc1 (i.e., the total sum of the adsorption forces of the convex portions 21b) can be adjusted. Moreover, by adjusting the holding force of the first holding piece 21, the holding force required for the first process can be maintained, and after joining (stacking) the first joining object Gc1 and the second joining object Gc2 in the joining step described later, it is easy to peel off the first holding piece 21 from the first joining object Gc1. Specifically, the holding force of the first holding piece 21 can be adjusted to an appropriate size to avoid damage to the joining portion when peeling off the first holding piece 21 from the first joining object Gc1. Therefore, by using such a first holding piece 21 to hold the first joining object Gc1, the first joining object Gc1 can be firmly held, and when peeling or transferring is required, it can be easily peeled or transferred without causing adverse effects (such as damage) to the joining portion or the like. Thus, even the thinned first joining object Gc1 is easy to operate.
[0052] In the first forming step, after holding the first joining object Gc1 by the first holding piece 21, in this state, a dicing portion Ec ( Figure 3 in the lower left figure) for singulating each of the plurality of chip regions Rc is formed in the first joining object Gc1. In the present embodiment, at least any one of a groove, a hole, and an internal defect is formed as the dicing portion Ec, so that the first joining object Gc1 can be broken when the stretchable sheet 23 is stretched in the stretching step described later (refer to Figure 11 ). Here, when it is intended to break an object (here, the first joining object Gc1) at a specified line (hereinafter, this specified line is referred to as the "predetermined breaking line"), the groove is a bottomed groove formed along the predetermined breaking line, and the hole is a through hole formed at multiple positions at the position of the predetermined breaking line. In addition, the internal defect is a defect such as a cavity, a crack, or a modified layer formed inside the object by using a laser at the position of the predetermined breaking line. Such an internal defect can be formed by focusing a laser inside the object. In addition, the technique of breaking an object by using an internal defect is called an invisible dicing technique.
[0053] Figure 5 The (A) of Figure 3 is a conceptual diagram showing an enlarged view in the thickness direction of the longitudinal section at the Za-Za line shown in the lower left figure of Figure 5 , and the (A) of
[0054] In addition, as the cutting portion Ec, holes or internal defects may be formed instead of the grooves. Figure 5 (B) of Figure 5 shows a case where a cavity as an internal defect is formed as the cutting portion Ec.
[0055] With such a structure of the cutting portion Ec, adjacent chip regions Rc remain connected, so that the position of the chip region Rc is not easily shifted during the conveyance and bonding of the first bonding object Gc1.
[0056] <Second forming step>
[0057] Figure 6 is a conceptual diagram showing the second forming step. In the second forming step, the second bonding object Gc2 thinned in the thinning step is held by the second holding sheet 22 ( Figure 6 the upper left and right figures of Figure 6 ). Here, similar to the first holding sheet 21, the second holding sheet 22 is a sheet mainly composed of at least a resin (such as polyolefin, silicone, etc.) or rubber (such as fluororubber, etc.) having an adsorption force, and can hold the second bonding object Gc2 by the adsorption force of the main component. In the present embodiment, the second holding sheet 22 further has the following structure.
[0058] Figure 7 is an enlarged perspective view of a part of the Yb region shown in the upper left figure of Figure 6 in the second holding sheet 22. As shown in this figure, in the second holding sheet 22, a plurality of convex portions 22b are provided on the holding surface 22a for holding the second bonding object Gc2. In addition, the convex portions 22b are made of the same component as the above-mentioned main component constituting the second holding sheet 22 and are integrally formed with the second holding sheet 22. Then, by the surface contact between the second bonding object Gc2 and the plurality of convex portions 22b, the second bonding object Gc2 is held on the second holding sheet 22 by the adsorption force of the convex portions 22b.
[0059] On the other hand, when the second bonding object Gc2 is held on the holding surface 22a, only the portion of the holding surface 22a where the convex portions 22b are provided is in surface contact with the second bonding object Gc2. Figure 7 In the example of Figure 7 , the convex portions 22b are formed such that the contact surface 22c with the second bonding object Gc2 is rectangular (including square). In addition, the shape of the contact surface 22c of the convex portions 22b may be appropriately changed to various shapes including polygons, circles, etc. as long as the second bonding object Gc2 can be held.
[0060] By using such a second holding piece 22, the contact area with the second joining object Gc2 can be adjusted by adjusting the number of convex portions 22b and the area of the contact surface 22c. As a result, the holding force of the second holding piece 22 on the second joining object Gc2 (i.e., the total sum of the adsorption forces of the convex portions 22b) can be adjusted. Moreover, by adjusting the holding force of the second holding piece 22, the holding force required for the first process can be maintained, and after joining (stacking) the first joining object Gc1 and the second joining object Gc2 through the joining step described later, it is easy to peel off the second holding piece 22 from the second joining object Gc2. Specifically, the holding force of the second holding piece 22 can be adjusted to an appropriate size to avoid damage to the joining portion when peeling off the second holding piece 22 from the second joining object Gc2. Therefore, by using such a second holding piece 22 to hold the second joining object Gc2, the second joining object Gc2 can be firmly held, and when peeling or transferring is required, peeling or transferring can be easily performed without causing adverse effects (such as damage) to the joining portion or the like. Thus, even the thinned second joining object Gc2 is easy to operate.
[0061] Furthermore, in the present embodiment, in the first holding piece 21 and the second holding piece 22 (refer to Figure 4 and Figure 7 ), on their holding surfaces 21a and 22a, the convex portions 21b and 22b are formed such that the contact area between the two when the second holding piece 22 holds the second joining object Gc2 is smaller than the contact area between the two when the first holding piece 21 holds the first joining object Gc1. With such a structure, by forming a difference in the size of the contact area, a difference in the holding force can be generated, and as a result, a difference in the peeling strength can also be generated (specifically, the smaller the contact area, the easier it is to peel). Therefore, in the case of peeling off the second holding piece 22 after joining (stacking) the first joining object Gc1 and the second joining object Gc2 through the joining step described later, the above-mentioned difference in the holding force (here, the difference in the peeling strength) can be utilized to easily peel off only the second holding piece 22.
[0062] In the second forming step, after holding the second joining object Gc2 by the second holding piece 22, in this state, in the same manner as in the first forming step, a cutting portion Ec for singulating each of the plurality of chip regions Rc is formed in the second joining object Gc2 ( Figure 6 lower left figure). Specifically, as the cutting portion Ec, at least any one of a groove, a hole, and an internal defect is formed in the same manner as the cutting portion Ec formed in the first joining object Gc1 (refer to Figure 5 (A) of Figure 5 and Figure 11) When the stretchable sheet 23 is stretched, the second bonding object Gc2 can be broken. With such a structure of the cutting portion Ec, the adjacent chip regions Rc remain connected, so the position of the chip region Rc is not easily offset during the transportation and bonding of the second bonding object Gc2.
[0063] <Joining steps>
[0064] Figure 8 (A) is a conceptual diagram showing the joining step. After both the first forming step and the second forming step are performed, in the joining step, the first joining object Gc1 and the second joining object Gc2 are joined in a manner in which their chip regions Rc are stacked in a prescribed positional relationship by moving the first retaining piece 21 and the second retaining piece 22 relative to each other.
[0065] In this embodiment, the first and second joining objects Gc1 and Gc2 are joined (stacked) by relatively moving the first and second holding pieces 21 and 22 so that the positioning marks Pt provided on the first and second joining objects Gc1 and Gc2 are aligned.
[0066] Figure 8 (B) is to Figure 8 A conceptual diagram showing an enlarged longitudinal section at the Yb-Yb line shown in (A) of FIG. 1 . As shown in the figure, according to the above-mentioned bonding step, the chip regions Rc are stacked in a predetermined positional relationship, so that the corresponding parts of the cut portions Ec formed in the first bonding object Gc1 and the second bonding object Gc2 are in the top view (viewed from above). Figure 8 (B) Align with the paper surface.
[0067] By adopting such a process, all chip regions Rc provided on the second bonding object Gc2 can be batch bonded (stacked) to all chip regions Rc provided on the first bonding object Gc1 through one bonding. Therefore, compared with the case where the chip regions Rc obtained by cutting out the second bonding object Gc2 are stacked one by one on the first bonding object Gc1, the manufacturing process of the stacked device 10 can be significantly simplified. Specifically, in the case where the first bonding object Gc1 and the second bonding object Gc2 are each provided with N chip regions Rc, when the chip regions Rc obtained by cutting out the second bonding object Gc2 are stacked one by one on the N chip regions Rc of the first bonding object Gc1, N stacking operations are required, while the above-mentioned process of the present embodiment only requires one bonding, so the manufacturing process of the stacked device 10 can be significantly simplified.
[0068] <Thinning step to bonding step repeated>
[0069] When the stacked device 10 to be manufactured is a device in which three or more chips Mc are stacked on the substrate 11 as in the present embodiment (see Figure 1 ), three or more chip base materials Gc need to be joined. In this case, before entering the transfer step described later, it is necessary to repeatedly execute the above-described thinning step, second forming step, and joining step according to the number of chip base materials Gc to be joined. The case where the thinning step to the joining step are repeatedly performed will be specifically described below.
[0070] After the above-described joining step, the second holding sheet 22 is peeled off from the second joining object Gc2. In the present embodiment, as described above, the holding force of the second holding sheet 22 is adjusted so that the joining portion between the first joining object Gc1 and the second joining object Gc2 is not damaged when the sheet is peeled off. Therefore, the above-described joining portion is hardly damaged due to the peeling of the second holding sheet 22. Further, in the first holding sheet 21 and the second holding sheet 22, on their holding surfaces 21a and 22a, the convex portions 21b and 22b are formed such that the contact area between the two when the second holding sheet 22 holds the second joining object Gc2 is smaller than the contact area between the two when the first holding sheet 21 holds the first joining object Gc1. Therefore, there is a difference in the holding force of the first holding sheet 21 and the second holding sheet 22 with respect to the joining object, and as a result, there is also a difference in the peeling strength (specifically, the second holding sheet 22 with a small contact area is more easily peeled off). Thus, in the present embodiment, it is possible to easily peel off only the second holding sheet 22 by utilizing such a difference in holding force (here, a difference in peeling strength).
[0071] Thereafter, another chip base material Gc is joined to the second joining object Gc2. Specifically, the first joining object Gc1 and the second joining object Gc2 joined in the above-described joining step (see Figure 8 (A)) are used as the new first joining object Gc1, and the above-described another chip base material Gc is used as the new second joining object Gc2, and then the thinning of the second joining object Gc2 and the above-described second forming step (see Figure 6 ) and joining step (see Figure 9 (A) and Figure 9 (B)) are executed.
[0072] Then, according to the number of chip base materials Gc to be joined, the peeling of the second holding sheet 22, the thinning of the second joining object Gc2, the second forming step (see Figure 6 ) and the joining step (see Figure 9 (A)) are repeatedly executed. In the present embodiment, since seven chip base materials Gc need to be joined, after the first joining step (see Figure 8 (A)), in order to further join the chip base materials Gc from the third layer to the seventh layer, the above-described repeated process is performed five times.
[0073] With such a repetitive process, even when the chip base material Gc (the object to be joined) is joined (stacked) to the first joining object Gc1 multiple times as the second joining object Gc2, for each such chip base material Gc, all the chip regions Rc provided on the chip base material Gc can be batch-joined (stacked) by a single joining operation. Specifically, when N chip regions Rc are provided on each of the first joining object Gc1 and another chip base material Gc (the second joining object Gc2), when the chip regions Rc singulated from the other chip base material Gc (the second joining object Gc2) are stacked three-dimensionally one by one on the N chip regions Rc of the first joining object Gc1 M times, N×M stacking operations are required, while only M joining operations are needed with the above-mentioned repetitive process. Therefore, the manufacturing process of the stacked device 10 can be significantly simplified.
[0074] <Transfer Step>
[0075] Figure 10 It is a conceptual diagram showing the transfer step. After joining the required number of chip base materials Gc, in the transfer step, the second holding sheet 22 is peeled off from the base material stack Hm which is a stack of the chip base materials Gc, and then the base material stack Hm is transferred to the stretchable sheet 23 which is another holding sheet. Additionally, in the transfer step, the first holding sheet 21 can also be peeled off from the base material stack Hm and then transferred to the stretchable sheet 23. Here, as described above, the holding forces of the first holding sheet 21 and the second holding sheet 22 (the holding forces on the joining objects) are adjusted to such magnitudes that the joining portions between the first joining object Gc1 and the second joining object Gc2 (here, it refers to the joining portions formed within the base material stack Hm) are not damaged when these sheets are peeled off. Therefore, almost no damage is caused to the above-mentioned joining portions due to the peeling of the first holding sheet 21 or the second holding sheet 22 (including the peeling required during transfer).
[0076] Here, the stretchable sheet 23 uses a sheet with an area larger than that of the chip base material Gc so that its end portions can be grasped and stretched. In addition, as the stretchable sheet 23, in order to make the sheet stretchable, a sheet mainly composed of a resin (such as polyolefin, silicone, etc.) or rubber (such as fluororubber, etc.) which has stretchability in addition to adsorption force can be used. At this time, on the stretchable sheet 23, similar to the first holding sheet 21 and the second holding sheet 22, a plurality of convex portions are provided on the surface to which the base material stack Hm is to be transferred.
[0077] <Stretching Step>
[0078] Figure 11 It is a conceptual diagram showing the stretching step. After the transfer step, in the stretching step, by stretching the stretchable sheet 23, the chip regions Rc adjacent to each other across the cutting portion Ec are separated.
[0079] By using the above-described transfer step and stretching step, through a simple single process of stretching the stretchable sheet 23, the plurality of chip base materials Gc (bonding objects) joined at the cutting portion Ec can be broken. Thereby, the plurality of chip regions Rc provided on the plurality of chip base materials Gc (bonding objects) can be singulated (cut into individual pieces) in a bonded (stacked) state, group by group (in this embodiment, one group includes seven layers of chip regions Rc), to form the chip stack 12.
[0080] By adopting such a first process, a remarkable simplification of the manufacturing process can be achieved in the manufacturing technology of the stacked device 10.
[0081] [1-2-2] Second process
[0082] In the second process, using a base material Gb provided with a plurality of regions (base regions Rb) serving as the base 11 and the chip stack 12 obtained by the first process, the Figure 1 stacked device 10 shown is completed. Here, the base material Gb is not particularly limited and is, for example, a semiconductor wafer.
[0083] Figure 12 is a conceptual diagram showing the second process. In the second process, the chip stacks 12 are joined one by one to the plurality of base regions Rb provided on the base material Gb in a predetermined positional relationship. At this time, the base material Gb can be held by the holding sheet 24. The holding sheet 24 can use the same sheet material as the first holding sheet 21 and the second holding sheet 22. Then, after the joining of the chip stacks 12 is completed in all the above base regions Rb, by performing a cutting process on the base material Gb ( Figure 12 in the example, the base material Gb is cut at the position of the single dotted line), the plurality of base regions Rb are singulated (cut into individual pieces). Thereby, the stacked device 10 is completed (refer to Figure 1 ).
[0084] [2] Variation
[0085] [2-1] First variation
[0086] In the first process of the above embodiment, instead of performing the transfer step, either the first holding sheet 21 or the second holding sheet 22 can be used as the stretchable sheet.
[0087] Specifically, at least one of the first holding piece 21 and the second holding piece 22 is a stretchable piece, and the area is larger than that of the chip base material Gc. In addition, in the above-described embodiment, when the first holding piece 21 and the second holding piece 22 are mainly composed of a resin (such as polyolefin, silicone, etc.) or rubber (such as fluororubber, etc.) that has stretchability in addition to adsorption force, these sheets inherently have stretchability in addition to adsorption force. In this case, it is only necessary to make the area of at least one of the sheets larger than the area of the chip base material Gc. Then, retain any one of the first holding piece 21 and the second holding piece 22 that is the stretchable piece and peel off the other holding piece, and then separate the chip regions Rc adjacent to each other across the cutting portion Ec by stretching the holding piece that is not peeled off and retained.
[0088] With such a structure, the first process can be further simplified.
[0089] [2-2] Second modification
[0090] In the above-described embodiment, the stacked device 10 to be manufactured can also be appropriately changed to a device in which chips Mc of different sizes are stacked on the substrate 11. In this case, in the first process of the above-described embodiment, by appropriately changing the width of the cutting portion Ec formed on the chip base material Gc in the first forming step and the second forming step, the sizes of the stacked chips Mc can be made different.
[0091] Figure 13 It is a conceptual diagram showing an example of a stack of chip base materials Gc obtained by the first bonding step in this modification, with its longitudinal section magnified in the thickness direction. Figure 13 The example shows a case where, in order to make the size of the chip Mc obtained from the second bonding object Gc2 smaller than the size of the chip Mc obtained from the first bonding object Gc1, the width of the groove (cutting portion Ec) formed in the second bonding object Gc2 is set to be larger than the width of the groove (cutting portion Ec) formed in the first bonding object Gc1.
[0092] [2-3] Other modifications
[0093] In the above-described embodiment, a dicing tape (scribing tape) can be appropriately used for the first holding piece 21 and the second holding piece 22.
[0094] In the above-described embodiment, the cutting portion Ec can also be appropriately changed to a structure formed by completely cutting the chip base material Gc using a blade, laser, plasma, etc.
[0095] The descriptions of the above embodiments and variations are examples in all respects and should not be construed as restrictive. The scope of the present invention is not limited by the above embodiments or variations. Moreover, the scope of the present invention is intended to include all changes within the meaning and scope equivalent to the claimed technical solution.
[0096] In addition, from the above embodiments and variations, as objects of the invention, several steps constituting the manufacturing method of the stacked device 10 can be partially extracted, or each step can be extracted separately.
[0097] Description of Reference Numerals
[0098] 10 Stacked device
[0099] 11 Substrate
[0100] 12 Chip stack
[0101] 21 First holding sheet
[0102] 22 Second holding sheet
[0103] 21a, 22a Holding surfaces
[0104] 21b, 22b Protrusions
[0105] 21c, 22c Contact surfaces
[0106] 23 Stretchable sheet
[0107] 24 Holding sheet
[0108] Ec Cutting portion
[0109] Gb Substrate base material
[0110] Gc Chip base material
[0111] Hm Base material stack
[0112] Mc Chip
[0113] Pt Alignment mark
[0114] Rb Substrate region
[0115] Rc Chip region
[0116] Gc1 First bonding object
[0117] Gc2 Second bonding object
Claims
1. A manufacturing method of a stacked device, characterized in that, Perform the following steps: A first forming step of forming, in a state where a first bonding object provided with a plurality of chip regions is held by a first holding sheet, a cutting portion for singulating each of the plurality of chip regions in the first bonding object; A second forming step of forming, in a state where a second bonding object provided with a plurality of chip regions is held by a second holding sheet, a cutting portion for singulating each of the plurality of chip regions in the second bonding object; and A bonding step of bonding the first bonding object and the second bonding object in such a manner that their chip regions are stacked in a predetermined positional relationship by relatively moving the first holding sheet and the second holding sheet with respect to each other.
2. The manufacturing method of the stacked device according to claim 1, characterized in that: After the bonding step, the second holding sheet is peeled off from the second bonding object, and another bonding object provided with a plurality of chip regions is bonded to the second bonding object again, using the first bonding object and the second bonding object bonded by the bonding step as a new first bonding object, and using the other bonding object as a new second bonding object, and performing the second forming step and the bonding step again.
3. The manufacturing method of the stacked device according to claim 1 or 2, characterized in that: Before the second forming step, a thinning step of thinning the second bonding object is also performed.
4. The manufacturing method of the stacked device according to any one of claims 1 to 3, characterized in that, The following steps are also performed: A transfer step of peeling off the first holding sheet or the second holding sheet after the bonding step, and then transferring the first bonding object and the second bonding object to a stretchable sheet serving as another holding sheet; and A stretching step of separating chip regions adjacent to each other across the cutting portion by stretching the stretchable sheet.
5. The manufacturing method of the stacked device according to any one of claims 1 to 3, characterized in that: At least one of the first holding sheet and the second holding sheet is a stretchable sheet, The method for manufacturing the stacked device further performs the following steps: A stretching step of, after the bonding step, retaining either one of the first holding sheet and the second holding sheet that is the stretchable sheet and peeling off the other holding sheet, and then separating chip regions adjacent to each other across the cutting portion by stretching the retained holding sheet that is not peeled off.
6. The manufacturing method of the stacked device according to claim 4 or 5, characterized in that: In the first forming step and the second forming step, as the cutting portion, at least any one of a groove, a hole, and an internal defect for breaking the first bonding object and the second bonding object when the stretchable sheet is stretched in the stretching step is formed.
7. The manufacturing method of the stacked device according to any one of claims 1 to 6, characterized in that: Both the first holding sheet and the second holding sheet are sheets mainly composed of at least a resin or rubber having an adsorbing force, and a plurality of convex portions are provided on the holding surfaces for holding the first bonding object and the second bonding object, respectively.
8. The manufacturing method of the stacked device according to claim 7, characterized in that: On the holding surface, the convex portions are formed such that the contact area between the second holding sheet and the second bonding object when the second holding sheet holds the second bonding object is smaller than the contact area between the first holding sheet and the first bonding object when the first holding sheet holds the first bonding object.
Citation Information
Patent Citations
Eddy current flaw detecting method
JP1983053754A
Coaxial plasma gun
JP1988091999A
JP1989085897A