Semiconductor device manufacturing methods
The method of forming an annular fixing portion under reduced pressure addresses the high cost and waste issues in semiconductor substrate manufacturing, enabling efficient handling and reducing defects in high-temperature processes.
Patent Information
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- TATSUMO KK
- Filing Date
- 2022-12-14
- Publication Date
- 2026-05-26
AI Technical Summary
The challenge in manufacturing semiconductor substrates using wide bandgap materials is the high cost and waste associated with polishing, and the difficulty in bonding semiconductor wafers to supports during high-temperature processes without using costly CMP methods.
A method involving forming a divided layer in a semiconductor wafer, creating an annular fixing portion under reduced pressure, and utilizing a pressure difference to bond the substrate to a support, allowing separation and handling without full-surface bonding.
Facilitates easier handling of semiconductor substrates post-formation by providing sufficient bonding force without full-surface adhesion, reducing waste and costs, and minimizing stress-induced defects during high-temperature processes.
Smart Images
Figure 0007865871000001 
Figure 0007865871000002 
Figure 0007865871000003
Abstract
Description
Technical Field
[0001] The present invention relates to a manufacturing technique for semiconductor devices.
Background Art
[0002] As a manufacturing technique for semiconductor devices, there is a process that includes forming a semiconductor substrate with a desired thickness (for example, a thickness of 150 μm or less) by polishing a semiconductor wafer (for example, a wafer with a thickness of about 300 μm).
[0003] In recent years, wide bandgap (WBG) materials such as GaAs, SiC, GaN, AlN, BN, and diamond have been attracting attention as semiconductor materials that enable normal operation of semiconductor devices even at high temperatures reaching 300°C. The WBG materials are also semiconductor materials that enable an increase in switching speed, and in that regard, are expected as materials for semiconductor devices for high-speed communication. Also, the WBG materials are semiconductor materials having resistance to high electric fields, and in that regard, are expected as materials for semiconductor devices for high voltages. Thus, the WBG materials have advantages as materials for semiconductor devices in various aspects.
[0004] On the other hand, the WBG materials are expensive materials. Nevertheless, when a semiconductor substrate of the WBG material is formed by polishing a semiconductor wafer, the portion removed by polishing in the semiconductor wafer becomes wasted, leading to a significant increase in manufacturing cost.
[0005] Therefore, in order to prevent such waste, a technique has been proposed in which a split layer is formed in the semiconductor wafer by implanting hydrogen ions from the surface of the semiconductor wafer, and the semiconductor substrate is formed by splitting the semiconductor wafer at the split layer (see, for example, Patent Document 1). According to such a technique, a thin semiconductor substrate can be manufactured without wasting the semiconductor material constituting the semiconductor wafer.
Prior Art Documents
Patent Documents
[0006] [Patent Document 1] Japanese Patent Publication No. 2007-250576 [Overview of the project] [Problems that the invention aims to solve]
[0007] In the aforementioned technology for forming semiconductor substrates by hydrogen ion implantation, it is necessary to hold the portion that will become the semiconductor substrate with a support (such as a glass substrate) when the semiconductor wafer is divided into separate layers. On the other hand, the semiconductor substrate after separation from the semiconductor wafer is exposed to a high-temperature atmosphere while still held by the support during processes such as circuit formation on the semiconductor substrate. For this reason, it is necessary to bond the semiconductor wafer and the support in order to hold the semiconductor substrate with the support, but it is difficult to select an adhesive as the bonding method.
[0008] Conventionally, as a means of joining a semiconductor wafer and a support, a method (direct bonding method) has been used in which both bonding surfaces are subjected to CMP (Chemical Mechanical Polishing), and then the bonding surfaces are directly bonded together, thereby bonding the entire bonding surface through interatomic or intermolecular forces.
[0009] However, bonding using CMP had problems such as (1) the high cost of the CMP process and (2) the fact that the semiconductor wafer and the support are strongly bonded, so if it is desired to remove the support from the semiconductor substrate after separation from the semiconductor wafer, a process such as polishing is required to remove the support.
[0010] Therefore, the objective of the present invention is to facilitate the handling of the semiconductor substrate after its formation in a semiconductor device manufacturing technology that includes a process for forming a semiconductor substrate using segmented layers. [Means for solving the problem]
[0011] A method for manufacturing a semiconductor device according to the present invention comprises a divided layer formation step, a fixing portion formation step, and a splitting step. In the divided layer formation step, a divided layer is formed at a predetermined depth from the surface of a first substrate, which is mainly composed of a semiconductor, to enable the separation of the semiconductor substrate having a thickness equal to the predetermined depth from the first substrate. After the divided layer formation step, in the fixing portion formation step, a fixing portion for fixing the portion of the first substrate that will become the semiconductor substrate to the second substrate is formed in an annular shape in an atmosphere of lower pressure than a predetermined pressure, thereby sealing the inside of the annular shape. After the fixing portion formation step, in the splitting step, the first substrate is split by the divided layer in an atmosphere of the predetermined pressure.
[0012] According to the above manufacturing method, in the fixing part formation step, by forming an annular fixing part in an atmosphere with a pressure lower than a predetermined pressure (e.g., atmospheric pressure), a pressure difference can be created between the pressure inside the annular fixing part (internal pressure) and the pressure outside the annular fixing part (external pressure) in an atmosphere with a predetermined pressure (the atmosphere used in the splitting step). As a result, an adsorption force due to the pressure difference can be generated between the first substrate and the second substrate. Therefore, in an atmosphere with a predetermined pressure, the first substrate and the second substrate can be firmly bonded by the bonding force of the fixing part and the adsorption force due to the pressure difference between the internal and external pressures. Thus, even without bonding the first substrate and the second substrate across the entire bonding surface as in the conventional method (i.e., without performing bonding using CMP), sufficient bonding force can be obtained to hold the semiconductor substrate separated from the first substrate in the splitting step to the second substrate. [Effects of the Invention]
[0013] According to the present invention, in a semiconductor device manufacturing technology that includes a process for forming a semiconductor substrate using segmented layers, the handling of the semiconductor substrate after its formation becomes easier. [Brief explanation of the drawing]
[0014] [Figure 1]It is a conceptual diagram showing the manufacturing method according to the embodiment in the order of processing. [Figure 2] It is a conceptual diagram showing the subsequent processing of FIG. 1 in the order of processing. [Figure 3] It is a plan view showing the shape of the fixing portion formed in the embodiment. [Figure 4] It is a conceptual diagram showing a part of the manufacturing method according to the first modification example in the order of processing. [Figure 5] (A) It is a conceptual diagram showing the split layer formation step executed in the manufacturing method according to the second modification example, and (B) it is a conceptual diagram showing a comparative example. [Figure 6] It is a conceptual diagram showing a part of the manufacturing method according to the fourth modification example in the order of processing. [Figure 7] It is a plan view showing the shape of the fixing portion formed in the fourth modification example. [Figure 8] (A) It is a conceptual diagram and (B) it is a plan view showing a further modification example of the manufacturing method of FIGS. 6 and 7. [Figure 9] It is a plan view showing a first example of the shape of the fixing portion Q formed in the fifth modification example. [Figure 10] (A) It is a plan view showing a second example and (B) it is a plan view showing a third example of the shape of the fixing portion Q formed in the fifth modification example. [Figure 11] (A) It is a conceptual diagram showing the fixing portion formation step and the cutting step executed in the manufacturing method according to the sixth modification example, and (B) it is a plan view showing the shape of the fixing portion formed in the sixth modification example. [Figure 12] It is a conceptual diagram showing the cutting step and the peeling step executed in the manufacturing method according to the eighth modification example.
Mode for Carrying Out the Invention
[0015] Hereinafter, the manufacturing method of the semiconductor device according to the present invention will be specifically described with respect to its embodiments and modification examples. The manufacturing method described below can be realized using various well-known devices.
[0016] [1] Embodiment FIG. 1 and FIG. 2 are conceptual diagrams showing the manufacturing method according to the embodiment in the processing order. In this manufacturing method, a dividing layer forming step S1, a fixing portion forming step S2, a dividing step S3, a device forming step S4, a cutting step S5, a peeling step S6, and an individual piece forming step S7 are executed in this order. Hereinafter, each step will be specifically described.
[0017] <Dividing layer forming step S1> In the dividing layer forming step S1 (see FIG. 1), a dividing layer 111 is formed at a position with a predetermined depth Dt from the surface 11a of the first substrate 11 mainly composed of a semiconductor. Here, the first substrate 11 is not particularly limited, but for example, it is a single crystal semiconductor wafer, and wide bandgap (WBG) materials such as GaAs, SiC, GaN, AlN, BN, and diamond can be used for its main component (semiconductor). Further, the dividing layer 111 is a layer that enables separation of the semiconductor substrate K having a thickness Td corresponding to the predetermined depth Dt from the first substrate 11, and can be formed by implanting hydrogen ions from the surface 11a of the first substrate 11. According to the method of implanting hydrogen ions, the dividing layer 111 can be formed at a position at a depth of about 10 μm or less from the surface 11a, and as a result, a semiconductor substrate K with a thin thickness Td can be obtained.
[0018] Here, instead of the method of implanting hydrogen ions, a method of irradiating laser light with the position at a predetermined depth Dt from the surface 11a as the focus can be used for forming the dividing layer 111. According to this method, since the position of the focus can be changed in the depth direction, the dividing layer 111 can be formed at a desired depth position. Therefore, according to the method of irradiating laser light, the dividing layer 111 can be formed at a deeper position (for example, a position at a depth of about 50 to 150 μm from the surface 11a) than the method of implanting hydrogen ions, and as a result, a semiconductor substrate K with a large thickness Td can be obtained.
[0019] The first substrate 11 is divided by a dividing layer 111 in the dividing step S3 described later, and as a result, a semiconductor substrate K having a thickness Td equal to a predetermined depth Dt is separated from the first substrate 11. Here, when dividing the first substrate 11, it is necessary to hold the portion that will become the semiconductor substrate K with a support. Therefore, in the next fixing part formation step S2, the second substrate 12, which will serve as the support, is bonded to the surface 11a of the first substrate 11.
[0020] <Step S2 for forming the fixed portion> The second substrate 12 is not particularly limited, but for example, it may be a polycrystalline semiconductor wafer, and its main component (semiconductor) may be a wide-bandgap (WBG) material such as GaAs, SiC, GaN, AlN, BN, or diamond. Alternatively, instead of a semiconductor wafer, the second substrate 12 may be a substrate formed from another material not limited to semiconductors (such as Si, sapphire, or quartz).
[0021] In the fixing portion formation step S2, a fixing portion Q for fixing the portion of the first substrate 11 that will become the semiconductor substrate K to the second substrate 12 is formed in an annular shape in an atmosphere with a pressure lower than a predetermined pressure Pt, thereby sealing the inside of the annular portion. Here, the predetermined pressure Pt is the atmospheric pressure used in the division step S3 described later, and is not particularly limited, but for example it is atmospheric pressure.
[0022] Specifically, in the bonding step S2 (see Figure 1), the metal layer formation step S21 and the laser light irradiation step S22 are performed in this order.
[0023] In the metal layer formation step S21, a metal layer 13 is formed over the entire surface 11a of the first substrate 11. Although not particularly limited, the metal layer 13 is formed to a thickness of 1 μm or less using a film deposition method such as vapor deposition. Furthermore, metals such as Cu, Al, Cr, Ti, Ta, and Au can be used as the main component of the metal layer 13.
[0024] Subsequently, in the laser irradiation step S22, the first substrate 11 and the second substrate 12 are placed facing each other with the metal layer 13 interposed between them. In this state, the portion of the metal layer 13 where the fixing portion Q is to be formed (in this embodiment, the portion on the peripheral edge 112 of the first substrate 11; details will be described later (see Figure 3)) is heated by irradiating it with laser light. At this time, the first substrate 11 and the second substrate 12 may be sandwiched between quartz plates or the like to increase the degree of contact with the portion of the metal layer 13 where the fixing portion Q is to be formed.
[0025] By irradiating the metal layer 13 with laser light, the metal layer 13 can be melted together with the first substrate 11 and the second substrate 12 at the laser irradiation site (the area where the bonding portion Q is to be formed), or the metal, which is the main component of the metal layer 13, can be diffused into the first substrate 11 and the second substrate 12. As a result, compounds (such as metal silicides) or alloys (such as metal-Si alloys) of the main components of the first substrate 11 and the second substrate 12 (such as semiconductors) and metal are formed at the respective interfaces between the first substrate 11 and the second substrate 12 and the metal layer 13. This forms a bonding portion Q that firmly bonds the first substrate 11 (the part that will become the semiconductor substrate K) and the second substrate 12.
[0026] Furthermore, since such a fixed portion Q is formed continuously and in an annular shape under an atmosphere of pressure lower than a predetermined pressure Pt, the inside of the annular portion is sealed while maintaining a pressure lower than the predetermined pressure Pt.
[0027] Figure 3 is a plan view showing the shape of the adhesive portion Q formed in this embodiment. As shown in Figure 3, in this embodiment, the adhesive portion Q is formed around the entire circumference of the peripheral edge 112 of the first substrate 11 (specifically, the peripheral edge of the portion of the first substrate 11 that will become the semiconductor substrate K; see also Figure 1). In the example in Figure 3, the first substrate 11 is disc-shaped, and the adhesive portion Q is formed in an annular shape along its peripheral edge 112.
[0028] Furthermore, the shape of the fixing portion Q is not limited to an annular shape, but can be appropriately changed to another shape (such as a polygon) depending on the peripheral shape of the first substrate 11. Also, when the fixing portion Q is formed on the peripheral portion 112 of the first substrate 11 as in this embodiment, in the metal layer formation step S21, the metal layer 13 may not be formed on the entire surface 11a of the first substrate 11, but only on the region of the surface 11a on the peripheral portion 112, or only on the portion of the surface 11a where the fixing portion Q is to be formed. Moreover, in the metal layer formation step S21, instead of forming the metal layer 13 on the surface 11a of the first substrate 11, the metal layer 13 may be formed on the surface 12a of the second substrate 12. In this case as well, in the metal layer formation step S21, the metal layer 13 may not be formed over the entire surface 12a of the second substrate 12, but may be formed only in the region of the surface 12a that will face the peripheral portion 112, or the metal layer 13 may be formed only in the portion of the surface 12a where the fixing portion Q is to be formed.
[0029] According to this bonding step S2, by forming an annular bonding portion Q in an atmosphere with a pressure lower than a predetermined pressure Pt (e.g., atmospheric pressure), a pressure difference can be created between the pressure inside the annular bonding portion Q (internal pressure) and the pressure outside the annular bonding portion Q (external pressure) in an atmosphere with the predetermined pressure Pt (the atmosphere used in the division step S3 described later). As a result, an adsorption force due to the pressure difference can be generated between the first substrate 11 and the second substrate 12. Therefore, in an atmosphere with the predetermined pressure Pt, the first substrate 11 and the second substrate 12 can be firmly bonded by the bonding force of the bonding portion Q and the adsorption force due to the pressure difference between the internal and external pressures. Thus, even without bonding the first substrate 11 and the second substrate 12 across the entire bonding surface as in the conventional method (i.e., without performing bonding using CMP), sufficient bonding force can be obtained to hold the semiconductor substrate K separated from the first substrate 11 in the division step S3 described later on the second substrate 12.
[0030] Therefore, when the fixing portion Q is formed in an annular shape along the peripheral portion 112 as in this embodiment, the peripheral portion 112 of the semiconductor substrate K can be held by the second substrate 12 by the bonding force of the fixing portion Q, while the portion of the semiconductor substrate K inside the fixing portion Q (the portion where multiple device regions Rd are provided) can be held by the second substrate 12 by the suction force caused by the pressure difference. As a result, the semiconductor substrate K can be held by the second substrate 12 with sufficient bonding force.
[0031] <Splitting step S3> In the splitting step S3 (see Figure 1), the atmospheric pressure is adjusted to a predetermined pressure Pt (for example, atmospheric pressure). As a result, the portion of the first substrate 11 that will become the semiconductor substrate K is held to the second substrate 12 with sufficient bonding force (bonding force due to the fixing portion Q and the adsorption force due to the pressure difference between the internal and external pressures). Then, in the splitting step S3, the first substrate 11 is split at the splitting layer 111 under the atmosphere of the predetermined pressure Pt.
[0032] As a result, the semiconductor substrate K, having a thickness Td equal to the depth (a predetermined depth Dt) to which the divided layer 111 was formed, is separated from the first substrate 11 while still being held by the second substrate 12. Then, the device formation step S4, described below, is performed on this semiconductor substrate K. Meanwhile, the remaining portion 11R of the first substrate 11 other than the semiconductor substrate K (the remaining portion after division) is reused for the manufacture of a new semiconductor substrate K and a semiconductor device.
[0033] <Device Formation Step S4> In the device formation step S4 (see Figure 2), at least a portion of the elements that the semiconductor device will have (hereinafter referred to as "element Gd") is fabricated in each of the multiple device regions Rd (see also Figure 3) provided on the semiconductor substrate K. Although not particularly limited, in the device formation step S4, element Gd such as a MOSFET or Schottky diode can be formed in each device region Rd. Before forming element Gd on the semiconductor substrate K, the surface Ka of the semiconductor substrate K (the surface exposed by the division at the division layer 111) may be polished to adjust the smoothness (surface roughness) of the surface Ka to the smoothness required for fabricating element Gd on the semiconductor substrate K.
[0034] Such device formation step S4 is often performed at high temperatures exceeding 1000°C. Now, consider the case where the thermal expansion coefficient of the semiconductor substrate K is different from that of the second substrate 12. In this case, if the semiconductor substrate K and the second substrate 12 were bonded across the entire bonding surface as in the conventional method (i.e., bonding using CMP), then, due to this bonding across the entire bonding surface, when device formation step S4 is performed at a high temperature, one of the semiconductor substrates, K or the second substrate 12, which has a larger thermal expansion coefficient, will try to pull the other substrate and cause it to expand beyond its thermal expansion coefficient, while conversely, the other substrate will try to prevent the expansion of the other substrate. As a result, stress is generated in the semiconductor substrate K and the second substrate 12, and this stress may manifest as defects such as strain in the semiconductor substrate K and the second substrate 12. Furthermore, the stress generated in the semiconductor substrate K may also cause defects in elements Gd such as MOSFETs and Schottky diodes formed on the semiconductor substrate K.
[0035] On the other hand, in this embodiment, the semiconductor substrate K and the second substrate 12 are not bonded to each other inside the fixing portion Q. Therefore, when the device formation step S4 is performed at a high temperature, inside the fixing portion Q, the substrate with the larger expansion coefficient of the semiconductor substrate K and the second substrate 12 can bend independently of the other substrate, and as a result, both substrates can expand by an amount corresponding to their own expansion coefficients. Therefore, stress is less likely to occur in either the semiconductor substrate K or the second substrate 12, and consequently, defects are less likely to occur in elements Gd such as MOSFETs and Schottky diodes formed on the semiconductor substrate K.
[0036] <Cutting step S5> In cutting step S5 (see Figure 2), the semiconductor substrate K and the second substrate 12 are held in the holding sheet 14 with the semiconductor substrate K facing the holding sheet 14. In this state, the second substrate 12 and the semiconductor substrate K are cut in an annular shape along the fixing portion Q at a position inside the annular fixing portion Q. Specifically, the cutting of the second substrate 12 and the semiconductor substrate K is performed so that the cutting line surrounds all the device regions Rd provided on the semiconductor substrate K. Blade dicing, plasma etching, laser ablation, etc., can be used as the cutting method at this time.
[0037] In this embodiment, the semiconductor substrate K and the second substrate 12 are not bonded to each other inside the adhesive portion Q. Therefore, cutting inside the annular adhesive portion Q separates the portions of the semiconductor substrate K and the second substrate 12 inside the cut from the adhesive portion Q. In other words, these inner portions are released from the bond by the adhesive portion Q. Furthermore, the sealing within the annular portion by the adhesive portion Q is released by the cutting, and as a result, the adsorption force generated by the pressure difference between the internal and external air pressures disappears. Consequently, the portion of the second substrate 12 inside the cut (hereinafter referred to as the "inner portion 121") becomes separated from both the adhesive portion Q and the semiconductor substrate K. In other words, the inner portion 121 of the second substrate 12 becomes easily detachable from the semiconductor substrate K.
[0038] <Peeling step S6> In peeling step S6 (see Figure 2), the inner portion 121 of the second substrate 12 is peeled from the semiconductor substrate K at the interface between the second substrate 12 and the metal layer 13. This exposes the metal layer 13 inside the annular portion 122 that remains after peeling (after peeling of the inner portion 121) of the second substrate 12. Subsequently, the exposed metal layer 13 is removed by methods such as polishing or etching, thereby exposing the semiconductor substrate K inside the annular portion 122.
[0039] Furthermore, if, in the aforementioned adhesive portion formation step S2, the metal layer 13 is formed on the surface 12a of the second substrate 12 instead of the surface 11a of the first substrate 11, then in the peeling step S6, the inner portion 121 of the second substrate 12 can be peeled off from the semiconductor substrate K at the interface between the metal layer 13 and the semiconductor substrate K. In this case, the semiconductor substrate K can be exposed by peeling off only the inner portion 121 of the second substrate 12. Also, if, in the metal layer formation step S21, the metal layer 13 is formed only on the peripheral portion 112 of the surface 11a of the first substrate 11, or on the region of the surface 12a of the second substrate 12 that faces the peripheral portion 112, then, since the metal layer 13 is not interposed between the semiconductor substrate K and the second substrate 12 inside the peripheral portion 112, the semiconductor substrate K can be exposed by peeling off only the inner portion 121 of the second substrate 12.
[0040] <Individualization step S7> In the individualization step S7 (see Figure 2), while the semiconductor substrate K is held in place by the holding sheet 14, the semiconductor substrate K is cut to separate the multiple device regions Rd provided on the semiconductor substrate K into individual pieces. Specifically, the semiconductor substrate K is cut along the boundary line Lb (see Figure 3) that separates the multiple device regions Rd, thereby separating the multiple device regions Rd into individual pieces. Methods such as blade dicing, plasma etching, and laser ablation can be used for this cutting. This process results in the manufacture of multiple semiconductor devices equipped with elements Gd, such as MOSFETs and Schottky diodes.
[0041] As described above, according to the manufacturing method of this embodiment, the first substrate 11 and the second substrate 12 can be firmly joined by the bonding force of the annular fixing portion Q and the adsorption force caused by the pressure difference between the air pressure inside the annular portion (internal air pressure) and the air pressure outside the annular portion (external air pressure). Therefore, even without bonding the first substrate 11 and the second substrate 12 over the entire bonding surface as in the conventional method (i.e., without performing bonding using CMP), sufficient bonding force can be obtained to hold the semiconductor substrate K separated from the first substrate 11 in the splitting step S3 to the second substrate 12. Furthermore, since the semiconductor substrate K and the second substrate 12 are not bonded to each other inside the fixing portion Q, the inner portion 121 of the second substrate 12 can be easily peeled off from the semiconductor substrate K by cutting inside the annular fixing portion Q in the cutting step S5, and the semiconductor substrate K can be exposed by peeling off the inner portion 121.
[0042] Thus, according to the manufacturing method of this embodiment, in a semiconductor device manufacturing technology that includes a process for forming a semiconductor substrate K using a divided layer 111, handling of the semiconductor substrate K after its formation (after the execution of the dividing step S3 in this embodiment) becomes easier.
[0043] [2] Variant [2-1] First variation Figure 4 is a conceptual diagram showing a part of the manufacturing method according to the first modified example in order of processing. In the manufacturing method described above, the reinforcing layer formation step S8 may be performed after the peeling step S6 and before the individualization step S7. Specifically, in the reinforcing layer formation step S8, a reinforcing layer 21 of a predetermined thickness Te is formed as a reinforcing substrate for the semiconductor substrate K on the back surface Kb of the semiconductor substrate K (the surface that was the surface 11a of the first substrate 11) located on the side opposite to the surface Ka of the semiconductor substrate K facing the holding sheet 14, inside the annular portion 122 remaining after peeling (after peeling of the inner portion 121) of the second substrate 12. More specifically, a reinforcing material (for example, a thermoplastic material) can be applied to the back surface Kb of the semiconductor substrate K inside the annular portion 122 of the second substrate 12 using a coating method such as spin coating or spray coating, thereby forming a reinforcing layer 21 of a predetermined thickness Te. The reinforcing material may be an electrical insulating material or a conductive material.
[0044] Then, in the individualization step S7, when the semiconductor substrate K is cut, the reinforcing layer 21 is also cut at the same position as the cutting position of the semiconductor substrate K when viewed from above, thereby individualizing each of the multiple device regions Rd provided on the semiconductor substrate K. This makes it possible to manufacture a semiconductor device supported by the reinforcing substrate 21S.
[0045] [2-2] Second variation Figure 5(A) is a conceptual diagram showing the divided layer formation step S1 performed in the manufacturing method according to the second modified example. In the manufacturing method described above, as the first substrate 11, a substrate with a rounded outer peripheral surface 11d (in other words, a convex outer peripheral surface 11d) may be used, as shown in Figure 5(A). On the other hand, if a first substrate 11 of such shape is used, and the division step S3 is performed immediately after the formation of the divided layer 111, a sharp edge 11e will be formed on the outer peripheral edge of the semiconductor substrate K formed in the division step S3, as shown in Figure 5(B). Such a sharp edge 11e may damage other components or chip, causing defects in the semiconductor substrate K during the semiconductor device manufacturing process.
[0046] Therefore, as shown in Figure 5(A), after forming the divided layer 111, the peripheral edge 112 (the rounded portion of the outer peripheral surface 11d) of the first substrate 11 may be cut off. Then, the process from the above-described fixing portion formation step S2 may be performed using the first substrate 11 from which the peripheral edge 112 has been cut off as a new first substrate 11. By cutting off the peripheral edge 112 (the rounded portion of the outer peripheral surface 11d) of the first substrate 11 in this way, it is possible to prevent the formation of sharp edges 11e on the outer peripheral edge of the semiconductor substrate K.
[0047] Furthermore, from the viewpoint of preventing the formation of sharp edges 11e as described above, instead of cutting off the peripheral edge 112 (the part where the outer surface 11d is rounded) of the first substrate 11 after forming the divided layer 111, the peripheral edge 112 (the part where the outer surface 11d is rounded) of the first substrate 11 may be cut off first, and then the divided layer 111 may be formed.
[0048] [2-3] Third variation In the manufacturing method described above, even if the device formation step S4 is performed at a high temperature, if there is no risk of problems caused by substrate expansion as described above (problems where stress caused by substrate expansion cannot be relieved and appears as defects in the substrate or element Gd) occurring (for example, when the expansion coefficient of the semiconductor substrate K and the expansion coefficient of the second substrate 12 are the same), the following process may be performed after the division step S3.
[0049] Here, the annular fixing portion Q formed in the fixing portion formation step S2 seals the inside of the annular portion while maintaining a pressure lower than the predetermined pressure Pt. Therefore, in an atmosphere of the predetermined pressure Pt or an atmosphere of a pressure higher than the predetermined pressure Pt, a pressure difference can be created between the pressure inside the annular portion Q (internal pressure) and the pressure outside the annular portion (external pressure). As a result, an adsorption force due to the pressure difference can be generated between the first substrate 11 and the second substrate 12. Accordingly, this adsorption force can cause the first substrate 11 and the second substrate 12 to adhere closely together, or, if a metal layer 13 is interposed between the substrates, to adhere closely to the metal layer 13.
[0050] Therefore, after the division step S3, while maintaining the aforementioned state of close contact (i.e., under an atmosphere of a predetermined pressure Pt or an atmosphere of higher pressure than said predetermined pressure Pt), heat treatment may be applied to the semiconductor substrate K and the second substrate 12 to generate compounds (such as metal silicide) or alloys (such as metal-Si alloy) at their interface between the main components of the first substrate 11 and the second substrate 12 (such as semiconductors) and the metal that is the main component of the metal layer 13. As a result, the semiconductor substrate K and the second substrate 12 may also be fixed in the region inside the fixing portion Q.
[0051] This method of fixing the inner region of the fixing part Q by utilizing the adsorption force caused by the pressure difference can also be achieved when heat treatment is performed at a low temperature of 1000°C or less (for example, 400°C) (e.g., eutectic bonding). In this case, even if the thermal expansion coefficient of the semiconductor substrate K is different from that of the second substrate 12, the inner region of the fixing part Q can be fixed without causing a large difference in their expansion amounts. In other words, the inner region of the fixing part Q can be fixed without generating large stresses caused by the difference in thermal expansion coefficients. Therefore, when it is desired to fix the semiconductor substrate K and the second substrate 12 over the entire bonding surface, and when fixing the entire bonding surface does not pose a problem in subsequent processes (for example, when the device formation step S4 is not performed at a high temperature), this method of fixing the inner region of the fixing part Q by heat treatment at a low temperature can be used.
[0052] [2-4] Fourth variation Figure 6 is a conceptual diagram showing a part of the manufacturing method according to the fourth modified example in processing order. Figure 7 is a plan view showing the shape of the adhesive portion Q formed in this modified example. In the manufacturing method described above, in the adhesive portion formation step S2, for each of the multiple device regions Rd provided in the portion of the first substrate 11 that will become the semiconductor substrate K, the adhesive portion Q may be formed in an annular shape around the entire circumference of the peripheral edge of the device region Rd. In this case, the adhesive portion Q is formed continuously and annularly under an atmosphere of lower pressure than a predetermined pressure Pt, thereby sealing the inside of the annular portion while maintaining a pressure lower than the predetermined pressure Pt. In the example of Figure 7, the shape of each device region Rd is rectangular, and the adhesive portion Q is formed in a rectangular annular shape along its peripheral edge. Note that the shape of the adhesive portion Q is not limited to a rectangular annular shape, but can be appropriately changed to another shape (such as a circle or polygon) depending on the peripheral shape of each device region Rd.
[0053] With the formation of such a fixing portion Q, under an atmosphere of a predetermined pressure Pt, a pressure difference can be created between the pressure inside the annular fixing portion Q (internal pressure) and the pressure outside the annular fixing portion (external pressure) for each device region Rd. As a result, multiple adsorption forces due to the pressure difference can be generated between the first substrate 11 and the second substrate 12. Therefore, under an atmosphere of a predetermined pressure Pt, the first substrate 11 and the second substrate 12 can be firmly bonded by the bonding force of the fixing portion Q and the adsorption force due to the pressure difference between the internal and external pressures. Thus, even without bonding the first substrate 11 and the second substrate 12 across the entire bonding surface as in the conventional method (i.e., without performing bonding using CMP), sufficient bonding force can be obtained to hold the semiconductor substrate K separated from the first substrate 11 in the splitting step S3 to the second substrate 12.
[0054] Then, after the division step S3 and the device formation step S4, in the cutting step S5, the semiconductor substrate K and the second substrate 12 are held in the holding sheet 14 and cut at a position between the fixing portions Q formed in two adjacent device regions Rd, respectively, thereby allowing each of the multiple device regions Rd provided on the semiconductor substrate K to be separated into individual pieces. As a result, a semiconductor device supported by a support substrate 12S (a substrate formed by cutting the second substrate 12) can be manufactured. At this time, each semiconductor device is firmly supported by the support substrate 12S by the bonding force of the annular fixing portion Q and the adsorption force caused by the pressure difference between the air pressure inside the annular portion (internal air pressure) and the air pressure outside the annular portion (external air pressure).
[0055] Figures 8(A) and 8(B) are conceptual and plan views showing further modifications of the manufacturing methods described in Figures 6 and 7. As shown in Figures 8(A) and 8(B), in addition to forming annular fixing portions Q on the periphery of each device region Rd, annular fixing portions Q may also be formed along the periphery 112 of the first substrate 11 (specifically, the periphery of the portion of the first substrate 11 that will become the semiconductor substrate K).
[0056] With this configuration, peeling and flapping of the substrate edge during cutting in cutting step S5 can be suppressed by the annular fixing portion Q formed along the peripheral portion 112.
[0057] [2-5] Fifth variation In the manufacturing method described above, in the fixing part formation step S2, the annular fixing part Q is not limited to the shape shown in Figure 3 (embodiment) or Figure 7 (fourth modified example), but may also be formed in the following shapes.
[0058] Figures 9, 10(A), and 10(B) are plan views showing three examples of the shape of the fixing portion Q formed in the fifth modified example. As shown in these figures, multiple device regions Rd may be divided into several groups, and for each group, the fixing portion Q may be formed in an annular shape so as to surround all the device regions Rd within that group.
[0059] Here, the example in Figure 9 shows a case where the fixing portion Q is formed to be an independent annular shape for each group. On the other hand, Figures 10(A) and 10(B) show a case where the fixing portion Q is formed by combining a circular portion and a linear portion so that a part of the fixing portion Q can be shared between groups. Specifically, Figure 10(A) shows a case where the fixing portion Q is formed by a circular portion formed annularly along the peripheral portion 112 and a linear portion that crosses the inside of the circular portion. In this case, the linear portion of the fixing portion Q is shared between groups, and as a result, the fixing portion Q has two annular portions that can enclose all the device regions Rd within each group. Also, Figure 10(B) shows a case where the fixing portion Q is formed by a circular portion formed annularly along the peripheral portion 112 and two linear portions that cross the inside of the circular portion and intersect each other (orthogonal in the example of Figure 10(B)). In this case as well, the linear portion of the fixing part Q is shared among the groups, and as a result, the fixing part Q is formed with four annular portions that can enclose all the device regions Rd within each group.
[0060] [2-6] Sixth variation Figure 11(A) is a conceptual diagram showing the fixing part formation step S2 and cutting step S5 performed in the manufacturing method according to the sixth modified example. Figure 11(B) is a plan view showing the shape of the fixing part Q formed in the sixth modified example.
[0061] In the manufacturing method described above, in the fixing part formation step S2, the fixing part Q may be formed on the boundary line Lb that separates the multiple device regions Rd provided on the first substrate 11. In the example in Figure 10(B), the fixing part Q is shown to be formed by a circular portion formed in an annular shape along the peripheral edge 112 and a plurality of straight portions that form a grid by crossing the inside of the circular portion through the boundary line Lb. By forming a grid with a plurality of straight portions in this way, multiple annular portions are formed on the fixing part Q, each surrounding a device region Rd.
[0062] When a fixed portion Q is formed on the boundary line Lb of the device region Rd in this manner, in the cutting step S5 (see Figure 11(A)), the fixed portion Q can be removed along the portion of the second substrate 12 and the semiconductor substrate K that is removed during cutting by cutting the second substrate 12 and the semiconductor substrate K along the boundary line Lb.
[0063] [2-7] Seventh variation In the manufacturing method described above, the method for forming the fixed portion Q in the fixed portion formation step S2 is not limited to forming the fixed portion Q by irradiating the metal layer 13 with laser light, but may be appropriately changed to a method in which the first substrate 11 and the second substrate 12 are in direct contact and the interface between them is irradiated with laser light to form the fixed portion Q.
[0064] [2-8] Eighth variation Figure 12 is a conceptual diagram showing the cutting step S5 and peeling step S6 performed in the manufacturing method according to the eighth modified example. In the manufacturing method according to the above embodiment, in the cutting step S5, only the second substrate 12 of the semiconductor substrate K may be cut in an annular shape along the annular fixing portion Q at a position inside the annular fixing portion Q. Specifically, only the cutting of the second substrate 12 is performed such that the cutting line surrounds all the device regions Rd provided on the semiconductor substrate K.
[0065] According to this cutting method, the inner portion 121 of the second substrate 12 is separated from the fixing portion Q, while the semiconductor substrate K remains connected to the fixing portion Q as a whole. In other words, the entire semiconductor substrate K is bonded to the annular portion 122 of the second substrate 12 that remains after delamination (after delamination of the inner portion 121) by the fixing portion Q.
[0066] Therefore, by peeling the inner portion 121 of the second substrate 12 from the semiconductor substrate K in peeling step S6, the annular portion 122 of the second substrate 12 can be used as an annular support, and the semiconductor substrate K supported by this support can be obtained.
[0067] [2-9] Other variations In the manufacturing method described above, if an adsorption force due to a pressure difference between the air pressure inside the ring of the fixing part Q (internal air pressure) and the air pressure outside the ring (external air pressure) is not required, the shape of the fixing part Q may be appropriately changed to various shapes, not limited to rings. For example, the fixing part Q may be formed as a point or an open linear shape (a linear shape with ends).
[0068] The above-described embodiments and modifications should be considered in all respects to be illustrative and not restrictive. The scope of the present invention is indicated by the claims, rather than by the above-described embodiments and modifications. Furthermore, the scope of the present invention is intended to include all modifications within the meaning and scope equivalent to the claims.
[0069] Furthermore, from the above-described embodiments and modifications, several steps constituting a method for manufacturing a semiconductor device may be partially extracted as the subject of the invention, or each step may be extracted individually. [Explanation of Symbols]
[0070] K Semiconductor Substrate Q. Adhesion 11. First circuit board 11a surface 11d Outer surface 11e Edge 11R remainder 12 Second board 12a surface 12S support board 13 Metal layer 14 Retaining Sheet 21 Reinforcement layer 21S Reinforcement board Dt predetermined depth Gd element Ka surface Kb back Lb border Pt (Prescribed atmospheric pressure) Rd device area Td thickness Te (predetermined thickness) 111 Split layer 112 Peripheral area 121 Inner part 122 Ring section S1 Layer formation step S2 Step for forming the fixed part S3 splitting step S4 Device Formation Step S5 Cutting Step S6 Peeling step S7 Separation step S8 Reinforcement layer formation step S21 Metal layer formation step S22 Laser beam irradiation step
Claims
1. A divided layer formation step, which involves forming a divided layer at a predetermined depth from the surface of a first substrate mainly composed of semiconductors, with a thickness of the predetermined depth, to enable the separation of the semiconductor substrate from the first substrate, After the divided layer formation step, a fixing portion for fixing the portion of the first substrate that will become the semiconductor substrate to the second substrate is formed in an annular shape in an atmosphere of lower pressure than a predetermined pressure, thereby sealing the inside of the annular portion, After the fixing portion forming step, a dividing step is performed in which the first substrate is divided at the dividing layer under the atmosphere of the predetermined pressure, Equipped with, In the fixing portion formation step, the fixing portion is formed on the peripheral edge of the portion of the first substrate that will become the semiconductor substrate, extending over the entire circumference of that peripheral edge. The aforementioned fixing part forming step is, A metal layer forming step in which a metal layer is formed on at least a region on the peripheral edge of the surface of the first substrate, or on at least a region on the surface of the second substrate that faces the peripheral edge, A laser beam irradiation step in which the metal layer is interposed between the first substrate and the second substrate, and in that state, a laser beam is irradiated onto the portion of the metal layer on the peripheral edge to heat that portion and form the fixed portion, A method for manufacturing semiconductor devices, including [the specified element].
2. Following the division step, a device formation step is performed in which at least a portion of the elements that will be provided by the semiconductor device are fabricated in each of the multiple device regions provided on the semiconductor substrate, After the device formation step, the semiconductor substrate and the second substrate are held in a holding sheet with the semiconductor substrate facing the holding sheet, and in that state, the cutting step is performed to cut at least the second substrate at a position inside the annular fixing portion. A method for manufacturing a semiconductor device according to claim 1, further comprising the above.
3. A divided layer forming step that enables the separation of a semiconductor substrate having a thickness of the predetermined depth from the surface of a first substrate mainly composed of a semiconductor, at a predetermined depth from the surface of the first substrate, After the divided layer formation step, a fixing portion for fixing the portion of the first substrate that will become the semiconductor substrate to the second substrate is formed in an annular shape in an atmosphere of lower pressure than a predetermined pressure, thereby sealing the inside of the annular portion, After the fixing portion forming step, a dividing step is performed in which the first substrate is divided at the dividing layer under the atmosphere of the predetermined pressure, A method for manufacturing semiconductor devices, comprising: In the fixing portion formation step, the fixing portion is formed on the peripheral edge of the portion of the first substrate that will become the semiconductor substrate, extending over the entire circumference of that peripheral edge. Following the division step, a device formation step is performed in which at least a portion of the elements that will be provided by the semiconductor device are fabricated in each of the multiple device regions provided on the semiconductor substrate, After the device formation step, the semiconductor substrate and the second substrate are held in a holding sheet with the semiconductor substrate facing the holding sheet, and in that state, the cutting step is performed to cut at least the second substrate at a position inside the annular fixing portion. A method for manufacturing semiconductor devices that further includes the following features.
4. After the cutting step, a peeling step is performed to peel off the inner portion of the annular fixed portion of the second substrate from the semiconductor substrate. A method for manufacturing a semiconductor device according to claim 2 or 3, further comprising the above.
5. After the peeling step, a reinforcing layer forming step is performed on the back surface of the semiconductor substrate, located on the side opposite to the surface of the semiconductor substrate facing the retaining sheet, on the inside of the annular portion of the second substrate remaining after peeling, as a reinforcing substrate for the semiconductor substrate. A method for manufacturing a semiconductor device according to claim 4, further comprising the above.
6. In the divided layer formation step, after forming the divided layer, the peripheral edge of the first substrate is cut off. The method for manufacturing a semiconductor device according to claim 1, wherein in the fixing portion formation step, the first substrate from which the peripheral edge portion was cut off in the divided layer formation step is used as a new first substrate, and the fixing portion is formed on the peripheral edge portion of the portion of the new first substrate that will become the semiconductor substrate, extending around the entire circumference of that peripheral edge portion.
7. A divided layer forming step of forming a divided layer at a predetermined depth from the surface of a first substrate mainly composed of a semiconductor, for the purpose of separating a semiconductor substrate having a thickness of the predetermined depth from the first substrate, After the divided layer formation step, a fixing portion for fixing the portion of the first substrate that will become the semiconductor substrate to the second substrate is formed in an annular shape in an atmosphere of lower pressure than a predetermined pressure, thereby sealing the inside of the annular portion, After the fixing portion forming step, a dividing step is performed in which the first substrate is divided at the dividing layer under the atmosphere of the predetermined pressure, Equipped with, In the divided layer formation step, after forming the divided layer, the peripheral edge of the first substrate is cut off. A method for manufacturing a semiconductor device, wherein in the fixing portion formation step, the first substrate from which the peripheral edge portion was cut off in the divided layer formation step is used as a new first substrate, and the fixing portion is formed on the peripheral edge portion of the portion of the new first substrate that will become the semiconductor substrate, extending around the entire circumference of that peripheral edge.
8. A divided layer forming step that enables the separation of a semiconductor substrate having a thickness of the predetermined depth from the surface of a first substrate mainly composed of a semiconductor, at a predetermined depth from the surface of the first substrate, After the divided layer formation step, a fixing portion for fixing the portion of the first substrate that will become the semiconductor substrate to the second substrate is formed in an annular shape in an atmosphere of lower pressure than a predetermined pressure, thereby sealing the inside of the annular portion, After the fixing portion forming step, a dividing step is performed in which the first substrate is divided at the dividing layer under the atmosphere of the predetermined pressure, Equipped with, A method for manufacturing a semiconductor device, comprising: after the division step, applying heat treatment to the semiconductor substrate and the second substrate in an atmosphere of a predetermined pressure or a pressure higher than the predetermined pressure, thereby fixing the semiconductor substrate and the second substrate in the region inside the fixing portion.
9. A divided layer forming step of forming a divided layer at a predetermined depth from the surface of a first substrate mainly composed of a semiconductor, for the purpose of separating a semiconductor substrate having a thickness of the predetermined depth from the first substrate, After the divided layer formation step, a fixing portion for fixing the portion of the first substrate that will become the semiconductor substrate to the second substrate is formed in an annular shape in an atmosphere of lower pressure than a predetermined pressure, thereby sealing the inside of the annular portion, After the fixing portion forming step, a dividing step is performed in which the first substrate is divided at the dividing layer under the atmosphere of the predetermined pressure, Equipped with, A method for manufacturing a semiconductor device, wherein in the step of forming the fixed portion, the fixed portion is formed around the entire circumference of the peripheral edge of each of the multiple device regions provided in the portion of the first substrate that will become the semiconductor substrate.
10. The aforementioned fixing part forming step is, A metal layer forming step of forming a metal layer on the surface of the first substrate or the surface of the second substrate, A laser beam irradiation step in which the metal layer is interposed between the first substrate and the second substrate, and in that state, laser light is irradiated onto the peripheral edges of each of the device regions of the metal layer to heat the portion and form the fixed portion, A method for manufacturing a semiconductor device according to claim 9, including the method described in claim 9.
11. Following the division step, a device formation step is performed in which at least a portion of the elements that will be provided by the semiconductor device are fabricated in each of the device regions provided on the semiconductor substrate, After the device formation step, the semiconductor substrate and the second substrate are held in a holding sheet with the semiconductor substrate facing the holding sheet, and in that state, the semiconductor substrate and the second substrate are cut at a position between the fixing portions formed in two adjacent device regions, respectively. A method for manufacturing a semiconductor device according to claim 9 or 10, further comprising the above.