METHODS AND SYSTEMS OF FORMING METAL INTERCONNECT LAYERS USING ENGINEERED TEMPLATES

Abstract

Described herein are methods and systems for forming metal interconnect layers (MILs) on engineered templates and transferring these MILs to device substrates. This “off-device” approach of forming MILs reduces the complexity and costs of the overall process, allows using semiconductor processes, and reduces the risk of damaging the device substrates. An engineered template is specially configured to release a MIL when the MIL is transferred to a device substrate. In some examples, the engineered template does not include barrier layers and/or adhesion layers. In some examples, the engineered template comprises a conductive portion to assist with selective electroplating. Furthermore, the same engineered template may be reused to form multiple MILs, having the same design. During the transfer, the engineered template and device substrate are stacked together and then separated while the MIL is transitioned from the engineered template to the device substrate.

Claims

1. A method of integrating a transferable metal interconnect layer (MIL) to a device substrate, the method comprising: transferring the transferable MIL from an engineered template to the device substrate by stacking the transferable MIL and the device substrate such that the transferable MIL contacts the device substrate, and separating the engineered template from the transferable MIL, wherein the transferable MIL is retained by the device substrate and while the transferable MIL maintains a MIL design on the device substrate set by the engineered template.

2. The method of claim 1, wherein, during the stacking, the transferable MIL adheres to the device substrate such that an adhesion strength between the transferable MIL and the device substrate is greater than an adhesion strength between transferable MIL and the engineered template.

3. The method of claim 1, wherein the engineered template comprises: a base portion, and template features, protruding into the base portion and determined by a MIL design of the MIL, the template features configured to receive the MIL, wherein a surface of the template features is specifically selected such that adhesion of the MIL to the engineered template is below a threshold, corresponding to the device substrate, thereby allowing to transfer the MIL from the engineered template to the device substrate.

4. The method of claim 3, wherein the engineered template comprises silicon dioxide, forming the surface of the template features.

5. The method of claim 3, wherein the engineered template is free from a barrier layer.

6. The method of claim 3, wherein the engineered template comprises a base portion and a conductive portion such that the template features are formed within the base portion and selectively protrude up to the conductive portion.

7. The method of claim 6, wherein the base portion of the engineered template comprises silicon oxide, and wherein the conductive portion of the engineered template comprises a low-resistivity silicon wafer.

8. The method of claim 1, further comprising forming an adhesion layer on the device substrate such that, when stacking the engineered template and the device substrate, the transferable MIL contacts the adhesion layer of the device substrate.

9. The method of claim 8, wherein the adhesion layer comprises (3-mercaptopropyl) tri methoxysilane.

10. The method of claim 8, wherein the adhesion layer comprises an adhesion promoter, which is selected from the group comprising hexamethyldisilazane, Ti-Prime, and a diphenylsilanediol-derivative.

11. The method of claim 8, wherein the adhesion layer is formed using (a) vapour-deposition or (b) nitrogen rinse followed by chemical binding.

12. The method of claim 8, wherein the adhesion layer has a thickness of between about 0.5 nm and 50 nm.

13. The method of claim 1, wherein: the engineered template and the transferable MIL form a stack further comprising a seed layer positioned between the engineered template and the transferable MIL, and after transferring the transferable MIL from the engineered template to the device substrate, the transferable MIL is covered by the seed layer.

14. The method of claim 13, wherein the method further comprises, after transferring the transferable MIL from the engineered template to the device substrate, removing the seed layer of the transferable MIL thereby exposing the transferable MIL.

15. The method of claim 14, wherein the seed layer and the transferable MIL are formed from substantially similar materials.

16. The method of claim 14, wherein the seed layer and the transferable MIL are formed from different materials.

17. The method of claim 1, further comprising forming a selective capping layer on the transferable MIL such that the device substrate remains free from the selective capping layer.

18. The method of claim 1, wherein, after the transferring, the transferable MIL protrudes above the device substrate.

19. The method of claim 1, prior to, stacking the transferable MIL and the device substrate, adhering the transferable MIL to an intermediate transfer layer (ITL), wherein: separating the engineered template from the transferable MIL is performed while the transferable MIL is adhered to the ITL, and stacking the transferable MIL and the device substrate is performed by separating the transferable MIL from the ITL.

20. The method of claim 19, further comprising reducing an adhesion strength between the transferable MIL and the ITL after stacking the transferable MIL and the ITL and before separating the transferable MIL from the ITL.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0019] FIGS. 1A is a schematic illustration of the overall process workflow, corresponding to “off-device” MIL forming followed by integrating the MIL into a device.

[0020] FIGS. 1B, 1C, and 1D are schematic illustrations of different tools and components used for MIL forming, transferring, and integration into a device.

[0021] FIG. 2 is a process flowchart corresponding to a method of forming a MIL for transferring to the device substrate, in accordance with some examples.

[0022] FIGS. 3A-3D are schematic representations of different processing stages of the method of FIG. 2, in accordance with some examples.

[0023] FIGS. 3E-3J are schematic representations of different processing stages of the method of FIG. 2, in accordance with some examples.

[0024] FIG. 3K is the scanning electron microscope (SEM) and FIG. 3L is the optical image of a MIL on a device substrate, formed in accordance with the operating steps shown in FIGS. 3E-3J.

[0025] FIGS. 4A-4D are schematic representations of different processing stages of the method of FIG. 2, in accordance with some examples.

[0026] FIGS. 4E-4F and, separately, FIGS. 4G-4H are schematic representations of different processing stages of the method of FIG. 2, in accordance with some examples.

[0027] FIG. 5 is a process flowchart corresponding to a method of integrating a MIL to a device substrate, in accordance with some examples.

[0028] FIGS. 6A-6G represent different stages of the method of FIG. 5.

[0029] FIG. 7A is a photograph of an engineered template used to manufacture a MIL shown in FIGS. 7A-7E.

[0030] FIG. 7B is a photograph of a MIL, manufactured using the engineered template of FIG. 7A, after removing the MIL from the engineered template and positioning it on a device substrate but before removing a seed layer.

[0031] FIG. 7C is a photograph of the MIL from FIG. 7B, after removing the seed layer.

[0032] FIGS. 7D and 7E are higher magnification photos of two portions of the MIL shown in FIG. 7C.

[0033] FIGS. 8A and 8B represent optical images of the transferred metal interconnect layer on the adhesive tape surface at different optical magnifications.

[0034] FIGS. 9A and 9B illustrate an optical image of a MIL on an adhesive tape surface and a surface profile measurement for this sample.

[0035] FIGS. 9C and 9D illustrate another image of a MIL a surface profile measurement for this sample.

DETAILED DESCRIPTION

[0036] In the following description, numerous specific details are outlined in order to provide a thorough understanding of the presented concepts. In some examples, the presented concepts are practiced without some or all of these specific details. In other instances, well-known process operations have not been described in detail to not unnecessarily obscure the described concepts. While some concepts will be described in conjunction with the specific examples, it will be understood that these examples are not intended to be limiting.

Introduction

[0037] MILs are used in different electronic and/or related devices, such as semiconductor chips, solar cells, displays, flexible electronics, and the like. For example, in addition to billions of transistors, an advanced chip comprises tens of miles of metal interconnects, positioned at multiple levels and interconnecting transistors and other components to each other. The chip performance depends, in large part, on signals and power transmission through these MILs. Often, a MIL becomes a limiting component in the chip or other electronic devices.

[0038] Conventionally, MILs are formed right on devices. This “on-device” approach requires subjecting the devices to many different process steps while forming MILs to interconnect these devices. First, each processing step adds to the overall device cost. Second, many of these steps involve harsh and stressful environments and conditions, which are potentially damaging to the resulting devices. Some examples of these steps include etching, polishing, high-temperature annealing or sintering, and the like. In some instances, these steps limit applications of MILs in various types of devices, e.g., an air-gap technology in dielectric layers.

[0039] Proposed methods and systems use engineered templates to form MILs, followed by transferring these metal interconnect layers to device substrates. These MILs may be referred to as transferable MILs. Specifically, a MIL is first formed on an engineered template, which comprises template features determined by the MIL design. Once formed, the MIL is transferred to the device substrate, e .g., to a specifically designated location on the device substrate and adhered to this device substrate. The MIL design and the transfer position of the MIL are determined by the position of devices on the substrate. With this “off-device” approach, the device substrate is not subjected to environments and conditions used to form MILs. This “off-device” approach not only preserves the integrity of the device substrate and any components positioned on the substrate but also helps to save on processing costs.

[0040] FIGS. 1A is a schematic illustration of the overall process workflow 100, corresponding to the “off-device” MIL forming followed by integrating the MIL into a device. Specifically, workflow 100 comprises two separate processes, i.e., MIL forming (method 110) and MIL circuit integration (method 150). In some examples, a device may include multiple MILs. Therefore, the MIL forming process may be performed multiple times as schematically shown with block 111. At the same time, several engineered templates of the same design may be used to provide asynchronous operations of the MIL formation and transfer.

[0041] FIGS. 1B-FIG. 1D are schematic illustrations of different components used for MIL forming, transferring, and integration into a device. Specifically, FIG. 1B illustrates engineered template 210 with MIL 250, formed using engineered template 210. FIG. 1B also illustrates device substrate 290 before transferring MIL 250 to device substrate 290. Various examples of forming MIL 250 using engineered template 210 are described below with reference to FIG. 2. In some examples, device substrate 290 comprises adhesion layer 292.

[0042] FIG. 1C illustrates engineered template 210 and MIL 250 adhered to device substrate 290 or, more specifically, to adhesion layer 292 of device substrate 290. Engineered template 210 and adhesion layer 292 are configured in such a way that the adhesion of MIL 250 to device substrate 290 is greater than to engineered template 210. Therefore, when engineered template 210 is separated from device substrate 290, as schematically shown in FIG. 1D, MIL 250 remains on device substrate 290 and is separated from engineered template 210. At this point, engineered template 210 is ready for forming another MIL.

[0043] It should be noted that the MIL forming process (methods 110) and the MIL device integration (methods 150) may be performed by different entities. Therefore, various aspects of the circuit design may not be known at least to one entity or, in some examples, to both. This information separation provided by the “off-device” approach provides additional benefits.

MIL Forming Examples

[0044] FIG. 2 is a process flowchart corresponding to method 110 of forming MIL 250 for transferring to device substrate 290, in accordance with some examples.

[0045] In some examples, method 110 comprises fabricating (block 112 in FIG. 2) engineered template 210, which is an optional operation. Engineered template 210 comprises base portion 214 and template features 212. Template features 212 are determined by the design of MIL 250. Base portion 214 may be formed from silicon or silicon oxide (e.g., a silicon or silicon oxide wafer). In other examples, base portion 214 is formed from a plastic or dielectric material, arranged as a plate or as a film. Overall, base portion 214 is suitable for patterning and MIL formation techniques, further described below.

[0046] In some examples, engineered template 210 is a silicon wafer, used to support many standard semiconductor operations. In other examples, engineered template 210 is a roll, used for roll-to-roll processing while forming MIL 250 and also while transferring MIL 250 to device substrates.

[0047] Template features 212 may be formed by preparing the surface (e.g., polishing, cleaning) of base portion 214, depositing a photoresist layer, lithographic exposure of the desired pattern (corresponding to the MIL design), developing the photoresist layer, clearing of undeveloped areas of the photoresist and clearing of openings, etching of the template in the openings to the desired depth, cleaning (e.g., ashing) of the remaining photoresist and cleaning of the formed pattern and template surface. This process may be repeated for dual damascene structures, which would include both trenches and vias. More advanced process steps may include forming etch stop layers.

[0048] It should be noted that the pattern of template features 212 on base portion 214 is a mirror image of the MIL pattern. As such, MIL 250, once is formed, is ready for the direct transfer to a device substrate as a mirror image. Alternatively, MIL 250 is ready for a two-step transfer as a direct image, e.g., with an intermediate-assisted transfer on a temporary substrate.

[0049] In some examples, template features 212 have non-one dimensional shapes (e.g., an L-shape, S-shape, or T-shape) to promote the transferability of MIL 250 to device substrate 290 as, for example, is schematically shown in FIGS. 4E-4F. Specifically, the shapes are selected to increase the contact surface between MIL 250 and device substrate 290 thereby increasing the adhesion bond between MIL 250 and device substrate 290 and ensuring that MIL 250 is fully retained on device substrate 290 when engineered template 210 is removed.

[0050] In some examples, template features 212 have trapezoid and/or triangular shapes, which are open to the outer (contact) surface of engineered template 210 to promote the transferability of MIL 250 to device substrate 290 as, for example, is schematically shown in FIGS. 4G-4H.

[0051] One example of engineered template 210 is shown in FIG. 3A. Engineered template 210 comprises base portion 214 having face surface 211. Engineered template 210 also comprises multiple template features 212, protruding into base portion 214 and away from face surface 211. Template features 212 correspond to the design/pattern of MIL 250, formed using engineered template 210. Template features 212 are defined by feature surfaces 213.

[0052] Another example of engineered template 210 is shown in FIG. 4A. In this example, engineered template 210 comprises base portion 214 and conductive portion 216. In this example, template features 212 are formed within base portion 214 and selectively protrude up to conductive portion 216. More specifically, conductive portion 216 forms bottoms of at least some of template features 212 as shown in FIG. 4A. Conductive portion 216 is used to assist with the electrofilling of template features 212 while forming MIL 250, as further described below.

[0053] In some examples, base portion 214 of engineered template 210 comprises at least one of silicon dioxide, silicon, plastic, dielectric, and other like materials. In the same or other examples, conductive portion 216 of engineered template 210 comprises at least one of low resistivity (doped) silicon.

[0054] In some examples, method 110 comprises forming (block 114 in FIG. 2) adhesion-control layer 215 on the surface of engineered template 210. Unlike barrier layers and adhesion-promoting layers used in conventional devices and techniques, adhesion-control layer 215 is used to reduce the adhesion between engineered template 210 and other layers formed on engineered template 210, such as seed layer 220. This adhesion reduction helps to maintain the integrity of MIL 250 when MIL 250 is separated from engineered template 210. In some examples, adhesion-control layer 215 is formed by depositing a silicon oxide layer over engineered template 210, chemical treatment, and/or plasma treatment of engineered template 210. In general, any surface-modifying treatment can be used to achieve the desired adhesion properties.

[0055] In some examples, method 110 comprises depositing (block 116 in FIG. 2) seed layer 220 on engineered template 210 as, e.g., is schematically shown in FIGS. 3B and 4B. As noted above, engineered template 210 comprises template features 212 determined by the design of MIL 250. Seed layer 220 is conformal and covers both feature surface 213 and face surface 211 of base portion 214. Seed layer 220 may be formed using a physical vapor deposition (PVD) technique. For example, a copper layer having a thickness from 50 to 100 nm may be used as seed layer 220. In some examples, seed layer 220 and MIL 250 are formed from substantially similar materials. For purposes of this disclosure, the term “substantially similar” refers to variations of the composition being less than 20% atomic or even less than 10% atomic. For example, both seed layer 20 and MIL 250 are formed from copper. Alternatively, seed layer 220 and MIL 250 are formed from different materials. In these examples, the composition of seed layer 220 and MIL 250 vary by more than 20% atomic. For example, seed layer 220 is formed from copper while MIL 250 is formed from gold.

[0056] In some examples, seed layer 220 is formed directly over and in contact with the base material (e.g., silicon dioxide) of engineered template 210. For purposes of this disclosure, the “base material” is defined as the main primary material of a structure such that at least 90% by volume of the structure is formed from the base material. The direct interface between the base material and the seed layer, in other words, there are no intermediate layers (e.g., diffusion barrier layers, adhesion layers, etc.) between seed layer 220 and engineered template 210. Unlike conventional substrates, which later become parts of devices, engineered template 210 is a reusable tool and is not integrated into any electronic devices. Therefore, there are no concerns with contamination of the base material (of engineered template 210, e.g., by metal diffusion into engineered template 210.

[0057] Similarly, there are no concerns with strong bonds between seed layer 220 and engineered template 210. In fact, the adhesion strength between seed layer 220 and engineered template 210 and, later, between MIL 250, formed on and/or incorporating seed layer 220, and engineered template 210 should be less than between MIL 250 and device substrate 290 to allow transfer of MIL 250 from engineered template 210 to device substrate 290.

[0058] Special consideration can be taken for controlling the adhesion strength between MIL 250 and engineered template 210 by forming specific layers (e.g., adhesion-control layer 215) on engineered template 210, i.e., between engineered template 210 and seed layer 220. In some examples the specific layers on engineered template 210, i.e., between engineered template 210 and seed layer 220 can be formed to promote the electroless deposition of metal (copper) so PVD seed layer formation can be replaced by electroless copper seed layer formation.

[0059] In some examples, method 110 comprises forming (block 118) blocking layer 230 over seed layer 220 as, e.g., schematically shown in FIG. 3G. In these examples, seed layer 220 comprises first portions 221, extending over the field of engineered template 210. Seed layer 220 also comprises second portions 222, covering the sidewalls of template features 212. Furthermore, seed layer 220 comprises third portions 222, covering the bottoms of template features 212 and extending between second portions 222. As shown in the FIG. 3G example, blocking layer 230 covers first portions 221 and third portions 223 of seed layer 220. Second portions 222 of seed layer 220 are free from blocking layer 230 and remain exposed depositing blocking layer 230 and before depositing MIL 250. In these examples, MIL 250 is selectively electroplated over second portions 222 of seed layer 220 and not over blocking layer 230 as, e.g., is schematically shown in FIG. 3H. In other words, MIL 250 is formed on the sidewalls of template features 212 and not on the bottoms of template features 212 or on the field of engineered template 210.

[0060] In other examples, blocking layer 230 does not extend into template features 212 and only covers the field of engineered template 210, extending between template features 212. In other words, in these examples, first portions 221 of seed layer 220 are covered with blocking layer 230, while second portions 222 and third portions 223 of seed layer 220 are free from blocking layer 230 and available for the MIL deposition. In these examples, MIL 250 completely fills each of template features 212, in a manner similar to FIG. 3C. Furthermore, blocking layer 230 prevents the deposition of the material (forming MIL 250) on the field of engineered template 210.

[0061] In some examples, method 110 comprises selectively electroplating (block 120 in FIG. 2) MIL 250 into template features 212. FIGS. 3C, 3H, and 4C illustrate engineered template 210 with MIL 250 at least partially filling template features 212. In some examples, e.g., FIGS. 3C and 4C, template features 212 are fully filled with MIL 250. In these examples, the width of template features 212 defines the width of MIL traces. Alternatively, e.g., FIG. 3H, template features 212 are partially filled with MIL 250. In this example, electroplating conditions (e.g., time, current, concentration, etc.) define the width of MIL traces. In some examples, the remaining portions of template features 212 (not filled with MIL 250) are later filled with an interlayer dielectric as, e.g., schematically shown in FIG. 3I. Overall, these alternative examples may yield much narrowed MIL traces as, e.g., shown FIGS. 3K and 3L.

[0062] As noted above, the adhesion strength between MIL 250 and engineered template 210 or, more specifically, between seed layer 220 and engineered template 210 should be less than the adhesion strength between MIL 250 and device substrate 290. This adhesion strength difference allows the transfer of MIL 250 from engineered template 210 to device substrate 290 while preserving the integrity of MIL 250 during the transfer. Overall, the adhesion strength of MIL 250 to engineered template 210 is below a set threshold, corresponding to device substrate 290.

[0063] In some examples, method 110 comprises annealing (block 128 in FIG. 2) MIL 250 on engineered template 210. Annealing is used to improve the quality of metal (copper) interconnects by advancing the grain structure of deposited material.

[0064] In some examples, method 110 comprises performing selective capping (block 129). This operation is optional. An example of capping layer 260 is shown in FIG. 3C. It should be noted that capping layer 260 is only formed over MIL 250. In this example, a portion of seed layer 220 extending between over MIL 250 remains free from capping layer 260. Capping layer 260 may help to improve the adhesion of MIL 250 to a device substrate, which is described below with reference to FIG. 5.

[0065] It should be noted that the final product of method 110 may be an assembly comprising MIL 250 and engineered template 210. Engineered template 210 is used to support MIL 250 prior to transferring MIL 250 to a device substrate, which may be performed in a different method by a different entity. However, one having ordinary skill in the art would understand that when MIL 250 is removed from engineered template 210, engineered template 210 may be reused in method 110 to form another MIL, having the same design. This reuse of engineered template 210 is schematically shown in FIGS. 3D, 3J, and 4D.

[0066] It should be also noted that the same engineered template 210 may support multiple MILs having the same or different designs. Each of these MILs may be removed individually from engineered template 210, e.g., for transferring to the corresponding device substrate. The alignment of the individual MILs to the corresponding device substrate is achieved during the transfer and is not limited by the initial positioning of these MILs on engineered template 210. As such, the position of different MILs on engineered template 210 can be selected, e.g., to achieve higher utilization of engineered template 210 and other factors.

Example of Integrating MILs to Device Substrates

[0067] FIG. 5 is a process flowchart corresponding to method 150 of integrating MIL 250 to device substrate 290, in accordance with some examples. Method 150 may be performed for the entire workpiece surface (e.g., matching of device substrate 290 and engineered template 210) as well as for any selected part of device substrate 290 (e.g., a die in the case of semiconductor devices). For example, one device substrate 290 may receive multiple MILs at different positions. As noted above, method 150 may be performed independently and by a different entity from method 110 thereby allowing more reduced cost, simplified integration, limited design sharing, and a wide range of usable materials for interconnects and interlayer dielectrics.

[0068] In some examples, method 150 comprises forming (block 160) adhesion layer 292 on device substrate 290 as, e.g., is schematically shown in FIG. 6A. Adhesion layer 292 comprises at least one of. adhesion promoters such are HMDS (hexamethyldisilazane), Ti-Prime and diphenylsilanediol-derivatives (AR 300-80). HMDS is vapour-deposited or applied from a nitrogen rinse onto the heated wafers (80-150° C.), followed by chemical binding of HMDS to the wafer surface as a monolayer. Diphenylsilanediol is dissolved in a solvent mixture and applied by spin coating. In the subsequent bake step (160-180° C.), the silicon-organic compound melts and forms a thin, adhesion-promoting film. Adhesion layer 292 has a thickness of between about 0.5 nm and 50 nm.

[0069] In some examples, adhesion layer 292 comprises a material with double functional groups, with one of the groups of high affinity to substrate material (silicon) and the second functional group providing high adhesion to interconnect material. One example of such materials is (3-Mercaptopropyl)trimethoxysilane (MPTS).

[0070] It should be noted that adhesion layer 292 is optional and, in some examples, MIL 250 directly contacts the bulk portion of device substrate 290. In other examples, an adhesive layer may be formed on the assembly of engineered template 210 and MIL 250. In more specific examples, an adhesive layer is selectively formed on MIL 250 (e.g., similar to a capping layer described above).

[0071] In some examples, method 150 comprises direct transferring (block 162) MIL 250 from engineered template 210 to device substrate 290. For example, this transferring operation may involve stacking engineered template 210, comprising MIL 250, and device substrate 290, comprising adhesion layer 292, such that MIL 250 contacts adhesion layer 292 or, directly, to device substrate 290. This stack is schematically shown in FIG. 6B. During this stacking operation, MIL 250 adheres to device substrate 290. As noted above, the adhesion strength between MIL 250 and device substrate 290 is greater than between MIL 250 and engineered template 210. More specifically, the adhesion strength between MIL 250 and device substrate 290 is greater than between seed layer 220 and engineered template 210, at least when seed layer 220 is positioned between MIL 250 and engineered template 210.

[0072] In some examples, seed layer 220 provides an envelope for the transfer of any structures (e.g., MIL 250 and potentially other structures), which are formed over seed layer 220. More specifically, these structures are separated from engineered template 210 by seed layer 220 thereby allowing the separation of the structures from engineered template 210. As noted above, seed layer 220 controls the adhesion to engineered template 210 and is responsible for the transfer of these structures from engineered template 210 to device substrate 290. It should be noted that any type of structures can be formed over seed layer 220, such as multi-layered structures, active elements (e.g., transistors, diodes, solar cells, and the like). This envelope can be transferred as a whole, thereby maintaining the internal arrangements within these structures and also the external arrangement among the structures.

[0073] The transferring operation then proceeds with separating engineered template 210 from device substrate 290 while MIL 250 is retained by adhesion layer 292 on device substrate 290 (because of the difference in the adhesion strength described above). This part of the operation is schematically shown in FIG. 6C. After the transferring operation, MIL 250 protrudes above device substrate 290. For this transfer method, the desired MIL pattern is converted into a mirror image, also with consideration that via structures in engineered template 210 protrudes upright after the pattern transfer.

[0074] In some examples, method 150 comprises transferring (block 161) MIL 250 from engineered template 210 to device substrate 290 using, e.g., an intermediate transfer layer (ITL). The ITL comprises a flexible or rigid substrate with an adhesive coating. In some examples, the adhesion of the ITL is chosen to be lower than the adhesion of device substrate 290. In other examples, the ITL is chosen from materials with a controlled adhesion strength, e.g., with adhesion strength variation provided by the application of UV radiation and/or heat. The adhesion strength between MIL 250 and the ITL could be highest when MIL 250 is transferred from engineered template 210 to the ITL and, later, significantly reduced for the transfer from the ITM to device substrate 290. For this method of MIL transfer, the template pattern is the same as desired MIL pattern on device substrate 290 (not a mirror image).

[0075] In some examples, method 150 comprises removing (block 164) seed layer 220 from MIL 250 as, e.g., is schematically shown in FIG. 6D. The seed layer removal may involve wet etch or dry etch. Any etching process is applicable after the selective electrodeposition, which prevents the metal deposition in the field area. In these examples, the undesired material between MIL 250 (between the lines) is very thin (e.g., the seed layer thickness) and can be easily removed before affecting the thickness of MIL 250.

[0076] In some examples, method 150 comprises coating (block 166) of the transferred MIL with a special material to improve the interconnect performance. An example of such improvements is the selective cobalt coating of copper lines, which may be also referred to as selective capping. Cobalt capping reduces the electromigration of copper in interconnects, also preventing copper diffusion into dielectric layers. Cobalt capping is formed, e.g., selectively on MIL 250 by an electroless deposition process. An example of capping layer 260 is shown in FIG. 6E.

[0077] In some examples, method 150 comprises forming (block 170) an interlayer dielectric (ILD) over MIL 250, while MIL 250 is disposed over device substrate 290. An example of ILD 270 is shown in FIG. 6F. The ILD formation, as a second step after the MIL transfer, enables using different methods and structures such as: (a) silicon oxide coating with the following CMP step if necessary; (b) Spin-on Dielectric fabrication with desired thickness to open the interlayer contact areas; (c) Low-k material deposition (d) formation air-gap structures.

[0078] In some examples, device substrate 290 comprises conductive features 294 as, e.g., is schematically shown in FIG. 6G. In these examples, conductive features 294 are aligned with the features of MIL 250 to form electrical connections between at least some of these features. In more specific examples, MIL 250 is disposed between two device substrates, each comprising conductive features, e.g., device substrate 290 with conductive features 294 and additional device substrate 291 with additional conductive features 293. In these examples, at least some conductive features of MIL 250 may form electrical connections with both at least a subset of conductive features 294 and at least a subset of additional conductive features 293 as, e.g., is schematically shown in FIG. 6G.

Examples of Experimental Verification of MIL Formation and Transfer

[0079] Various experiments were performed to show the feasibility of methods of forming MILs on engineered templates and transferring these MILS to a device substrate. In one experiment, a template was formed on a 200-mm silicon oxide wafer (base) by etching SEMI 854 pattern structures using a lithography process. A 75-nm thick copper seed layer was deposited using a PVD process. The copper seed layer was selectively coated at the field areas with a blocking layer, which may e also referred to as a protective agent. This protective agent prevents the electrochemical deposition of copper in the areas covered by the agent and facilitates the “trench only” deposition. The template was subjected to an electrochemical deposition of copper on limited 2-inch diameter surface areas. The seed layer of copper was etched back (between the MILs features) at least partially by the application of the reverse polarity at the end of the deposition. This seed layer removal resulted in the exposure of the underlayer template surface.

[0080] FIG. 7A is a photograph of an engineered template used in this experiment. This engineered template comprises 4 separate visible dies. It should be noted that a typical wafer can contain any number of dies, each of which can be considered as a different MIL and transferred separately with the precise alignment at the end of the transfer on the device substrate. The MIL template is not shown in this photograph. FIG. 7A identifies two areas (D1 and D2) on the engineered template, which have distinctive designs. FIG. 7B is a photograph of a MIL, manufactured using the engineered template of FIG. 7A. Specifically, FIG. 7B illustrates the MIL after removing the MIL from the engineered template and positioning the MIL on a device substrate, which is a transparent plastic film in this example. The MIL, shown in FIG. 7B, is still with a seed layer. FIG. 7C is a photograph of the MIL from FIG. 7B, after removing the seed layer. FIG. 7C shows a clear separation between the different features of the MIL. Furthermore, FIGS. 7C also identifies two areas (D1 and D2), corresponding to the areas on the engineered template identified in FIG. 7A. The distinctive designs of the engineered template are visible in the corresponding features of the MIL. FIGS. 7D and 7E are higher magnification photos of two portions of the MIL shown in FIG. 7C.

[0081] FIGS. 8A and 8B represent optical images of a transferred MIL on an adhesive tape surface at different optical magnifications. The copper seed layer was removed from this MIL (before the transfer), and FIG. 8B shows clearly separated copper lines. This image confirms an acceptable quality of the MIL in the micron size range on a flexible substrate.

[0082] FIG. 9A is an optical image of a MIL on an adhesive tape surface. FIG. 9B is a corresponding surface profile, obtained by a laser confocal microscope. The seed layer of copper was not removed from the MIL before the transfer. FIG. 9C is an optical image of another MIL on an adhesive tape surface. FIG. 9D is a corresponding surface profile, obtained by a laser confocal microscope. This data also demonstrates a good quality of MIL formation in the micron size range on a flexible substrate.

Conclusion

[0083] Although the foregoing concepts have been described in some detail for purposes of clarity of understanding, it will be apparent that certain changes and modifications may be practiced within the scope of the appended claims. It should be noted that there are many alternative ways of implementing processes, systems, and apparatus. Accordingly, the present examples are to be considered illustrative and not restrictive.