Method for separating at least two substrates along a selected interface

Abstract

A process for separating at least two substrates comprising at least two separation interfaces along one of the interfaces includes, before inserting a blade between the substrate, damaging at least one portion of a peripheral region of a chosen one of the interfaces, then inserting the blade and partially parting the substrates, and applying a fluid in a space between the parted substrates while the blade remains inserted therebetween, and decreasing a rupture energy of the chosen interface by stress corrosion involving breaking of siloxane bonds present at the interface.

Claims

1. A method for separating at least two substrates forming part of a structure comprising at least two separation interfaces extending parallel to main faces of the structure, along one interface selected from the at least two separation interfaces, the method comprising: before separating the at least two substrates, damaging at least one portion of a peripheral region of the one interface such that a rupture energy in the at least one portion of the peripheral region is lower than a rupture energy of other interface regions of the at least two separation interfaces; partially parting the at least two substrates along the one interface in the damaged region; and applying a fluid in a space between the partially parted at least two substrates and decreasing the rupture energy of the one interface by stress corrosion involving breaking of siloxane bonds present at the one interface.

2. The method of claim 1, wherein the peripheral region of the one interface is in the form of a ring.

3. The method of claim 1, wherein the damaging of the at least one portion of the peripheral region of the one interface in an insertion region is performed prior to assembly of the at least two substrates to form the structure.

4. The method of claim 3, wherein the damaging of the at least one portion of the peripheral region is performed by laser irradiation of the one interface.

5. The method of claim 3, wherein the damaging of the at least one portion of the peripheral region is performed by chemical etching of the one interface.

6. The method of claim 1, wherein the damaging of the at least one portion of the peripheral region of the one interface is performed after assembly of the at least two substrates to form the structure.

7. The method of claim 6, wherein the damaging of the at least one portion of the peripheral region is performed by laser irradiation of the one interface.

8. The method of claim 6, wherein the damaging of the at least one portion of the peripheral region is performed by chemical etching of the one interface.

9. The method of claim 1, wherein the damaging of the at least one portion of the peripheral region is performed by laser irradiation of the one interface.

10. The method of claim 1, wherein the damaging of the at least one portion of the peripheral region is performed by chemical etching of the one interface.

11. The method of claim 10, wherein the chemical etching is carried out by applying hydrofluoric acid to the at least one portion of the peripheral region.

12. The method of claim 1, wherein the structure comprises a semiconductor-on-insulator structure including a support substrate, a buried silicon oxide layer and a silicon layer, and wherein the one interface comprises an interface between the oxide layer and the silicon layer.

13. The method of claim 1, wherein the rupture energy of each of the at least two separation interfaces is greater than 1 J/m.sup.2 before partially parting the at least two substrates along the one interface in the damaged region.

14. The method of claim 13, wherein the rupture energy of each of the at least two separation interfaces is greater than 1.5 J/m.sup.2 before partially parting the at least two substrates along the one interface in the damaged region.

15. The method of claim 14, wherein the rupture energy in the peripheral region of the one interface is less than or equal to 1 J/m.sup.2 at the time of the partially parting of the at least two substrates along the one interface in the damaged region.

16. The method of claim 1, wherein the rupture energy in the peripheral region of the one interface is less than or equal to 1 J/m.sup.2 at the time of the partially parting of the at least two substrates along the one interface in the damaged region.

17. The method of claim 16, wherein the rupture energy in a remainder of the one interface outside the peripheral region of the one interface is greater than or equal to 1 J/m.sup.2 at the time of the partially parting of the at least two substrates along the one interface in the damaged region.

18. The method of claim 17, wherein the rupture energy in the remainder of the one interface outside the peripheral region of the one interface is greater than or equal to 1.5 J/m.sup.2 at the time of the partially parting of the at least two substrates along the one interface in the damaged region.

19. The method of claim 1, wherein the fluid applied in the space between the partially parted at least two substrates is selected from the group consisting of deionized water, ethanol, water vapor, aqueous ammonia and hydrazine.

20. The method of claim 1, wherein the partially parting of the at least two substrates along the one interface in the damaged region comprises inserting a blade into the structure along the one interface in the damaged region between the at least two substrates, and wherein applying the fluid in the space between the partially parted at least two substrates comprises applying the fluid in the space while the blade remains inserted between the at least two substrates.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) Other features and advantages of the disclosure will emerge from the detailed description that follows, with reference to the appended drawings in which:

(2) FIG. 1 is a cross-sectional view of a structure to be separated;

(3) FIGS. 2A and 2B illustrate, as top views, two methods of damaging the interface;

(4) FIGS. 3A and 3B illustrate the successive steps of separating the structure;

(5) FIG. 4 is a structure of silicon-on-insulator type that may be separated according to one exemplary embodiment of the disclosure; and

(6) FIG. 5 illustrates the insertion of the blade with a view to separating the structure from FIG. 4.

DETAILED DESCRIPTION

(7) With reference to FIG. 1, a structure S to be separated comprises two substrates S1 and S2.

(8) At least one of these substrates is intended to be used in electronics, optics, optoelectronics and/or photovoltaics.

(9) The structure S furthermore comprises two separation interfaces I1, I2 respectively having rupture energies 1, 2 (expressed in J/m.sup.2).

(10) As mentioned above, at least one of the interfaces I1 and I2 may be a bonding interface, while the other interface is an interface of another type (for example, resulting from epitaxy, deposition, etc.).

(11) Alternatively, the interfaces I1 and I2 may both be bonding interfaces.

(12) For example, the substrates S1 and S2 may have been bonded along the interface I2, while the interface I1 is an interface formed during the epitaxy of a material on a support, the material and the support together forming the substrate S1.

(13) It is considered that the separation must take place along the interface I1.

(14) Naturally, the structure S could comprise more interfaces without departing from the scope of the present disclosure. The teaching relating to the interface I2 will then be applied to each of the other interfaces.

(15) The interface I1 is advantageously an interface susceptible to stress corrosion.

(16) More specifically, the chosen materials on either side of the interface I1 generate siloxane bonds, which are capable of being broken by a fluid under the action of a parting force for parting the substrates S1 and S2.

(17) The materials on either side of the interface I1 may be identical or different, as long as there are siloxane bonds between the materials.

(18) The interfaces that contain such siloxane bonds are interfaces that involve, in particular, silicon oxide (SiO.sub.2), whether it is native or formed intentionally on a support (by deposition, oxidation, etc.), silicon, when it is assembled by a hydrophilic bonding, and/or silicon oxynitrides.

(19) There are very many means for generating an interface comprising siloxane bonds including, in particular, the processes of bonding, of deposition of an oxide layer, of oxidation of the silicon, of treatment of the silicon with oxygen plasma, or else of implantation of oxygen.

(20) Thus, the interface I1 may be a bonding interface, that is to say, along which two materials have been bonded via molecular adhesion during the fabrication of one of the substrates S1, S2 or of the structure S.

(21) For example, the interface I1 may be formed by bonding two layers of silicon, each covered with a native oxide layer via which they are in contact.

(22) As a variant, the interface I1 may be formed by bonding one layer of silicon, optionally covered with a native oxide layer, and one layer of silicon oxide.

(23) Alternatively, the interface I1 may be formed by a technique other than bonding.

(24) For example, the interface I1 may be formed by weakening a layer of a material containing siloxane bonds, for example, by ion implantation or laser illumination.

(25) Furthermore, according to one advantageous embodiment of the disclosure, the interface I1 has a high rupture energy, that is to say, a rupture energy greater than or equal to 1 J/m.sup.2, preferably greater than or equal to 1.5 J/m.sup.2.

(26) The Maszara method mentioned above for measuring the bonding energy may be applied more generally to the measurement of the rupture energy of an interface.

(27) With regard to the other interface I2, along which it is desired that the separation does not take place, it may or may not be sensitive to stress corrosion.

(28) Furthermore, this other interface I2 advantageously has a high rupture energy that is either higher or lower than the rupture energy of the interface I1 chosen for carrying out the separation.

(29) Before carrying out the separation of the structure S, it is ensured that the interface I1 is weakened in a peripheral region comprising the blade insertion region.

(30) This weakening (which results in a localized reduction of the rupture energy, making it possible, for example, to attain a rupture energy of less than or equal to 1 J/m.sup.2) is obtained by localized damaging at the periphery of the interface I1.

(31) According to one embodiment, illustrated in FIG. 2A, a damaged region R1 of the interface I1 is within a sector of a peripheral ring of the interface I1.

(32) Preferably, the angular amplitude of this sector is between 2 and 30.

(33) According to another embodiment, the damaged region R1 of the interface I1, the width of which in the radial direction is preferably between 0.3 mm and 10 mm, is in the form of a peripheral ring.

(34) Various treatments make it possible to damage the region R1 of the chosen interface I1.

(35) A person skilled in the art will be able to choose to carry out the chosen treatment before or after the assembly of the substrates S1, S2 forming the structure S, in particular, by taking into account the practical conditions for carrying out the process for fabrication of the structure S.

(36) Advantageously, the treatment is carried out before the formation of the structure S; the disclosure thus makes it possible to separate a structure S, even if its fabrication process does not contain any step dedicated to the localized damaging of the interface I1.

(37) According to one embodiment, the damaging of the region R1 is obtained by laser irradiation of the chosen interface I1.

(38) The laser beam is chosen so as to selectively heat the interface I1 to be weakened, causing damaging of the interface I1, and consequently the reduction of its rupture energy.

(39) This damaging may be, for example, the thermal decomposition of a material present at the interface I1 to give a gas phase.

(40) Such is the case, in particular, when one of the materials present at the interface I1 is a polymer or a ceramic.

(41) Alternatively, the damaging of the region R1 is obtained by chemical etching of the chosen interface I1.

(42) Chemical etching is carried out by means of an etchant that makes it possible to selectively etch one of the materials present at the interface I1, without etching the materials present at the interface I2.

(43) A person skilled in the art is able to select the appropriate etchant as a function of the nature of the materials of the interface I1.

(44) The application of ultrasonic waves may also be envisaged for locally reducing the rupture energy of the interface I1.

(45) Irrespective of the damaging treatment chosen, at the end of this treatment, the rupture energy in the region R1 of the interface I1 is lower than the rupture energy in the remainder of the interface I1, and may be considered to be a low enough rupture energy so as not to risk causing fracture of the substrates S1, S2 during the insertion of the blade and the start of separation.

(46) Furthermore, the rupture energy of the interface I1 in the region R1 is lower than the rupture energy of the interface I2, at least in the region provided for the insertion of the blade.

(47) This makes it possible to ensure that, during the insertion of the blade, the start of separation indeed takes place along the interface I1 (which locally has the lowest rupture energy) and not along the interface I2.

(48) With reference to FIG. 3A, the separation comprises inserting a blade B, preferably that is thick, between the two substrates S1, S2 of the structure S, from the periphery thereof, and in applying a parting force to the bevels of the substrates S1, S2.

(49) The term thick is understood to mean that the blade B enables a sizeable parting of the substrates S1, S2, so as to enable the physical separation thereof without coming into contact with the front faces (i.e., the faces of the substrates S1, S2 located at the interface I1) in order to avoid damaging them.

(50) Furthermore, the blade B must be inserted between the substrates S1, S2 along a plane parallel to the plane of the separation interface.

(51) During the separation and in order to avoid any rupture of the substrates S1, S2, the substrates S1, S2 are held by a support arranged so that at least one of the substrates is capable of being deformed.

(52) Thus, according to one preferred embodiment, the structure S is positioned vertically in a separation device that comprises, in its lower part, a structure-holding member and, in its upper part, a separation member comprising the blade B that can move vertically in translation in the axis of the holding member.

(53) The holding member comprises a groove that has a base and inclined edges on either side of the base. The base of the groove is wide enough to receive the assembled structure without exerting stress thereon, while the edges are high enough to prevent the substrates S1, S2 from falling out after their separation.

(54) The displacement of the blade B in the direction of the inside of the structure causes a wedge effect and the parting of the two portions thereof along the interface I1 (see FIG. 3B, in which the interface I2 has not been represented).

(55) This parting of the two portions over a length L has the effect of initiating the formation of a separation wave.

(56) After the parting of the substrates S1, S2 has started, a fluid F that promotes stress corrosion is applied in the space between the substrates S1, S2.

(57) Under the combined effect of this fluid F and the parting force exerted by the blade B, the siloxane bonds of the interface I1 break, which results in a significant reduction in the rupture energy of the interface I1.

(58) In particular, the rupture energy becomes low enough to prevent any risk of the substrates S1, S2 breaking during application of the parting force.

(59) Advantageously, the structure S is held in a vertical position during the separation.

(60) This is because this position favors the flow of the fluid F used for the stress corrosion along the interface I1.

(61) In this case, the blade B is preferably vertically oriented and introduced at the top of the structure S so that the separation wave moves downward, becoming horizontal as it moves away from the insertion point of the blade B.

(62) Among the fluids that promote stress corrosion, non-limiting mention may be made of deionized water, ethanol, water vapor, aqueous ammonia and hydrazine.

(63) The fluid F may be introduced between the substrates S1, S2 in various ways.

(64) Thus, according to one embodiment, the structure S may be partially submerged in a bath of the fluid F promoting stress corrosion.

(65) Alternatively, the fluid F may be sprayed, preferably continuously, onto the structure S, in particular, at the blade insertion region, once the blade B has been introduced between the substrates S1, S2.

(66) In the case where the interface I2 is also sensitive to stress corrosion, it is important to initiate the opposite, dry separation, i.e., separation in the absence of any fluid that promotes stress corrosion in the blade insertion region, and wait for the separation to be started in order to bring the chosen interface into contact with the fluid that promotes stress corrosion.

(67) This is because, if the various interfaces sensitive to stress corrosion were brought into contact with the fluid in the blade insertion region, as soon as the blade was inserted, the fluid would have the effect of reducing the rupture energy of each of these interfaces, thus leading to an equalizing of the rupture energies of the interfaces.

(68) On the other hand, dry initiation makes it possible to prevent such equalizing and to initiate the separation along the chosen interface, which has, at least locally, the lowest rupture energy.

(69) Once the separation is started, bringing the chosen interface I1 into contact with a fluid that promotes stress corrosion makes it possible to facilitate and accelerate the separation by reducing the rupture energy of the interface.

(70) The insertion of the blade in the presence of the fluid is continued until the substrates are completely separated.

(71) The disclosure is particularly suitable for separation along interfaces of silicon/silicon oxide type, and for any other interface that is sensitive to stress corrosion and that has a high rupture energy.

(72) The separation described above may be obtained on structures of all dimensions.

(73) In particular, the structure may consist of substrates of large diameter, for example, having a diameter of 300 mm.

Exemplary Embodiment

(74) FIG. 4 illustrates a substrate S1 to be separated, the substrate being a structure of silicon-on-insulator (SOI) type.

(75) The substrate S1 successively comprises a support substrate 1, a buried silicon oxide layer 2, sometimes denoted by the term BOX (acronym for Buried OXide), and a thin silicon layer 3, referred to as an active layer, which is generally intended to receive components for electronics, optics, optoelectronics and/or photovoltaics.

(76) The processes that enable the fabrication of such a substrate are well known to a person skilled in the art.

(77) Mention will especially be made of layer transfer processes and, more particularly, of the SMART CUT process.

(78) According to one exemplary embodiment, the SMART CUT process typically comprises: the provision of a donor substrate comprising the layer of silicon to be transferred to the support substrate; the formation of an oxide layer at the surface of the donor substrate (for example, by thermal oxidation); the introduction of atomic species (for example, by implantation) into the donor substrate, so as to create a weakened zone that delimits the layer of silicon to be transferred to the support substrate; the bonding via molecular adhesion of the oxide layer to the support substrate; the breaking of the donor substrate along the weakened zone, resulting in the transfer of the layer of silicon to the support substrate; optional steps for finishing the transferred layer of silicon (annealing, polishing, etc.).

(79) The substrate S1, therefore, comprises two interfaces: the interface I1 between the silicon layer 3 and the oxide layer 2 (which is an interface resulting from the oxidation of the silicon), and the interface I2 between the oxide layer 2 and the support substrate 1 (which is a bonding interface).

(80) The silicon/oxide interface I1 has a high rupture energy, of the order of 1.6 J/m.sup.2.

(81) The energy of this interface I1 may be locally reduced so as to be lower than the rupture energy of the interface I2 in the region R1 (see FIG. 3A) of the start of the separation.

(82) For example, it is possible to locally weaken the interface I1 by laser irradiation or by an ultrasound treatment.

(83) As illustrated in FIG. 5, the substrate S1 is bonded, by means of the thin layer 3, to a second substrate S2, which provides a second support bevel for the blade B used for the separation. A third interface I3 is thus formed between the thin layer 3 and the substrate S2.

(84) The rupture energy of interface I3 is chosen so as to be greater than the rupture energy of the interface I1 in the region R1, in order to prevent, during the insertion of the blade B, the separation from taking place along the interface I3.

(85) The insertion of the blade B between the substrates S1 and S2, in the damaged region R1 of the interface I1, makes it possible to start the separation along interface I1, then deionized water is applied, for example, by spraying, in the space between the substrates S1 and S2, until the two substrates are completely detached.

(86) This method, therefore, makes it possible to detach the active layer of an SOI, even when no steps have been taken during its fabrication in order to make it detachable.