Separation method for a layer system comprising a wafer, use of a polymethacrylate disc in the method and apparatus for carrying out the method
Patent Information
- Authority / Receiving Office
- DE · DE
- Patent Type
- Patents
- Current Assignee / Owner
- NISSAN CHEM CORP
- Filing Date
- 2008-12-23
- Publication Date
- 2026-07-30
Smart Images
Figure 00000000_0000_ABST
Abstract
Description
[0001] The invention relates to a method for mechanically separating a layered composite from a first support, wherein the layered composite It comprises a wafer and a second, stretchable support. It also concerns the use of a specific separating aid for separating a Layer composite and a device for carrying out the method according to the invention. State of the art
[0002] In the semiconductor industry, there is a constant need for ever thinner components or wafers from which these components are individually produced. Layered composite solutions are used in the production and especially the thinning of wafers to create the layer to be thinned. to protect and mechanically stabilize wafers during the thinning process. These layer systems also perform the aforementioned functions in further processing steps. In particular, however, they serve to process thinned wafers, which, due to their small thickness, are mechanically particularly challenging. to stabilize sensitive materials. Layered systems made of materials best suited to the specific purpose, such as... are used for this purpose. Films, additional wafers, or glass plates. These components of the layered systems can in turn be combined with wax, elastomers, or other materials. Plastics.
[0003] If in such layer systems a glass plate or a disk with mechanical properties comparable to the wafer to be thinned Properties – such as those naturally possessed by another wafer – are used as a carrier, which in particular serves to To stabilize the entire layer system, it is necessary that a layer be placed between the surface to be protected and this carrier. is present, which connects the wafer to be processed and the substrate. This layer must, on the one hand, provide sufficient adhesion. between the carrier and the wafer to be processed; on the other hand, it must be able to maintain the regularly occurring surface to compensate for the topographic irregularities present on the wafer to be processed. These topographic irregularities include, for example, caused by electronic components on the active side of the wafer, which of course needs protection in the case of thin materials, as well as, for example, by... Contacts such as bumps. The connecting layer can of course also be a layered system comprising several different layers. which, with regard to their properties, such as adhesion, elasticity, hardness, thermal conductivity, chemical resistance, etc., are in the sense of advantageously complements the intended use.
[0004] In the production of very thin wafers, the thinning process places a high mechanical stress on the wafer being thinned. exercised. Since the wafer to be thinned has often already undergone a large number of manufacturing steps and, in particular, has frequently already had the electronics processed. Since the components are located on its active side (front side), it is also of great economic importance that during the thinning process... Wafer breakage and thus rejects are avoided as much as possible. To this end, the wafer is protected before high mechanical stress occurs. still in its unthinned state, it is connected to the substrate. Of course, the back side to be thinned (i.e., the side that the (Excluding electronic components) remain free. Connecting the wafer to its substrate is also called bonding.
[0005] After bonding, the wafer is thinned from its back side, which – as already indicated above – results in mechanical stability is lost. This is compensated for during thinning and subsequently by the support structure. Accordingly, the The thinner the wafer is at the end of the thinning process, the more stabilizing the support structure becomes. Accordingly, at relatively low thicknesses... Thick wafers can also serve as carriers for the final product, even when the wafer is thin. In principle, it is possible that the wafer is already divided into its individual components when it is thin. Components are separated. This is desirable in many cases, as it eliminates a subsequent separation step. Separation can be achieved, for example, by... This can be achieved, for example, by providing deep depressions on the active side of the wafer between the individual building elements (dices). are sufficient that these indentations are already touched during the thinning of the back sides and are therefore continuous.
[0006] The company DISCO HI-TEC offers a process called “Dicing by Grinding” in which the wafer is cut by means of Thinning can also be done in stages. In this process, structures are ground or scored onto the front side of the wafer before the carrier is applied. or etched. These structures have a depth greater than the final thickness of the wafer to which it is thinned. Thinning causes Accordingly, as described above, the structures were opened and the wafer was thus separated.
[0007] Singling leads to an overall reduction in the risk of breakage, but the singled components themselves are in turn more vulnerable must be protected from mechanical stresses.
[0008] A problem with thinning is always that after thinning, the support must be separated from the wafer. In particular, the The carrier must therefore be separated because it restricts access to the electronic components on the active side. A relatively thick (only slightly thinned) wafer is still mechanically resilient to a certain extent, and – the more mechanically resilient the The thinner the wafer is at its final thickness, the more flexible the substrates that can be used. Accordingly, flexible substrates, e.g., in film form, can be easily used. mechanically removed.
[0009] With less flexible substrates, such as glass plates or other wafers, such peeling is of course not possible. Here, the Separation is particularly difficult because hard substrates are used especially in highly thinned wafers, resulting in high mechanical stresses. must be avoided at all costs.
[0010] For this purpose, layers are often inserted between the wafer and the support, which are chemically or physically modified to The adhesive strength between the wafer and the substrate decreases or disappears. An example of such a layer is wax, which is affected by heat. The wax softens, thus facilitating separation. For this purpose, the wax is heated until it is sufficiently liquid so that the substrate can be separated. The wafer can be moved. A disadvantage of this method is that it particularly affects the sensitive wafer surface, which contains the electronic components. The components, or rather the contacts to them, must subsequently be cleaned.
[0011] In principle, the use of special adhesives is also possible, which in turn achieve their properties through thermal or radiation action. Reduce adhesion. One such method is offered by the manufacturer 3M, in which a layer of adhesion promoter is removed from under the The use of laser energy is achieved.
[0012] Here again, it is problematic that parts of the adhesion-promoting layer remain on the wafer surface, which are difficult to clean. must be.
[0013] Since cleaning involves additional effort, mechanically and often also chemically stresses the thinned wafer and / or Since it is very difficult to guarantee that all adhesive residues are sufficiently removed, purely mechanical methods are not suitable for many applications. Solutions are preferred in which a separating layer is mechanically peeled off from the active side of the wafer, leaving no residue. remain.
[0014] Such a separating layer is disclosed in WO 2004 / 051708. This separating layer is in turn disclosed in WO 2007 / 099146, further developed thinning processes are used. In the latter patent application, after thinning the wafer, the (thinned) A second support is provided on the reverse side to assist the separation process from the first support. The separation takes place between the separation layer. and the active side of the wafer. WO 2007 / 099146 discloses a mechanical separation process in which the wafer composite and The material is guided over a roller on a second carrier, so that the cutting process is mechanically effected. This second carrier is often a saw-cutting film. In this method, the second support must be flexible, as it needs to be able to follow a curve. The problem with this method is that... The stretchable beam, when separated, can itself undergo a deformation in addition to the curvature caused by the roller, so that The separation process is not sufficiently controlled. This is particularly true for very thin wafers, which are mechanically very sensitive. The method disclosed in WO 2007 / 099146, when using the deflection pulley, too frequently results in excessively strong forces perpendicular to the wafer plane, leading to... Wafer breakage can occur because the stretchable carrier does not fully follow the curvature of the roller surface. This results in, that the separation front is not directly under the roller's support surface, but is slightly shifted laterally. This increases the vertical component of the force exerted on the wafer.
[0015] The object of the present invention was therefore to provide a method by which the safest and gentlest possible separation of the Wafers are made possible by the stabilizing support during thinning.
[0016] This problem is solved by a method for mechanically separating a layered composite from a first support, wherein the The layered composite comprises a wafer and a second, stretchable support, including the following steps: a) Providing a layered system encompassing the first carrier, the wafer and the second carrier, b) providing a separating aid, c) fixing the separating aid to the second carrier, so that during the separation process the second support remains fixed to the separation aid immediately behind a separation front created during separation, andd) mechanical separation of the layered composite from the first support using a separation front.
[0017] The wafer may preferably be silicon wafers, which may optionally be doped. Wafers made of AlSb, AlAs, AlN, AlP, BN, BP, BAs, GaSb, GaAs, GaN, GaP, InSb, InAs, InN or InP may be preferred for many applications.
[0018] Unless otherwise indicated in the following text, the wafer can regularly be replaced by the The separation process also includes the components that are already isolated from a wafer.
[0019] In this context, a stretchable support is a support that, under the mechanical separation conditions of the carried out The separation process can deform and, in particular, stretch the material. This stretching is especially problematic because it can cause the... Controlling a separation front (see below) is not sufficiently possible if the expansion movement is not controlled as such. Preferred Stretchable carriers include the saw foils commonly used in industry, but also other foils that are adhesive on one side or otherwise sticky.
[0020] In the context of this application, “fixing” means that a connection is made by a force acting perpendicularly The force acts between the two areas to be fixed. Preferred fixing methods are gluing, suction by negative pressure, and electrostatic bonding. Holders. The method of guiding over a roller, as shown in WO 2007 / 099146, is expressly not considered fixing within the meaning of this application. of tensile force.
[0021] In the context of this text, the separation front is the area in which the separation takes place. This means that, ideally, it is A fall around a line where the adhesive force between the areas to be separated drops abruptly from 100% to 0%. In practice, the separation front is always Let the area be a surface, where the adhesive force ranges from < 100% to > 0. The end of the separating front is, of course, the The area where the adhesive force reaches 0, the beginning where it is just < 100%. The separation front shifts across the area during a typical separation process. Separation plane, until the surfaces to be separated are separated and therefore no longer exert any adhesive force against each other.
[0022] The adhesive force (adhesion strength) between the areas can be determined by a person skilled in the art according to DIN 971 / 1:1996 / 09 and it is defined as the “totality of the binding forces between a coating and its substrate”.
[0023] Step c) of the method according to the invention is carried out in such a way that the second support is forced in the area which is directly behind the separating front, i.e. adjacent to it, but in which the separation between the wafer and the first carrier has already taken place, the to follow the movements of the separating aid precisely. In other words, the (surface-based) fixation results in the elasticity of the second support being affected. does not affect the position of the dividing front. In the method disclosed in WO 2007 / 099146, contrary to the invention, In this process, a flexible second support can detach (albeit slightly) from the deflection pulley because the adhesive force between the wafer and the first one is too great. The carrier counteracts an ideal following of the deflection pulley movement (or along the circumference of the deflection pulley).
[0024] The core of the invention is to eliminate this negative effect. It is necessary to ensure that the second, stretchable support so that it is fixed to the separating aid that the position of the separating front is not changed by the elasticity of the support. As already indicated, this is necessary. It is crucial that the carrier is fixed directly behind the cutting edge in such a way that it must follow the movements of the cutting aid. This fixing can be temporary during the cutting process, for example by a roller that is electrostatically fixed directly behind its contact surface. Force is applied to fix the second support over its entire surface. Alternatively, and preferably, the second support is already fixed over its entire surface before the separation process begins. The separating aid is fixed in place, and this fixation is not released during the separating process. However, this is not necessary for the implementation of the invention. Absolutely necessary; it has only advantages in terms of practicality.
[0025] Any object that has a (partially) convex surface, or that is convex, can serve as a separating aid. The surface can be deformed. At the same time, the separating aid must be designed in such a way that the fixing of the second support is at least This is ensured immediately behind the cutting edge. Suitable cutting aids can be, for example, rollers, especially those with a large circumference. or flexible, bendable sheets, such as plexiglass sheets, other plastic sheets or thin metal sheets.
[0026] Of course, it is possible that in the method according to the invention, a or several other layers are present. The layer composite to be separated may also contain additional layers besides the wafer and the comprise a second (stretchable) support. However, what is crucial for the application of the present invention is that any further supports that may be present The layers in the layered composite to be separated are materially equipped in such a way that they a) provide the decisive factor for the formation of the separation front b) follow the curvature of the separating aid (which may only be generated during the separating process) and b) that it determines the position of the separating front. not affecting them by causing them to deform.
[0027] It should also be noted that fixing the second support to the separating aid “immediately behind the separating The resulting separation front refers in terms of its position to the plane perpendicular to the separation front: The adhesion between the separation aid and the second The carrier is of course not located in the separating plane.
[0028] As already indicated above, it is crucial for the method according to the invention that the fixing of the second carrier to the separating aid This is done in such a way that the position of the separating front is not affected by the material properties of the second support. The fixation can be permanent. or be temporary, and it does not have to encompass the entire already detached part of the second support. Only the area is decisive. behind the dividing front. Here, "area" in the sense of fixation means that perpendicular to the dividing direction in the dividing plane (or in the plane emerging from the (Separation plane, protruding curvature of the separating aid) the fixing surface has a depth greater than 0, wherein the width of the planar fixing preferably determined by the dividing front.
[0029] The mechanical separation in step d) of the method according to the invention is carried out by utilizing a curvature which is either in the The separating aid is already present or can be created within it. If the separating aid is, for example, a roller, the mechanical separation is achieved by Unroll the roll over the surface of the second carrier facing away from the wafer. The second carrier must first be fixed to the roll. Preferably, the fixation takes place in an area where the second carrier protrudes beyond the wafer (loose end). After fixing, the roller is guided in a rolling motion over the surface of the layer system, so that a layer is applied to the layer system. Shear force acts (also due to the fixation of the second support to the roller) and thus – with appropriate design of the layer system – the layered composite of wafer and second, stretchable support from the rest of the layered system, in particular the first support is separated. As already described several times, it is crucial in this context that the second support is fixed to the The separating aid (roller) must always be present directly behind the separating front. This fixation must therefore be ensured during the separating process, especially in the case of a roller. This will occur immediately after separating each section of the second beam (and thus the layer composite to be separated). Naturally, the second carrier and also the wafer connected to it are subject to the curvature of the separating aid.
[0030] In an alternative and preferred separation process, the separation aid is flat. Here, a large-area [device] can be applied before separation in step d). Large sections or the entire surface of the second support are fixed to the separating aid. Separation is achieved by bending the Separation aid, so that it forms a convex shape in the area of the separation front (or initially where the separation front is created) relative to the layer system. assumes a curvature (analogous to a large-circumference roller surface). Here too, the second support, due to its fixation, becomes necessary. forced to follow the convex curvature of the separating aid and with it – again with appropriate design of the layer system – the wafer. Naturally, care must be taken during all separation processes to ensure that the first carrier is not subject to the forced separation. This results in convex deformation of the layered composite, creating a shear force between the layered composite and the first support.
[0031] It should be noted once again that the method according to the invention allows for particularly good control of the separation front. is. In fact, especially with a large radius of convex curvature in the region of the dividing front, the dividing front becomes in The separation direction is widened, meaning that the area in which the adhesive force is just less than 100% (i.e., just less than the adhesive force without) The force vector acting on the separating aid is increased until complete separation (adhesion force = 0). This results in the force vector acting on the The layer system has a relatively small portion perpendicular to the separation plane. This means, in particular, that the undesired Folding, as is known, for example, when unrolling adhesive tape, is avoided. In other words, due to the invention, it is The process makes it possible to limit the particularly critical mechanical stress, i.e., the force acting perpendicular to the parting line, and to control. This can be achieved, for example, by using suitable materials for the (flexible) separating aid, by using appropriate surface geometries (such as e.g. roller diameter) and of course through suitable material selection for the carrier (as well as for the entire layer system) become.
[0032] A preferred method according to the invention is wherein the separating aid is fixed to the second carrier by means of a vacuum or by means of electrostatic charging occurs. These forms of fixation have the advantage that they can be applied locally (if desired), i.e., only to a part of the Separation aids can be used in a precisely controllable manner; in particular, they have the advantage that the fixing is achieved by switching off the negative pressure. or can be easily reversed by removing the electrostatic charge.
[0033] As already indicated, the method according to the invention makes it possible to produce very thin wafers or layer composites that have a very thin wa to separate the wafer from a first support. Accordingly, a preferred method according to the invention is one in which the wafer a thickness of ≤ 400 µm, preferably ≤ 150 µm, further preferably ≤ 80 µm, particularly preferably ≤ 20 µm and entirely especially preferred ≤ 10 µm.
[0034] A preferred method according to the invention is wherein the layer system provided in step a) is positioned between the wafer and the first The carrier has one, two, three, four or more layers, and that the separation occurs in one of these layers, between two adjacent surfaces these layers or between the surface of the wafer and the adjacent layer or between the surface of the first support and the adjacent layer.
[0035] In this context, it is of course preferred that the separation between the surface of the wafer (active side of the wafer) and the adjacent layer. This makes it possible to access the active side without leaving any residue that still needs cleaning on it. remain on the wafer.
[0036] As already indicated above, it is particularly important when producing very thin wafers that the first support has a high possesses mechanical strength. Accordingly, for the method according to the invention, it is preferred that the first support be a glass plate or a silicon wafer (with a greater thickness than the thinned wafer). The preferred material for the glass plate is borosilicate glass or quartz glass. Provided that If the first support is made of inflexible materials, fixing the first support during the separation process is not necessarily required. required because it does not conform to the convex curvature of the layer composite (consisting at least of the (thinned) wafer and second support) during the separation process. Nevertheless, for reasons of procedural reliability, it is preferred to also use an inflexible first support during the separation. to fix the separation.
[0037] The first carrier can also be material that is applied to the wafer surface (directly or indirectly) by It is applied by spin coating, spray coating, laminating and / or compression coating. This material can be cross-linked in the process. or otherwise modified in its consistency to influence the desired mechanical properties.
[0038] A preferred method according to the invention is wherein at least one separating layer is located between the first support and the wafer is located. A separation layer has the task of deliberately creating a predetermined breaking line for the separation, i.e., a plane in which the separation front runs during the To enable and / or ensure separation.
[0039] It should also be noted that the separation plane is not continuous, depending on the surfaces being separated. It must be even. This is especially true when the separation between a structured area of the front (active) side of the wafer and a layer above it.
[0040] It is preferred for the method according to the invention that at least one of the separating layers is a plasma polymer layer.
[0041] The advantage of plasma-polymer layers is that they can be applied in a contour-following manner, thus optimally adapting the surface of the active side of a wafer can be protected and their adhesive properties can be optimally adjusted to the separation requirements.
[0042] A further preferred method according to the invention is in which the layer system provided in step a) is between the first The carrier and the wafer comprise a layer of a cured, partially cured or curable elastomer material.
[0043] This elastomer material has advantages, in particular, when bonding the (possibly structured) wafer surface onto which A (plasma-polymer) separating layer may optionally be applied, with a rigid, optionally flat, first support.
[0044] Accordingly, a method according to the invention for separating a wafer is particularly preferred, wherein the layer system the The following layers, in the specified order, comprise: a first support, a layer of elastomeric material as described above, a plasma polymer separation layer, a wafer and a second support.
[0045] Such a layer system is described in the above-mentioned WO 2007 / 099146, which is fully available by reference This application will include, in particular, the parts that define the design of the separating layer and the Describe the elastomer layer. Also incorporated by reference is WO 2004 / 051708, which also It contains information on suitable separating layers. In particular, this information is incorporated into this application by reference.
[0046] According to the invention, a method can also be preferred in which the layer system between the first support and in the layer is made of one of The hardened, partially cured, or curable elastomer material contains a second separating layer. This second separating layer makes it possible to to free the first carrier from the remaining (residual) layer system after separation, so that the first carrier can be easily reused. can be.
[0047] A preferred method according to the invention is one in which the layer system comprises a separating layer as described above and in the separating layer or between one of the two surfaces of the separating layer and the surface adjacent to that surface (of another layer or of the wafer) exhibits a lower adhesive strength than in all other layers and between all other adjacent layers of the A shift system is present.
[0048] In this way, the separation plane that is used during the separation process is predetermined.
[0049] According to the invention, it is also preferred to charge the first carrier preferably by means of a vacuum or electrostatic charging during to fix the separation process on a holding device.
[0050] According to the invention, a method in which the separating aid comprises a surface that is at least partially convex is also preferred or the surface of the separating aid can be deformed, at least partially, into a convex surface. Preferably, the (existing) or during The deformation is achieved by adjusting the curvature of the surface so that the wafer does not break when it is removed from the first support. Such a setting The person skilled in the art can easily determine this when carrying out the method according to the invention.
[0051] For the method according to the invention, the surface of the separating aid to which the second carrier is attached, at least during the separating process, is preferred. is fixed to the surface of a cylindrical shell or a cylindrical section. This is also the case when the separation is achieved using a Separation aid is provided, which is convexly deformed in the area of the separation front.
[0052] A preferred method according to the invention is wherein the separation aid is a roller and the roller is used to separate the wafer from the first The carrier is rolled over the side of the second carrier facing away from the wafer, and the second carrier is immediately behind the separation front on the roller. is fixed.
[0053] A method according to the invention is particularly preferred in which the separating aid is a flexible plate.
[0054] In a particularly preferred method according to the invention, the flexible plate is removed by applying a Force that acts at least partially perpendicular to the surface of the flexible plate and is directed away from the first support, at least partially (from (Incipient separation in the area of the separation front) convexly deformed. Initially, only the distance of a part of the flexible plate from the first Carrier enlarged.
[0055] A preferred method according to the invention is wherein the first support is fixed on a holding means and a flexible plate serves as a separating aid. is used and the holding agent and the flexible plate each have at least one shape and which prevent separation (and deformation of the The force acting between at least one shape of the flexible plate and at least one shape of the holding element (flexible plate) acts (or attacks there).
[0056] In this case, the shapes are preferably protrusions in accordance with the last described preferred method of the invention. (Protrusions) to which the force causing the separation can be applied. Preferably, one or more shapes of the flexible plate and / or a bridge on the bracket. It is particularly preferred that the bridges are positioned opposite each other in such a way that a pushing force is exerted between them. It can be created that ultimately causes the separation.
[0057] A preferred method according to the invention is wherein the separating aid is an optically transparent plate, since in this case the The separation process can be particularly well monitored visually. The following materials are especially suitable: polymethyl methacrylate. (abbreviation PMMA, trade names Friacryl®, Vitroflex®, AcrylglasPlexiglas®, Limacryl® or Piacryl) or polycarbonate (abbreviation PC, (Trade names Makrolon, Calibre or Lexan). The selection of materials for the release aid is made by the specialist taking into account the Requirements for separation include the load on the wafer or the requirements of the separation device.
[0058] Preferably, the separating aid, being a Plexiglas plate, has a thickness of 0.5–5 mm, preferably 1–4 mm, more preferably 2–4 mm. mm or 3–4 mm, depending on the set adhesive strength of the separating layer.
[0059] Alternatively or additionally, in a method according to the invention, the curvature of the surface of the separating aid can also be achieved by the This causes the swelling of an erectile tissue. Such an erectile tissue must be properly fixed to the separating aid so that it can... It can cause curvature. Of course, the separating aid itself can also be a hemorrhoid.
[0060] A preferred method according to the invention is in which the second, stretchable carrier is a saw film (also referred to as blue tage). Preferred saw foils are UV-release tapes used as standard in the industry, e.g.: Lintec Adwill D.
[0061] Such a saw-cut foil (or a comparable material) is laminated onto the back of the wafer after it has been thinned. For this purpose, the following are available: The specialist has a wide range of options available. Typically, such a saw blade film has one side that is protected by an applied coating. The adhesive on the back of the wafer adheres to the substrate. The adhesive strength of the second carrier to the wafer can preferably be increased by thermal energy and / or UV light. Radiation is modified to potentially remove the second carrier from the wafer. Saw films (as second carriers) with such These adhesives are also referred to as “thermorelease films” or “UV release films”.
[0062] Saw films are stretchable within the meaning of the present application.
[0063] Typically, and preferably also for methods preferred according to the invention, the saw film extends over the edge of the lamination process. The wafer or the layer system to be separated is extended and held at its edge by a ring. This ring, which is made of metal or a A frame made of plastic is usually called a saw frame or "dicing frame". The frame, together with the film, forms a type of drum, whose membrane is the second support and wherein the rest of the layer system starting with the back of the wafer is on this Membrane (saw foil) is in place.
[0064] In a particularly preferred method according to the invention, the layer system comprising the saw foil is now placed in a device brought, in which the first carrier is fixed to a chuck (preferably by means of negative pressure).
[0065] Now, on the exposed back side of the saw foil (second carrier), another chuck is formed, again preferably by means of a vacuum. fixed. Both chucks are preferably positioned so that they are laterally aligned with each other when viewed from above. The chuck is... to which the saw foil is now fixed, is made of a bendable (flexible) material and is of a thickness that allows it to perform the function according to the invention. can fulfill the separation aid to be used.
[0066] Preferably, the two chucks (holding means and release aid) have at least on one surface of the device an area which extends over the extends beyond the surface of the wafer or layer system. With appropriate fixation, a force can now be exerted between these areas. which results in the chuck (separation aid) holding the saw foil (the second carrier) separating on one side from the other chuck (holding device) is removed. This causes the separating aid to bend, creating the desired separating front.
[0067] In this case, it is very possible to precisely control the applied force so that the wafer is not subjected to excessive stress during the separation process. is claimed.
[0068] Provided that this preferred method is combined with a separating layer and / or the further preferred layer system described above When carried out, it is possible to achieve a clean separation of even very thin wafers without excessive mechanical stress (breakage). to ensure this. It is also possible to measure the applied force and / or adjust it accordingly, so that a maximum force is not exceeded. It is also possible to use the force to set a fixed distance between the two points / lines / surfaces where the force is applied. This allows A separating aid that is originally flat but flexible creates a continuous separating front due to its deformation, since the deformation of the separating aid It spreads over a large area until complete separation has occurred and the separating aid can return to its original form.
[0069] As already described above, it is preferred that the two supports are fixed by means of a vacuum, but there are also other options available. Other options include fixing with adhesives.
[0070] If a saw frame is used in a separation process according to the invention, the saw frame is preferably designed such that The surface curvature of the cutting aid can also be transferred to the saw frame. Thus, the saw frame and the clamped saw film follow the... Curvature in the area of the separation front, wherein, according to the invention, they are fixed directly behind the separation front on the separation aid. The wafer in turn follows the bending (curvature) of the saw foil (second support), so that a shear force is created between the wafer and the first support.
[0071] As already indicated above, an advantage of the method according to the invention, particularly in its preferred embodiments, is that the separation front represents a relatively large area, so that the shear forces perpendicular to the wafer surface are kept low can.
[0072] Without being bound to a theory, a force distribution gradient arises within the due to the method according to the invention. Separation front (i.e., a force distribution gradient sinks) right towards the line of the actual separation (the end of the separation front), i.e. the line where the adhesive force between the surfaces to be separated is 0), whereby the depth of the separation front (and thus the gradient slope) can be optimally controlled within the framework of the inventive method, so that the mechanical stress on potentially very thin wafers can be kept as low as possible.
[0073] Accordingly, a particularly preferred method according to the invention is one in which the layer system comprises the following The layers in the specified order comprise: – a first support consisting of a glass plate or wafer, – a layer of elastomer material as described above, – a plasma-polymer separating layer as described above – a Wafer and – a saw-cut foil as a second carrier.
[0074] Part of the invention is also the use of a Plexiglas plate which is arranged so that a vacuum is used to create a shape as described above. The second support described can be fixed to it, as a separating aid in the mechanical separation of a layered composite (comprising the second Carrier and a wafer) from a first carrier. This naturally includes, for the use of the layer system according to the invention, The layer composite and the first support are preferably designed as for one of the preferred methods described above.
[0075] Part of the invention is also a device which is set up to carry out the method according to the invention. In particular Such a device comprises a means for fixing the first support and a separating aid as described above. This separating aid It also serves as a means for fixing the second support. A device according to the invention preferably comprises means for effecting the Fixation by means of electrostatic or negative pressure force, a holding means for fixing the first carrier and preferably a means by which a force can be exerted only on a part of the separating aid, so that the force causes the separating aid to have a convex curvature away from the first Carrier experiences.
[0076] Preferably, the separating aid is a chuck consisting of a plexiglass plate, which is further preferably subjected to negative pressure. can.
[0077] The invention will now be explained in more detail with reference to drawings and an example. The drawings and examples serve but not to restrict the invention.
[0078] Fig. 1 shows a layer system to be separated, comprising a wafer after thinning.
[0079] Fig. 2 shows a layer system to be separated, which is already fixed for the separation process and
[0080] Fig. 3 shows the layer system to be separated during the separation process. Reference symbol list 1 first carrier 2 elastomeric layer 3 Separating layer 4 structured front side of the wafer 5 back of the wafer 6 part of the layer system to be separated, comprising the first carrier 7 Wafer (front and back) 18 liability mediating side of the second carrier 19 Layer system to be separated (without a second support) 20 second carrier 21 Saw frame 28 Holding device for fixing the first support 29 Separation aid 30, 31 Alternatives for the direction of action and point of application of the force causing the separation 33 dividing front 34 End of the dividing line More detailed description of the characters
[0081] Fig. 1 shows the layer system 19, 20 to be separated. A first support 1, preferably designed as a glass plate, is shown. is connected to an elastomer layer 2, which in turn is connected to a separating layer 3. The separating layer 3 is connected to the active The front side of wafers 4 and 5 is connected. The adhesive force between the separating layer 3 and wafers 4 and 5 is the lowest within the entire Layer system. The second carrier 20 is attached to the back of the wafer 5. This can be designed as a saw-cut foil, which is cut into a Saw frame 21 is clamped in place.
[0082] Fig. 2 shows the layer system 19, 20 to be separated in the state prepared for separation. The first support is on a Means for fixing the first carrier 28 are fixed. The layer system to be separated, which comprises the first carrier 6, is still fully covered with the Wafer (front and back) 7 connected. The separating aid 29 is connected over its entire surface to the second support 20 and parallel to the second support and also aligned with the first support. Arrows 30 and 31 show possible directions of action and points of application for the force causing the separation. on.
[0083] Fig. 3 shows the separation process in which the layer system 19 , 20 to be separated is being separated into the part of the layer system to be separated, comprehensively, the first support 6 and the layered composite 7, 20 are separated. This is achieved by the connection in the area of arrow 31. Force that causes the separating aid 29 to move away from the holding device for fixing the first support 28 and thus also from the first support 1 on one side is moved. In the process, the separating aid undergoes a curvature. The second carrier 20 follows this curvature completely, and with it the wafer 7. This creates a shearing effect in the layer system 19 to be separated, which manifests itself in the region of the separation front 33. At the end of the separation front 34 is the adhesive force between the part of the layer system 6 to be separated and the wafer 0. In the present example, this is determined by the Separation layer 3 ensures that the separation plane runs between wafer 7 and separation layer 3.
[0084] By fixing the second support 20 over a flat area to the separating aid 29 (also) directly adjacent to the separating front 33, a Optimal force transmission is ensured: The vertical forces acting on the wafer are relatively small and within the limits of a Gradients across the separation front 33 increase until the end of the separation front 34. This results in a gradient particularly in the region of the separation front 33. A virtual curvature angle is created that approaches infinity, thus putting as little stress on wafer 7 as possible. Example
[0085] The front side of a wafer, i.e. the side on which the electronic components are located, is coated with a PECVD by means of a The surface is coated with a plasma-polymerized release layer. This release layer is adjusted with regard to its adhesion properties so that it adheres to the surface of the wafer. The side facing the wafer has relatively low adhesion, while the adhesion is high compared to the side facing away from the wafer. applied layer. In this example, the latter layer is a silicone elastomer layer, which is first applied to a glass plate as The first carrier is applied. Subsequently, the wafer with the already applied release layer is pressed onto the glass plate, release layer side facing the glass plate. The applied elastomer layer is pressed (bonded). A layered system now exists comprising the wafer, the release liner, the elastomer layer, and the glass plate as a substrate.
[0086] Bonding can be carried out using typical separators known from the prior art. For details on the creation of the described For an example of the shift system, reference is made to WO 2007 / 099146.
[0087] Subsequently, the wafer is thinned on its exposed back side. The thinning is carried out using methods known in the prior art. Process. The wafer is thinned to a thickness of 50 µm. After thinning, the layer system comprising the wafer is split into a The wafer back is transferred to a processing plant where it undergoes further treatment by etching. Etching is primarily used to remove surface damage. To compensate for imperfections on the wafer back side that arise during thinning. The etching process is carried out according to the state of the art, e.g., wet chemically using Etching bath or dry chemical process using plasma. Etching, in particular, causes damage in the form of chips and cracks. completely eliminated and the wafer gains higher mechanical and thermal resistance.
[0088] Subsequently, a saw film, e.g., Lintec Adwill D, is laminated onto the layer system with its adhesive side. Now the The layer system is placed in a separating device in which the back of the glass plate is fixed by means of a vacuum chuck so that it does not move. more can be postponed.
[0089] A special vacuum chuck, made of a 3 mm thick Plexiglas plate, is now fixed to the back of the saw foil as a cutting aid. consists.
[0090] After the vacuum has been established at both vacuum chucks, a force perpendicular to the plane is applied to one side of the Plexiglas plate. of the wafer. This force causes the flexible Plexiglas sheet to bend, with the bending forming a separating front in the layered system. This creates a separation front that runs in the plane between the front of the wafer and the release layer. The curvature of the plexiglass sheet is convex relative to the first support. During the continuous bending of the Plexiglas sheet (vacuum chuck) of the wafer and the second support. This is not the case for the rest of the layer system, which is fixed to the first support. The resulting shear force causes the Separation of the wafer from the release layer.
[0091] The force can be adjusted so that even very thin wafers, such as the 50 µm thickness in the present example, are not damaged by the The first beam can be separated. This is achieved by moving the separation front at a controllable speed across the entire separation surface. running.
[0092] Finally, the front side of the wafer is freed from the remaining layer system. After separation, the vacuum between the two can be released. The vacuum chucks are switched off. Now the wafer, laminated onto a saw film, can be removed and subjected to further processing steps, such as... B. the Singulation by sawing, feeding in. QUOTES INCLUDED IN THE DESCRIPTION
[0093] This list of documents cited by the applicant was generated automatically and is solely for the better information of the Readers' contributions were included. The list is not part of the German patent or utility model application. The DPMA assumes no liability whatsoever. for any errors or omissions. Cited patent literature
[0094] - WO 2004 / 051708 [0014, 0045] - WO 2007 / 099146 [0014, 0014, 0014, 0020, 0023, 0045, 0086] Cited non-patent literature
[0095] - DIN 971 / 1:1996 / 09
[0022]
Claims
[1] Method for mechanically separating a layered composite (7 , 20 ) from a first support (1 ), wherein the layered composite (7 , 20 ) has a comprising wafer (7 ) and a second, stretchable carrier (20 ), including the following steps: a) Providing a layer system (1 , 7 , 20 ) comprising the first carrier (1 ), the wafer (7 ) and the second carrier (20 ), b) Providing a separation aid (29 ) c) Fixing the separating aid (29) to the second support (20) so that during the separating process the second support (20) is immediately behind a The separating front (33) created during the separation process remains fixed to the separating aid (29), and d) mechanical separation of the layered composite (7 , 20 ) from the first support (1 ) by exploiting a separation front (33 ). [2] Method according to claim 1, wherein the fixing of the separating aid (29) to the second support (20) is carried out by means of a vacuum or by means of electrostatic force. Charging is in progress. [3] Method according to one of claims 1 or 2, wherein the wafer (7 ) has a thickness of ≤ 150 µm, preferably ≤ 80 µm, particularly preferably of ≤ 20 µm and most preferably has a thickness of ≤ 10 µm. [4] Method according to any of the preceding claims, wherein the layer system between the wafer (7) and the first support (1) is a, comprising two, three, four or more layers and the separation in one of these layers, between two adjacent surfaces of this layers or between the surface of the wafer (7) and the adjacent layer or between the surface of the first support (1) and the adjacent layer. [5] Method according to any of the preceding claims, wherein at least one separating layer (3 ) is located. [6] Method according to claim 5, wherein at least one of the separating layers is a plasma polymer layer. [7] Method according to one of the preceding claims, characterized in that the layer system between the first carrier (1 ) and the wafer (7) comprises a layer (2) of a cured, partially cured or curable elastomer material. [8] Method according to claim 1 for separating a wafer, wherein the layer system comprises the following layers in the specified order includes: a first carrier (1 ) a layer (2 ) of elastomeric material as described in claim 7, a plasma polymer separation layer (3 ) a wafer (7 ) and a second support (20 ). [9] Method according to one of claims 7 or 8, wherein the layer system between the first carrier (1) and the layer (2) is made of a a second separating layer comprises cured, partially cured or curable elastomer material. [10] Method according to any one of the preceding claims, wherein the layer system comprises a separating layer (3) as in any one of claims 5, 6 or 8 defined includes and in the separating layer (3) or between one of the two surfaces of the separating layer and that surface The adjacent surface exhibits a lower adhesive force than in all other layers and between all other adjacent layers of the A shift system is present. [11] Method according to one of the preceding claims, wherein the first support (1 ) is fixed on a holding means (28 ). [12] Method according to claim 11, wherein the fixing is carried out by means of a vacuum or by means of electrostatic charging. [13] Method according to any of the preceding claims, wherein the separating aid comprises a surface that is at least partially convex or which The surface of the separating aid (29) can be deformed at least partially into a convex surface. [14] Method according to claim 13, wherein the curvature of the convex surface of the separating aid (29) is such that the wafer is separated from the first support does not break. [15] Method according to one of the preceding claims, wherein the surface of the separating aid (29) is fixed to the second support, which Surface of a cylindrical shell or a cylindrical section. [16] Method according to any of the preceding claims, wherein the separating aid is a roller. [17] Method according to claim 16, wherein the roller for separating the wafer from the first carrier via the side facing away from the wafer of the The second carrier is rolled and the second carrier is fixed directly behind the separating front on the roller. [18] Method according to any one of claims 1 to 14, wherein the separating aid (29) is a flexible plate. [19] Method according to claim 18, wherein to detach the wafer (7) the flexible plate (29) is at least partially removed by applying a force acts perpendicular to the surface of the flexible plate (29) and is directed away from the first support (1), is at least partially convexly deformed and the The distance of at least part of the flexible plate (29) to the first support (1) is increased. [20] Method according to one of claims 18 or 19, wherein the first support (1 ) is fixed on a holding means (28 ), the flexible plate (29 ) and the retaining means (28) each have at least one shape and wherein the force between at least one shape of the flexible plate (29) and at least one embodiment of the holding agent (28 ) acts. [21] Method according to claim 20, wherein one or more of the shapes of the flexible plate (29) and the retaining means (28) each have a projection or is an exception. [22] Method according to one of claims 20 or 21, wherein one or more of the shapes of the flexible plate (29) and the retaining means (28) is a jetty. [23] Method according to any one of claims 18 to 22, wherein the separating aid (29 ) is a Plexiglas sheet. [24] Method according to any one of claims 1 to 12, wherein the separation aid includes a swelling element and the separation is effected by swelling of the erectile tissue. [25] Method according to one of the preceding claims, wherein the second support (20 ) is a saw foil. [26] Method according to claim 1 for separating a wafer (7) , wherein the layer system (1 , 7 , 20 ) comprises the following layers in the The order listed includes: a first support (1 ) consisting of a glass plate or a wafer, a layer (2 ) of elastomeric material as defined in claim 7, a plasma polymer separation layer (3 ) a wafer (7 ) and a saw film (20 ). [27] Use of a Plexiglas plate (29) which is configured to support a second support (20) as in any of the preceding claims described as being able to be fixed to it by means of negative pressure, as a separation aid (29) in the mechanical separation of a layered composite comprising the second carrier (20 ) and a wafer (7 ) from a first carrier (1 ). [28] Device for carrying out a method according to one of claims 1 to 26, comprising a holding means (28 ) for fixing the first carrier (1) and a holding means for fixing the second carrier (20), wherein the holding means (28) for fixing the second carrier (20) is configured as The separating aid (29) is designed as defined in one of the preceding claims. [29] Device according to claim 28, wherein the separating aid (29) is designed as a flexible plate and wherein the device comprises a means by which one force of which can only be exerted on a part of the separating aid (29 ), causing the separating aid (29 ) to have a convex curvature away from the first carrier (1 ).