Apparatus and method for joining materials

Abstract

A method for preparing a surface for bonding utilizes multiple processing techniques to increase bonding strength. The method can include performing a first processing technique on a surface of a material using a first plasma process to promote diffusion into a first surface of a diffusion layer deposited after the first processing technique, where the first processing technique uses a capacitively coupled plasma (CCP) or a combination of both CCP and inductively coupled plasma (ICP); forming the diffusion layer on the surface of the material; performing a second processing technique to increase diffusion of the diffusion layer into the surface of the material using a second plasma process, where the second processing technique uses CCP; and performing a third processing technique to form dangling bonds on the diffusion layer using a third plasma process, where the third processing technique uses ICP or a combination of both CCP and ICP.

Description

Technical Field

[0001] The implementation of the principles in this case generally involves semiconductor processing of semiconductor substrates. Background Technology

[0002] In semiconductor manufacturing, many processes require bonding materials together. Poor bonding between materials leads to delamination, which can result in poor device performance, leading to reduced reliability and yield. To avoid poor surface bonding, an adhesive layer can be incorporated between the two materials. However, the inventors observed that the adhesive layer bonding strength is often very low, leading to bonding failure, especially after thermal cycling. For low thermal budget components, attempts to increase bonding strength (such as using high temperatures for diffusion enhancement) are often not feasible.

[0003] Thus, the inventors provide an apparatus and method for improving the bonding strength between materials while maintaining compatibility with low thermal budgets. Summary of the Invention

[0004] This article provides methods and apparatus for improving the bonding properties of materials while maintaining a low thermal budget.

[0005] In some embodiments, a method for preparing a surface for bonding may include: performing a first processing step on a first surface of a first material using a first plasma process to promote diffusion into the first surface of a first diffusion layer deposited after the first processing step, wherein the first processing step uses capacitively coupled plasma (CCP) or a combination of both CCP and inductively coupled plasma (ICP); forming a first diffusion layer on the first surface of the first material; performing a second processing step using a second plasma process to increase the diffusion of the first diffusion layer into the first surface of the first material, wherein the second processing step uses CCP; and performing a third processing step using a third plasma process to form dangling bonds on the first diffusion layer, wherein the third processing step uses ICP or a combination of both CCP and ICP.

[0006] In some embodiments, the method may further include: rotating a first material about a vertical axis during a first or third processing step; using an argon-based gas, a nitrogen-based gas, an oxygen-based gas, a hydrogen-based gas, or a fluorocarbon-based gas (C). x F y ) gas, or nitrogen fluoride-based (NF) xThe first plasma process for forming a gas-generated plasma; a first processing process, a second processing process, and a third processing process, each having a duration of approximately 30 seconds to approximately 5 minutes; a first processing process using a source power of approximately 1 kW to approximately 3 kW when using CCP and approximately 100 W to approximately 10 kW when using ICP; a first diffusion layer formed using physical vapor deposition (PVD), atomic layer deposition (ALD), chemical vapor deposition (CVD), thermal evaporation, or electron beam evaporation; a first diffusion layer formed using a source power of approximately 5 kW to approximately 60 kW at a pressure of approximately 0.5 mTorr to approximately 10 mTorr; and a first diffusion layer having a diameter of approximately 0.5 nm to approximately 5 nm. A first diffusion layer with a thickness of nm; a first diffusion layer being a molybdenum-based, titanium-based, tantalum-based, cobalt-based, tungsten-based, copper-based, silicon-based, nitride-based, silicide-based, or carbide-based material; a second or third plasma process using argon-based, nitrogen-based, oxygen-based, or hydrogen-based gas to form plasma; a second processing process executed at a pressure of approximately 0.5 mTorr to approximately 10 mTorr and a power of approximately 1 kW to approximately 3 kW; and a second processing process executed at a pressure of approximately 0.5 mTorr to approximately 20 mTorr and a power of approximately 100 kW. A third processing step performed at a source power of approximately 10 kW; a seed layer formed directly on the first diffusion layer after the third processing step, or an adhesive layer formed first on the first diffusion layer after the third processing step and subsequently a seed layer formed on the adhesive layer; the method is repeated on a second material, which is then bonded to the first material, wherein the first diffusion layer and the second diffusion layer on the second material serve as an intermediate layer between the bonded first and second materials; the first material is a glass-based material, and the method is performed on a glass-through-glass via (TGV) in the glass-based material; and / or a method performed in a single processing chamber, an integrated tool, or a separate processing chamber.

[0007] In some embodiments, a non-transitory computer-readable medium storing instructions thereon, when executed, causes a method for preparing surfaces for bonding, the method comprising: performing a first processing step on a first surface of a first material using a first plasma process to promote diffusion into the first surface of a first diffusion layer deposited after the first processing step, wherein the first processing step uses capacitively coupled plasma (CCP) or a combination of both CCP and inductively coupled plasma (ICP); forming a first diffusion layer on the first surface of the first material; performing a second processing step using a second plasma process to increase the diffusion of the first diffusion layer into the first surface of the first material, wherein the second processing step uses CCP; and performing a third processing step using a third plasma process to form dangling bonds on the first diffusion layer, wherein the third processing step uses ICP or a combination of both CCP and ICP.

[0008] In some embodiments, the method on a non-transitory computer-readable medium may further include: using argon-based gas, nitrogen-based gas, oxygen-based gas, hydrogen-based gas, or fluorocarbon-based gas. x F y ) gas, or nitrogen fluoride-based (NF) x The first plasma process for forming plasma using a gas; a first processing process, a second processing process, and a third processing process, each having a duration of approximately 30 seconds to approximately 5 minutes; a first processing process using a source power of approximately 1 kW to approximately 3 kW when using CCP and approximately 100 W to approximately 10 kW when using ICP; a first diffusion layer having a thickness of approximately 0.5 nm to approximately 5 nm; a diffusion layer being a molybdenum-based material, a titanium-based material, a tantalum-based material, a cobalt-based material, a tungsten-based material, a copper-based material, a silicon-based material, a nitride-based material, a silicide-based material, or a carbide-based material; a second plasma process or a third plasma process using an argon-based gas, a nitrogen-based gas, an oxygen-based gas, or a hydrogen-based gas to form plasma; a second processing process executed at a pressure of approximately 0.5 mTorr to approximately 10 mTorr and a power of approximately 1 kW to approximately 3 kW; and / or at a pressure of approximately 0.5 mTorr to approximately 20 mTorr and a power of approximately 100 W to approximately 10 kW. The third processing technology is performed on a source power of kW.

[0009] In some embodiments, an apparatus for preparing surfaces for bonding may include: a processing chamber having a rotating substrate support; a processing volume enclosed by an upper processing shield, a lower processing shield, and a rotatable top shield; and a plurality of selectable cathodes positioned on the top of the processing chamber, the cathodes being connected to a dual-power target power supply configured to generate RF power, DC power, or pulsed DC power, one of the cathode locations being an inductively coupled plasma (ICP) source; the dual-power power supply electrically connected to electrodes in the rotating substrate support and configured to generate capacitively coupled plasma (CCP) within the processing volume; and the ICP source including the dual-power power supply electrically connected to an antenna and configured to generate ICP within the processing volume, wherein the antenna is connected to the rotatable top shield. The system is positioned to be exposed to a processing volume; and a controller is configured to rotate a rotatable top shield to select a cathode for deposition or an antenna for ICP generation within the processing volume and to perform a method comprising performing a first processing process on the surface of a material using a first plasma process to promote diffusion into the surface of a diffusion layer deposited after the first processing process, wherein the first processing process uses CCP or a combination of both CCP and ICP to form a diffusion layer on the surface of the material; performing a second processing process using a second plasma process to increase the diffusion of the diffusion layer into the surface of the material, wherein the second processing process uses CCP; and performing a third processing process using a third plasma process to form dangling bonds on the diffusion layer, wherein the third processing process uses ICP or a combination of both CCP and ICP.

[0010] In some embodiments, the device may further include 20 dual power supplies, each including a first RF power supply having a first frequency of approximately 60 MHz and a second RF power supply having a second frequency of approximately 2 MHz.

[0011] Other and further implementation methods are disclosed below. Attached Figure Description

[0012] The embodiments of this principle, briefly summarized above and discussed in more detail below, can be understood by referring to the illustrative embodiments of this principle depicted in the accompanying drawings. However, the drawings only show common embodiments of this principle and are therefore not intended to be limiting, as other equivalent and effective embodiments are permissible.

[0013] Figure 1 A cross-sectional view of a processing chamber with joint surface treatment capability according to some embodiments of this principle is depicted.

[0014] Figure 2A bottom-up view depicts a top shield and cathode assembly configuration according to some embodiments of this principle.

[0015] Figure 3 This refers to a method for preparing surfaces for bonding based on some embodiments of this principle.

[0016] Figure 4 Cross-sectional views of the surface of the process treatment and diffusion layer formation according to some embodiments of this principle are depicted.

[0017] Figure 5 Cross-sectional views of material deposition onto a treated surface with an adhesive layer, according to some embodiments of this principle, are depicted.

[0018] Figure 6 Cross-sectional views of material deposition onto a treated surface are depicted according to some embodiments of this principle.

[0019] Figure 7 A cross-sectional view of two treated surfaces joined together according to some embodiments of this principle is depicted.

[0020] Figure 8 A cross-sectional view of a glass substrate with glass perforations according to some embodiments of this principle is depicted, the glass perforations having a surface treatment and an adhesive layer.

[0021] Figure 9 A cross-sectional view of a glass substrate with glass perforations according to some embodiments of this principle is depicted, the glass perforations having a surface treatment and not having an adhesive layer.

[0022] Figure 10 Cross-sectional views of different treated surfaces that can be joined together are depicted according to some embodiments of this principle.

[0023] Figure 11 A graph depicts the bonding strength of this method compared with other processes, based on some embodiments of this principle.

[0024] Figure 12 A top schematic diagram depicts an integration tool based on some implementations of this principle.

[0025] Figure 13 A top schematic diagram of another integration tool depicting some implementations based on this principle is shown.

[0026] Figure 14 A bottom-up view depicts alternative top shield and cathode assembly configurations according to some embodiments of this principle.

[0027] To facilitate understanding, the same reference numerals have been used where possible to identify common elements in the figures. The figures are not drawn to scale and have been simplified for clarity. Elements and features of one embodiment may be advantageously incorporated into other embodiments without further description. Detailed Implementation

[0028] Methods and apparatus provide improved bonding performance between similar and dissimilar materials. These techniques enhance the bonding process by using an interlayer of reinforcing diffusion and / or adhesive layers positioned between the two materials. This interlayer bonds the two materials together with a bonding strength exceeding 600% compared to conventional processes. The increased bonding strength also improves component reliability by reducing performance degradation attributable to material delamination after thermal cycling. This technique improves bonding performance without increasing the roughness of the bonding surfaces, thereby improving signal and power integrity in high-frequency applications such as, but not limited to, data center applications, 5G base station applications, and the like. The enhanced bonding process can be used for any materials, such as, but not limited to, glass-to-metal, glass-to-glass, and / or hybrid bonding, and is easily adaptable based on the materials being bonded. The process can be performed in multiple chambers, dedicated surface-treated chambers, or integrated chambers. This technique can be used for substrate-to-substrate, die-to-die, die-to-substrate, and / or through-hole bonding, such as, but not limited to, glass-through-hole (TGV) material bonding. This technique can also be used for interlayer bonding between substrates and the like.

[0029] Traditional bonding processes attempt to improve adhesion between materials by increasing interface roughness through techniques such as glass polishing and wet chemical etching, which results in signal integrity loss and additional power consumption. This technology utilizes a vacuum-based plasma device to improve adhesion by increasing the bond strength between the first material and the diffusion layer at the interface and between the diffusion layer and the second material through surface activation of the diffusion layer. This process and apparatus allow for enhanced adhesion strength between similar and dissimilar materials and are compatible with current processes, such as, for example, glass metallization processes for forming TGVs and hybrid bonding processes using dielectric and / or conductive materials. In some embodiments, the adhesion-enhancing process may include surface treatments to enhance subsequent diffusion layer formation, forming the diffusion layer on the treated surface, further diffusion-enhancing treatments of the diffusion layer, and surface activation of the diffusion layer to form dangling bonds on the diffusion layer surface. The enhanced adhesion process increases the reliability of TGV metallization, increases the adhesion between the adhesive layer and the materials (such as, but not limited to, glass and the like), and reduces peeling and delamination that can occur between the adhesive layer and the materials when subjected to stress accumulation, thermal cycling, and / or moisture, and the like.

[0030] This method can be used in independent chambers or in integrated tools (e.g., Figure 12 In a separate chamber of the integrated tool 1200, or in a surface preparation chamber having both inductively coupled plasma (ICP) and capacitively coupled plasma (CCP) capabilities along with sputtering deposition capabilities (such as, for example, in...). Figure 1 The processing is performed in a processing chamber 100. The processing chamber 100 provides the capability to complete all surface treatments and diffusion layer formation within a single chamber. The processing chamber 100 has an electrically grounded chamber body 102 and provides a processing volume 116 for processing a substrate 106 on a substrate support 104. The processing volume 116 is enclosed by a processing sleeve comprising a lower shield 120, an upper shield 118, and an electrically grounded top shield 122. The top shield 122 can be rotated 124 about a vertical axis 188 via a central shaft 126. The top shield 122 has an opening 128 for selecting a cathode or ICP source and shielding other sources / cathodes for protection during processing (see, for example...). Figure 2 and Figure 14 ).

[0031] A pumping system 130 provides a vacuum within a processing volume 116 for processing substrate 106. In some embodiments, the pumping system 130 may provide a pressure of approximately 0.5 mTorr to approximately 20 mTorr within the processing volume 116. An electromagnet 136 surrounding the chamber body 102 is configured to tune the plasma shape for more uniform film deposition onto the substrate 106. A gas supply 132 is fluidly connected to the processing volume 116 via a gas diffuser 134. The gas supply 132 may be configured to contain multiple gases, which may be controlled by a system controller 170 during processing of the substrate 106. In some embodiments, the gas supply 132 may supply one or more gases, such as, but not limited to, argon, nitrogen, oxygen, hydrogen, methane (CH4), carbon dioxide (CO2), and fluorocarbons (C). x F y ), nitrogen fluoride (NF) x ), and / or sulfur fluoride (S x F y ), and similar ones.

[0032] The substrate support 104 is surrounded by a cover ring 112 and a deposition ring 114 outside the cover ring 112. The substrate support 104 can rotate 110 about a vertical axis 188 during processing of the substrate 106 to allow for more uniform surface treatment and / or deposition when ICP is used to generate plasma in the processing volume 116. The substrate support 104 has an embedded plasma electrode 108 for generating CCP plasma within the processing volume 116. The embedded plasma electrode 108 is electrically connected to a CCP power supply 144 and generates an electromagnetic field in the vicinity of the substrate 106. The CCP power supply 144 includes a first matching network 138 electrically connected to a first RF power supply 140 and a second RF power supply 142. The first RF power supply 140 and the second RF power supply 142 can operate at the same frequency or different frequencies. In some embodiments, the CCP power supply 144 may use high-frequency RF to provide a high plasma density in the processing volume 116 and simultaneously use low-frequency RF to simultaneously increase the plasma energy in the processing volume 116. In some embodiments, the first RF power supply 140 can operate at 60 MHz and the second RF power supply 142 can operate at 2 MHz. The CCP power supply 144 is used to capacitively generate plasma within the processing volume 116 between the embedded plasma electrode 108 and the grounded chamber body 102 and the processing shielding assembly. In some embodiments, the CCP power supply 144 can generate approximately 1 kW to approximately 2 kW of power.

[0033] Processing chamber 100 includes a plurality of cathode assemblies 184 for a plurality of targets 182. Each of the cathode assemblies 184 is electrically connected to a sputtering power supply 158. The sputtering power supply 158 includes a second matching network 160 electrically connected to an RF sputtering power supply 162 and / or a DC sputtering power supply 164. The DC sputtering power supply 164 provides constant DC power and / or pulsed DC power to the cathode assemblies 184 and the targets 182. In some embodiments, the RF sputtering power supply 162 may operate at frequencies of, but not limited to, 13.56 MHz and the like. The targets 182 may be metallic or non-metallic. Metallic targets are sputtered using DC power. Non-metallic targets are sputtered using RF power. In some embodiments, one or more locations of the plurality of cathode assemblies 184 may be replaced by an ICP source 146 including an antenna 148. In some embodiments, the antenna 148 may be a planar coil or a cylindrical coil and the like. ICP source 146 is electrically connected to ICP power supply 150, which includes a matching network 152 electrically connected to a first RF ICP power supply 154 and a second RF ICP power supply 156. The first RF ICP power supply 154 and the second RF ICP power supply 156 may operate at the same and / or different frequencies. ICP source 146 generates inductively coupled plasma within processing volume 116. In some embodiments, ICP power supply 150 can provide approximately 100 watts to approximately 5 kilowatts of power. In some embodiments, the substrate support 104 rotates when inductively coupled plasma is generated within processing volume 116 due to the offset of ICP source 146 (offset relative to the vertical axis 188 passing through substrate support 104). In some embodiments, both ICP power supply 150 and CCP power supply 144 may operate simultaneously.

[0034] Processing chamber 100 may also include a system controller 170, which controls the operation of processing chamber 100 using direct control or alternatively by controlling a computer (or controller) associated with processing chamber 100. In operation, system controller 170 performs data collection and feedback from processing chamber 100 to optimize the performance of processing chamber 100. System controller 170 typically includes a central processing unit (CPU) 172, memory 174, and support circuitry 176. CPU 172 may be any form of general-purpose computer processor that can be used in industrial settings. Support circuitry 176 is conventionally coupled to CPU 172 and may include cache, frequency circuitry, input / output subsystems, power supplies, and the like. Software routines (such as those described below) may be stored in memory 174 and, when executed by CPU 172, transform CPU 172 into a dedicated computer (system controller 170). Software routines can also be stored and / or executed via a second controller (not shown) located remotely in the processing chamber 100.

[0035] Memory 174 takes the form of a computer-readable storage medium containing instructions that, when executed by CPU 172, facilitate the operation of semiconductor processes and devices. The instructions in memory 174 take the form of a program product, such as a program implementing the methods of this principle. The program code may conform to any of several different programming languages. In one instance, this disclosure may be implemented as a program product stored on a computer-readable storage medium for use with a computer system. The program product defines the functionality of aspects of the program product (including the methods described herein). Illustrative computer-readable storage media include, but are not limited to: non-writable storage media on which information is permanently stored (e.g., read-only memory elements within a computer, such as CD-ROM discs readable by a CD-ROM drive, flash memory, ROM chips, or any type of solid-state non-volatile semiconductor memory); and writable storage media on which variable information is stored (e.g., floppy disks or hard disk drives within a disk drive or any type of solid-state random access semiconductor memory). Such computer-readable storage media are aspects of this principle when carrying computer-readable instructions that direct the functionality of the methods described herein.

[0036] Figure 2 This is a bottom-up view 200 of the top shield 122. Figure 2In the example depicted, the top shield 122 has been rotated such that opening 128 exposes the ICP source 146, while multiple available targets 182A to 182E are covered or shielded by the top shield 122. During deposition, the top shield 122 can be rotated 202 such that opening 128 in the top shield 122 exposes one of the multiple available targets 182A to 182E as needed. For the multiple available targets 182A to 182E, the target material composition may be the same or different. The target material used for diffusion layer formation may include, but is not limited to, molybdenum-based materials, titanium-based materials, tantalum-based materials, cobalt-based materials, tungsten-based materials, copper-based materials, silicon-based materials, alloys thereof, nitride-based materials, silicide-based materials, and / or carbide-based materials, and the like. The ICP source 146 is offset within the chamber, and the plasma generation density may vary across the processing volume 116. To provide a more uniform treatment of the substrate 106, the rotatable substrate support 104 is configured to uniformly expose the surface of the substrate 106 to inductively coupled plasma (see [link to documentation]). Figure 1 In some embodiments, the top shield 122 of the chamber body 102, the cathode assembly 184, the target 182, and the ICP source 146 may be arranged as follows: Figure 14 The alternative configuration and shape are depicted in view 1400. Fewer cathode assemblies 182A and 182B are used in conjunction with an ICP source 146 having an antenna 148.

[0037] The terms "bonding strength" or "adhesive strength" are used interchangeably throughout this document. Bonding or adhesive strength refers to the strength required for a solid to separate two linked materials. The two materials can be the same or different. Bonding can occur through direct physical contact, through chemical interaction, or by depositing one material onto another. Bonding other materials to, for example, glass with high adhesive strength is particularly difficult to achieve. For the sake of brevity, the examples in this document may be based on bonding metallic materials to glass, but bonding metallic materials to glass is not intended to be limiting. This method can be used to enhance the bonding strength between any type of materials or between materials of the same type.

[0038] The proliferation of mobile devices and the Internet of Things (IoT) has made RF communication increasingly demanding. More frequency bands with higher frequencies have been introduced, requiring minimal power loss even at frequencies in the gigahertz range. Significant efforts have been made to expand interposer technologies for 2.5D or 3D-IC stacking by utilizing glass substrates for advanced packaging. Glass offers adjustable dielectric constants, high resistivity, and low electrical losses, especially at high frequencies. Its relatively high stiffness and ability to adjust the coefficient of thermal expansion (CTE) allow for warpage management in glass core substrates used in glass-through applications. Glass forming processes allow for panel-like formation even at thicknesses as low as 100 μm. Currently available glass forming processes reduce or eliminate time-consuming and expensive thinning or polishing processes, making glass a cost-effective substrate. Glass is unaffected by conductive anodic filament (CAF) formation (a failure mechanism in organic core substrates).

[0039] Glass through-hole (TGV) / surface metallization is crucial for glass-core substrates and TGV applications. Prior to metallization, glass substrates allowed for glass through-holes in TGVs without the use of barrier layers or additional dielectric layers. However, the reliability of TGVs is problematic due to the weak adhesion between the adhesive layer and the glass. The inventors have discovered that delamination and peeling occur between the adhesive layer and the glass when subjected to stress accumulation, thermal cycling, and moisture erosion. Therefore, the inventors have found that increasing the adhesion strength to the glass is critical for accepting glass-core substrates in semiconductor manufacturing. Previous attempts to increase adhesion through surface roughening have resulted in signal integrity degradation. Further previous attempts to increase interfacial diffusion by subjecting the glass to temperatures of 450°C and above are incompatible with low thermal budget semiconductor devices. Adding additional adhesive layers leads to increased system complexity and higher costs, while reducing throughput.

[0040] This method provides enhanced bonding / adhesion strength between materials (such as, but not limited to, metals and glass) through interface processing. In the exemplary case of glass, a film / glass diffusion layer at the interface and film / glass bonding through surface activation are increased. The method may include promoting the diffusion layer and adding dangling bond formation on glass and other surfaces to enhance the film / glass or film / material adhesion of glass and other material metallization (e.g., a substrate core with TGV). The techniques of this method keep the interface roughness low, thereby providing high signal integrity and power integrity for packaged components. The resulting material interface produces high bonding / adhesion strength while having low roughness (e.g., Ra less than about 0.3 nm), low-temperature interface diffusion (e.g., temperatures less than about 150 degrees Celsius, without the use of high-temperature annealing for diffusion layer enhancement), and the ability to bond materials in a processing chamber (see [link to relevant documentation]). Figure 1 This allows for increased performance and throughput while remaining cost-effective.

[0041] In short, the surface preparation for bonding includes surface treatment for diffusion prior to the formation of the diffusion layer, the formation of the diffusion layer, diffusion enhancement treatment, and dangling bond formation treatment via surface activation. The method is compatible with many material types and substantially improves the reliability of, for example, on-substrate packaged devices (particularly glass-based devices for high-frequency communications and the like). Figure 3 Method 300 provides a process in which surfaces are prepared for bonding, whether the bonding is to a deposited layer (e.g., a seed layer in a TGV) or through contact or solid bonding between materials under compression (e.g., substrate-to-substrate bonding, grain-to-substrate bonding, grain-to-grain bonding, or bonding stacking of grains or substrates). Reference may be made in the discussion of method 300. Figure 4 In box 302, a first material 402 is provided, such as... Figure 4 Depicted in view 400A. The first material may be in the form of a substrate, grain, or other structure. In some embodiments, the first material may be glass or other dielectric materials. In some embodiments, in block 304, an optional degassing process may be performed on the first material to remove any moisture and the like. In block 306, if desired, an optional pre-cleaning process may also be performed on the first material to prepare a surface for bonding. The pre-cleaning process removes surface contaminants. In some embodiments, argon gas may be used to sputter the material surface to clean it. Degassing and pre-cleaning can be used to ensure that the material is free of moisture and particles before treating the surface for bonding.

[0042] In block 308, a first processing step 404 is performed on the first material 402 to treat the surface to facilitate the diffusion of a subsequently deposited diffusion layer using a first plasma process, such as... Figure 4Depicted in view 400B. The first processing step 404 reduces the surface density of the first material 402, thereby making the surface more conducive to accepting the diffusion layer without increasing the surface roughness of the first material 402. The first processing step 404 uses capacitively coupled plasma or a combination of capacitively coupled plasma and inductively coupled plasma (increasing plasma density to increase process speed). If using Figure 1 In the processing chamber 100, when ICP is incorporated into the first processing process 404, the substrate support 104 rotates 110 around the vertical axis 188 during the process to facilitate uniform processing above the surface. In some embodiments, the first plasma process uses argon-based gas, nitrogen-based gas, oxygen-based gas, hydrogen-based gas, or fluorocarbon-based gas. x F y ) gas, or nitrogen fluoride-based (NF) x ( ) gas to form plasma.

[0043] In some embodiments, the reactant gas is used to cause a chemical reaction with the surface of the first material to promote diffusion. In some embodiments, the first processing process 404 has a duration of approximately 30 seconds to approximately 5 minutes. In some embodiments, when using a CCP, the first processing process 404 can use a CCP power of approximately 1 kW to approximately 3 kW (e.g., Figure 1 The CCP power supply 144), and when using ICP, the first processing technology can use approximately 100 W to approximately 10 kW of ICP power ( Figure 1 The ICP power supply is 150. The CCP power uses a high-frequency RF power supply (e.g., Figure 1 A first RF power supply 140 is used to provide a high plasma density, and a low-frequency RF power supply (e.g., a second RF power supply 142) is used simultaneously to increase plasma energy for more efficient surface treatment. In some embodiments, the high-frequency RF power supply may operate at 60 MHz and the low-frequency RF power supply may operate at 2 MHz.

[0044] In frame 310, a first diffusion layer 406 is formed on the treated surface of the first material 402, such as Figure 4The first diffusion layer 406 is depicted in view 400C. It may be the same material as the subsequently deposited adhesive layer or a different material. The first diffusion layer 406 provides a transition between the first material 402 and subsequently deposited materials or other materials bonded to the first material. The first diffusion layer 406 may be formed using physical vapor deposition (PVD), atomic layer deposition (ALD), chemical vapor deposition (CVD), thermal evaporation, or electron beam evaporation processes and the like. In some embodiments, the first diffusion layer 406 is formed using a sputtering source power of approximately 5 kW to approximately 60 kW at a pressure of approximately 0.5 mTorr to approximately 10 mTorr (e.g., Figure 1 The first diffusion layer 406 is formed by a sputtering power supply 158. In some embodiments, the first diffusion layer 406 may be a molybdenum-based material, a titanium-based material, a tantalum-based material, a cobalt-based material, a tungsten-based material, a copper-based material, a silicon-based material, a nitride-based material, a silicide-based material, or a carbide-based material and the like.

[0045] In some embodiments, the first diffusion layer 406 introduces some metallic material into the first material 402, which may be primarily dielectric. In some embodiments, the first diffusion layer 406 may introduce a dielectric material into the first material 402, such as, but not limited to, silicon and the like. In some embodiments, the first diffusion layer 406 may have a thickness 414 of about 0.5 nm to about 5 nm. The thicker the first diffusion layer 406, the less second material can be deposited at a later point in a process such as that used for TGV. TGV has a limited volume in which a seed layer is to be formed. A thick diffusion layer will reduce the volume that the seed layer can occupy. In addition, a thick diffusion layer will be more likely to break internally than a thin diffusion layer, resulting in reduced bonding strength (the diffusion layer becomes a weak point in the bond). If the first diffusion layer 406 is too thin, the first diffusion layer 406 will not have sufficient thickness to cover the surface in order to achieve sufficient bonding strength at the bonding interface.

[0046] In block 312, a second processing step 408 is performed on the first material 402 to further diffuse the first diffusion layer 406 into the first material 402 using a second plasma process, such as... Figure 4Depicted in view 400D. The second processing process 408 uses capacitively coupled plasma. In some embodiments, the second plasma process uses an argon-based gas, a nitrogen-based gas, an oxygen-based gas, and / or a hydrogen-based gas to form the plasma. In some embodiments, the second processing process 408 has a duration of approximately 30 seconds to approximately 5 minutes. In some embodiments, the second processing process 408 can use a CCP power of approximately 1 kW to approximately 3 kW and a process pressure of approximately 0.5 mTorr to approximately 10 mTorr. The CCP power uses a high-frequency RF power supply to provide a high plasma density, and simultaneously uses a low-frequency RF power supply to increase plasma energy for more efficient surface treatment. In some embodiments, the high-frequency RF power supply can operate at 60 MHz and the low-frequency RF power supply can operate at 2 MHz. The second processing process 408 is used to further diffuse the first diffusion layer 406 into the first material.

[0047] In block 314, a third processing step 416 is performed on the first diffusion layer 406 to form dangling bonds 412 by surface activation using a third plasma process, such as... Figure 4 The third processing step 416 is depicted in views 400E and 400F. It uses inductively coupled plasma or inductively coupled plasma combined with capacitively coupled plasma. If using... Figure 1 In the processing chamber 100, the substrate support 104 rotates 110 around the vertical axis 188 during the third processing process 416 to promote uniform processing above the surface when inductively coupled plasma is used in the third processing process. In some embodiments, the third plasma process uses an argon-based gas, a nitrogen-based gas, an oxygen-based gas, or a hydrogen-based gas to form the plasma. In some embodiments, the third processing process 416 has a duration of approximately 30 seconds to approximately 5 minutes. In some embodiments, the third processing process 416 can use an ICP power of approximately 100 W to approximately 10 kW and a process pressure of approximately 0.5 mTorr to approximately 20 mTorr. The third processing process 416 is used to form dangling bonds 412, which have enhanced bonding strength with other similarly processed surfaces or subsequently formed adhesive layers. Furthermore, the third processing process 416 further reduces the water contact angle (WCA) of the surface after the second processing process 408 by approximately 5 degrees or more.

[0048] After the surface treatment process described above has been completed, the actual bonding can be used for the adhesive layer or another material. In frame 316, depending on the situation, the adhesive layer 504 can be formed on the first diffusion layer 402, such as... Figure 5As depicted in cross-sectional view 500, the adhesive layer 504 interacts with the dangling bonds 412 of the first diffusion layer 406 and forms a high-strength bond with the first diffusion layer 406 and subsequently with the underlying first material 402. In block 318, a second material 502 is then formed on the adhesive layer 504, thereby bonding the second material 502 to the adhesive layer 504 and subsequently to the underlying first material 402. Alternatively, block 318 can be performed after a surface treatment process has been completed, such as... Figure 6 The second material 502 is depicted in cross-sectional view 600. It is formed directly on the first diffusion layer 406 and interacts with the dangling bonds 412, thereby achieving a high-strength bond with the first diffusion layer 406 and subsequently with the underlying first material 402. In this case, for example, the first material 402 may be a glass substrate, and the second material 502 may be a metal seed layer for glass perforation. Figure 8 Cross-sectional view 800 depicts a TGV 804 having a glass core 802, which has undergone the aforementioned surface treatments to form a first diffusion layer 406 on the wall 806 of the TGV and an adhesive layer 504 on the first diffusion layer 406. A second material 502 has subsequently been deposited to fill the TGV. Figure 9 In cross-sectional view 900, a TGV 804 with a glass core 802 is depicted, which has undergone the above surface treatment to form a first diffusion layer 406 on the wall 806 of the TGV. A second material 502 has subsequently been deposited directly on the first diffusion layer 406 (without using an adhesive layer) to fill the TGV.

[0049] In alternative applications, after the above surface treatment process has been completed, bonding can be performed on another material that has undergone the same surface treatment to bond the first diffusion layer 406 to the second diffusion layer 406A, such as... Figure 7 The cross-sectional view 700 depicts this. In block 320, previous blocks 308 to 314 (at least blocks 302 to 306 are optional) are repeated on a third material 702 that may be different from or the same as the first material 402. In block 322, a first diffusion layer 406 of the first material 402 is directly bonded to a second diffusion layer 406A of the third material 702. In this case, for example, the first material 402 may be a substrate or a die, and the third material 702 may be a substrate or a die, such that substrate-to-substrate bonding (see, for example, view 1000A), die-to-substrate bonding (see, for example, view 1000B), or die-to-die bonding (see, for example, view 1000C) may occur in processes such as hybrid bonding and the like. Figure 10 The surface treatment process can also be extended to substrate stacking (see, for example, view 1000D) and die stacking (see, for example, view 1000E), as depicted in the corresponding views. Figure 10 As depicted in the corresponding view.

[0050] Parameters such as pressure, temperature, duration, and similar parameters for the above surface treatments (boxes 308 to 314) can be adjusted based on the material type and also on the diffusion layer material type. The above surface treatments (boxes 308 to 314) can be performed at temperatures below approximately 150 degrees Celsius, thus preserving the low thermal budget of the semiconductor structure. Figure 11 As depicted in Figure 1100, the surface treatment produces a bond strength approximately six times greater than that of conventional processes. Left axis 1102 represents the increase in bond strength in the upward direction. Bar 1104 represents the conventional bonding process. Bar 1106 represents the bonding process without the third processing step of this method. Bar 1108 represents the method of performing all three processing steps.

[0051] The method described in this article can be used in a specialized chamber (e.g., see [link]). Figure 1 It can be executed independently in a separate processing chamber or as part of a cluster tool, such as those described below. Figure 12 The described integration tool 1200 (i.e., clustering tool) and the following about Figure 13 The integration tool 1300 is described. An advantage of using the integration tool 1200 is the absence of vacuum disruption between chambers, and thus, degassing and pre-cleaning of the substrate are not required prior to processing within the chambers. For example, in some embodiments, the methods described above can be advantageously performed in the integration tool such that there is limited or no vacuum disruption between processes, thereby limiting or preventing substrate contamination, such as oxidation and the like. The integration tool 1200 includes a vacuum-sealed processing platform 1201, a factory interface 1204, and a system controller 1202. The processing platform 1201 includes multiple processing chambers, such as 1214A, 1213B, 1214C, 1214D, 1214E, and 1214F, operatively coupled to vacuum substrate transfer chambers (transfer chambers 1203A, 1203B). The factory interface 1204 is connected via one or more load-locked chambers (two load-locked chambers, such as...). Figure 12 The 1206A and 1206B shown are operatively coupled to the transfer chamber 1203A.

[0052] In some implementations, the factory interface 1204 includes at least one docking station 1207 and at least one factory interface robot 1238 to facilitate the transfer of semiconductor substrates. The docking station 1207 is configured to accept one or more front-opening unified pods (FOUPs). Four FOUPs (such as 1205A, 1205B, 1205C, and 1205D) are... Figure 12The embodiment is illustrated. A factory interface robot 1238 is configured to transfer a substrate from a factory interface 1204 to a processing platform 1201 via load-locking chambers (such as 1206A and 1206B). Each of the load-locking chambers 1206A and 1206B has a first port coupled to the factory interface 1204 and a second port coupled to a transfer chamber 1203A. The load-locking chambers 1206A and 1206B are coupled to a pressure control system (not shown) that evacuates and discharges the load-locking chambers 1206A and 1206B to facilitate substrate transfer between the vacuum environment of the transfer chamber 1203A and the substantially surrounding (e.g., atmospheric) environment of the factory interface 1204. The transfer chambers 1203A and 1203B have vacuum robots 1242A and 1242B disposed within the respective transfer chambers 1203A and 1203B. Vacuum robot 1242A is capable of transferring substrate 1221 between load locking chambers 1206A, 1206B, processing chambers 1214A and 1214F and cooling station 1240 or pre-cleaning station 1242. Vacuum robot 1242B is capable of transferring substrate 1221 between cooling station 1240 or pre-cleaning station 1242 and processing chambers 1214B, 1214C, 1214D and 1214E.

[0053] In some embodiments, processing chambers 1214A, 1214B, 1214C, 1214D, 1214E, and 1214F are coupled to transfer chambers 1203A and 1203B. Processing chambers 1214A, 1214B, 1214C, 1214D, 1214E, and 1214F may include, for example, atomic layer deposition (ALD) processing chambers, physical vapor deposition (PVD) processing chambers, chemical vapor deposition (CVD) chambers, annealing chambers, or the like. Chambers may include any chamber suitable for performing all or part of the methods described herein, as discussed above. In some embodiments, one or more optional service chambers (illustrated as 1216A and 1216B) may be coupled to transfer chamber 1203A. Service chambers 1216A and 1216B may be configured to perform other substrate processes, such as degassing, orientation, substrate metering, cooling, and the like.

[0054] System controller 1202 controls the operation of integration tool 1200 using direct control of processing chambers 1214A, 1214B, 1214C, 1214D, 1214E, and 1214F, or alternatively, by controlling a computer (or controller) associated with processing chambers 1214A, 1214B, 1214C, 1214D, 1214E, and 1214F and integration tool 1200. In operation, system controller 1202 collects and provides feedback on data from the respective chambers and systems to optimize the performance of integration tool 1200. System controller 1202 typically includes a central processing unit (CPU) 1230, memory 1234, and support circuitry 1232. CPU 1230 can be any type of general-purpose computer processor that can be used in industrial settings. Support circuitry 1232 is conventionally coupled to CPU 1230 and may include cache, frequency circuitry, input / output subsystem, power supply, and the like. Software routines (such as those described above) may be stored in memory 1234 and, when executed by CPU 1230, transform CPU 1230 into a dedicated computer (system controller) 1202. Software routines may also be stored and / or executed via a second controller (not shown) located remotely to integration tool 1200.

[0055] Another integration tool 1300 can also be used to perform the methods described above. In some embodiments, the integration tool 1300 may be configured to handle workpieces including high aspect ratio features, such as high aspect ratio through-holes (e.g., TGVs), as described herein. In some embodiments, the workpiece may be a substrate, such as a silicon or other semiconductor substrate, or an interposer, such as an interposer containing organic materials or a glass interposer and the like. The integration tool 1300 typically includes an EFEM 1302 for loading workpieces into the integration tool 1300, a first load-locking chamber 1304 coupled to the EFEM 1302, a transfer chamber 1306 coupled to the first load-locking chamber 1304, and a plurality of other chambers coupled to the transfer chamber 1306, as described in detail below. The EFEM 1302 typically includes one or more robots 1305 configured to transfer workpieces from FOUP 1303 to at least one of the first load-locking chamber 1304 or the second load-locking chamber 1320.

[0056] Advancing counterclockwise from the first load-locking chamber 1304 around the transfer chamber 1306, the integration tool 1300 includes a first dedicated degassing chamber 1308, a first pre-cleaning chamber 1310, a first deposition chamber 1312, a second pre-cleaning chamber 1314, a second deposition chamber 1316, a second dedicated degassing chamber 1318, and a second load-locking chamber 1320. In some embodiments, the first deposition chamber 1312 and the second deposition chamber 1316 may be PVD chambers, ALD chambers, CVD chambers, or combinations thereof. In some embodiments, the transfer chamber 1306 and each chamber coupled to the transfer chamber 1306 are maintained under vacuum. As used herein, the term “vacuum” may refer to a pressure less than 760 Torr and will typically be maintained at around 10. -5 Torr (i.e., ~10) -3 Under pressures of (Pa). However, some high vacuum systems can operate at pressures below nearly 10 Pa. -7 Torr (that is, ~10) -5 The operation is carried out at a pressure of 100 Pa. In some embodiments, the vacuum is generated using a roughing pump and / or a turbomolecular pump coupled to each of the transfer chamber 1306 and one or more processing chambers (e.g., processing chambers 1308 to 1318). However, other types of vacuum pumps may also be used.

[0057] System controller 1328 controls the operation of integration tool 1300 using direct control of the processing chamber or alternatively by controlling a computer (or controller) associated with the processing chamber and integration tool 1300. In operation, system controller 1328 collects and provides feedback on data from the respective chambers and systems to optimize the performance of integration tool 1300. System controller 1328 typically includes a central processing unit (CPU) 1330, memory 1334, and support circuitry 1332. CPU 1330 can be any type of general-purpose computer processor that can be used in industrial settings. Support circuitry 1332 is conventionally coupled to CPU 1330 and may include cache, frequency circuitry, input / output subsystems, power supplies, and the like. Software routines (such as those described above) can be stored in memory 1334 and, when executed by CPU 1330, transform CPU 1330 into a dedicated computer (system controller) 1328. Software routines can also be stored and / or executed via a second controller (not shown) located remotely to the integration tool 1300.

[0058] Implementations based on this principle can be carried out in hardware, firmware, software, or any combination thereof. Implementations can also be carried out using instructions stored on one or more computer-readable media, which can be read and executed by one or more processors. The computer-readable media may include any mechanism for storing or transmitting information in a machine-readable form (e.g., a computing platform or a “virtual machine” running on one or more computing platforms). For example, the computer-readable media may include any suitable form of volatile or non-volatile memory. In some implementations, the computer-readable media may include non-transitory computer-readable media.

[0059] Although the foregoing relates to implementations of this principle, other and further implementations of this principle may be designed without departing from its basic scope.

Claims

1. A method for preparing surfaces for bonding, the method comprising: A first processing step is performed on a first surface of a first material using a first plasma process to facilitate diffusion into the first surface of a first diffusion layer deposited after the first processing step, wherein the first processing step uses capacitively coupled plasma (CCP) or a combination of both CCP and inductively coupled plasma (ICP). The first diffusion layer is formed on the first surface of the first material; A second processing step is performed using a second plasma process to increase the diffusion of the first diffusion layer to the first surface of the first material, wherein the second processing step uses CCP; as well as A third processing step is performed using a third plasma process to form dangling bonds on the first diffusion layer, wherein the third processing step uses ICP or a combination of both CCP and ICP.

2. The method of claim 1, wherein the first material is rotated about a vertical axis during the first processing or the third processing.

3. The method of claim 1, wherein the first plasma process uses an argon-based gas, a nitrogen-based gas, an oxygen-based gas, a hydrogen-based gas, or a fluorocarbon-based gas (C). x F y ) gas, or nitrogen fluoride-based (NF) x ( ) gas to form plasma.

4. The method of claim 1, wherein the first processing step, the second processing step, and the third processing step each have a duration of approximately 30 seconds to approximately 5 minutes.

5. The method of claim 1, wherein when using CCP, the first processing process uses a power of about 1 kW to about 3 kW and when using ICP, the first processing process uses a source power of about 100 W to about 10 kW.

6. The method of claim 1, wherein the first diffusion layer is formed using a physical vapor deposition (PVD) process, an atomic layer deposition (ALD) process, a chemical vapor deposition (CVD) process, a thermal evaporation process, or an electron beam evaporation process.

7. The method of claim 1, wherein the first diffusion layer is formed using a source power of approximately 5 kW to approximately 60 kW at a pressure of approximately 0.5 mTorr to approximately 10 mTorr.

8. The method of claim 1, wherein the first diffusion layer has a thickness of about 0.5 nm to about 5 nm.

9. The method of claim 1, wherein the first diffusion layer is based on a molybdenum-based material, a titanium-based material, a tantalum-based material, a cobalt-based material, a tungsten-based material, a copper-based material, a silicon-based material, a nitride-based material, a silicide-based material, or a carbide-based material.

10. The method of claim 1, wherein the second plasma process or the third plasma process uses an argon-based gas, a nitrogen-based gas, an oxygen-based gas, or a hydrogen-based gas to form a plasma.

11. The method of claim 1, wherein the second processing step is performed at a pressure of about 0.5 mTorr to about 10 mTorr and a power of about 1 kW to about 3 kW.

12. The method of claim 1, wherein the third processing is performed at a pressure of about 0.5 mTorr to about 20 mTorr and a source power of about 100 W to about 10 kW.

13. The method of claim 1, wherein the seed layer is formed directly on the first diffusion layer after the third processing step, or the adhesive layer is first formed on the first diffusion layer after the third processing step and then the seed layer is formed on the adhesive layer.

14. The method of claim 1, wherein the method is repeated on a second material, the second material being subsequently bonded to the first material, wherein the first diffusion layer and the second diffusion layer on the second material serve as an intermediate layer between the bonded first material and the second material.

15. The method of claim 1, wherein the first material is a glass-based material and the method is performed on a glass-based perforation (TGV) in the glass-based material.

16. The method of claim 1, wherein the method is performed in a single processing chamber, an integrated tool, or a separate processing chamber.

17. A non-transitory computer-readable medium having instructions stored thereon, which, when executed, cause to perform a method for preparing surfaces for bonding, the method comprising: A first plasma process is used to perform a first treatment process on the surface of a material to facilitate diffusion into the surface of a diffusion layer deposited after the first treatment process, wherein the first treatment process uses capacitively coupled plasma (CCP) or a combination of both CCP and inductively coupled plasma (ICP). The diffusion layer is formed on the surface of the material; A second processing step is performed using a second plasma process to increase the diffusion layer to the surface of the material, wherein the second processing step uses CCP; as well as A third processing step is performed using a third plasma process to form dangling bonds on the diffusion layer, wherein the third processing step uses ICP or a combination of both CCP and ICP.

18. The non-transitory computer-readable medium of claim 17, wherein the method further comprises at least one of a, b, c, d, e, f, g, or h. a) The first plasma process described herein uses argon-based gas, nitrogen-based gas, oxygen-based gas, hydrogen-based gas, or fluorocarbon-based gas (C). x F y ) gas, or nitrogen fluoride-based (NF) x () gas to form plasma; b) wherein the first processing step, the second processing step, and the third processing step each have a duration of approximately 30 seconds to approximately 5 minutes; c) Wherein, when using CCP, the first processing technology uses a power of approximately 1 kW to approximately 3 kW and when using ICP, the first processing technology uses a source power of approximately 100 W to approximately 10 kW; d) The diffusion layer has a thickness of approximately 0.5 nm to approximately 5 nm; e) wherein the diffusion layer is based on a molybdenum-based material, a titanium-based material, a tantalum-based material, a cobalt-based material, a tungsten-based material, a copper-based material, a silicon-based material, a nitride-based material, a silicide-based material, or a carbide-based material; f) wherein the second plasma process or the third plasma process uses an argon-based gas, a nitrogen-based gas, an oxygen-based gas, or a hydrogen-based gas to form plasma; g) The second processing step is performed at a pressure of about 0.5 mTorr to about 10 mTorr and a power of about 1 kW to about 3 kW. or h) The third processing step is performed at a pressure of about 0.5 mTorr to about 20 mTorr and a source power of about 100 W to about 10 kW.

19. An apparatus for preparing surfaces for bonding, the apparatus comprising: The processing chamber has a rotating substrate support, a processing volume enclosed by an upper processing shield, a lower processing shield, and a rotatable top shield, and a plurality of selectable cathodes positioned on the top of the processing chamber, the cathodes being connected to a dual-power target power supply configured to generate RF power, DC power, or pulsed DC power, one of the cathode locations being an inductively coupled plasma (ICP) source. A dual-power power supply, electrically connected to electrodes in the rotating substrate support and configured to generate capacitively coupled plasma (CCP) within the processing volume; The ICP source includes a dual-power power supply electrically connected to the antenna and configured to generate ICP within the processing volume, wherein the antenna is positioned to be exposed to the processing volume by rotating the rotatable top shield. as well as A controller configured to rotate the rotatable top shield to select either a cathode for deposition or an antenna for ICP generation within the processing volume and to execute a method comprising: A first treatment process is performed on the surface of a material using a first plasma process to facilitate diffusion into the surface of a diffusion layer deposited after the first treatment process, wherein the first treatment process uses CCP or a combination of both CCP and ICP. The diffusion layer is formed on the surface of the material; A second processing step is performed using a second plasma process to increase the diffusion layer to the surface of the material, wherein the second processing step uses CCP; and A third processing step is performed using a third plasma process to form dangling bonds on the diffusion layer, wherein the third processing step uses ICP or a combination of both CCP and ICP.

20. The device of claim 19, wherein the dual power supply comprises a first RF power supply having a first frequency of approximately 60 MHz and a second RF power supply having a second frequency of approximately 2 MHz.

Images