Device composite, method for separating a device from the device composite, and method for manufacturing a device composite

Abstract

A device composite (1) is proposed, which has a plurality of devices (2) and a carrier (5), wherein the devices are fixed to the carrier by means of a connecting layer (4) and the connecting layer forms at least one support structure (41) for each device, and the connecting layer adjoins the device on the support structure. A sacrificial layer (3) is provided locally between the device and the connecting layer. A part of the devices belongs to a first group (2A) and another part of the devices belongs to a second group (2B), wherein the devices of the first group and the devices of the second group differ in terms of the occupation by the sacrificial layer. In addition, a method for separating the devices from the device composite and a method for manufacturing the device composite are proposed.

Description

Field of the Invention

[0001] The present application relates to a device composite having a plurality of devices, a method for separating a device from the device composite, and a method for manufacturing a device composite. Background Art

[0002] In order to manufacture an electronic or optoelectronic device, the devices to be processed can be provided on a carrier and transferred into the corresponding device. However, with the increasing miniaturization of devices, the requirements for the transfer process also increase, for example, in terms of positioning accuracy. Summary of the Invention

[0003] One object is to improve the reliability of transferring devices.

[0004] This object is achieved in particular by the device composite according to the invention, the method for separating a device from the device composite, and the method for manufacturing a device composite. Other design solutions and suitable solutions are the subject of the following description.

[0005] A device composite having a plurality of devices is proposed, wherein the devices are arranged on a carrier.

[0006] The devices are, for example, electronic, microelectromechanical or optoelectronic devices, for example light-emitting diodes. The side length of the devices is, for example, at most 100 μm or at most 50 μm, for example, between 4 μm and 30 μm, including the boundary values.

[0007] According to at least one embodiment of the device composite, the devices are fixed to the carrier by means of a connecting layer. In particular, the connecting layer directly adjoins the devices locally. In addition, the connecting layer directly adjoins the carrier, for example. However, one or more layers may also be present between the connecting layer and the devices and / or between the carrier and the connecting layer. In a top view of the device composite, the connecting layer may completely cover the carrier.

[0008] According to at least one embodiment of the device composite, the connecting layer forms at least one support structure for each device, and the connecting layer adjoins the device at the support structure. The support structure is configured to mechanically support the respectively associated device. For example, each device is associated with exactly one support structure. For example, in a top view of the device composite, the support structure overlaps with the center of gravity of the associated device. However, in a top view of the device composite, the support structure may also overlap with two or more components. In this case, appropriately, each device is associated with two or more support structures.

[0009] According to at least one embodiment of the device composite, a sacrificial layer is provided locally between the device and the connection layer. The sacrificial layer is configured to be removed before separating the device from the device composite, for example, by chemical means. The support structure is surrounded by the sacrificial layer in the transverse direction, for example, respectively, especially along its entire circumference.

[0010] The direction extending parallel to the main extension plane of the device composite is regarded as the transverse direction.

[0011] According to at least one embodiment of the device composite, a part of the device is assigned to a first group and another part of the device is assigned to a second group. In particular, the devices of the first group and the devices of the second group are of the same type themselves. That is to say, the devices of one group are not distinguishable from the devices of the other group within the manufacturing tolerances.

[0012] Suitably, all the devices of the device composite are assigned to exactly one of these groups. For example, the total number of groups is between two and six, including the boundary values.

[0013] According to at least one embodiment of the device composite, the devices of the first group and the devices of the second group are different in terms of the occupation by the sacrificial layer. In other words, the sacrificial layer is configured differently for the devices of the first group and for the devices of the second group. This involves a targeted difference that exceeds the undesired fluctuations caused by manufacturing.

[0014] In particular, the sacrificial layer is configured such that the devices of the first group can be separated by a preset separation force after partial removal of the sacrificial layer, while the devices of the second group cannot be separated by the same separation force. For example, the force required to separate the inseparable devices, such as the devices of the second group, is at least 20% or at least 100% greater than the separation force for separating the devices of the first group.

[0015] In particular, the following devices are separable. In the case of these devices, after removal of the part of the sacrificial layer, only the connection layer remains between the device and the carrier in the vertical direction, especially above the entire bottom surface of the corresponding device in a top view. In other words, at this stage, the connection layer only adjoins the separable device.

[0016] In at least one embodiment of the device composite, the device composite has a carrier and a plurality of devices, wherein the devices are fixed to the carrier by means of a connection layer. The connection layer forms at least one support structure for each device, at which the connection layer adjoins the device. Locally, a sacrificial layer is provided between the device and the connection layer. A part of the device is assigned to a first group and another part of the device is assigned to a second group. The devices of the first group and the devices of the second group are different in terms of the occupation by the sacrificial layer.

[0017] Unlike the device composite that extends over all devices with the same shape as the sacrificial layer, the sacrificial layer can be removed in a multi-step process such that only the first set of devices is separable after the first step of removing the material of the sacrificial layer. Subsequently, other materials of the sacrificial layer can be removed in order to form the second set of devices, and if necessary, other sets of devices can be iteratively formed in a separable manner. Thus, the device composite is set up for a multi-step transfer process, in particular in which the material of the sacrificial layer is removed in at least two steps, and in which devices, such as all devices of the first set, have been transferred between these two steps.

[0018] According to at least one embodiment of the device composite, the devices of the first set are arranged in a regular pattern. Suitably, the devices of the second set and possibly other sets of devices are also arranged in a regular pattern.

[0019] According to at least one embodiment of the device composite, the device has a bottom surface, and the bottom surface has a basic polygonal shape. For example, the bottom surface is triangular or quadrilateral, especially rectangular. The basic polygonal shape especially means in this context that the corners of the basic shape can be rounded. The basic shape can also have more than four corners. For example, the nearest neighbors of the devices of the first set along all edges of the device do not belong to the devices of the first set.

[0020] According to at least one embodiment of the device composite, at least some of the devices of the first set have nearest neighbors that do not belong to the devices of the first set only along two mutually perpendicular spatial directions. For example, the device composite has exactly two sets of devices, and these devices are arranged alternately along two mutually perpendicular spatial directions. In other words, the devices of the first set and the devices of the second set are arranged in a checkerboard pattern.

[0021] According to at least one embodiment of the device composite, at least some of the devices of the first set are surrounded by nearest neighbors that do not belong to the devices of the first set along the entire circumference. Suitably, in a similar manner, at least some of the devices of the second set are surrounded by nearest neighbors that do not belong to the devices of the second set along the entire circumference. For example, the devices of the device composite are divided into four sets, and the devices of the four sets are arranged alternately such that all devices of one set are surrounded by nearest neighbors that do not belong to this set along the entire circumference. At least some devices respectively represent at least 50% or at least 80% of the devices of each group, for example.

[0022] According to at least one embodiment of the device composite, the devices of the first set and the second set differ by at least 10% in terms of the average thickness of the sacrificial layer between the respective device and the carrier, for example, with respect to the thickness of the sacrificial layer between the devices of the first set and the carrier.

[0023] The thickness herein represents the extension of the sacrificial layer in the vertical direction, i.e., perpendicular to the main extension plane of the device composite. The thickness of the sacrificial layer can be set such that it determines how quickly the material of the sacrificial layer can be removed at the corresponding position through a chemical process. The material removal rate in the lateral direction is greater for a thicker sacrificial layer compared to a thinner one, or vice versa. For example, when the etching process proceeds under diffusion control, the etching rate in the lateral direction increases with the increase in layer thickness. In this case, the etching speed is basically determined by mass transport. In a reaction-controlled situation, the chemical reaction itself determines the etching speed, such that a thin layer is etched faster than a thick layer.

[0024] According to at least one embodiment of the device composite, the sacrificial layer has a first sublayer of a first material composition and a second sublayer of a second material composition different from the first material composition. The first group of devices and the second group of devices differ from each other in terms of the occupancy of the material of the first sublayer. For example, at least one material component is present in the first material composition in a share that is at least 10 wt% more than in the second material composition.

[0025] For example, the material of the first sublayer can be removed at a greater removal rate compared to the material of the second sublayer by means of a specific etching method. In this case, a group of devices with a greater occupancy of the first sublayer compared to other groups can be separated earlier compared to the devices of the other groups.

[0026] Accordingly, when there is still a large amount of material of the sacrificial layer under another group or multiple groups of devices such that these devices cannot be separated, the material of the sacrificial layer under the first group of devices can already be completely removed.

[0027] According to at least one embodiment of the device composite, islands are formed by means of the first sublayer, and the islands are surrounded along their lateral circumference by at least 80% or at least 90%, for example completely, by the second sublayer. Thus, the material of the islands can only be removed in the lateral direction when the second sublayer has been removed along its circumference. For example, the islands are surrounded by a circumferential sub-region of the second sublayer.

[0028] Furthermore, a method for separating devices from a device composite is proposed. For this purpose, in particular, a device composite is provided that has at least one of the features described above for the device composite.

[0029] Therefore, the features described in connection with the device composite can also be used for the method of separating devices and vice versa.

[0030] According to at least one embodiment of the method, the method includes the step of locally removing the material of the sacrificial layer. Thus, there is still material of the sacrificial layer after this step. For example, the material of the sacrificial layer is removed such that the first group of devices is separable.

[0031] According to at least one embodiment of the method, at least one device of the first group is separated. In particular, all devices of the first group are separated. The devices of the second group are not yet separable at this point in time.

[0032] "Not separable" for a group of devices in this context particularly means that the separating force that conventionally separates the devices of the first group is not yet sufficient to separate the devices of this group.

[0033] According to at least one embodiment of the method, the method comprises the step in which other materials of the sacrificial layer are removed. This step is particularly carried out after all the devices to be separated of the first group have been separated. For example, other materials of the sacrificial layer are removed such that the devices of the second group are separable.

[0034] According to at least one embodiment of the method, the method comprises the step in which at least one device of the second group is removed.

[0035] Thus, the removal of the material of the sacrificial layer is carried out in a multi-step process. In particular, the other materials of the sacrificial layer are removed at the point in time at which the devices have been separated from the device composite.

[0036] According to at least one embodiment of the method, the devices of the second group are not separable after the material of the sacrificial layer has been removed locally and before the other materials of the sacrificial layer are removed by means of the separating force used to separate the devices of the first group. In other words, only the devices of the first group are separable before the other materials of the sacrificial layer are removed.

[0037] According to at least one embodiment of the method, the separation of the devices is carried out by means of a transfer body. For example, the transfer body is a post of a punch. Suitably, a plurality of devices, in particular a plurality of devices of the same group, are separated simultaneously.

[0038] According to at least one embodiment of the method, the transfer body has a bottom surface that is at least 20% larger than the bottom surface of one of the devices to be separated. In other words, the transfer body is larger than the device to be separated in a top view of the device composite. Suitably, the bottom surface of the transfer body is at most as large such that the transfer body only overlaps with the devices to be separated at this point in time. Additionally, the transfer body can overlap with one or more non-separable devices.

[0039] According to at least one embodiment of the method, the main lateral extension axis of the transfer body is rotated with respect to the main lateral extension axis of one of the separated devices. For example, the angle between the main extension axes is between 20° and 70°, including the boundary values, in particular between 40° and 70°, including the boundary values. In the case of a quadrilateral device and transfer body, the main extension axes are the diagonals of the transfer body or the device, respectively. The transfer body and the device to be separated can, however, also have other, for example other polygonal or at least partially curved and in particular also different from one another basic shapes.

[0040] According to at least one embodiment of the method, the sacrificial layer has a first sublayer of a first material composition and a second sublayer of a second material composition different from the first material composition, wherein the removal of the material of the sacrificial layer is carried out by means of a chemical method before separating the devices of the first group, the chemical method being selective with respect to the first and second material compositions. For example, the material of the second material composition is not removed or is removed only at a significantly lower removal rate than the material of the first material composition. This can be achieved particularly reliably by means of different materials for sub-regions of the sacrificial layer and the associated selective etching method, such that the devices of the first group are already separable when the devices of the second group are not yet separable with the same separating force.

[0041] Alternatively, the entire sacrificial layer can be formed from only one material composition. In this case, for example, the removal rate of the material of the sacrificial layer in the lateral direction at this location can be set via the local thickness of the sacrificial layer.

[0042] Furthermore, a method for manufacturing a device composite is proposed. The described method is particularly suitable for manufacturing the aforementioned device composite. Thus, the features described in connection with the device composite can also be used for the method of manufacturing a device composite and vice versa.

[0043] In at least one embodiment of the method, a plurality of devices are provided. A sacrificial layer is formed on the devices, wherein the occupation of the sacrificial layer is different for the devices of the first group and for the devices of the second group. The devices are fixed to the carrier by means of a connecting layer.

[0044] The method steps are preferably carried out in the order listed above.

[0045] The devices are provided, for example, on an initial carrier. The initial carrier can subsequently be removed, in particular after fixing the devices to the carrier.

[0046] The sacrificial layer is applied, for example, by means of a deposition method, such as by chemical vapour deposition (CVD) or physical vapour deposition (PVD), such as by evaporation or sputtering. Thus, the sacrificial layer is in direct contact with the device in part.

[0047] According to at least one embodiment of the method, the connecting layer is at least partially flowable at the time of fixing the device. For example, the material for the connecting layer is applied in a flowable state. In particular, the material of the connecting layer can flow into the voids of the sacrificial layer and be in direct contact with the device at these locations. Subsequently, if necessary, the material for the connecting layer can be cured.

[0048] According to at least one embodiment of the method, the sacrificial layer is formed by means of a first sublayer and a second sublayer, wherein the first sublayer is structured lithographically before the second sublayer is applied. The second sublayer is also structured lithographically, for example. For example, the structuring of the first sublayer and the second sublayer is carried out such that the first sublayer and the second sublayer only overlap locally. In this way, the sacrificial layer can be simply manufactured, which has different thicknesses at different locations. Additionally, the first sublayer and the second sublayer can be different from each other in terms of their material composition.

[0049] This application is particularly based on the idea that a multi-stage transfer process generally results in higher reliability. In particular, with the described design of the sacrificial layer, the device is configured to be only partially transferable during the transfer process, such that in the first step only a defined share, for example, the devices of the first group, are removed in a preset grid. If all transferable devices have been removed, then by further removing the material of the sacrificial layer, new devices, i.e., the devices of the second group, and possibly other groups of devices in an iterative manner can be made transferable.

[0050] The risk of adjacent devices becoming loose due to misalignment of the transfer body during the transfer of the device is reduced.

[0051] In addition, the transfer body can have a larger lateral extension than the device to be removed that is relevant. The method can thus also be carried out reliably for particularly small devices.

[0052] The device composite and the related method are suitable, for example, for optoelectronic devices, such as light-emitting diodes, for example for manufacturing video walls, displays or projectors or for use in so-called augmented reality. In principle, the device can however also be other optoelectronic devices, such as detectors, purely electronic devices or microelectromechanical devices. Description of the Drawings

[0053] Other design solutions and suitable solutions can be obtained from the description of the embodiments in the following with reference to the drawings.

[0054] The drawings show:

[0055] Figure 1 A schematic cross-sectional view showing an embodiment of a device composite;

[0056] Figures 2A to 2F An embodiment of a method for separating a device is shown by means of intermediate steps each shown in a schematic cross-sectional view;

[0057] Figures 3A to 3D Embodiments of the device composite are shown in schematic top views respectively ( Figure 3A ) and an embodiment of the related separation method is shown according to the intermediate steps;

[0058] Figure 4 An embodiment of the device composite is shown in a schematic top view and an embodiment of the related separation method is shown according to the intermediate steps;

[0059] Figures 5A to 5C An embodiment of a method for separating a device is shown in a schematic top view according to the intermediate steps;

[0060] Figures 6A to 6C An embodiment of a method for separating a device is shown in a schematic top view according to the intermediate steps;

[0061] Figure 7 、 8 、 9 and 10 show embodiments of the device composite in schematic cross-sectional views respectively;

[0062] Figures 11A to 11C An embodiment of a method for separating a device is shown according to the intermediate steps respectively shown in schematic cross-sectional views;

[0063] Figure 12 A schematic cross-sectional view showing an embodiment of the device composite;

[0064] Figures 13A to 13D A method for manufacturing a device composite is shown according to the intermediate steps respectively shown in schematic cross-sectional views; and

[0065] Figures 14A to 14D A reference example of a method for separating a device is shown.

[0066] Identical, of the same type, or having the same function elements are respectively provided with the same reference numerals in the drawings.

[0067] The drawings are respectively schematic diagrams and thus not necessarily to scale. More precisely, individual elements and especially layer thicknesses may be shown exaggerated for better view or for better understanding. Detailed implementation mode

[0068] In Figure 1 An embodiment of a device composite 1 having a plurality of devices 2 is shown. The accompanying drawings show a partial cross-sectional view of the device composite having four devices 2, and the devices are exemplarily classified into four groups of devices. The devices of the first group 2A, the second group 2B, the third group 2C, and the fourth group 2D are respectively fixed to the carrier 5 of the device composite by means of the connection layer 4.

[0069] The connection layer 4 forms a support structure 41 for each device 2 respectively. The support structure 41 is arranged centrally with respect to the device 2, that is to say, in a top view of the device composite, the support structure overlaps with the corresponding center of gravity of the associated device 2.

[0070] The support structure 41 is directly adjacent to the device 2 respectively. A sacrificial layer 3 is locally provided between the connection layer 4 and the device 2. The sacrificial layer 3 has a first sub-layer 31 and a second sub-layer 32. The first sub-layer 31 and the second sub-layer 32 only partially overlap with each other. This results in the following regions in which the first sub-layer 31 and the second sub-layer are arranged one above the other, for example, between the devices of the first group 2A and the carrier 5.

[0071] Only the first sub-layer 3 of the sacrificial layer 3 exists between the devices of the second group 2B and the carrier 5. Only the second sub-layer 32 exists between the devices of the fourth group 2D and the carrier 5. Thus, for different devices 2, different occupancies of the sacrificial layer 3 in the form of different average layer thicknesses of the sacrificial layer 3 for different groups of devices are obtained. Therefore, the first sub-layer 31 and the second sub-layer 32 are particularly used to constitute different average layer thicknesses under different groups. The expression "under" the device or device group respectively refers to the region between the carrier and the corresponding device or device group.

[0072] The first sub-layer 31 and the second sub-layer 32 can be the same with respect to the materials used in this embodiment.

[0073] For example, the sacrificial layer 3, the first sub-layer 31, and / or the second sub-layer 32 contain silicon, silicon nitride, silicon oxide, germanium, molybdenum, or spin-on glass materials.

[0074] In such a device composite 1, the devices 2A, 2B, 2C, 2D of each group can be sequentially and separably formed.

[0075] This will become clear below according to the embodiment described in Figures 2A to 2F In Figure 2A The provided device composite 1 shown in Figure 1is configured as described. By means of a chemical method, such as the action of wet chemical or dry chemical etching, the removal of the material of the sacrificial layer 3 starts from the intermediate space between the devices 2. The sacrificial layer adjacent to the devices 2 is removed by laterally undercutting the devices 2. If the etching process is carried out in a diffusion-controlled manner, the greater the thickness of the sacrificial layer 3 below the devices 2, the faster the lateral undercutting (laterale ) takes place. Accordingly, the devices of the first group 2A do not initially have the sacrificial layer 3 on the side facing the carrier 5 and are only still lying flat on the support structure 41.

[0076] In this phase, the devices of the first group 2A can be separated by means of the transfer body 7. For example, the transfer body 7 is the post 71 of the punch 72. For example, polydimethylsiloxane is suitable as the material for the post.

[0077] In the following embodiments, one or more posts are shown as examples of the transfer body. However, in principle, other designs of the transfer body are also suitable.

[0078] By subsequently removing the remaining material of the sacrificial layer 3, the devices of the second group 2B can be separated. This method phase is illustrated in Figure 2B FIG. The transfer body 7 in the form of the post 71 overlaps the devices of the second group 2B and can lift them. At this point in time, the devices of the first group 2A have already been removed.

[0079] Advantageously, a plurality of devices of a group are separated simultaneously. For example, the punch 71 has a plurality of posts, where the posts respectively overlap the devices from just one group of devices.

[0080] After separating all the devices of the first group 2A and the second group 2B, only the devices of the third group 2C and the fourth group 2D remain ( Figure 2C ). There is still a residual part of the sacrificial layer 3 below these devices, so that these devices are not yet separable.

[0081] In a subsequent etching step, the remaining material of the sacrificial layer 3 is removed until there is no longer any material of the sacrificial layer below the devices 2 of the third group 2C ( Figure 2D ). However, there is still material of the sacrificial layer 3 below the devices of the fourth group 2D, so that these devices are not yet separable.

[0082] After separating the devices of the third group 2C, only the devices of the fourth group 2D remain ( Figure 2E ). By means of a further etching step, the sacrificial layer 3 is also completely removed below the devices of the fourth group 2D, so that now the devices of the fourth group are separable ( Figure 2F ).

[0083] Therefore, the removal of the sacrificial layer 3 is carried out in a multi-stage process, wherein a group of components is respectively removed between two etching steps. It goes without saying that the etching step can have two or more sub-steps. Therefore, usually n etching steps are performed when there are n groups of components, and in particular the components are respectively separated between these etching steps.

[0084] By means of the described method, components can be sequentially separably formed so that the components to be separated are each surrounded at most by components that were not yet separable in the corresponding method stage. This reduces the risk that incorrect adjustment of the transfer body 7 will inadvertently cause the separation of adjacent components during the transfer step.

[0085] In contrast, Figures 14A to 14D A reference method based on a reference composite 10 is shown in FIG. 1 . In the reference composite, the sacrificial layer 3 is formed in the same manner below all components 2. Figure 14B As schematically shown in FIG. 1 , the sacrificial layer 3 is removed uniformly across all devices 2 so that all devices 2 are separable ( Figure 14C ). In this case, a slight misalignment of the pillar 71 may result in not only the removal of the desired component, but also the jamming of the adjacent component by the pillar 71.

[0086] exist Figure 3A 2 shows a component assembly 1 having components of a first group 2A and a second group 2B. The components are arranged in a regular pattern. Along the edges 21 of the components of the first group 2A, only components of the second group 2B are adjacent. The components of the first group 2A and the components of the second group 2B are thus arranged in a checkerboard manner.

[0087] As in Figure 3B As shown in FIG. 1 , the bottom surface 70 of the pillar 71 may have substantially the same size as the devices to be separated of the first group 2A.

[0088] However, when the transfer body overlaps only one separable component, the transfer body can also have a larger bottom surface than the component to be separated. By increasing the bottom surface, the reliability of separation is increased, especially for particularly small components.

[0089] exist Figure 3C In the embodiment shown in FIG. 2 , the transverse main extension axis 75 of the column 71 is rotated relative to the transverse main extension axis 25 of the components of the first group 2A, exemplarily by 45°. By means of said rotation, the bottom surface of the transfer body can be roughly doubled without the transfer body overlapping two separable components, exemplarily two components of the first group 2A, at the same time.

[0090] In the illustrated embodiment, the device 2 and the transfer body in the form of the post 71 are each configured to have a square basic shape. However, other basic shapes can also be used, such as other polygonal basic shapes, such as triangular or quadrilateral or hexagonal basic shapes, or locally curved basic shapes, such as circular or oval basic shapes. In addition, the basic shapes of the transfer body 7 and the device 2 to be separated can be different from each other.

[0091] For example, in Figure 3D an embodiment is shown in which the transfer body is configured in the form of a circular post 71. Compared to the square transfer body oriented with respect to the device as shown in Figure 3A , the area of the transfer body is increased by a factor of π / 2 here.

[0092] In Figure 4 the illustrated embodiment, the devices of the device composite 1 are assigned to a first group 2A, a second group 2B, a third group 2C, and a fourth group 2D. Along each second row, the devices of the first group 2A and the second group 2B are alternately arranged. Along the rows located therebetween, the devices of the third group 2C and the fourth group 2D are alternately arranged. This results in the following arrangement, in which each device of one group, for example a device of the first group 2A, is surrounded only by devices assigned to other groups, such as the second group 2B, the third group 2C, and the fourth group 2D. In this arrangement, the bottom surface of the transfer body 7, for example the bottom surface of the post 71, is approximately nine times as large as the bottom surface of the device to be separated of the first group 2A. However, each post 71 only overlaps with the separable device.

[0093] Thus, the maximum dimension of the transfer body is no longer limited by the center distance between adjacent devices, but can be significantly larger than the bottom surface of the device to be separated by forming a corresponding large number of groups of devices and a suitable design of the sacrificial layer, without the transfer body overlapping two separable devices at the same time.

[0094] In other words, the spacing between the closest separable devices determines the maximum extent of the transfer body in that direction and does not determine the center distance of the adjacent devices themselves.

[0095] In Figures 5A to 5C the illustrated embodiment of the method for separating devices basically corresponds to that in connection with Figures 2A to 2FThe described embodiments. In contrast thereto, the devices are grouped into a first group 2A, a second group 2B, and a third group 2C. A group of devices are respectively located in a hexagonal grid. The devices of the first group 2A, the second group 2B, and the third group 2C are respectively configured to have a circular bottom surface. The devices of the first group 2A can be separated by partially removing the sacrificial layer 3 such that the devices of the first group can be separated via the posts 71. The maximum radius of the posts 71 is obtained by subtracting the radius of the device from the center-to-center distance between two devices in the same group. In this case, the bottom surface of the posts 71 can also be significantly larger than the bottom surface of the device to be separated, and the posts do not overlap with the two separable devices simultaneously.

[0096] Figure 5B The following method steps are shown in which, after removing other materials of the sacrificial layer, the devices of the second group 2B are separable and separated by means of the posts 71. In a further etching step, the devices of the third group 2C can be separated ( Figure 5C ).

[0097] In Figures 6A to 6C the embodiment shown basically corresponds to the embodiment described in connection with Figures 5A to 5C . In contrast thereto, the transfer body 7 and the device 2 have different basic shapes from each other. Thus, the transfer body 7 is exemplarily configured to have a triangular bottom surface. The device 2 exemplarily has a circular bottom surface.

[0098] The embodiment of the device composite 1 shown in Figure 7 differs from the device composite described in connection with Figure 1 especially in that the first sublayer 31 and the second sublayer 32 of the sacrificial layer 3 have different material compositions from each other.

[0099] For example, the first sublayer 31 contains silicon and the second sublayer 32 contains silicon nitride. By a chemical method selective with respect to the materials used, the devices of each group can be configured to be separable with different degrees of difficulty. For example, silicon nitride is etched at a lower etching rate than silicon in a fluorine plasma or a XeF2 plasma.

[0100] In Figure 7 the embodiment shown, mainly the material of the first sublayer 31 is present under the devices of the first group 2A and the devices of the second group 2B. The thickness of the sacrificial layer 3 is greater under the devices of the first group 2A than under the devices of the second group 2B such that first the devices of the first group 2A are separable.

[0101] Between the devices of the third group 2C and the carrier 5 there is also material of the first sublayer 31 and material of the second sublayer 32. A part of the first sublayer 31 is configured as an island 35, which is completely surrounded in the lateral direction by the material of the second sublayer 32. Thus, the material of the island 35 can only be removed when the surrounded material of the second sublayer 32 has been removed. Thus, the devices of the third group 2C can only be separated after the devices of the second group 2B.

[0102] Below the devices of group 2D there is almost only material of the second sublayer 32. Thus, the devices can only be separated after an additional etching step.

[0103] Thus, the method according to this embodiment utilizes not only the different etching rates due to different layer thicknesses, but also the different etching rates due to different materials.

[0104] In Figure 8 the embodiment shown basically corresponds to the embodiment described in connection with Figure 7 However, the thickness of the sacrificial layer 3 is the same for all groups 2A, 2B, 2C, 2D of the devices. The components of the devices differ from each other with respect to the occupation with material of the first sublayer 31. The greater the occupation with material of the first sublayer, the earlier the devices of the relevant group can be separated.

[0105] The support structures 41 are each surrounded by material of the first sublayer 31 in the region adjacent to the devices 2.

[0106] In Figure 9 the embodiment shown basically corresponds to the embodiment described in connection with Figure 8 However, the support structures 41 are surrounded by material of the second sublayer 32. The devices of the group with the greatest occupation with material of the first sublayer 31 can again be separated first, i.e., the devices of the first group 2A.

[0107] In Figure 10 an embodiment of the device composite 1 is shown, in which the support structures 41 each overlap at least two adjacent devices 2. The support structures 41 do not extend along the entire circumference of the devices 2 in a top view of the device composite 41, such that the sacrificial layer 3 is locally accessible for chemical material removal in the intermediate space between the adjacent devices. With this design of the support structures 31, the risk of the devices 2 tipping over during separation by means of a transfer body can be minimized as much as possible. Appropriately, each device overlaps with two or more spaced-apart support structures.

[0108] This design of the support structures 41 can also be applied to the remaining embodiments.

[0109] In Figures 11A to 11C an embodiment of a method for separating devices is shown, in which a device composite is provided as described in connection withFigure 10 as described. Due to the intermediate space between adjacent devices 2, the material of the sacrificial layer 3, i.e., the material of the first sub-layer 31, is removed in the first step. Thereby, the devices of the first group 2A can be separated. The removal of the first sub-layer 31 is preferably carried out by a chemical method that does not act on the second sub-layer 32 or acts on the second sub-layer 32 only with a significantly reduced etching rate compared to the first sub-layer 31.

[0110] After separating the devices of the first group 2A, in a further step, other materials of the sacrificial layer, i.e., the material of the second sub-layer 32, can be removed. Thereby, the devices of the second group 2B can be separated and can be transferred accordingly ( Figure 11C ).

[0111] The method is particularly insensitive to fluctuations in the etching rate or etching duration because an overly long duration of the first etching step does not cause significant material removal of the second sub-layer, such that when the devices of the first group 2A are already separable, the devices of the second group 2B remain reliably inseparable.

[0112] The embodiment of the device composite shown in Figure 12 basically corresponds to the embodiment described in connection with Figure 7 . The groups of devices 2A, 2B, 2C, 2D also differ in their occupation of the materials of the first sub-layer 31 and the second sub-layer 32. Only the material of the first sub-layer 31 exists between the devices of the first group 2A and the connection layer 4. There is a surrounding sub-region 33 of the second sub-layer 32 below the devices of the second group 2B, and this sub-region serves as a delay element. The surrounding sub-region is formed, for example, in a frame shape or an annular shape. In the first etching step, the material below the devices of the first group 2A is completely removed, while the surrounding sub-region 33 below the devices of the second group 2B, the third group 2C, and the fourth group 2D protects the laterally surrounding material of the first sub-layer from material removal, such that only the devices of the first group are separable. The surrounding devices can then be removed together with the adjacent material of the first sub-layer 31. Thereby, the devices of the second group 2B can be separated.

[0113] For the devices of the third group 2C, there is a surrounding second sub-region 331 within the surrounding sub-region 33.

[0114] For the devices of the fourth group 2D, there is a surrounding third sub-region 332 within the surrounding second sub-region.

[0115] By repeating the etching step iteratively, these devices can be made separable in sequence.

[0116] By means of this interleaving of non-etchable structures or at least structures that can only be etched with difficulty, the following sequence can be defined in a reliable manner, according to which the individual groups of the device can be separated successively.

[0117] In Figures 13A to 13D an embodiment of a method for manufacturing a device composite is shown, in which, by way of example, a device composite as described in Figure 1 is manufactured. However, the method is in principle also suitable for manufacturing device composites according to other embodiments. For example, a plurality of devices 2 ( Figure 13A ) are provided on an initial carrier 29.

[0118] A sacrificial layer 3 is formed on the device 2. This is done in the illustrated embodiment by forming a first sub-layer 31 and a second sub-layer 32 of the sacrificial layer ( Figure 13B and 13C ). The sacrificial layer is configured such that the occupation with the sacrificial layer is different for the devices of the first group 2A and for the devices of the second group 2B.

[0119] In particular, before applying the second sub-layer 32, the first sub-layer 31 is formed in a lithographically structured manner.

[0120] The sacrificial layer 31 is applied, for example, by evaporation or sputtering on the device, where different deposition methods can also be applied for the different sub-layers.

[0121] Subsequently, the device with the sacrificial layer 3 is fixed to the carrier 5 by means of a connecting layer 4. For example, the material for the connecting layer 4 is applied in a flowable state to the sacrificial layer 3 and fills the intermediate spaces of the sacrificial layer. Subsequently, the connecting layer 4 can be cured if necessary, for example thermally. The initial carrier 29 can then be removed.

[0122] Figure 13D The finished device carrier composite 1 is shown.

[0123] This application claims the priority of German Patent Application 10 2019 126 862.1, the disclosure of which is incorporated herein by reference.

[0124] The invention is not limited by the description of the embodiments. Rather, the invention includes any new features and any combination of features, which in particular includes any combination of the features in the embodiments, even if such features or combinations are not themselves described in detail in the embodiments.

[0125] List of reference numerals

[0126] 1 device composite

[0127] 10 reference composite

[0128] 2 device

[0129] 2A Devices of the first group

[0130] 2B Devices of the second group

[0131] 2C Devices of the third group

[0132] 2D Devices of the fourth group

[0133] 21 Edge

[0134] 25 Transverse main extension axis

[0135] 29 Initial carrier

[0136] 3 Sacrificial layer

[0137] 31 First sublayer

[0138] 32 Second sublayer

[0139] 33 Surrounding sub-region

[0140] 331 Surrounding second sub-region

[0141] 332 Surrounding third sub-region

[0142] 35 Island

[0143] 4 Connecting layer

[0144] 41 Support structure

[0145] 5 Carrier

[0146] 7 Transfer body

[0147] 70 Bottom surface

[0148] 71 Column

[0149] 72 Punch

[0150] 75 Transverse main extension axis

Claims

1. A device composite, having a plurality of devices and a carrier, wherein - the devices are fixed to the carrier by means of a connection layer; - the connection layer forms at least one support structure for each device, at which the connection layer adjoins the device; - a sacrificial layer is locally provided between the device and the connection layer; - a part of the devices belongs to a first group; - another part of the devices belongs to a second group; and - the sacrificial layer is configured differently for the devices of the first group than for the devices of the second group.

2. The device composite according to claim 1, wherein the devices of the first group are arranged in a regular pattern.

3. The device composite according to claim 1 or 2, wherein at least some of the devices of the first group have, along two mutually perpendicular spatial directions, only nearest neighbors that do not belong to the devices of the first group.

4. The device composite according to claim 1 or 2, wherein the devices have a bottom surface with a basic shape of a polygon and, for a device of the first group, the nearest neighbors along all edges of the device do not belong to the devices of the first group.

5. The device composite according to claim 1 or 2, wherein at least some of the devices of the first group are surrounded along the entire circumference by nearest neighbors that do not belong to the devices of the first group.

6. The device composite according to claim 1 or 2, wherein the devices of the first group and the second group differ in terms of the average thickness of the sacrificial layer between the respective devices and the carrier.

7. The device composite according to claim 1 or 2, wherein the sacrificial layer has a first sublayer of a first material composition and a second sublayer of a second material composition different from the first material composition, and wherein the devices differ from each other in terms of the occupation by the material of the first sublayer.

8. The device composite according to claim 7, wherein islands are formed by means of the first sublayer, and the islands are surrounded by the second sublayer by at least 80% along their lateral circumference.

9. A method for separating devices from a device composite, having the following steps: a) providing a device composite according to any one of claims 1 to 8; b) locally removing the material of the sacrificial layer; c) separating at least one device of the first group; d) removing the other material of the sacrificial layer; e) separating at least one device of the second group.

10. The method according to claim 9, wherein the devices of the second group are inseparable by the separating force used for separation in step c) after step b) and before step d).

11. The method according to claim 9 or 10, wherein the separation of the devices is carried out by means of a transfer body.

12. The method according to claim 11, wherein the transfer body has a bottom surface that is at least 20% larger than the bottom surface of one of the separated devices.

13. The method according to claim 11, wherein the lateral main extension axis of the transfer body is rotated with respect to the lateral main extension axis of one of the separated devices.

14. The method according to claim 9 or 10, wherein the sacrificial layer has a first sub-layer of a first material composition and a second sub-layer of a second material composition different from the first material composition, and wherein step b) is carried out by means of a chemical method which is selective with respect to the first and second material compositions.

15. A method for manufacturing a device composite, comprising the following steps: a) providing a plurality of devices; b) forming a sacrificial layer on the devices, wherein a part of the devices belongs to a first group and another part of the devices belongs to a second group, and wherein the sacrificial layer is formed differently for the devices of the first group than for the devices of the second group; c) fixing the devices to a carrier by means of a connecting layer.

16. The method according to claim 15, wherein the connecting layer is at least partially flowable at the time of fixing the devices.

17. The method according to claim 15 or 16, wherein the sacrificial layer is formed by means of a first sub-layer and a second sub-layer, and wherein the first sub-layer is structured lithographically before applying the second sub-layer.

18. The method according to claim 15 or 16, wherein a device composite according to any one of claims 1 to 8 is manufactured.

Images