WO2013039869A1

WO2013039869A1 - Peeling method and peeling device

Info

Publication number: WO2013039869A1
Application number: PCT/US2012/054585
Authority: WO
Inventors: Kazuta Saito; Shinya Nakajima
Original assignee: 3M Innovative Properties Company
Priority date: 2011-09-12
Filing date: 2012-09-11
Publication date: 2013-03-21
Also published as: TW201320176A; JP2013060489A

Abstract

To allow a first rigid member and a second rigid member that are adhered through an adhesive layer to be peeled quickly and easily, using a simpler method, when peeling apart. A member-peeling method, including a step of elastically flexing at least a portion of a first rigid member (14) in a direction away from a second rigid member (16); and a step of partially peeling an outer edge region (12a) of an adhesive layer (12) from the first rigid member (14) or the second rigid member (16), between at least the portion of the first rigid member (14) that has been flexed elastically and the second rigid member (16). In a member-peeling device (10), a chamber (20) for containing the second rigid member (16) is formed using a jig (24) having a wall (18) that contacts a front face (14a) of the first rigid member (14), and the pressure in the chamber (20) is reduced by a vacuum device (26) to less than an air pressure (P1) that acts on a back face (14b) of the first rigid member (14). Essentially, the entirety of the first rigid member (14) is flexed uniformly to be pressed into the chamber (20), by this differential pressure.

Description

PEELING METHOD AND PEELING DEVICE

TECHNICAL FIELD

The present invention relates to a member-peeling method and a member-peeling device for peeling apart first and second rigid members that are adhered together through an adhesive layer.

BACKGROUND

In semiconductor chip manufacturing processes, there is a known process wherein, after the fabrication of circuits on one surface of a wafer of a particular thickness, the thickness of the wafer is reduced uniformly through grinding the back face, which is an opposite side from the surface on which the circuits are fabricated (hereinafter termed the "circuit face"), in a state wherein a protective film or a supporting member, such as a glass plate or the like, has been adhered to the circuit face. The wafer that is thinned by grinding of the back face typically is peeled from the supporting member prior to being transferred to a subsequent process, such as dicing.

For example, Japanese Unexamined Patent Application Publication 2006-303180 discloses a securing method for securing a substrate, used in a semiconductor device, to a supporting member at the time of grinding of the back face. This securing method is described as including "a preparing step of preparing a film; a first adhering step of adhering together, using an adhesive, the substrate and a first primary face of the film that has been prepared; and a second adhering step of adhering together, using wax, a second primary face of the film and the supporting member. "Moreover, there is also the following description: "First (1) the substrate to be ground and an adhesive-coated surface of a film sheet are adhered together, after which (2) a supporting member and the surface of the film sheet that is not coated with the adhesive are adhered together using wax. In this manner, the substrate becomes strongly secured to the supporting member. After this, (3) the supporting member is secured and the substrate is ground to a desired thickness. Once the grinding is completed, (4) the wax is melted to release the film sheet from the supporting member, after which (5) the adhesiveness of the adhesive is reduced and the film sheet is peeled from the substrate. After this, (6) the substrate is rinsed with purified water and dried, making it possible to obtain a substrate with a better ground state." As a specific configuration for the peeling step, there is a description of "After the grinding, the supporting member 40, to which the substrate 10 that has completed the grinding illustrated in FIG. 3a has been secured, is placed on a hot plate. The back face of the supporting member 40 is heated to 80°C, and then the film 20 is peeled from the supporting member 40, while the wax 30 is in the melted state (Step S105).... After this, as illustrated in FIG. 3b, ultraviolet irradiation is performed for 5 seconds from the second primary face 212 side of the film 21 under conditions of a total exposure dose of 300 mJ/cm2 using a 1 kW metal halide lamp that produces ultraviolet radiation. Doing so hardens the adhesive 22, reducing the adhesive strength. In this state with the ground face side of the substrate 10 secured by drawing a vacuum, the film sheet 20 is peeled slowly from an edge (Step SI 06). This makes it possible to obtain a substrate 10 without cracks and chips, as illustrated in FIG. 3c."

When peeling a first and second rigid member that are adhered to each other by an adhesive, such as in a step of peeling a wafer that has undergone back face grinding from a supporting member, it is desirable to be able to perform this peeling quickly and safely using a simpler method.

SUMMARY

One aspect of the present invention is a member-peeling method for peeling apart a first rigid member and a second rigid member that are adhered together through an adhesive layer, including: a step of elastically flexing at least a portion of the first rigid member in a direction away from the second rigid member; and a step of partially peeling an outer edge region of the adhesive layer from the first rigid member or the second rigid member, between at least a portion of the elastically flexed first rigid member and the second rigid member.

Another aspect of the present invention is member-peeling device for peeling apart a first rigid member and a second rigid member that are adhered together through an adhesive layer, including: a differential pressure generating mechanism for applying a difference between an air pressure that acts on a front face of the first rigid member, to which the second rigid member is adhered, and an air pressure that acts on a back face of the first rigid member, which is an opposite side from the front face; wherein the differential pressure generating mechanism causes at least a portion of the first rigid member to flex elastically in a direction away from the second rigid member, to partially peel an outer edge region of the adhesive layer from the first rigid member or the second rigid member, between at least a portion of the elastically flexed first rigid member and the second rigid member. Yet another aspect of the present invention is a member-peeling device for peeling apart a first rigid member and a second rigid member that are adhered together through an adhesive layer, including: a pressing mechanism for applying a first pressing force to a front face of the first rigid member, to which the second rigid member is adhered, in a direction to separate the front face from the second rigid member and, simultaneously, applying a second pressing force to a back face of the first rigid member, which is an opposite side from the front face, in a direction that is opposite from that of the first pressing force;

wherein the pressing mechanism causes at least a portion of the first rigid member to flex elastically in a direction away from the second rigid member, to partially peel an outer edge region of the adhesive layer from the first rigid member or the second rigid member, between at least a portion of the elastically flexed first rigid member and the second rigid member.

The member-peeling method according to the first aspect of the present invention makes it possible to produce partial peeling, which serves as a starting point for complete peeling, between the outer edge region of the adhesive layer and the first or the second rigid member by merely flexing at least a portion of the first rigid member in the direction away from the second rigid member, in these first and second rigid members that are adhered to each other with an adhesive layer therebetween. It is possible to completely peel the first and second rigid members from each other by moving the surfaces of the two relatively in a direction pulling the surfaces apart, after this partial peeling of the adhesive layer has been produced. Consequently, there is no need for processes for melting the adhesive layer or reducing the adhesive strength, and it is possible to peel the first and second rigid members from each other quickly and safely using a simpler process.

The member-peeling device according to another aspect of the present invention makes it possible to flex essentially the entirety of the first rigid member quickly and safely to produce a bowed shape of a uniform desired curvature without damaging the front face or the back face of the first rigid member. Moreover, by providing the differential pressure generating mechanism of a simple structure, it is possible to reduce equipment costs. This makes it possible to peel the first and second rigid members from each other quickly and safely through a simpler process.

The member-peeling device according to yet another aspect of the present invention makes it possible to flex a desired position of the first rigid member quickly and safely to produce a bowed shape of a desired magnitude. By providing the pressing mechanism of a simple structure, it is possible to reduce equipment costs. This makes it possible to peel the first and second rigid members from each other quickly and safely through a simpler process.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a cross-sectional view schematically illustrating a member-peeling method and a member-peeling device according to one embodiment of the present invention, wherein FIG. la illustrates a state prior to peeling of the members from each other and FIG. lb illustrates a state after peeling of the members from each other.

FIG. 2 is a plan view schematically illustrating the member-peeling device of FIG. 1 FIGS. 3a and 3b are cross-sectional views schematically illustrating a member- peeling method and a member-peeling device according to another embodiment of the present invention, wherein FIG. 3a illustrates a state prior to peeling of the members from each other and FIG. 3b illustrates a state after peeling of the members from each other.

FIG. 4 is a plan view illustrating another example of members to which the member- peeling methods of FIG. 1 to FIG. 3 can be applied.

FIG. 5 is a front view schematically illustrating one example of an adhering step of a pair of members to which the member-peeling methods of FIG. 1 through FIG. 3 can be applied, wherein FIG. 5a shows a state prior to adhering the members to each other and FIG. 5b shows a state after adhering the members to each other. DETAILED DESCRIPTION

Embodiments according to the present invention will be explained in detail below, referencing the appended figures. Throughout all of the figures, corresponding structural elements will be assigned common reference codes. FIG. 1 illustrates a member-peeling method according to one embodiment of the present invention, and a member-peeling device 10 according to one embodiment of the present invention, which is able to perform this member-peeling method. The member-peeling method and member-peeling device 10 that are illustrated are for peeling apart a first rigid member 14 and a second rigid member 16 that are adhered to each other with an adhesive layer 12 therebetween, and can be applied to a step of, for example, peeling a wafer that has undergone back face grinding from a supporting member, in a semiconductor chip manufacturing process. However, the application of the member-peeling method related to this aspect of the present invention and the application of the member-peeling device related to other aspects of the present invention are not limited thereto.

In the structure illustrated, the first rigid member 14 is a plate-shaped element having a flat front face 14a and a flat back face 14b that extend essentially in parallel with each other, and an annular outer peripheral surface 14c between the front face 14a and the back face 14b. The first rigid member 14 can support the second rigid member 16 stably on the front face 14a, without any deformation to the member itself. On the other hand, the first rigid member 14 has a rigidity such that it is possible for at least a portion thereof to flex elastically up to an extent established in advance through an external force within a range that is established in advance. The first rigid member 14 may be a substrate made from, for example, a ceramic such as glass or the like, or a plastic such as Bakelite or the like.

In the structure illustrated, the second rigid member 16 is a flat plate element having a flat front face 16a and a flat back face 16b that extend essentially in parallel with each other, and an annular outer peripheral surface 16c between the front face 16a and the back face 16b. The second rigid member 16 has sufficient rigidity to allow the form of the flat plate shape to be maintained against the machining forces of, for example, grinding, polishing, or the like of the back face 16b, and external forces such as flexing forces propagated from the first rigid member 14, through the adhesive layer 12, when on the front face 14a of the first rigid member 14. The second rigid member 16 may be a wafer or a substrate made from, for example, silicon, gallium arsenide, quartz, glass, or the like, and, in particular, may be a wafer or a substrate having a relatively high hardness such as sapphire, silicon carbide (SiC), or lithium tantalate (LiTa03), or the like.

When the first rigid member 14 has a disk- shaped form, a diameter of the first rigid member 14 can be, for example, more than about 70 mm and less than about 300 mm. A thickness of the first rigid member 14 can be, for example, more than about 0.5 mm and less than about 5 mm, and can be more than about 1 mm and less than about 2 mm. When the second rigid member 16 has a disk-shaped form, a diameter of the second rigid member 16 can be less than the diameter of the first rigid member 14, for example, more than about 50 mm and less than about 280 mm. Furthermore, a thickness of the second rigid member 16 can be, for example, more than about 0.3 mm and less than about 2.0 mm.

When the member-peeling method illustrated is executed in a step of peeling a wafer that has undergone back face grinding from a supporting member, in a semiconductor chip manufacturing process, the first rigid member 14 becomes a supporting member both for covering the circuit face of the wafer and for supporting the wafer stably at the time of the back face grinding, and at least the front face 14a is a smooth surface capable of improving the degree of parallelism of the surface being ground relative to the circuit face of the wafer. On the other hand, the second rigid member 16 is a wafer that has undergone back face grinding, which is a source material for semiconductor chips, with the front face 16a being a circuit face in which a specific circuit pattern has been formed, and the back face 16b being a surface after grinding (namely, the surface that has been ground). A thickness of the wafer is, for example, from about 0.5 mm to about 1 mm, and is standardized together with the diameter. Moreover, a thickness of the semiconductor chips that have undergone back face grinding is, for example, from about 50 μιη to about 100 μιη, and is anticipated to become even thinner.

The second rigid member 16 illustrated typically has a disk-shaped form, as a wafer, but may also have a square, plate-shaped form, for example. Similarly, the first rigid member 14 illustrated has a disk-shaped form, similar to that of the second rigid member 16, but may also have, for example, a square, flat, plate-shaped form, and may have a form different from that of the second rigid member 16. There are no particular limitations on the materials, shapes, dimensions, and the like of the first and second rigid members 14 and 16 aside from at least a portion of the first rigid member 14 being able to produce a specific elastic flexure, and the front face 14a of the first rigid member 14 having a dimension to be able to extend outside of the outer peripheral surface 16c of the second rigid member 16 (which, in the case of a disk shape, implies that the diameter of the first rigid member 14 is larger than the diameter of the second rigid member 16). Note that when a radiation-curable adhesive, such as an ultraviolet-curable adhesive or the like, is used for the adhesive layer 12, preferably the first rigid member 14 will have adequate permeability. Permeability herein indicates permeability within a specific range of the electromagnetic spectrum that is applicable for a curable adhesive, such as the ultraviolet spectrum, for example.

The adhesive layer 12 may be made from, for example, a curable adhesive, a solvent-type adhesive, a thermoplastic resin (including hot melt-type adhesives), an aqueous dispersion-type adhesive, or the like, that is cured or hardened to exhibit adhesive strength to hold strongly, in a secured state, the front face 16a of the second rigid member 16 to the front face 14a of the first rigid member 14. Herein, the curable adhesive is a liquid adhesive that is cured through irradiation with energy, such as heat, ultraviolet radiation, or the like, the solvent-type adhesive is a liquid adhesive that is hardened through the evaporation of a solvent, and the hot melt-type adhesive is an adhesive that is melted through heating and hardens through cooling. Moreover, the aqueous dispersion-type adhesive is an adhesive where the adhesive component is dispersed in water and is hardened through evaporation of the water. As curable adhesives, while not particularly limited, examples include single- component thermally curable adhesives based on epoxy, urethane, or acryl; dual-liquid mixed reaction adhesives based on epoxy, urethane, or acryl; and ultraviolet-curable and electron beam-curable adhesives based on acryl or epoxy. Moreover, as solvent-type adhesives, while not particularly limited, examples include rubber adhesives, wherein rubber, elastomers, or the like are dissolved in solvents.

The adhesive layer 12 is formed at an essentially uniform thickness throughout an entire space between the front face 14a of the first rigid member 14 and the front face 16a of the second rigid member 16, and these front faces 14a and 16a are tightly adhered in a state wherein air bubbles have been excluded, for example by an adhering method for the first and second rigid members 14 and 16 as described below. The thickness of the adhesive layer 12 may be, for example, greater than 0.001 mm and less than about 0.2 mm. As illustrated, the outer edge region 12a of the adhesive layer 12 may be positioned to extend farther to the outside than the outer peripheral surface 16c of the second rigid member 16. In this case, the outer edge region 12a of the adhesive layer 12 may extend to the outside to a degree so as to cover the outer peripheral surface 16c and reach about halfway to the thickness of the second rigid member 16. Conversely, the outer edge region 12a of the adhesive layer 12 may be positioned between the two front faces 14a and 16a (namely, farther toward the inside from the outer peripheral surface 16c of the second rigid member 16).

In a case where the second rigid member 16 is a wafer, the adhesive used in the adhesive layer 12 may be the same as the adhesive used in adhering the supporting plate (the first rigid member 14), for protecting and supporting the circuit face (the front face 16a) at the time of the back face grinding process on the wafer. For example, the ultraviolet- curable liquid adhesive LC-3200, which can be obtained from Sumitomo 3M Co., Ltd. (Tokyo) can be used.

In the structure illustrated, the disk- shaped first rigid member 14 and the disk- shaped second rigid member 16, which has a diameter that is smaller than that of the first rigid member 14, are layered together, arranged so that the respective faces 14a and 16a face each other and are essentially parallel to each other, arranged to be essentially coaxial, and are adhered together by the adhesive layer 12 that is interposed between these front faces 14a and 16a (FIG. la). The member-peeling method illustrated, for this group of structural elements, is structured including a step of elastically flexing at least a portion of the first rigid member 14 in a direction away from the second rigid member 16, and a step of partially peeling the outer edge region 12a of the adhesive layer 12 from the first rigid member 14 or the second rigid member 16, between at least the portion of the first rigid member 14 that has been flexed elastically and the second rigid member 16.

In the step of elastically flexing at least a portion of the first rigid member 14, at least a portion of the first rigid member 14 is flexed in the direction away from the second rigid member 16 through the application of an external force, such as air pressure, water pressure, a mechanical pressing force, or the like, to a required location of the first rigid member 14 with the second rigid member 16 essentially not flexing (FIG. lb). Because most of the front face 14a of the first rigid member 14 is adhered to the front face 16a of the second rigid member 16 through the adhesive layer 12, in practice causing at least a portion of the first rigid member 14 to flex in the direction away from the second rigid member 16 means that an outside force is applied to an outer periphery region of the first rigid member 14 at a position that is farther to the outside than the outer peripheral surface 16c of the second rigid member 16. As a result, by flexing at least a portion of the first rigid member 14 in the direction away from the second rigid member 16, stresses will concentrate in the outer edge region 12a of the adhesive layer 12 that exists between at least the portion of the first rigid member 14 that has been flexed elastically and the second rigid member 16.

By concentrating stresses in the outer edge region 12a of the adhesive layer 12, in the subsequent step, the outer edge region 12a of the adhesive layer 12 begins to partially peel from the first rigid member 14 or the second rigid member 16 at the outermost edge 12b of the adhesive interface with the front face 14a or 16a (FIG. lb). As the amount of flexure of the first rigid member 14 increases, the peeling of the outer edge region 12a of the adhesive layer 12 progresses steadily toward the inside from the outermost edge 12b of the adhesive interface with the front face 14a or 16a. At this stage where the peeling of the outer edge region 12a of the adhesive layer 12 from the first or second rigid member 14 or 16 has advanced a desired distance toward the inside along the adhesive interface, the flexure of the first rigid member 14 is released to allow the first rigid member 14 to return elastically to the plate-shaped form. Even when the first rigid member 14 is allowed to return to the plate form after the peeling has advanced, the part of the adhesive layer 12 that is already peeled (namely, the peeled portion) does not re-adhere to the first rigid member 14. This is because an adhesive that generally does not re-adhere, as described above, is used as the adhesive layer 12. Given this, in the peeled portion, by moving relatively the two front faces 14a and 16a of the first rigid member 14 and the second rigid member 16 in a direction pulling the surfaces apart, peeling of the adhesive layer 12 can be caused to extend from the peeled portion across an entirety of the adhesive interface, to peel the first rigid member 14 and the second rigid member 16 from each other. At the time of this relative motion, the first and second rigid members 14 and 16 can be moved together in a parallel state.

The member-peeling method as set forth above is able to produce a partial peeling between the outer edge region 12a of the adhesive layer 12 and the first or second rigid member 14 and 16, which serves as a starting point for a complete peeling, by merely flexing at least a portion of the first rigid member 14 in the direction away from the second rigid member 16. After the partial peeling of the adhesive layer 12 has been produced, it is possible to peel the first and second rigid members 14 and 16 completely from each other by merely producing relative motion in the direction peeling apart the two front faces 14a and 16a. Consequently, there is no need for processes for melting the adhesive layer 12 or reducing the adhesive strength, and it is possible to peel the first and second rigid members 14 and 16 from each other quickly and safely using a simpler process.

In the member-peeling method set forth above, after completion of the peeling of the first and second rigid members 14 and 16 from each other, the adhesive layer 12, preferably an entirety thereof, will remain in a state adhered to the front face 14a of the first rigid member 14 or to the front face 16a of the second rigid member 16. Conversely, the adhesive layer 12 may be divided appropriately in a structure whereby the layer remains partially on the front faces 14a and 16a of the first and second rigid members 14 and 16.

When the member-peeling method illustrated is performed in a step of peeling a wafer that has undergone back face grinding from a supporting member, in a semiconductor chip manufacturing process, it is useful to cause the entirety of the adhesive layer 12 to remain on the supporting member (the first rigid member 14) that supports the wafer (the second rigid member 16), from the perspective of transferring the wafer smoothly to subsequent steps. Consequently, the step wherein the outer edge region 12a of the adhesive layer 12 is partially peeled from the first or second rigid member 14 or 16 may be performed to partially peel the outer edge region 12a of the adhesive layer 12 from the second rigid member 16 while still adhered to the first rigid member 14.

As is illustrated, by extending the outer edge region 12a of the adhesive layer 12 farther to the outside than the outer peripheral surface 16c of the second rigid member 16, stresses that are concentrated in the outer edge region 12a of the adhesive layer 12 are caused, by the flexure of at least a portion of the first rigid member 14 in the direction away from the second rigid member 16, to act dominantly on the adhesive interface having a relatively small surface area (namely, the adhesive interface between the outer edge region 12a of the adhesive layer 12 and the outer peripheral surface 16c of the second rigid member 16), more than on the adhesive interface having a relatively large surface area

(namely, the adhesive interface between the outer edge region 12a of the adhesive layer 12 and the front face 14a of the first rigid member 14). Therefore, the outer edge region 12a is caused pull away from the outer peripheral surface 16c. The peeling that starts at the adhesive interface between the outer edge region 12a of the adhesive layer 12 and the outer peripheral surface 16c of the second rigid member 16 will advance smoothly between the adhesive layer 12 and the front face 16a of the second rigid member 16 (FIG. lb). As a result, the first and second rigid members 14 and 16 can be peeled from each other in a state where the entirety of the adhesive layer 12 remains on the front face 14a of the first rigid member 14.

A variety of means may be adopted to cause the adhesive strength of the adhesive layer 12 relative to the front face 14a of the first rigid member 14 to be greater than the adhesive strength of the adhesive layer 12 relative to the front face 16a of the second rigid member 16, in order to enable partial peeling of the outer edge region 12a of the adhesive layer 12 from the second rigid member 16 while still adhered to the first rigid member 14. One example of such means is the selection of the materials for the first and second rigid members 14 and 16. For example, the first rigid member 14 may be formed from a plastic, such as Bakelite, or the like, and the second rigid member 16 may be made from a wafer member, such as sapphire, or the like. Moreover, as another example of such means, a surface treatment may be performed in advance on the front face 14a in order to increase the adhesive strength of the adhesive layer 12 relative to the front face 14a of the first rigid member 14, and a surface treatment may be performed in advance on the front face 16a in order to reduce the adhesive strength of the adhesive layer 12 relative to the front face 16a of the second rigid member 16. In the member-peeling method set forth above, the step of elastically flexing at least a portion of the first rigid member 14 in the direction away from the second rigid member 16 may be performed by applying a difference between an air pressure that acts on the front face 14a of the first rigid member 14, to which the second rigid member 16 is adhered, and an air pressure that acts on the back face 14b of the first rigid member 14, which is an opposite side from the front face 14a. For example, as illustrated in FIG. 1, the step may include a step of placing a chamber 20 that contains the second rigid member 16 and that also has a wall 18 that contacts the front face 14a of the first rigid member 14, and a step of reducing the pressure of the chamber 20 to be less than an air pressure PI that acts on the back face 14b of the first rigid member 14 (namely, to an air pressure P2 (<P1)).

The member-peeling device 10, illustrated in FIG. 1 and FIG. 2, is equipped with a differential pressure generating mechanism 22 for applying a difference between the air pressure that acts on the front face 14a of the first rigid member 14, to which the second rigid member 16 is adhered, and the air pressure that acts on the back face 14b of the first rigid member 14, which is an opposite side from the front face 14a. The differential pressure generating mechanism 22 is provided with a jig 24 that has the wall 18, and a vacuum device (for example, a vacuum pump) 26 that is connected to the chamber 20. The jig 24 has a plate-shaped end wall 28 and a cylindrical wall 18, which stands on one surface along the outer peripheral edge of the end wall 28. The wall 18 extends to a uniform height from the end wall 28, and an entirety of the distal end 18a thereof makes uniform contact with the front face 14a of the first rigid member 14. Moreover, the wall 18 and the end wall 28 have dimensions to be able to contain the second rigid member 16 within a recessed portion that is formed between the two walls 18 and 28, without contacting the second rigid member 16. The chamber 20, which is sealed air-tightly and which contains the second rigid member 16, is defined between the jig 24 and the first rigid member 14, in a state where the distal end 18a of the wall 18 makes uniform contact with the front face 14a of the first rigid member 14 (FIG. la).

The jig 24 has sufficient rigidity to be able to maintain the individual shape thereof against the reduced pressure of the chamber 20. The jig 24 illustrated assumes that the first and second rigid members 14 and 16 have a disk-shaped form, and this jig 24 has a structure having a disk-shaped end wall 28, which resembles the first and second rigid members 14 and 16, and a cylindrical wall 18 that can be arranged at a uniform distance from the outer peripheral surface 16c of the second rigid member 16 (FIG. 2). Conversely, the jig 24 may have a shape that does not correspond to the shapes of the first and second rigid members 14 and 16. The jig 24 may be made from a typical structural material such as metal, plastic, or the like, and the wall 18 and the end wall 28 may be formed as a single unit, or a wall 18 and an end wall 28 that are formed separately may be joined together. There are no particular limitations on the materials, shapes, dimensions, or the like, of the jig 24, except for the ability to contain the second rigid member 16, without contact, in a recessed portion that is formed between the wall 18 and the end wall 28, and being able to maintain air-tight contact between the front face 14a and the distal end 18a of the wall 18 at the time of elastic flexure of the first rigid member 14.

In the member-peeling device 10, as a preliminary operation, the jig 24 is placed in a state where the second rigid member 16 is contained, without contact, in a recessed portion formed between the wall 18 and the end wall 28, and also the distal end 18a of the wall 18 makes uniform contact the front face 14a of the first rigid member 14, to form the chamber 20, which is sealed airtightly and which contains the second rigid member 16, between the jig 24 and the first rigid member 14. At this time, the distal end 18a of the wall 18 contacts the portion of the front face 14a of the first rigid member 14 that is in proximity with an entirety of the outer peripheral surface 14c (FIG. la). In this state, the vacuum device 26 is actuated, and the member-peeling method set forth above begins.

In the jig 24, when the pressure within the chamber 20 is reduced through the action of the vacuum device 26, essentially an entirety of the first rigid member 14 is flexed uniformly so as to be pushed into the chamber 20 through the differential pressure between the air pressure PI outside of the chamber 20 and the air pressure P2 within the chamber 20, with required portions of the first rigid member 14 supported stably and without the wall 18 deforming. At this time, the second rigid member 16 is not supported directly by the jig 24, and thus maintains an individual shape, without being flexed by the differential pressure.

As a result, the first rigid member 14 flexes elastically in a direction away from the second rigid member 16 uniformly, particularly at the part that is positioned to the inside of the wall 18. Moreover, the outer edge region 12a of the adhesive layer 12 partially peels from the second rigid member 16 between the part of the first rigid member 14 that flexes elastically and the second rigid member 16. The form of the peeling of the adhesive layer 12 at the start and as the peeling advances is as described above.

The member-peeling method and the member-peeling device 10, set forth above, which flex elastically at least a portion of the first rigid member 14 through a differential pressure, make it possible to quickly and safely flex essentially the entirety of the first rigid member 14, to achieve a bowed shape with a uniform and desired curvature, without damaging the front face 14a or back face 14b of the first rigid member 14. Moreover, the member-peeling device 10 provided with the differential pressure generating mechanism 22 is able to achieve a reduction in equipment costs by providing a jig 24 with a simple structure and a common vacuum device 26. Consequently, the member-peeling device 10 makes it possible to peel the first and second rigid members 14 and 16 from each other quickly and safely through a simpler method.

The member-peeling method set forth above that elastically flexes at least a portion of the first rigid member 14 through differential air pressure can use a method where a second chamber (not shown) is placed so that the back face 14b of the first rigid member 14 is taken as part of the wall, and compressed air is provided into the second chamber to increase the internal pressure within the second chamber to higher than the internal pressure in the chamber 20 or to higher than the air pressure that acts on the front face 14a of the first rigid member 14. This method may be used in addition to, or instead of, the method set forth above whereby the pressure within the chamber 20 that contains the second rigid member 16 is reduced to below the air pressure PI that acts on the back face 14b of the first rigid member 14. With this configuration as well, essentially the entirety of the first rigid member 14 is flexed to achieve a uniform bowed shape without damaging the front face 14a or the back face 14b of the first rigid member 14.

FIG. 3 illustrates a member-peeling method according to another embodiment of the present invention, as well as a member-peeling device 30 according to another embodiment of the present invention, which is able to perform this member-peeling method. As with the member-peeling method and member-peeling device 10 illustrated in FIG. 1, the member- peeling method and member-peeling device 30 are for peeling apart a first rigid member 14 and a second rigid member 16 that are adhered to each other through an adhesive layer 12, and can be applied to a step of, for example, peeling a wafer that has undergone back face grinding from a supporting member, in a semiconductor chip manufacturing process.

The member-peeling method illustrated in FIG. 3 is structured including a step of elastically flexing at least a portion of the first rigid member 14 in a direction away from the second rigid member 16, and a step of partially peeling the outer edge region 12a of the adhesive layer 12 from the first rigid member 14 or the second rigid member 16, between at least the portion of the first rigid member 14 that has been flexed elastically and the second rigid member 16. The principles, actions, and the like, in these basic steps are the same as in the member-peeling method illustrated in FIG. 1.

In the member-peeling method illustrated in FIG. 3, the step of elastically flexing at least a portion of the first rigid member 14 in the direction away from the second rigid member 16 may include a step of simultaneously applying a first pressing force Fl to the front face 14a of the first rigid member 14 to which the second rigid member 16 is adhered in a direction that separates the front face 14a from the second rigid member 16, and applying a second pressing force F2 in a direction opposite to the first pressing force Fl to the back face 14b of the first rigid member 14, which is an opposite side from the front face 14a.

The member-peeling device 30 illustrated in FIG. 3 is provided with a pressing mechanism 32 for simultaneously applying a first pressing force Fl to the front face 14a of the first rigid member 14 to which the second rigid member 16 is adhered in a direction that separates the front face 14a from the second rigid member 16, and applying a second pressing force F2 in the direction opposite to the first pressing force Fl to the back face 14b of the first rigid member 14, which is an opposite side from the front face 14a. The pressing mechanism 32 includes a pressing member 34 that contacts the front face 14a of the first rigid member 14, a driving portion 36 that drives the pressing member 34 to apply a first pressing force Fl onto the front face 14a from the pressing member 34, a movable supporting member 38 that supports, by a second pressing force F2, the back face 14b of the first rigid member 14, a driving portion 40 that drives the movable supporting member 38 to change the position of the second pressing force F2 that is applied to the back face 14b, and a stationary supporting member 42 that supports, by a third pressing force F3, the front face 14a of the first rigid member 14 (FIG. 3a).

The pressing member 34 contacts a desired position in an area of the front face 14a of the first rigid member 14 that is to the outside than the outer peripheral surface 16c of the second rigid member 16, at a position not contacting the second rigid member 16. In particular, it is effective for the pressing member 34 to contact a position that is near to the outer peripheral surface 14c of the first rigid member 14, as illustrated. The pressing member 34 itself is driven by the driving portion 36 to act to apply a first pressing force Fl at the position of contact with the front face 14a, to move the first rigid member 14 in the direction of peeling away from the second rigid member 16 (FIG. 3b). Note that the driving portion 36 may be structured from a common driving device that uses electric power or hydraulic or pneumatic pressure.

The movable supporting member 38 contacts the back face 14b of the first rigid member 14 at a desired position within the area where the second rigid member 16 is adhered to the front face 14a side, at a position not contacting the second rigid member 16. The movable supporting member 38 acts to produce a required flexure in a portion of the first rigid member 14 that extends between a position of application of the first pressing force Fl and a position of application of the second pressing force F2, by supporting the back face 14b of the first rigid member 14 with the second pressing force F2 against the first pressing force Fl that is applied to the front face 14a of the first rigid member 14 by the pressing member 34 (FIG. 3b). This movable supporting member 38 itself is driven by the driving portion 40 to move along the back face 14b of the first rigid member 14, to change the spacing between the position of application of the first pressing force Fl and the position of application of the second pressing force F2 (FIG. 3b). Note that the driving portion 40 may be structured from a common driving device that uses electric power or hydraulic or pneumatic pressure.

The stationary supporting member 42 contacts a desired position of the front face 14a of the first rigid member 14 in an area that is farther to the outside than the outer peripheral surface 16c of the second rigid member 16, on an opposite side from the position of application of the first pressing force Fl, with the second rigid member 16 held therebetween, at a position not contacting the second rigid member 16. The stationary supporting member 42 acts to reliably produce a specific flexure of the portion of the first rigid member 14 that extends between the position of application of the first pressing force Fl and the position of application of the second pressing force F2, by supporting the front face 14a of the first rigid member 14 through a third pressing force F3. This third pressing force F3 supports the front face 14a of the first rigid member 14 against the output of the first pressing force Fl that results from a lever action with the position of application of the second pressing force F2 acting as a fulcrum (FIG. 3b).

The pressing member 34, the movable supporting member 38, and the stationary supporting member 42 have rigidity to be able to maintain the individual shapes thereof against the respective first pressing force Fl, second pressing force F2, and third pressing force F3. In the structure illustrated, the pressing member 34, the movable supporting member 38, and the stationary supporting member 42 have tapered tip ends 34a, 38a, and 42a that are able to apply the first pressing force Fl, the second pressing force F2, and the third pressing force F3, locally, onto the front face 14a or back face 14b of the first rigid member 14. The tip ends 34a, 38a, and 42a of the pressing member 34, the movable supporting member 38, and the stationary supporting member 42 are each structured with an edge that extends across a desired length, extending in a direction perpendicular to the plane of the paper in FIG. 3. Alternatively, instead of the edge, the pressing member 34, the movable supporting member 38, and the stationary supporting member 42 may be structured to contact the front face 14a or the back face 14b of the first rigid member 14 with a flat end surface. In particular, the movable supporting member 38 that contacts the back face 14b of the first rigid member 14 may be structured as a table-shaped member having a flat supporting surface that produces the second pressing force F2. The pressing member 34, the movable supporting member 38, and the stationary supporting member 42 may be fabricated from a common structural material, such as metal, plastic, or the like, or may be fabricated from a relatively soft plastic, such as Teflon™, or the like, for example, in order to avoid damaging the first rigid member 14. There are no particular limitations on the materials, shapes, dimensions, and the like of the pressing member 34, the movable supporting member 38, and the stationary supporting member 42, except for the

requirement that the first pressing force Fl, the second pressing force F2, and the third pressing force F3 must be reliably applied at desired locations on the front face 14a or the back face 14b of the first rigid member 14.

In the member-peeling device 30, as a preliminary operation, the pressing member 34 is caused to contact a position of the front face 14a of the first rigid member 14, which is in proximity to a portion of the outer peripheral surface 14c, the movable supporting member 38 is caused to contact a portion of the back face 14b of the first rigid member 14, which is in proximity to a portion of the outer peripheral surface 16c of the second rigid member 16 on the front face 14a side (position facing the pressing member 34), and the stationary supporting member 42 is caused to contact a position of the front face 14a of the first rigid member 14, which is on the opposite side from the pressing member 34 (FIG. 3a). In this state, the driving portions 36 and 40 are actuated, and the member-peeling process set forth above begins.

The first pressing force Fl is applied to the front face 14a by moving the pressing member 34 at the velocity VI in the direction pulling apart the first rigid member 14 from the second rigid member 16 (downward direction in the figure) through the driving of the driving portion 36. When this occurs, the first rigid member 14, which is supported by the second pressing force F2 and the third pressing force F3 from the movable supporting member 38 and the stationary supporting member 42, flexes at the portion that extends between the position of application of the first pressing force Fl and the position of application of the second pressing force F2. At this time, the second rigid member 16 is not supported directly by the movable supporting member 38 or the stationary supporting member 42, and thus maintains the individual shape, without being flexed by the first pressing force Fl through the third pressing force F3. As a result, the first rigid member 14 will elastically flex in the direction away from the second rigid member 16, particularly at the portion that is in proximity to the outer peripheral surface 14c of that rigid member 14 and to the outer peripheral surface 16c of the second rigid member 16. Additionally, the outer edge region 12a of the adhesive layer 12 partially separates from the second rigid member 16 between the second rigid member 16 and the portion of the first rigid member 14 that has flexed elastically. The form of the peeling of the adhesive layer 12 at the start and as the peeling advances is as described above.

In the member-peeling device 30, the driving portion 40 may be driven in synchronization with a velocity VI of the motion of the pressing member 34 (or somewhat more slowly) to move the movable supporting member 38 at a velocity V2 in the direction away from the pressing member 34 (toward the left in the figure). In this way, it is possible to steadily increase the interval between the position of application of the first pressing force Fl and the position of application of the second pressing force F2, and to cause the peeling of the adhesive layer 12 to advance smoothly to a desired location.

The member-peeling method and the member-peeling device 30, set forth above, which flex elastically at least a portion of the first rigid member 14 through the application of a mechanical pressing force, make it possible to quickly and safely flex essentially the entirety of the first rigid member 14, to form a bowed shape of a desired magnitude. While in the member-peeling method and member-peeling device 10 that use differential pressure it is necessary to flex the first rigid member 14 while maintaining the chamber 20 in an airtight state, the member-peeling method and member-peeling device 30 that use a mechanical pressing force make it possible to peel the first and second rigid members 14 and 16 reliably from each other even in cases where it would be difficult to maintain the chamber 20 due to, for example, the shape of the first rigid member 14. Moreover, the member-peeling device 30, equipped with the pressing mechanism 32 is equipped with a pressing member 34, a movable supporting member 38, and a stationary supporting member 42, which have simple structures, as well as common driving portions 36 and 40, and thus is able to reduce equipment costs. Consequently, the member-peeling device 30 makes it possible to peel the first and second rigid members 14 and 16 from each other quickly and safely through a simpler method.

The member-peeling methods of the embodiments illustrated and the member- peeling devices 10 and 30, for example as illustrated in FIG. 4, are able to peel single first rigid members 14' from a plurality of second rigid members 16 that are adhered to individual first rigid members 14' through the adhesive layer 12. The member-peeling method and member-peeling device 30 that uses mechanical pressing forces can reliably separate the first rigid member 14' and the individual second rigid members 16, even in the case wherein the use of the jig 24 would be difficult due to narrow spacing of adjacent second rigid members 16.

FIG. 5 shows one example of an adhering step of adhering together, through an adhesive layer 12, first and second rigid members 14 and 16 to which the member-peeling methods of the embodiments illustrated and the member-peeling devices 10 and 30 can be applied. The adhering process illustrated has the following steps:

The first rigid member 14 is placed on a flat supporting face 40a of a static platform 40 oriented so that the front face 14a is exposed, and is secured at a specific position of the supporting face 40a, for example, by a jig, not shown. The second rigid member 16 is supported on a flat supporting face 44a of a press 44 oriented so that the front face 16a is exposed, and secured by, for example, drawing a vacuum. The first rigid member 14 and the second rigid member 16 are arranged at positions where the two front faces 14a and 16a are facing, parallel to each other, and with a gap therebetween. A specific amount of a liquid adhesive (ultraviolet-curable type) 12' is provided in an area on the front face 14a of the first rigid member 14 facing essentially the center of the front face 16a of the second rigid member 16 (FIG. 5a for the above). The press 44 is moved toward the static platform 40, to press the front face 16a of the second rigid member 16 onto the front face 14a of the first rigid member 14, to layer together the first and second rigid members 14 and 16 while causing the adhesive 12' to spread out between these front faces 14a and 16a. The press 44 applies pressure until an appropriate amount of the adhesive 12' flows out from the outer peripheral surface 16c of the second rigid member 16 (FIG. 5b). At the point that the adhesive 12' forms a layer of a specific thickness, the second rigid member 16 is released from the press 44, and the press 44 is separated from the static platform 40. The first and second rigid members 14 and 16, which are layered together with the adhesive 12' therebetween, are exposed to ultraviolet radiation to cure the adhesive 12' and form the adhesive layer 12 (FIG. 1).

The following tests were performed in order to validate the effectiveness of the member-peeling methods illustrated in FIG. 1 through FIGS. 3a-3b and the member-peeling devices 10 and 30.

Test 1

Using the adhering method illustrated in FIG. 5, an adhesive 12' made from the ultraviolet-curable liquid adhesive LC-3200 (obtained from Sumitomo 3M Co., Ltd.

(Setagaya-ku, Tokyo)) was used to bond together a second rigid member 16, which was made from a sapphire wafer with a thickness of 1 mm and a diameter of 100 mm, and a first rigid member 14, which was made from a glass plate with a thickness of 1 mm and a diameter of 300 mm (Young's modulus of glass: 7.10 x 1010 Pa), to obtain first and second rigid members 14 and 16 adhered together through an adhesive layer 12 with a thickness of 30 μιη. Back face grinding was performed on the second rigid member 16 in a state where the first rigid member 14 had been secured, to reduce the thickness of the second rigid member 16 to 0.5 mm. After the back face grinding, the jig 24 of the member-peeling device 10 was placed on the first rigid member 14 with the relative positional relationships of the preliminary operation described above. In this manner, the second rigid member 16 is received in the chamber 20 that is formed between the jig 24 and the first rigid member 14. Continuing, the attaching device 26 is actuated to reduce the pressure in the chamber 20. The jig 24 used had an inner diameter of 122 mm for the wall 18. Peeling from the second rigid member 16 began at the outer edge region 12a of the adhesive layer 12 at the point where the first rigid member 14 was flexed until the outermost edge thereof had moved 0.5 mm from when flat. The reduction in pressure of the chamber 20 was continued, and then once the peeling of the adhesive layer 12 had advanced 30 mm toward the inside, in the radial direction, from the outer peripheral surface 16c of the second rigid member 16, the first rigid member 14 was flexed until the outermost edge thereof had moved

approximately 1.0 mm from the position thereof when flat. At this point, the vacuum device 26 was stopped and the pressure in the chamber 20 was returned to atmospheric pressure, and the first and second rigid members 14 and 16 were removed from the jig 24. A vacuum device having a suction plate with a diameter of 30 mm was used to pull the second rigid member 16 away from the first rigid member 14, and the first and second rigid members 14 and 16 were peeled from each other.

Test 2

Using the adhering method illustrated in FIG. 5, an adhesive 12' made from the ultraviolet-curable liquid adhesive LC-3200 was used to bond together a second rigid member 16, which was made from a sapphire wafer with a thickness of 1 mm and a diameter of 100 mm, and a first rigid member 14, which was made from a Bakelite disk with a thickness of 2 mm and a diameter of 300 mm (Young's modulus of Bakelite: 0.47 x 1010 Pa, obtained from Sumitomo 3M Co., Ltd. (Shinagawa-ku, Tokyo)), to obtain first and second rigid members 14 and 16 adhered together through an adhesive layer 12 with a thickness of 30 μιη. Back face grinding was performed on the second rigid member 16 in a state where the first rigid member 14 had been secured, to reduce the thickness of the second rigid member 16 to 0.5 mm. After the back face grinding, the jig 24 of the member- peeling device 10 was placed on the first rigid member 14 with the relative positional relationships of the preliminary operation described above. In this manner, the second rigid member 16 is received in the chamber 20 that is formed between the jig 24 and the first rigid member 14. Continuing, the attaching device 26 is actuated to reduce the pressure in the chamber 20. The jig 24 used had an inner diameter of 122 mm for the wall 18. Peeling from the second rigid member 16 began at the outer edge region 12a of the adhesive layer 12 at the point where the first rigid member 14 was flexed until the outermost edge thereof had moved 0.5 mm from when flat. The reduction in pressure of the chamber 20 was continued, and then once the peeling of the adhesive layer 12 had advanced 30 mm toward the inside, in the radial direction, from the outer peripheral surface 16c of the second rigid member 16, the first rigid member 14 was flexed until the outermost edge thereof had moved approximately 1.0 mm from the position thereof when flat. At this point, the vacuum device 26 was stopped and the pressure in the chamber 20 was returned to atmospheric pressure, and the first and second rigid members 14 and 16 were removed from the jig 24. A vacuum device having a suction plate with a diameter of 30 mm was used to pull the second rigid member 16 away from the first rigid member 14, and the first and second rigid members 14 and 16 were peeled from each other.

Test 3

Using the adhering method illustrated in FIG. 5, an adhesive 12' made from the ultraviolet-curable liquid adhesive LC-3200 was used to bond together a second rigid member 16, which was made from a sapphire wafer with a thickness of 1 mm and a diameter of 100 mm, and a first rigid member 14, which was made from a glass plate with a thickness of 1 mm and a diameter of 300 mm (Young's modulus of glass: 7.10 x 1010 Pa), to obtain first and second rigid members 14 and 16 adhered together through an adhesive layer 12 with a thickness of 30 μιη. Back face grinding was performed on the second rigid member 16 in a state where the first rigid member 14 had been secured, to reduce the thickness of the second rigid member 16 to 0.5 mm. The pressing member 34, the movable supporting member 38, and the stationary supporting member 42 of the member-peeling device 30 were positioned against the first and second rigid members 14 and 16 that had undergone back face grinding, with the relative positional relationships of the preliminary operation described above, and the driving portions 36 and 40 were actuated to move the pressing member 34 and movable supporting member 38. A pressing member 34 having an edge length of 20 mm and movable supporting member 38 and stationary supporting member 42 having edge lengths of 100 mm were used. Peeling from the second rigid member 16 began at the outer edge region 12a of the adhesive layer 12 at the point where the first rigid member 14 was flexed until the outermost edge thereof had moved 0.5 mm from when flat. The movement of the pressing member 34 and of the movable supporting member 38 was continued, and then once the peeling of the adhesive layer 12 had advanced 30 mm toward the inside, in the radial direction, from the outer peripheral surface 16c of the second rigid member 16, the first rigid member 14 was flexed until the outermost edge thereof had moved approximately 1.0 mm from the position thereof when flat. At this point, the driving portions 36 and 40 were stopped to stop the movement of the pressing member 34 and the movable supporting member 38, and the pressing member 34, the movable supporting member 38, and the stationary supporting member 42 were removed from the first and second rigid members 14 and 16. A vacuum device having a suction plate with a diameter of 30 mm was used to pull the second rigid member 16 away from the first rigid member 14, and the first and second rigid members 14 and 16 were peeled from each other.

Results of Testing

In each of the tests 1 through 3, the time required to completely peel the first and second rigid members was within 1 minute or less from the time that the vacuum device 26 or the driving portions 36 and 40 were actuated. Moreover, in each of these tests 1 through 3, the adhesive layer 12 after peeling remained on only the front face 14a of the first rigid member 14. The remaining adhesive layer 12 was removed easily using an arbitrary manual tool. It was possible to use a solvent, or the like, to clean the front face 14a of the first rigid member 14 as needed. It was confirmed that extremely useful effects were obtained in the case where the second rigid member 16 was a wafer, such as a sapphire wafer or the like.

Claims

WHAT IS CLAIMED IS:

A member-peeling method for peeling apart a first rigid member and a second rigid member that are adhered together through an adhesive layer, comprising:

a step of elastically flexing at least a portion of the first rigid member in a

direction away from the second rigid member; and

a step of partially peeling an outer edge region of the adhesive layer from the first rigid member or the second rigid member, between at least a portion of the elastically flexed first rigid member and the second rigid member.

The member-peeling method according to Claim 1 , wherein the step of elastically flexing includes a step of applying a difference between an air pressure that acts on a front face of the first rigid member, to which the second rigid member is adhered, and an air pressure that acts on a back face of the first rigid member, which is an opposite side from the front face.

The member-peeling method according to Claim 2, wherein the step of elastically flexing comprises: a step of providing a chamber that contains the second rigid member and that also has a wall that contacts the front face of the first rigid member; and a step of reducing a pressure of the chamber to less than the air pressure that acts on the back face of the first rigid member.

The member-peeling method according to Claim 1 , wherein the step of elastically flexing comprises a step of applying a first pressing force to a front face of the first rigid member, to which the second rigid member is adhered, in a direction to separate the front face from the second rigid member and, simultaneously, applying a second pressing force to a back face of the first rigid member, which is an opposite side from the front face, in a direction that is opposite from that of the first pressing force.

The member-peeling method according to any one of Claims 1 through 4, wherein the step of partially peeling comprises a step of partially peeling the outer edge region of the adhesive layer from the second rigid member, while still adhered to the first rigid member.

The member-peeling method according to any one of Claims 1 through 5, wherein a single aforementioned first rigid member and a plurality of aforementioned second rigid members, which are each adhered to the single aforementioned first rigid member through the adhesive layer, are peeled from each other.

A member-peeling device for peeling apart a first rigid member and a second rigid member that are adhered together through an adhesive layer, comprising:

a differential pressure generating mechanism for applying a difference between an air pressure that acts on a front face of the first rigid member, to which the second rigid member is adhered, and an air pressure that acts on a back face of the first rigid member, which is an opposite side from the front face;

wherein the differential pressure generating mechanism causes at least a portion of the first rigid member to flex elastically in a direction away from the second rigid member, to partially peel an outer edge region of the adhesive layer from the first rigid member or the second rigid member, between at least a portion of the elastically flexed first rigid member and the second rigid member.

a pressing mechanism for applying a first pressing force to a front face of the first rigid member, to which the second rigid member is adhered, in a direction to separate the front face from the second rigid member and, simultaneously, applying a second pressing force to a back face of the first rigid member, which is an opposite side from the front face, in a direction that is opposite from that of the first pressing force;

wherein the pressing mechanism causes at least a portion of the first rigid

member to flex elastically in a direction away from the second rigid member, to partially peel an outer edge region of the adhesive layer from the first rigid member or the second rigid member, between at least a portion of the elastically flexed first rigid member and the second rigid member.