Method for preparing a stack for bonding

Abstract

The invention relates to a method for preparing a stack (1), which method comprises providing a stack that comprises a substrate (10) on which a plurality of structures (1000) is mounted. Each structure comprises a pad (100) comprising a top (101) and at least one flank (103), and a first layer (200) mounted on the substrate, the first layer covering a portion of the flank of the pad, a second layer (300) covering at least the top of the pad. The structure comprises a removable section (500) that covers an upper portion (104) of the flank of the pad and is formed by at least one of the first layer and the second layer, and a covering section (600) that covers a lower portion (105) of the flank and extends from the removable section. The removable section is then removed leaving the covering section in place to form a protrusion.

Description

[0001] “Process for preparing a stack for bonding”

[0002] TECHNICAL FIELD OF THE INVENTION

[0003] The present invention relates to the preparation of a substrate, in particular with a view to bonding it with another substrate. It is particularly relevant to the manufacture of optoelectronic and microelectronic devices. It finds, for example, a particularly advantageous application in the field of assembling a substrate having copper-based elements with a substrate covered with a copper film.

[0004] STATE OF THE ART

[0005] Joining two substrates is a very common operation in microelectronics and optoelectronics. Such joining can notably aim to transfer elements, for example photo-elements such as pixels, from one substrate to another.

[0006] A common technique for joining two substrates illustrated in Figures 1A to 1C is based on direct bonding between protrusions 200' based on a dielectric, for example SiC>2, located on the surface of a first of the two substrates, and the second substrate 2'. Other elements of the first substrate T, typically copper pillars 100', are set back relative to the SiC>2 protrusions 200'. During the step of joining the two substrates, a heating step causes the copper pillars to expand. The expansion of the SiC>2 protrusions occurring in parallel is less significant than that of the copper pillars. Therefore, with the correct dimensioning of the withdrawal of the pillars in relation to the SiC>2 protrusions, the pillars are at the end of the heating step level with the SiC>2 protrusions and in contact with the surface of the second substrate.

[0007] While this process is functional in certain contexts, it is not applicable to all substrates, particularly when the latter contain materials whose melting temperature is lower than the temperature at which the material of the recessed elements can undergo volume expansion. For example, the annealing temperature of copper is 350°C, while some materials used that may require such assemblies have melting temperatures of around 250°C. It is therefore impossible to implement this solution when these materials are involved.

[0008] The heating step can also cause problems of distortion and strong mechanical stresses within the substrates when the latter contain materials with thermal expansion coefficients that are too different from each other. This is particularly the case when the second substrate comprises a glass-based support covered with a thin copper film.

[0009] Another disadvantage of this method lies in the precision required in the dimensioning of the shrinkage of the copper pillars in relation to the SiC>2 protrusions. Indeed, a variation in the value of this shrinkage, for example due to manufacturing hazards, could cause, at the time of expansion of the copper pillars, untimely detachments within the substrates.

[0010] US 2022 / 173161 A1, US 2022 / 352442 A1 and US 2017 / 069804 A1 disclose methods of bonding a first substrate to a second substrate via protrusions formed on the first substrate. These methods are, however, not optimal.

[0011] An objective of the present invention is therefore to provide a method for preparing a substrate for assembly with another substrate, this method not requiring the substrate to be subjected to high temperatures.

[0012] SUMMARY OF THE INVENTION To achieve this objective, according to one embodiment, a method for preparing a stack is provided comprising the following steps:

[0013] - providing a stack comprising a substrate topped with a plurality of structures separated by trenches, each structure comprising:

[0014] • a plot comprising a top and at least one side,

[0015] • a first layer based on a first material and covering the substrate, the first layer covering at least part of the side of the pad,

[0016] • a second layer based on a second material and covering at least the top of the pad,

[0017] • a withdrawal portion, covering, preferably by being in direct contact with, an upper part of the side of the pad, the upper part of the side extending from the top of the pad, the withdrawal portion being formed by at least one of the first layer and the second layer,

[0018] • a covering portion, covering, preferably by being in direct contact with, a lower part of the side of the pad and extending from the withdrawal portion, the covering portion being formed by the first layer,

[0019] - remove the second layer,

[0020] - remove the withdrawal portion while leaving the covering portion in place so as to expose the upper part of the side of the pad, the upper part of the side of the pad forming a protrusion beyond the covering portion.

[0021] The protrusion thus formed is suitable for bonding to another surface, for example the upper face of a transfer support.

[0022] A second aspect of the invention relates to a stack comprising a substrate surmounted by a plurality of structures separated by trenches, each structure comprising:

[0023] - a plot comprising a top and at least one side,

[0024] - a first layer based on a first material and covering the substrate, the first layer covering at least part of the side of the pad,

[0025] - a covering portion formed by the first layer, covering, preferably by being in direct contact with, a lower part of the side of the pad and leaving uncovered an upper part of the side of the pad, the upper part of the pad forming a protrusion relative to the first layer.

[0026] A third aspect of the invention relates to a method of bonding or transferring at least one pad included in a stack according to the second aspect of the invention onto a receiving stack:

[0027] - Providing a stack according to the second aspect of the invention,

[0028] - Provide a receiving stack with an upper face,

[0029] - Carry out direct bonding of the top of at least one pad of the stack onto the upper face of the receiving stack.

[0030] The protrusions formed during the preparation process in fact allow, during the bonding process, effective bonding. In particular, it is not necessary to carry out a heating step which risks causing distortions and / or strong mechanical stresses within the stacks, as was the case in the prior art. It is possibly possible to carry out a consolidation annealing after the direct bonding to improve the adhesion between the tops of the pads and the upper face of the receiving stack, but such annealing is carried out at temperatures lower than the temperatures at which mechanical deformations can occur.

[0031] Furthermore, this bonding process does not require dimensional control greater than that commonly available in industry.

[0032] BRIEF DESCRIPTION OF THE FIGURES

[0033] The aims, objects, as well as the characteristics and advantages of the invention will emerge more clearly from the detailed description of an embodiment thereof which is illustrated by the following accompanying drawings in which:

[0034] Figures 1A to 1C represent a method of assembling two substrates according to the prior art. Figure 1A illustrates a stack according to the prior art comprising a substrate surmounted by pillars set back relative to protrusions based on a dielectric. The protrusions laterally cover the pillars.

[0035] Figure 1B illustrates the turning over of the stack according to the prior art and its placement opposite a receiving stack.

[0036] Figure 1C illustrates the assembly of the two stacks and their bonding by heating. Figure 1C illustrates in particular the expansion of the pillars.

[0037] Figures 2A to 2J illustrate a first embodiment of the method according to the invention. Figure 2A represents the provision of a primary stack comprising a substrate and a first primary layer overlying the substrate.

[0038] Figure 2B illustrates the formation of a first etching mask on the first primary layer.

[0039] Figure 2C shows the formation of cavities in the first primary layer.

[0040] Figure 2D illustrates the formation of a third primary layer in the cavities and on the first primary layer.

[0041] Figure 2E shows the thinning of the third primary layer so as to form pads extending into the cavities.

[0042] Figure 2F shows the formation of a second primer layer on the pads and the first primer layer.

[0043] Figure 2G shows the formation of a second etching mask on the second primary layer.

[0044] Figure 2H illustrates the formation through the second etch mask of trenches extending into the second primary layer and the first primary layer. A plurality of second layers are thus formed from the second primary layer and a plurality of first layers from the first primary layer.

[0045] Figure 21 illustrates the removal of the second etching mask.

[0046] Figure 2J illustrates the removal of the second layers and a portion of the first layers extending, prior to the removal, from the second layers.

[0047] Figures 3A to 3J illustrate a second embodiment of the method according to the invention. Figure 3A represents the provision of a stack comprising a substrate, a first primary layer overlying the substrate and a first primary sub-layer overlying the first primary layer.

[0048] Figure 3B represents the formation of a first etching mask on the first primary sub-layer.

[0049] Figure 3C shows the formation of secondary cavities in the first primary layer and in the first primary sub-layer through the first etching mask. Figure 3D shows the formation of a third primary layer in the secondary cavities and on the first primary sub-layer.

[0050] Figure 3E illustrates the thinning of the third primary layer so as to form pads extending into the secondary cavities and flush with the first primary sub-layer.

[0051] Figure 3F illustrates the formation of a second primary sub-layer on the pads and on the first primary sub-layer. The first primary sub-layer and the second primary sub-layer form a second primary layer.

[0052] Figure 3G shows the formation of a second etching mask on the second primary layer.

[0053] Figure 3H illustrates the formation through the second etch mask of trenches extending into the second primary layer and the first primary layer. A plurality of second layers are thus formed from the second primary layer and a plurality of first layers from the first primary layer.

[0054] Figure 31 illustrates the removal of the second etching mask.

[0055] Figure 3J illustrates the removal of the second layers, thus forming a protrusion.

[0056] Figures 4A to 4H illustrate a third embodiment of the method according to the invention. Figure 4A represents the provision of a stack comprising a substrate and a first primary layer overlying the substrate. The first primary layer comprises cavities.

[0057] Figure 4B illustrates the formation of a third primary layer in the cavities and on the first primary layer.

[0058] Figure 4C illustrates the thinning of the third primary layer so as to distinguish the latter into individual pads protruding from the first primary layer.

[0059] Figure 4D shows the formation of a second primary layer on the pads and on the first primary layer.

[0060] Figure 4E shows the formation of a second etch mask on the second primary layer.

[0061] Figure 4F illustrates the formation through the second etch mask of trenches extending into the second primary layer and the first primary layer. A plurality of second layers are thus formed from the second primary layer and a plurality of first layers from the first primary layer.

[0062] Figure 4G illustrates the removal of the second etch mask.

[0063] Figure 4H illustrates the removal of the second layers, thus forming a protrusion.

[0064] Figures 5A and 5B are top views of the steps shown in Figures 21, 31 and 4G. Figure 5A illustrates an embodiment in which the pads are cylindrical with a square base.

[0065] Figure 5B illustrates the case of circular cylindrical plots.

[0066] The drawings are given as examples and are not limiting of the invention. They constitute schematic representations of principle intended to facilitate the understanding of the invention and are not necessarily to the scale of practical applications. In particular, the dimensions are not representative of reality.

[0067] DETAILED DESCRIPTION OF THE INVENTION

[0068] Before commencing a detailed review of the embodiments of the invention, optional features which may possibly be used in combination or alternatively are set out below:

[0069] According to one example, the substrate has a lower face extending mainly along a transverse plane defined by a first direction and a second direction, the protrusion having a height E pro t along a third direction perpendicular to said transverse plane, such that E pro t < 0.5*Efian c, preferably E pro t < 0.3*Efi an c and preferably E pro t < 0.1*Efianc, Efianc being the height of the flank taken in the Z direction. E pro t is thus measured, along Z, between one end of the covering portion and the top of the plot.

[0070] According to an example, E pro t > 5 nm, preferably E pro t ^ 10 nm. Preferably, E pro ts 50 nm,

[0071] E P rot < 30 nm.

[0072] In one example, the removal of the second layer and the removal of the removal portion are performed in a single etching step.

[0073] According to one embodiment, the withdrawal portion is formed solely by the first layer.

[0074] In one example, prior to the removal step, the first layer extends to the top of the pad. In particular, the second layer may not cover the sidewall.

[0075] According to one example, the removal of the removal portion is carried out by time control, as well as preferably by the etching selectivity of the different materials.

[0076] According to one embodiment, the withdrawal portion is formed at least in part by the second layer. Thus, the second layer covers part of the sidewall.

[0077] In one example, the shrinkage portion is formed solely by the second layer. This allows for very precise control of the height of the protrusion obtained by removing the shrinkage layer. This allows for the use of only very common and well-mastered steps in the industry. This also limits the number of steps required to remove the shrinkage portion. This results in time and cost savings.

[0078] According to one example, removing the removal portion comprises etching the second layer selectively at the first layer. The removal of the removal portion is carried out, for example, by etching the second layer with a stop at the first layer.

[0079] According to a particular embodiment of the invention, the provision of the stack comprises the following steps:

[0080] - Form the first layer,

[0081] - Form the plot,

[0082] - Form the second layer, and the formation of the second layer includes the following sub-steps:

[0083] - After the formation of the first layer and prior to the formation of the pad, form a first sub-layer of the second layer, the first sub-layer surmounting the first layer,

[0084] - After forming the pad, form a second sub-layer of the second layer, the second sub-layer at least surmounting the pad.

[0085] According to a particular embodiment of the invention, the provision of the stack comprises the following steps:

[0086] - Form the first layer,

[0087] - Form the plot,

[0088] - Form the second layer, and the second layer is deposited on the pad and the first layer after the pad is formed.

[0089] In one example, the pads are formed from a third material, the third material being one of copper, aluminum, tungsten, and titanium. Generally, the third material is an electrically conductive material, preferably a metallic material.

[0090] According to one example, the first material is based on one of: SiO2, SiN and Si. According to one example, the second material is based on one of SiN, SiO2, and Si. The first and second materials may each be based on a dielectric material. It is specified that, in the context of the present invention, the terms "on", "overcomes", "covers", "underlying", "facing" and their equivalents do not necessarily mean "in contact with". Thus, for example, the deposition, transfer, bonding, assembly or application of a first layer on a second layer does not necessarily mean that the two layers are in direct contact with each other, but means that the first layer at least partially covers the second layer by being either directly in contact with it or by being separated from it by at least one other layer or at least one other element.

[0091] A layer can also be composed of several sub-layers of the same material or of different materials.

[0092] A substrate, a layer, a device, "based" on a material M, is understood to mean a substrate, a layer, a device comprising this material M only or this material M and possibly other materials, for example alloying elements, impurities or doping elements. Thus a material based on a III-N material may comprise a III-N material with added dopants. Similarly, a GaN-based layer typically comprises GaN and alloys of AIGaN or InGaN.

[0093] The term "III-V material" refers to a semiconductor composed of one or more elements from columns III and V of Mendeleev's periodic table. Elements in column III include boron, gallium, aluminum, and indium. Column V contains, for example, nitrogen, arsenic, antimony, and phosphorus.

[0094] The term "selective etching with respect to" or "etching exhibiting selectivity with respect to" means etching configured to remove a material A or a layer A with respect to a material B or a layer B, and exhibiting an etching rate of the material A greater than the etching rate of the material B. The selectivity is the ratio between the etching rate of the material A and the etching rate of the material B. The selectivity between A and B is denoted SA:B.

[0095] The roughness of a surface is a parameter characterizing the state of this surface. It is most often associated in tribology with the characteristic depth of the striations furrowing the surface. The roughness of a surface is noted Ra and is typically expressed in pm (1 pm = 10' 6 m) or nm (1 nm = 10' 9 m).

[0096] A reference frame, preferably orthonormal, comprising the axes X, Y, Z is shown in figures 1 A, 2A, 3A, 4A, 5A and 5B. This reference frame is applicable by extension to the other figures.

[0097] In this patent application, we will preferably speak of thickness for a layer and height for a structure or device. The height is taken perpendicular to the transverse plane XY. The thickness is taken in a direction normal to the main extension plane of the layer. Thus, a layer typically has a thickness along Z, when it extends mainly along the transverse plane XY, and a projecting element, for example an isolation trench, has a height along Z. The relative terms "on", "under", "underlying" preferably refer to positions taken in the Z direction.

[0098] The terms "substantially", "approximately", "in the order of" mean "to within 10%, preferably to within 5%".

[0099] A first step of the method according to the invention consists of providing a stack 1 comprising a substrate 10 and a plurality of structures 1000 surmounting the substrate 10 and separated by the trenches 400.

[0100] The substrate 10 has a lower face 12 extending mainly in a transverse plane XY defined by a first direction X and a second direction Y. The substrate 10 may also comprise an emitting layer 15 comprising a plurality of photoemitting elements. The photoemitting elements may for example be active nanowires or microwires forming LEDs or micro-LEDs. They may for example be standard two-dimensional LEDs (2D LEDs). The photoemitting elements typically have an elongated shape in a third direction Z perpendicular to the transverse plane XY. The emitting layer 15 is typically based on one or more semiconductor materials.

[0101] The structures 1000 have a cylindrical shape, preferably that of a right cylinder whose generating lines extend in the third direction Z. The structures 1000 typically have a rectangular section, preferably square, in a plane parallel to the transverse plane XY, as shown in FIGS. 5A and 5B.

[0102] The structures 1000 are separated from each other by trenches 400. These trenches are typically rectilinear. The trenches 400 may for example comprise a first series of trenches 400a and a second series of trenches 400b, the trenches 400b of the second series being perpendicular to the trenches 400a of the first series. The trenches 400a of the first series extend for example in the first direction X and those of the second series in the second direction Y, as shown in FIGS. 5A and 5B.

[0103] Each of the structures 1000 surmounting the substrate 10 comprises a pad 100 having a top 101 and at least one side 103 having a height Efl anc according to the third direction Z. Efl anc is advantageously greater than 500 nm, typically substantially equal to 1 pm.

[0104] The top 101 of the pad 100 advantageously extends in a so-called upper plane 501, preferably parallel to the transverse plane XY.

[0105] The pad 100 typically has the shape of a cylinder, preferably a right cylinder whose generating lines defining one or more lateral faces of the pad 100 preferably extend in the third direction Z. The pad 100 may for example have a polygonal shape in the transverse plane XY. The pad 100 then has several lateral faces. In the particular case of a pad 100 having a rectangular or even square shape in the transverse plane XY, the pad 100 has four lateral faces 103a, 103b, 103c, 103d, as shown in FIG. 5A. The pad 100 may also have a circular cylindrical shape, as illustrated in FIG. 5B. It then has only one lateral face. In the case of a cylindrical pad, the at least one flank 103 of the pad 100 corresponds to one or the lateral face of the pad 100.

[0106] The pad 100 is preferably made of a metallic material, for example copper, aluminum, tungsten or titanium.

[0107] Each pad 100 advantageously covers at least one photoemitting element included in the emitting layer 15. The photoemitting elements are thus located at the level of the pads 100 in projection along the transverse plane XY. They may be in direct contact with the pads 100, possibly via a layer called contact layer 16 which may for example be based on aluminum, titanium or titanium nitride. Each of the structures further comprises a first layer 200 having an upper face 201 preferably extending in a plane parallel to the transverse plane XY. According to one example, the first layer 200 directly surmounts the substrate 10, that is to say that its lower face 202 is in contact with the substrate 10. It has a thickness E200 along the third direction Z. The first layer 200 covers at least a portion of the flank 103 of the pad 100.It is thus in contact with the plot 100 on a non-zero height Erecovery less than or equal to the height of the flank Enanc.

[0108] Each of the structures 1000 further comprises a second layer 300 covering at least the top 101 of the pad 100. It may also cover at least part of the first layer 200. The second layer 300 is preferably in direct contact with the top 101 of the pad 100 and possibly the first layer 200. It has an upper face 301 preferably extending in a plane parallel to the transverse plane XY. The stack 1 provided during the method according to the invention may for example be obtained according to the steps below:

[0109] - Provide a primary stack 1a comprising, as illustrated in Figure 2A:

[0110] • Substrate 10,

[0111] • A first primary layer 200a based on the first material and covering the substrate 10.

[0112] - Forming in the first primary layer 200a cavities 250 passing entirely through the first primary layer 200a in the third direction Z. This step is illustrated by figures 2B and 2C.

[0113] - Fill the cavities 250 with the third material, thus forming the pads 100 (figures 2E, 4C).

[0114] - Form a second primary layer 300a based on the second material and surmounting the pads 100 and the first layers 200 (figures 2F, 3A, 3F, 4D).

[0115] - Forming the trenches 400 completely crossing the second primary layer 300a and the first primary layer 200a in the third direction Z. The formation of the trenches 400 makes it possible to single out and form the structures 1000, as illustrated in figures 2H, 3J, 4F. They separate in particular the first layers 200 formed in the first primary layer 200a, and the second layers 300 formed in the second primary layer 300a, which will each be part of a structure 1000.

[0116] It is understood, however, that stack 1 can be obtained according to any other embodiment.

[0117] Following the provision of the stack 1, a step of removing a removal portion 500 is implemented at the level of each structure 1000.

[0118] The withdrawal portion 500 is formed by at least one of the first layer 200 and the second layer 300. It covers an upper part 104 of the flank 103 of the pad 100. In other words, it is adjacent, in a plane parallel to the XY plane, to the flank 103 of the pad 100, and this at any point of the upper part 104 of the flank 103. The withdrawal portion 500 is preferably in direct contact with the upper part 104 of the flank 103 of the pad 100. In the particular case of a pad 100 having several flanks, the upper part 104 advantageously extends into each of the flanks 103 of the pad 100.

[0119] The upper part 104 of the flank 103 extends from the top 101 of the pad 100. More precisely, the upper part 104 extends from the edge(s) of the pad 100 delimiting its top 101 and its flank 103. It has a height E 4 in the third direction Z. The recessed portion 500 extends in the third direction Z over a distance less than the height of the flank 103. We thus have E 104 <Eflanc.

[0120] The withdrawal portion 500 can thus extend, in the third direction Z, between the upper plane 501 in which the top 101 of the pads 100 extends, and a lower plane 502 extending, in the third direction Z, between the transverse plane XY and the upper plane 501.

[0121] In the step of removing the removal portion 500, a covering portion 600 is left in place.

[0122] The covering portion 600 is formed by the first layer 200. It covers a lower part 105 of the flank 103 of the pad 100. In other words, it is adjacent, in a plane parallel to the XY plane, to the flank 103 of the pad 100, and this at any point of the lower part 105 of the flank 103. The covering portion 600 is preferably in direct contact with the lower part 105 of the flank 103 of the pad 100.

[0123] The covering portion 600 extends from the withdrawal portion 500, that is to say that the upper part 104 and the lower part 105 of the flank 103 of the pad 100 are in contact.

[0124] The lower part 105 of the flank 103 has a height E 5 along the third direction Z. According to one example, the withdrawal portion 500 and the covering portion 600 together cover the flank 103 of the pad 100 over the entire height. We then have E104+E105 = Enanc.

[0125] The lower part 105 of the flank 103 is, when the withdrawal portion 500 is removed, left covered by the covering portion 600.

[0126] Once the withdrawal portion 500 has been removed, the upper part 104 of the flank 103 forms a protrusion beyond the covering portion 600. The pad 100 thus projects in the third direction Z relative to at least one remaining part of the first layer 200 corresponding to the covering portion 600. The protrusion formed has a height E pro t according to the third direction. Advantageously, Eprot is greater than or equal to 5 nm.

[0127] Different embodiments of the method according to the invention will now be described in more detail.

[0128] First example of implementation of the method according to the invention

[0129] A first embodiment of the method according to the invention will now be described with reference to Figures 2A to 2J.

[0130] A first step consists of providing a primary stack 1a comprising the substrate 10 and a first primary layer 200a overlying the substrate 10 (FIG. 2A). A first etching mask 50 is then deposited on the upper face 201a of the first primary layer 200a (FIG. 2B) and a removal of certain parts of the first primary layer 200a is carried out through openings 55 in the first etching mask 50, then, preferably, the latter is removed (FIG. 2C). This removal makes it possible to form cavities 250 in the first primary layer 200a. Preferably, the shrinkage in the first primary layer 200a is done over its entire thickness in the third direction Z. In other words, preferably, the cavities 250 extend to the upper face 11 of the substrate 10. This gives an assembly comprising the substrate 10 surmounted by the first primary layer 200a comprising the cavities.

[0131] The cavities 250 are then filled with the third material. It is for example possible to deposit a third primary layer 100a of third material extending into the cavities 250, directly above the cavities 250 in the third direction Z and above the first primary layer 200a (FIG. 2D). A portion of the third primary layer 100a can then be removed, for example by chemical mechanical polishing (CMP), so as to make it level in the third direction Z with the first primary layer 200a (FIG. 2E). The remaining portions of the third primary layer 100a, separated by the first primary layer 200a, constitute the pads 100. This method of proceeding, by deposition then polishing, makes it possible to reduce or even avoid any possible protrusion or removal of the pads 100 relative to the first primary layer 200a.

[0132] As illustrated in Figure 2F, a second primary layer 300a is then deposited on the upper face 201a of the first primary layer 200a and the top 101 of the pads 100. The second primary layer 300a has a thickness Esooa along the third direction Z. Advantageously, E3ooa is less than or equal to 200 nm and / or greater than or equal to 20 nm, preferably substantially equal to 60 nm.

[0133] A second etching mask 60 is then formed on the second primary layer 300a (FIG. 2G). The second etching mask 60 has openings 65 located directly above the first primary layer 200a.

[0134] Following the deposition of the second etching mask 60, trenches 400 are formed through the openings 65 in the second primary layer 300a, the first primary layer 200a, and possibly a portion of the substrate 10 (FIG. 2H). Preferably, the second etching mask 60 is then removed. The trenches 400 separate the structures 1000 described previously. Preferably, the trenches pass through the entire thickness along the third direction Z of the second primary layer 300a and the first primary layer 200a. Advantageously, at least some of the trenches 400 extend into the substrate 10. This can make it possible, depending on the intended applications, to single out different areas of the substrate 10, for example to electrically isolate them from each other.

[0135] The formation of the trenches 400 makes it possible to divide the first primary layer 200a into a plurality of first layers 200 and the second primary layer 300a into a plurality of second layers 300. This gives the stack 1 comprising the substrate 10 surmounted by the structures 1000, each structure 1000 comprising a pad, a first layer 200 and a second layer 300.

[0136] In this embodiment, in each structure 1000, the first layer 200 and the pad 100 are substantially level along the third direction Z. The second layer 300 does not cover the flank 103 of the pad 100. Thus, as illustrated in FIG. 2I, the withdrawal portion 500 is only formed by a part of the first layer 200. The upper part 104 of the flank 103 of the pad 100 is only covered by the first layer 200. The lower part 105 of the flank 103 is advantageously covered only by the first layer 200.

[0137] Advantageously, in this embodiment, when removing the removal portion 500, a portion of the second layer 300 is also removed. It is understood that the removal of the second layer 300 is understood as the removal of the second layers 300 of each structure 1000. This remark extends to the mention of all the other elements common to all the structures 1000.

[0138] The removal of the removal portion 500 is done for example during the same step as the removal of the second layer 300. It is for example possible to implement an etching of the first material and of the second material selective with respect to the third material forming the pads 100. This etching also preferably has a selectivity between the first material and the second material. If we denote Vi the etching speed of the first material, V2 the etching speed of the second material and V3 the etching speed of the third material, we thus preferably have: V2 > V1 and V2 > V3. Preferably, V2 > 10*Vi and / or V2 > 100*Vs.

[0139] The selectivity between the first, second and third materials has as a direct consequence the height E pro t of the protrusion formed during the withdrawal of the withdrawal portion 500. Thus, the height E pro t can be controlled by the choice of these materials.

[0140] This embodiment is characterized by its simplicity of implementation, in particular because relatively few steps are necessary for the formation of protrusions.

[0141] All the steps described in the context of this first embodiment can be implemented in other embodiments of the method according to the invention. The advantages presented by these steps are also transposable to other embodiments. It is in particular possible to use selective etching between the first, second and third material as described above for the formation of the protrusion during other variants of the method according to the invention.

[0142] Second example of implementation of the method according to the invention

[0143] A second embodiment of the method according to the invention will now be described with reference to Figures 3A to 3J.

[0144] A first step consists of providing a stack comprising the substrate 10, a first primary layer 200a overlying the substrate 10 and a first primary sub-layer 310a overlying the first primary layer 200a (FIG. 3A). The first primary sub-layer 310a is preferably in direct contact with the first primary layer 200a.

[0145] A first etching mask 50 having openings 55 is then deposited on the upper face 311a of the first primary sub-layer 310a (FIG. 3B). The parts of the first primary sub-layer 310a and of the first primary layer 200a located at the openings 55 are removed (FIG. 3C), then, preferably, the etching mask 50 is removed. This removal makes it possible to form secondary cavities 250'. The secondary cavities 250' preferably extend to the upper face 11 of the substrate 10.

[0146] Following the formation of the secondary cavities 250' in the first primary layer 200a and in the first primary sub-layer 310a, the latter are filled with the third material. As illustrated in FIGS. 3F and 3G, it is for example possible to form a third primary layer 100a extending into the secondary cavities 250', directly above the secondary cavities 250' in the third direction Z and on the first primary sub-layer 310a (FIG. 3D). The deposition of the third primary layer 100a is preferably conformal. The third primary layer 100a can then be thinned, for example by CMP, so that it is level with the first primary sub-layer 310a (FIG. 3E). A second primary sub-layer 320a is then formed on, and preferably in direct contact with, the upper face 311A ​​of the first primary sub-layer 310a and on, and preferably in direct contact with, the top 101 of the pads 100 (FIG. 3F).The assembly consisting of the first primary undercoat 310a and the second primary undercoat 320a is called the second primary coat 300a.

[0147] Trenches 400 are then formed in the second primer layer 300a and the first primer layer 200a. The trenches 400 pass through the second primer layer 300a over its entire thickness. They preferably also pass through the first primer layer 200a over its entire thickness.

[0148] The formation of the trenches 400 makes it possible to divide the first primary layer 200a into a plurality of first layers 200 and the second primary layer 300a into a plurality of second layers 300. More precisely, the first primary sub-layer 310a is divided into a plurality of first sub-layers 310 of the second layers 300, and the second primary sub-layer 320a is divided into a plurality of second sub-layers 320 of the second layers 300. Advantageously, at least some of the trenches 400 extend into the substrate 10. This can make it possible, depending on the intended applications, to single out different zones of the substrate 10, for example to electrically insulate them from each other.

[0149] We thus obtain the stack 1 comprising the substrate 10 surmounted by the structures 1000, each structure 1000 comprising a pad, a first layer 200 and a second layer 300 (figure 3I).

[0150] As described in the context of the first embodiment, the trenches 400 may be formed through openings 65 of a second etching mask 60 deposited on the second primary layer 300a (FIGS. 3G and 3H). This second etching mask 60 is advantageously removed after the formation of the trenches 400.

[0151] In the context of the second embodiment, the second layer 300 - more precisely the first sub-layer 310 of the second layer 300 - covers a portion of the flank 103. Prior to the removal step, the first layer 200 extends over only a portion of the flank 103 of the pad 100 and does not extend to the top 101 of the pad 100. Thus, for a structure 100, the removal portion 500 is at least formed by the first sub-layer 310 of the second layer 300. According to an advantageous example illustrated in FIG. 3I, the removal portion 500 extends only in the first sub-layer 310 of the second layer 300. However, it is also possible to remove a portion of the first layer 200 to form the protrusion.

[0152] The withdrawal portion 500 is selectively withdrawn relative to the pad 100 after the formation of the trenches 400.

[0153] As illustrated by the transition from Figure 3I to Figure 3J, the removal of the removal portion 500 advantageously comprises the removal of the entirety of the second layer 300, always selectively with respect to the pad 100.

[0154] In this second embodiment, the pads 100 are projecting relative to the first primary layer 200 as soon as they are formed. This comes from the fact that the pads 100 are formed in the secondary cavities 250', delimited, in any plane parallel to the transverse plane XY, on the one hand by the first primary layer 200a and on the other hand by the first primary sub-layer 310a. The height E prot of the protrusion obtained after removal of the removal portion 500 is therefore typically defined by the thickness Es a, measured along the third direction Z, of the first primary sub-layer 310a. In particular, if, for a structure 1000, the removal layer 500 corresponds to the first sub-layer 310 of the third layer 300, we have substantially Eprot — EsiOa-

[0155] An advantage of the second embodiment is the excellent control of the height E pro t of the protrusion. As mentioned previously, this height can be defined from the deposition of the first primary undercoat 310a, but the thickness of a layer during its deposition is a parameter that is very well controlled in the industry. We can therefore precisely control Es a and therefore Eprot-

[0156] This embodiment can also make it possible to form a protrusion having a height E prot particularly important. The removal of the second material forming the removal portion 500 selectively to the third material forming the pad 100 can in fact be carried out without difficulty over a significant height, typically from 50 to 200 nm.

[0157] All the steps described in the context of this second embodiment can be implemented in other embodiments of the method according to the invention. The advantages presented by these steps are also transposable to other embodiments. It is in particular possible to use the principle of deposition of the second layer 300 in two stages (first sub-layer 310 and second sub-layer 320) and of the formation of the pad 100 between the formation of the first primary sub-layer 310a and the formation of the second primary sub-layer 320a during other variants of the method according to the invention.

[0158] Third example of implementation of the method according to the invention

[0159] A third example embodiment of the method according to the invention will be described with reference to Figures 4A to 4H.

[0160] The steps for obtaining the substrate 10 topped with the first primary layer 200a comprising cavities 250 (FIG. 4A) may be similar to those described for the first embodiment with reference to FIGS. 2A to 2C.

[0161] A third primary layer 100a is then formed in the cavities 250 and on the upper face 201a of the second primary layer 200a. The third primary layer 100a is preferably conformal. Its upper face 101a preferably extends in a plane parallel to the transverse plane XY.

[0162] After the formation of the third primary layer 100a, the portions of the third primary layer 100a above the first primary layer 200a are removed. This makes it possible to expose the upper face 201a of the first primary layer 200a. This removal also makes it possible to separate the third primary layer 100a into pads 100.

[0163] This removal can be carried out by thinning, for example by CMP, of the first primary layer 200a of the third primary layer 100a. A CMP method can in fact make it possible to thin the first primary layer 200a selectively to the third primary layer 100a, by means of a judicious choice of the CMP slurry (English term meaning “mud”) and the CMP pad (commonly referred to by the English term “pad”).

[0164] According to another example, this removal is carried out, for example by wet etching, through a mask having openings overhanging the first primary layer 200a (not shown in the figures).

[0165] As illustrated in FIG. 4D, a second primary layer 300a is then formed on the pads 100 and the first primary layer 200A. The second primary layer 300a is preferably in contact with the top 101 of the pads 100 and the upper face 201A of the first primary layer 200A.

[0166] Trenches 400 are then formed in the second primary layer 300a and the first primary layer 200a (Figure 4G). The trenches 400 pass through the second primary layer 300a over its entire thickness. They preferably also pass through the first primary layer 200a over its entire thickness.

[0167] The formation of the trenches 400 makes it possible to divide the first primary layer 200a into a plurality of first layers 200 and the second primary layer 300a into a plurality of second layers 300. More precisely, the first primary sub-layer 310a is divided into a plurality of first sub-layers 310 of the second layers 300, and the second primary sub-layer 320a is divided into a plurality of second sub-layers 320 of the second layers 300.

[0168] The stack 1 is thus obtained comprising the substrate 10 surmounted by the structures 1000, each structure 1000 comprising a pad, a first layer 200 and a second layer 300 (FIG. 4G). As described in the context of the first embodiment, the trenches 400 can be formed through openings 65 of a second etching mask 60 deposited on the second primary layer 300a (FIGS. 4E and 4F). This second etching mask 60 is advantageously removed after the formation of the trenches 400.

[0169] In the context of the third embodiment, the second layer 300 covers a portion of the flank 103. Prior to the removal step, the first layer 200 extends over only a portion of the flank 103 of the pad 100 and does not extend to the top 101 of the pad 100.

[0170] Thus, for a structure 100, the withdrawal portion 500 is at least formed by a portion of the second layer 300. According to an advantageous example illustrated in FIG. 4G, the withdrawal portion 500 extends only in the second layer 300. It is however also possible to remove a portion of the first layer 200 to form the protrusion. The withdrawal portion 500 is removed selectively with respect to the pad 100 after the formation of the trenches 400.

[0171] As illustrated by the transition from Figure 4G to Figure 4H, the removal of the removal portion 500 advantageously comprises the removal of the entirety of the second layer 300, always selectively with respect to the pad 100.

[0172] This embodiment can also make it possible to form a protrusion having a height E prot particularly important, potentially even more important than by implementing the second embodiment described previously. The removal of the second material forming the removal portion 500 selectively to the third material forming the pad 100 can in fact be carried out without difficulty over a significant height, typically 50 to 200 nm.

[0173] All the steps described in the context of this second embodiment can be implemented in other embodiments of the method according to the invention. The advantages presented by these steps are also transposable to other embodiments. It is in particular possible to use the principle of thinning by CMP to form pads 100 projecting relative to the third primary layer 100a and then a protection of the top 101 of the pads 100 during other variants of the method according to the invention.

[0174] In all the embodiments previously described, during the formation of the trenches 400, the top 101 of the pads 100 is covered by the second primary layer 300a. This has the advantage of protecting this top 101 and in particular of making it possible to maintain its level of surface roughness, which could be significantly impacted during the formation of the trenches 400. The surface roughness obtained after thinning of the second primary layer 300a or of the pads 100, depending on the embodiment, can be controlled by adjusting certain thinning parameters. In particular, the CMP suspension, sometimes called CMP slurry, CMP paste or more commonly CMP slurry, used during the thinning step can be chosen according to the surface roughness targeted for the pads 100.This choice can be made in particular according to the characteristic dimension, often nanometric, of the abrasive particles contained in the CMP suspension. The CMP suspension is advantageously chosen so that the surface roughness of the pads 100 is compatible with their bonding with a specific object, according to the intended applications. It can for example be a thin film based on the third material.

[0175] The roughness R i of the top 101 of the pads 100 is advantageously less than 2 nm, preferably less than 1.5 nm.

[0176] Furthermore, another characteristic common to the embodiments described is the fact that the step of thinning the third primary layer 100A making it possible to obtain the pads 1 (transition from FIG. 2D to FIG. 2E in the first embodiment, from FIG. 3D to FIG. 3E in the second embodiment and from FIG. 4B to 4C in the third embodiment) takes place before the step of forming the trenches 400. This makes it possible to prevent, during a CMP step subsequent to a step of singularizing the structures 1000 (i.e. a step of forming the trenches 400), residues of CMP suspension from remaining in the trenches 400. These residues are in fact very difficult to remove and constitute pollution for the rest of the process. They could in particular harm the good adhesion of the pads 100 with a surface on which one would seek to bond them.

[0177] The previously mentioned material removals (formation of cavities 250, formation of trenches 400, removal of the removal portion 500, removal of the second layer 300, etc.) can be carried out by wet etching. The removal of the various etching masks 50, 60 can in particular be carried out by stripping, more commonly referred to by the English term “stripping”.

[0178] Through the various embodiments described above, it is clear that the invention proposes a method for preparing a stack for assembly with another stack or substrate that is simple to implement. It makes it possible to place, at the level of each structure 1000, a portion of the pad 100 in protrusion relative to the first layer. These protrusions can be used for bonding or transferring the stack onto another stack or substrate. In particular, as mentioned previously, the top 101 of the pads 100 is protected during the method and retains a roughness suitable for direct bonding.

[0179] Through the various embodiments described above, it is clear that the invention proposes a stack and a method for manufacturing this stack allowing an efficient transfer of structures from one stack to another. The presence of protrusions according to the invention in fact allows an efficient bonding of one or more pads or structures to a receiving stack, not requiring dimensional control greater than the control commonly accessible in the industry, and not requiring a heating step risking causing distortions and / or strong mechanical stresses within the stacks.

[0180] The invention is not limited to the embodiments previously described and extends to all embodiments covered by the invention.

Claims

CLAIMS Method for preparing a stack (1) comprising the following steps: • providing a stack (1) comprising a substrate (10) surmounted by a plurality of structures (1000) separated by trenches (400), each structure comprising: a pad (100) comprising a top (101) and at least one side (103), a first layer (200) based on a first material and surmounting the substrate (10), the first layer (200) covering at least part of the side (103) of the pad (100), a second layer (300) based on a second material and covering at least the top (101) of the pad (100), a recessed portion (500), covering an upper part (104) of the side (103) of the pad (100), the upper part (104) of the side (103) extending from the top (101) of the pad (100), the recessed portion (500) being formed by at least one of the first layer (200) and the second layer (300), a covering portion (600), covering a lower part (105) of the flank (103) of the pad (100) and extending from the withdrawal portion (500),the covering portion (600) being formed by the first layer, • remove the second layer (300), • removing the removal portion (500) while leaving the covering portion (600) in place so as to expose the upper part (104) of the flank (103) of the pad (100), the upper part (104) of the flank (103) of the pad (100) forming a protrusion beyond the covering portion (600). Method according to the preceding claim in which the substrate (10) has a lower face (12) extending mainly along a transverse plane (XY) defined by a first direction (X) and a second direction (Y), the protrusion having a height E pro t along a third direction (Z) perpendicular to said transverse plane (XY), such that E pro ts 0.5*Efi an c, preferably E pro ts 0.3*Efianc and preferably E pro t < 0.1*Efi an c, Efi ancbeing the height of the side (103) of the pad (100) taken in the third direction (Z). A method according to either of the preceding claims, wherein E prots 5 nm, preferably Eprot — 10 nm. Method according to any one of the preceding claims, wherein the removal of the second layer (300) and the removal of the removal portion (500) are carried out during a single etching step. Method according to any one of the preceding claims, wherein the removal portion (500) is formed solely by the first layer (200). Method according to the preceding claim, wherein, prior to the removal step, the first layer (200) extends to the top (101) of the pad (100). Method according to any one of the two preceding claims, wherein the removal of the removal portion (500) is carried out by time control. Method according to any one of claims 1 to 4, wherein the removal portion (500) is formed at least in part by the second layer (300).A method according to the preceding claim wherein the removal portion (500) is formed solely by the second layer (300). A method according to either of the preceding claims wherein removing the removal portion (500) comprises etching the second layer (300) selectively to the first layer (200). A method according to any one of claims 8 to 10 wherein providing the stack (1) comprises the following steps:. • Form the first layer (200), • Form the plot (100), • Forming the second layer (300), and wherein forming the second layer (300) comprises the following substeps: • After the formation of a first primary layer (200a) intended to form the first layer (200) and prior to the formation of the pad (100), form a first primary sub-layer (310a) of a second primary layer (300a) intended to form the second layer (300), the first primary sub-layer (310a) surmounting the first primary layer (200a), the first primary sub-layer (310a) being shaped so that after formation of the trenches (400) separating the structures (1000) it forms a first sub-layer (310) of the second layer (300) covering a part of the flank (103) of the pad (100) and forming the withdrawal portion (500), • After forming the pad (100), forming a second primary sub-layer (320a) of the second primary layer (300a), the second primary sub-layer (320a) covering the first primary sub-layer (310a), the second primary sub-layer (320a) being shaped so that after forming the trenches (400) separating the structures (1000) it forms a second sub-layer (320) of the second layer (300) surmounting at least the pad (100). Method according to any one of claims 8 to 10 wherein providing the stack (1) comprises the following steps: • Form the first layer (200), • Form the plot (100), • Forming the second layer (300), and in which a second primary layer (300a), shaped so as to form the second layer (300) after formation of the trenches (400) separating the structures (1000), is deposited: • on the plot (100) and • on a first primary layer (200a) shaped so as to form the first layer (200) after the formation of the trenches (400) after the formation of the pad (100). Method according to any one of the preceding claims wherein the pads (100) are formed from a third material, the third material being based on one of copper, aluminum, tungsten and titanium. Method according to any one of the preceding claims wherein the first material is based on one of: SiC>2, SiN and Si. Method according to any one of the preceding claims wherein the second material is based on one of SiN, SiC>2 and Si. Stack (1) comprising a substrate (10) surmounted by a plurality of structures (1000) separated by trenches (400), each structure comprising: • a pad (100) comprising a top (101) and at least one side (103), • a covering portion (600) formed by a first layer (200) based on a first material and surmounting the substrate (10), covering a lower part (105) of the flank (103) of the pad (100) and leaving uncovered an upper part (104) of the flank (103) of the pad (100), the upper part (104) of the pad (100) forming a protrusion relative to the first layer (200). Stack (1) according to the preceding claim in which the protrusion has a height E pro t greater than 5 nm, preferably greater than 10 nm, the height E pro t being taken according to a third dimension (Z) perpendicular to a transverse plane (XY) in which a lower face (12) of the substrate (10) mainly extends. Method for bonding or transferring at least one pad (100) included in a stack (1) according to any one of the two preceding claims onto a receiving stack (20): • Providing a stack (1) according to any one of the two preceding claims, • Provide a receiving stack (20) having an upper face (21), • Carry out direct bonding of the top (101) of at least one pad (100) of the stack (1) to the upper face (21) of the receiving stack (20).

Images