Method of manufacturing device using multi-layer stack

By adopting three-layer structure process and undercut profile technology in quantum computing device manufacturing, the problem of photoresist layer hardening and dry etching erosion is solved, and purer metallic layer edges and higher device performance are achieved.

CN118946242APending Publication Date: 2024-11-12GOOGLE LLC
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410868162.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2017-08-31
Filing Date
2017-12-07
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

Prior Art In the manufacturing of quantum computing devices, the photoresist layer is hardened during dry etching, resulting in residue remaining at the edge of the metallic layer, affecting the uniform formation and electromagnetic characteristics of the subsequent layer, and the dry etching erosion leads to the material. Deposition affects device performance.

Method used

Using a three-layer structure process, first deposit the protective layer, then deposit the support layer and photoresist layer. The undercut contour technology prevents direct contact between the photoresist and the metallic layer, and uses protective layer to prevent hardening during the etching process. , improve edge protection.

Benefits of technology

Effectively avoid the hardening of photoresist residues on the edge of the metallic layer, improve the purity of the patterned edges and device performance, reduce the loss caused by dielectric residues, improve the coherence time of the qubit and the quality of the resonator. Factor.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118946242A_ABST
    Figure CN118946242A_ABST
Patent Text Reader

Abstract

A method of manufacturing a device is presented. The method includes forming a multilayer stack (101 ', 102', 103 ') on a substrate (10', 100 ') having a major surface. The multi-layer stack comprises a support layer (102 ') formed over the main surface of the substrate and a photoresist layer (103') formed on the support layer; patterning the multilayer stack to form at least one opening such that the photoresist layer is undercut by the support layer; and dry etching the substrate anisotropically.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] This application is a divisional application of Chinese invention patent application 201780094508.4, whose international application date is December 7, 2017, the date of entering the Chinese national phase is March 2, 2020, and the invention name is “Method of manufacturing devices using multilayer stacking”. Technical Field

[0002] The present disclosure relates to methods of manufacturing devices using multilayer stacks. Background Art

[0003] Quantum computing is a new method of computing that exploits quantum mechanical phenomena such as the superposition of two quantum states and the entanglement between quantum states belonging to separate and distant entities. Compared to digital computers that use "bits" configured to be in two bistable states (e.g., "0" and "1") to store and manipulate information, quantum computing systems aim to manipulate information using "qubits" configured in a superposition of quantum states (e.g., a|0>+b|1>). The quantum states of each qubit can be entangled with each other, that is, the measurement results of one qubit are closely related to the measurement results of another qubit. These characteristics provide a key advantage over classical computers, namely that the speed of quantum computers is exponentially related to the number of qubits. Summary of the invention

[0004] Generally, in some aspects, the subject matter of the present disclosure can be embodied in a method of manufacturing a device, the method comprising forming a multilayer stack on a substrate having a main surface, the multilayer stack comprising a support layer formed above the main surface of the substrate, and a photoresist layer formed on the support layer; the method of manufacturing the device also includes: patterning the multilayer stack to form at least one opening so that the photoresist layer is undercut by the support layer; and anisotropically dry etching the substrate.

[0005] Implementations of the method may include one or more of the following features. For example, in some implementations, the multilayer stack may further include a first layer formed on a major surface of the substrate, the support layer formed on the first layer.

[0006] In some implementations, a support layer can be formed on the first layer.

[0007] In some implementations, the support layer may be overcut by the first layer.

[0008] In some implementations, the photoresist layer and the first layer can have respective edges that are aligned.

[0009] In some implementations, the first layer is a polymeric material.

[0010] In some implementations, the polymeric material can be sensitive to electron beam radiation.

[0011] In some implementations, the polymeric material can be methyl methacrylate.

[0012] In some embodiments, the polymeric material may be polymethyl methacrylate.

[0013] In some embodiments, the polymeric material may be methacrylic acid.

[0014] In some embodiments, the multi-layer stack may be a double-layer stack.

[0015] In some implementations, patterning the multilayer resist stack can include developing the photoresist layer using a developer.

[0016] In some implementations, the support layer is soluble in the developer.

[0017] In some implementations, the developer can be a base.

[0018] In some implementations, the support layer is a polymeric material.

[0019] In some implementations, the polymeric material can be sensitive to electron beam radiation.

[0020] In some implementations, the polymeric material can be polydimethylglutarimide (PMGI).

[0021] In some implementations, the polymeric material can be methyl methacrylate.

[0022] In some implementations, the polymeric material can be polymethyl methacrylate.

[0023] In some implementations, the polymeric material can be methacrylic acid.

[0024] In some implementations, patterning the multilayer stack can include performing plasma ashing.

[0025] In some implementations, the device may be a quantum information processing device.

[0026] In some implementations, the substrate can include an aluminum layer disposed on a silicon substrate.

[0027] In some implementations, the device may be a device obtainable by any of the above methods or implementations.

[0028] Implementations may include one or more of the following advantages. For example, in some implementations, the method may provide another layer of protection between the metallic layer to be patterned and the support layer described above. This may improve the protection around the edge of the patterned feature by providing a barrier to residues and allowing the solvent for the removal step to more easily enter.

[0029] For the purpose of this disclosure, a superconductor (or, superconducting) material may be understood as a material that exhibits superconducting properties at or below a superconducting critical temperature. Examples of superconductor materials include aluminum (e.g., a superconducting critical temperature of 1.2 Kelvin), niobium (e.g., a superconducting critical temperature of 9.3 Kelvin), and titanium nitride (e.g., a superconducting critical temperature of 5.6 Kelvin).

[0030] The details of one or more implementations are set forth in the accompanying drawings and the description below.Other features and advantages will be apparent from the description, drawings, and claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figures 1A-1H 1J-1K show a process for patterning a metallic layer on a substrate using a three-layer structure.

[0032] Figure 2 is a process flow diagram of a method for patterning a metallic layer on a substrate.

[0033] Figure 3A and Figure 3B A two-layer structure is shown before and after development, which can be used as an alternative to a three-layer structure. DETAILED DESCRIPTION

[0034] Quantum computing provides coherent processing of quantum information stored in multiple qubits of a quantum computer. In order to achieve maximum computing speed, ideally, the qubits are entangled with each other in a controllable manner so that the quantum state of each qubit immediately affects the corresponding quantum state of other qubits. Superconducting quantum computing is a promising implementation of quantum computing technology, in which quantum circuits are at least partially formed of superconductor materials on a substrate. In certain implementations, quantum circuit elements are preferably operated at low temperatures so that thermal fluctuations do not interfere with the coherence of the qubits or cause losses in circuit elements at the operating frequency. Other factors that may cause loss or decoherence are material defects, such as two-level states (TLS) and undesirable radiation coupling.

[0035] Therefore, metallic layers that exhibit superconducting behavior below a critical temperature can be used to form various quantum circuit elements and components, such as Josephson junctions, coplanar waveguides, LC oscillators, superconducting quantum interference devices (SQUIDs), inductors, capacitors, or other elements and components. Superconducting quantum computing devices can be multi-layer systems, but typically only the first metallic layer forms the core of the computing basis.

[0036] Routinely, for the metallic layer on the patterned substrate, a photoresist layer is deposited on the metallic layer and developed. These steps expose the part of the metallic layer to be etched. Dry etching such as ion milling can be used to etch the exposed area of ​​the metallic layer. Once the metallic layer has been etched, dry etching can also be used to etch the substrate below. After the dry etching, the photoresist layer can be removed with a solvent.

[0037] The disadvantage of this method is that, due to exposure to the etchant, the patterned photoresist can harden during the dry etching step. As a result, even after removing the photoresist layer with a solvent, the part of the hardened photoresist can also be retained on the metallic layer. These photoresist residues are particularly problematic around the edge of the patterned feature, where the hardened photoresist directly contacts the metallic layer. The residues remaining around the edge of the metallic feature can prevent the uniform formation of subsequent layers on the patterned metal structure, which is crucial for the manufacture of quantum information devices. They may also affect the electromagnetic properties of the patterned metal structure.

[0038] To solve this problem, an undercut profile is used. A support layer (also referred to herein as an "undercut layer") is deposited on the metallic layer, and a photoresist layer can be deposited on the support layer. The support layer is characterized in that, when developed, the support layer recedes laterally from the inner wall of the window formed by photolithography. Thus, the photoresist layer and the support layer form a hanging feature (or "undercut profile"). The materials used for the support layer are commercially available, such as polydimethylglutarimide (PMGI), LOR, and the like. TM and LOL TM .

[0039] Using a support layer, when the metal / substrate is dry etched, the exposed portion of the photoresist layer (i.e., the portion exposed to etching and hardened) is not in direct contact with the metallic layer. Therefore, using a support layer can alleviate the problem of hardened photoresist residues.

[0040] However, due to two reasons, this method still can not provide perfect protection to metallic layer.First, although ion milling and other forms of dry etching can be anisotropic (in other words, directionally vertical in etching), ions can still bounce back in patterned window by reflecting from the wall limited by photoresist layer and supporting layer.Therefore, the supporting layer itself contacted with the metallic layer can harden.Secondly, dry etching, particularly ion milling, is a kind of aggressive process, because the material from metallic layer and substrate layer ejects from substrate, and is deposited around the exposed edge of the window formed in the metallic layer.

[0041] The present disclosure relates to solving this problem, and proposes a method: the metallic layer on the patterned substrate simultaneously blocks or even avoids the dielectric / metallic residue near the edge of the patterned metallic layer. Specifically, the method provides another layer (here referred to as "first layer", "base layer" or "protective layer") for protection between the metallic layer to be patterned and the supporting layer described above. By providing a barrier to residue, and allowing the solvent for the removal step to enter more easily, this improves the protection around the edge of the patterned feature. Especially for quantum information devices, the purity level reached by this method will have a significant impact. In some implementations, the dielectric residue contacted with the metallic structure causes a loss at the operating frequency of several GHz, which affects the key parameters of the device, such as the coherence time of a quantum bit or the quality factor of a resonator.

[0042] Figures 1A-1K An example of a process for photolithographically patterning a metallic layer on a substrate is shown. Specifically, Figures 1A-1K The example process depicted in FIG. 1 shows a cross-sectional view of forming three metallic regions from a metallic layer deposited on a substrate by forming two trenches. Figure 2 Describe this method.

[0043] Reference Figure 1A , metallic layer 100 can be deposited on substrate 10.Metallic layer 100 may not need any tackifier to adhere to substrate 10, and a stable layer may be provided on substrate 10.Metallic layer 100 may be in the form of an aluminum layer, and substrate 10 may be in the form of silicon, such as, for example, p-type (100) silicon.The combination of aluminum and silicon may be used to realize a quantum information processing device including a superconducting qubit.However, considering that the current method may be used for any other metallic layer and substrate, the selection of material is not limited to these.For example, substrate 10 may be a sapphire wafer.When metallic layer 100 is an aluminum layer, metallic layer 100 may be approximately 100nm thick and may be polycrystalline.

[0044] Reference Figure 1B and Figure 2, an optional first layer (or "protective layer") 101 can be deposited on the metallic layer 100 (step S1). As an example, the protective layer 101 can be in the form of a poly (methyl methacrylate) (PMMA) layer. In order to form the protective layer or protective layer 101, PMMA can be dissolved in a solvent such as anisole and spin-coated on a substrate including the silicon substrate 10 and the metallic layer 100. After spin coating, the protective layer 101 can be baked to harden before depositing the undercut layer or support layer 102.

[0045] The support layer 102 is deposited on the protection layer 101 (if present) or on the metallic layer 100 (step S2).

[0046] As described above, examples of the support layer 102 may include PMGI, LOL TM and LOR TM When the protective layer 101 is a PMMA layer, these examples do not require any adhesion promoter to adhere to the protective layer 101. The support layer 102 is also spin-coated and baked above the glass transition temperature of about 185 degrees to evaporate the solvent in the layer. Alternatively, the support layer 102 can be baked at a temperature below the glass transition temperature (such as 160 degrees) for 5 minutes without causing significant mixing with the support layer 102 and the protective layer 101.

[0047] A photoresist layer 103 is deposited on the support layer 102 (step S3 ).

[0048] The thickness of the protective layer 101 may be about 100 nm and may be between 10 nm and 1 μm. In the case of using PMGI, the thickness of the support layer 102 may be about 200 nm and may be between 10 nm and 10 μm, regardless of the resin composition of the support layer 102. The thickness of the photoresist layer 103 may typically be between 800 nm and 7 μm. However, depending on process parameters, such as the extent of reactive ion etching into the silicon substrate 10, the thickness of the photoresist layer 103 may be in the range of 100 nm to 100 μm.

[0049] Reference Figure 1C and Figure 2 , using a photomask 104, the photoresist layer 103 and the support layer 102 are exposed (step S4). The spectral response of these two layers is near the wavelength range of 240 to 290nm (which is within the deep UV (DUV) range). Therefore, both the photoresist layer 103 and the support layer 102 can be exposed with DUV full-sheet exposure at the same time. Alternatively, the photoresist layer 103 can be completely patterned, and only the exposed portion of the support layer 102 can be exposed with DUV full-sheet exposure. Alternatively, the two layers 102, 103 can be exposed with an electron beam. For example, PMGI at 1 to 5 joules / cm2 of DUV2 range, and the electron beam dose is 30 to 40 μC / cm 2 The dose used for PMGI is 1 / 5 to 1 / 3 of the dose used for PMMA. Taking these parameters into account, the exposure dose can be optimized and determined accordingly.

[0050] Figure 1C An example is shown in which the photoresist layer 103 is a positive photoresist. However, a negative photoresist may be used for the photoresist layer. Support layer 102, PMGI, LOR TM and LOL TM Can be positive photoresist and need aqueous solvent.When photoresist layer 103 is positive photoresist, near the interface between photoresist and PMGI, slight degree of mixing can occur, because although PMGI is insoluble in most conventional photoresist solvents, the solvent cyclopentanone that is usually used for PMGI dissolves most materials used as photoresist layer.However, thoroughly baking the supporting layer can help reduce mixing.The example of positive photoresist includes the mixture of diazonaphthoquinone (DNQ) and novolac resin (phenol formaldehyde resin).The example of negative photoresist includes SU-8 photoresist polymer and non-stoichiometric thiol-ene (OSTE) polymer.

[0051] The material used for the photoresist layer 103 is generally designed to provide a non-isotropic slope near the edge for clean stripping after development. The material used for the support layer 102 is generally designed to provide an isotropic slope and a controllable degree of undercut or depression after development. Therefore, the materials used for the support layer 102 and the photoresist layer 103 cannot be used interchangeably. However, if the degree of undercut or depression can be reproducibly controlled and does not cause a large amount of interlayer mixing with the subsequent photoresist layer 103, any material that can be used as the photoresist layer 103 can be used as the support layer 102.

[0052] In this step, the protective layer 101, such as a PMMA layer, may also be exposed to a certain extent. However, if the protective layer 101 is a negative photoresist, the protective layer 101 tends to be insoluble in an aqueous developer when both the photoresist layer and the support layer are positive photoresists. Even if the protective layer 101 is a positive resist and is dissolved by the developer, the thickness and exposure dose of the protective layer 101 can be controlled so that mainly only the top two layers are exposed. In addition, when the protective layer 101 is a PMMA layer, the dose of DUV light that causes PMMA cross-linking is substantially greater than the dose required to expose the support layer 102 and the photoresist layer 103. Therefore, the effect of DUV full-sheet exposure on the protective layer 101 can be ignored.

[0053] Also refer to Figure 1D and Figure 2 , if the photoresist layer 103 is of a type that can be developed with a developer (such as TMAH or KOH solution) used to develop the support layer 102, the top two layers 102, 103, i.e., the photoresist layer 103 and the support layer 102, can be developed simultaneously. Alternatively, the photoresist layer 103 and the support layer 102 can be developed using two separate development steps. Due to the presence of the protective layer 101, the metallic layer 100 may not be exposed to the developer during the development stage (step S5).

[0054] At this stage, the patterned support layer 102' can be set back from the sidewalls defined by the patterned photoresist layer 103', thereby forming a hanging feature together with the photoresist. The extent of undercut can be reproducibly controlled by the degree of exposure. The extent of undercut in the support layer 102 can also be controlled by the baking temperature, the type of developer, and the resin composition within the support layer 102.

[0055] One of the parameters that can be used to control the degree of undercutting is the UV exposure. For example, to maximize undercutting, Figure 1C In step S4 shown, after the support layer 102 may be formed and baked, the entire support layer 102 may be flood exposed using DUV flood exposure before depositing the photoresist layer 103. The support layer 102 and the photoresist layer 103 may be developed simultaneously. This is an exemplary process for maximizing the exposure directly on the support layer 102 to maximize the degree of undercut.

[0056] Alternatively, after the support layer 102 is deposited and baked, the photoresist layer 103 may be deposited and baked. When the photoresist layer 103 is exposed to image, the support layer 102 may be exposed at the same time, such as in Figure 1C As explained in step S4 shown in the figure. In this case, the support layer 102 is exposed to a smaller extent than in the example where the entire support layer 102 is exposed, so the extent of undercutting will be smaller.

[0057] As another option, after the support layer 102 is deposited and baked, the photoresist layer 103 may be deposited, baked, exposed, and developed. After the photoresist layer 103 is completely patterned, the exposed portion of the support layer 102 may be additionally exposed with a DUV full-sheet exposure to control the extent of the undercut. The exposed area of ​​the support layer 102 is developed at a higher rate than the unexposed area of ​​the support layer 102, thereby forming an undercut of the support layer 102 that recedes toward the unexposed area of ​​the support layer 102.

[0058] To keep the degree of undercutting to a minimum, DUV light exposure on the support layer 102 can be eliminated. In other words, when exposing the photoresist layer 103, the dose and thickness of the photoresist layer 103 can be controlled so that the support layer 102 is not exposed to any DUV light. Even in this case, a certain amount of undercutting is expected.

[0059] The resin composition of the support layer 102 is also an important factor in reproducibly controlling the degree of undercut. For example, when using a particular type of support layer 102 such as NANO LOR resist, a very large degree of undercut can be achieved without any DUV flood exposure.

[0060] The degree of undercutting can also be determined by the baking time and temperature, because the amount of residual solvent in the support layer 102 determines the dissolution rate in the development stage. For example, in the case where the support layer 102 is PMGI, the baking temperature should be 185° C. or higher up to about 250° C. to remove any residual solvent in the support layer 102. However, when the support layer 102 is baked at a temperature below 185° C., more solvent will remain in the support layer 102. This results in a greater degree of undercutting due to the higher dissolution rate.

[0061] Another parameter that can be used to control the degree of undercutting is the developer type.Depending on the developer type, the dissolution rate of the support layer 102 can vary.

[0062] Therefore, the degree of undercutting can be reproducibly controlled by optimizing the following four parameters: degree of DUV flood exposure, bake temperature, resin composition and developer type.The dissolution rate of the support layer 102 and thus the degree of undercutting can be primarily controlled by the bake temperature.

[0063] Also refer to Figure 1E and Figure 1F as well as Figure 2 , protective layer 101 can be etched by reactive ion etching (RIE). If protective layer 101 is made up of an organic material such as PMMA, oxygen plasma can be used to selectively remove the organic protective layer without etching into metallic layer 100 (step S6). Because protective layer 101 can be etched so that the patterned edge of patterned protective layer 101' is aligned with the patterned edge of patterned photoresist layer 103', a high degree of etching directionality or anisotropy can be expected. This feature can provide additional protection from pseudo residues for the edge of the feature in metallic layer 100. For anisotropic reactive ion etching, an inductively coupled plasma (ICP) system equipped with an asymmetric configuration of electrodes can be used in the case of low-pressure oxygen in the range of 10 mTorr.

[0064] Even when the reactive ion etching is arranged to be anisotropic, the protective layer 101, especially when it is a PMMA layer, can retreat slightly, also resulting in an undercut structure. The degree of the depression or the degree of the undercut can be controlled by the pressure of the oxygen. The depression can be between a few nanometers and a few hundred nanometers.

[0065] Reference Figure 1G and Figure 2 , metallic layer 100 can be dry etched by ion milling, so that the pattern of patterned protective layer 101 'can be transferred to the metallic layer 100 below (step S7). Non-reactive gases such as argon (Ar) can be used for ion milling. Because the ion milling process is not selective, and involves the bombardment of heavy ions, once the metallic layer is etched, the substrate can be etched. Especially when the metallic layer 100 is deposited using a sputtering method, a considerable amount of debris from the metallic layer 100 can accumulate at the edge of the patterned feature. However, the method for etching in this step is not limited to dry etching such as ion milling. As long as the pattern defined by the photoresist layer 103 'and the protective layer 101 'can be transferred to the metallic layer 100 below in a substantially anisotropic manner, any other etching method can be used to be arranged to provide this step of anisotropic etching, such as reactive ion etching with inductively coupled plasma.

[0066] Advantages of the protective layer 101 may be as follows.

[0067] First, the presence of the protection layer 101 can prevent a continuous hardened film from being formed on the top of the metallic layer 100 , and make the removal of the photoresist layer 103 easier because the solvent can easily dissolve the undamaged portion of the photoresist layer 103 .

[0068] Secondly, although heavy ions reflected from the walls in the patterned holes can also harden the protective layer 101, the thickness of the protective layer 101 can be kept small, for example 100 nm for a PMMA layer. Therefore, only a small portion of the sidewalls exposed in the pattern can be hardened. This is in contrast to the typical thickness of the photoresist layer 103, which is about 1 μm.

[0069] Third, as regards Figure 1E and Figure 1F And step S6 discussed ground, the reactive ion etching step of removing protective layer 101 makes the sidewall of protective layer 101 retreat slightly.This recessed degree can be controlled so that protective layer 101 is not seriously exposed to the bombardment of heavy ions and protects edge from pseudo-particle influence simultaneously.Therefore, protective layer 101 can protect the edge of patterned feature, and the hardened part of protective layer 101 is not contacted with metallic layer 100.

[0070] Also refer to Figure 1H and Figure 2 , the photoresist layer 103' bombarded by heavy ions causes hardened photoresist regions 104 (or " covering ") at the exposed surface of patterned photoresist layer 103'. However, these regions 104 are spaced apart from patterned metallic layer 100' by patterned supporting layer 102' and patterned protective layer 101' (if present). And the edge of the feature in patterned metallic layer 100' is protected by protective layer 101. This is in contrast to the situation without protective layer, in which the retreating part of supporting layer exposes the edge of the feature in metallic layer 100. Therefore, the edge of the feature in patterned metallic layer 100' is not exposed to possible dielectric residue or metal residue.

[0071] Reference Figure 1J and Figure 2 , the patterned protective layer 101 ', the patterned support layer 102 ' and the patterned photoresist layer 103 ' can be removed using a solvent (step S8). The use of a three-layer structure enables the solvent to more easily enter the resist layer and facilitates a cleaner stripping of the layer. A suitable stripping method can be used, such as isotropic oxygen plasma. Heated N-methyl-2-pyrrolidone (NMP) can be used for this step.

[0072] Figure 1K A patterned metallic layer 100 ′ is shown following the above steps. The metallic layer 100 may be less susceptible to hardened photoresist residues or particles from the metallic layer 100 and the substrate 10 .

[0073] Reference Figure 3A As mentioned above, a two-layer structure can be used instead of a three-layer structure. The support layer 102 can be directly deposited on the metallic layer 100. The photoresist layer 103 can be deposited on the support layer 102. In this case, a two-layer structure 102, 103 is formed instead of Figure 1B The three-layer structure 101, 102, 103 shown. The two-layer structure can be formed using the same materials and the same process as those described above, so it will not be described in detail here. However, process parameters such as layer thickness and / or baking time can be adapted to adapt to the two-layer structure. After being formed, the two-layer structure can be processed using the same or similar process as used to process the three-layer structure. However, process parameters such as dose and / or development time can be adapted to adapt to the two-layer structure. Compared with the three-layer structure, there may be significant mixing between the protective layer 101 and the photoresist layer 103 without a supporting layer 102 between the protective layer 101 and the photoresist layer 103.

[0074] Reference Figure 3B ,from Figure 3AStarting from the two-layer structure 102, 103 shown, the above reference Figures 1A to 1K and Figure 2 After the oxygen plasma treatment step (step S6) and before the dry etching step (step S7), the patterned support layer 102' and the patterned photoresist layer 103' are formed as shown in FIG. Figure 3B The two-layer undercut profile shown, instead of Figure 1F In this case, the edges of the features in metallic layer 300 are not protected from redeposition of material and / or resist residues.

[0075] As mentioned above, the metallic layer 100 that can show superconducting behavior below critical temperature can be the main layer that comprises the key element and parts that contribute to quantum information processing.These elements and parts (for example, Josephson junction, coplanar waveguide, LC oscillator, superconducting quantum interference device (SQUID), inductor and capacitor) all or part of are limited to the pattern on the metallic layer, and ideally, the pattern has clean and well-defined edge.For example, coplanar waveguide can be limited to the strip in the metallic layer 100.Technique described herein can be used for providing cleaner edge on these strips, and this will greatly reduce the loss caused by the defect near the edge of these strips.In addition, technique described herein can prevent residue from staying near the edge of pattern, and residue stays near the edge of pattern and can cause loss or decoherence because of material defect such as secondary state (TLS).The clean and well-defined edge obtained from technique described herein also prevents the adverse effect to subsequent layers.Technique described herein may need to deposit one or more materials, such as superconductor, dielectric and / or metal. Depending on the materials selected, these materials can be deposited using deposition processes such as chemical vapor deposition, physical vapor deposition (e.g., evaporation or sputtering), or epitaxial techniques or other deposition processes. The processes described herein may also require the removal of one or more materials from the device during manufacturing. Depending on the materials to be removed, the removal process may include, for example, wet etching techniques, dry etching techniques, or stripping processes.

[0076] Other materials can be used for the metallic layer and substrate. For example, the metallic layer can be another superconducting metal such as niobium or tin, or a superconducting alloy. The substrate can be sapphire or a suitable dielectric material. The substrate can be a multilayer substrate, such as silicon on insulator.

[0077] The implementation of quantum themes and quantum operations described in this specification can be implemented in suitable quantum circuits, or more generally in a quantum computing system including the structures disclosed in this specification and their structural equivalents, or in a combination of one or more of them. The term "quantum computing system" may include, but is not limited to, a quantum computer, a quantum information processing system, a quantum cryptographic system, or a quantum simulator.

[0078] The terms quantum information and quantum data refer to information or data carried, held or stored by a quantum system, wherein the smallest non-trivial system is a qubit, e.g., a system defining a unit of quantum information. It is understood that the term "qubit" encompasses all quantum systems that can be appropriately approximated as a two-level system in the corresponding context. Such quantum systems may include, for example, multi-level systems having two or more energy levels. For example, such systems may include atoms, molecules, electrons, photons, ions, quantum dots, or superconducting qubits. In many implementations, the computational basis states are identified with the ground state and the first excited state, but it is understood that other arrangements in which the computational state is identified with an excited state of a higher energy level are feasible. It is understood that quantum memories are devices that can store quantum data for a long time with high fidelity and efficiency, such as light-matter interfaces, in which light is used for transmission and matter is used for storage and preservation of quantum features (such as superposition or quantum coherence) of quantum data.

[0079] Quantum circuit elements can be used to perform quantum processing operations. That is, quantum circuit elements can be configured to perform operations on data in a non-deterministic manner using quantum mechanical phenomena such as superposition and entanglement. Certain quantum circuit elements such as qubits can be configured to represent information and operate on information in more than one state at the same time. Examples of superconducting quantum circuit elements that can be formed using the processes disclosed herein include circuit elements such as coplanar waveguides, quantum LC oscillators, qubits (e.g., flux qubits or charge qubits), superconducting quantum interference devices (SQUIDs) (e.g., RF-SQUIDs or DCSQUIDs), inductors, capacitors, transmission lines, ground planes, or other circuit elements.

[0080] In contrast, classical circuit elements generally process data in a deterministic manner. Classical circuit elements can be configured to collectively execute the instructions of a computer program by performing basic arithmetic, logic, and / or input / output operations on data, where the data is represented in either analog or digital form.

[0081] In some implementations, classical circuit elements can be used to send data to and / or receive data from quantum circuit elements via electrical or electromagnetic connections. Examples of classical circuit elements that can be formed using the processes disclosed herein include fast single flux quantum (RSFQ) devices, reciprocal quantum logic (RQL) devices, and ERSFQ devices, which are energy-efficient versions of RSFQ that do not use bias resistors. Other classical circuit elements can also be formed using the processes disclosed herein.

[0082] During operation of a quantum computing system using superconducting quantum circuit elements and / or superconducting classical circuit elements, such as the circuit elements described herein, the superconducting circuit elements are cooled within a cryostat to a temperature that allows the superconducting material to exhibit superconducting properties.

[0083] Although this specification contains many specific implementation details, these details should not be interpreted as limitations on the scope of what is claimed, but rather as descriptions of features that may be unique to a particular implementation. Certain features described in this specification in the context of separate implementations may also be implemented in combination with a single implementation. Conversely, various features described in the context of a single implementation may also be implemented separately in multiple implementations or in any suitable subcombination.

[0084] Furthermore, although features may be described above as functioning in certain combinations and even initially claimed as such, in certain cases one or more features of the claimed combination may be deleted from the combination, and the claimed combination may be directed to sub-combinations or variations of the sub-combinations.

[0085] Similarly, although operations are depicted in a particular order in the accompanying drawings, this should not be understood as requiring that such operations be performed in the particular order shown or in a continuous order, or that all of the operations shown be performed to achieve the desired result. For example, the actions recited in the claims can be performed in a different order and still achieve the desired result. In some cases, multitasking and parallel processing can be advantageous. Moreover, the separation of the various components in the above implementations should not be understood as requiring such separation in all implementations.

[0086] A number of implementations have been described. However, it will be appreciated that various modifications may be made without departing from the spirit and scope of the invention. Other implementations are within the scope of the following claims.

Claims

1. A method for manufacturing a device, comprising: providing a substrate, wherein the substrate comprises a metal layer on a major surface of the substrate; A multi-layer stack is formed on the metal layer, the multi-layer stack comprising: a first polymer layer; a second polymer layer on the first layer, wherein the second polymer layer is in contact with a surface of the first polymer layer; and a third polymer layer on the second polymer layer, wherein the third polymer layer is in contact with a surface of the second polymer layer; patterning the multilayer stack to form an opening through the multilayer stack such that within the opening, the third polymer layer is undercut by the second polymer layer and the second polymer layer is overcut by the first polymer layer, wherein patterning the multilayer stack to form the opening comprises exposing the third polymer layer and the second polymer layer, developing the third polymer layer and the second polymer layer, wherein during the development of the second polymer layer, the second polymer layer recedes relative to the third polymer layer to form an undercut profile relative to the third polymer layer, dry etching the first polymer layer; dry etching the metal layer to form a patterned metal layer of the device; and removing the multi-layer stack, The device comprises a quantum computing circuit element. 2 . The method of claim 1 , wherein the third polymer layer and the second polymer layer are exposed simultaneously. 3 . The method of claim 1 , wherein the third polymer layer is exposed and developed before the second polymer layer is exposed and developed. 4 . The method of claim 1 , wherein exposing the third polymer layer and the second polymer layer comprises exposing the third polymer layer and the second polymer layer with an electron beam. 5 . The method of claim 1 , wherein the first polymer layer is polymethyl methacrylate (PMMA) and the second polymer layer comprises polydimethylglutarimide (PMGI). The method of claim 1 , wherein the second polymer layer and the third polymer layer are different materials. 7 . The method of claim 6 , wherein the second polymer layer and the third polymer layer are simultaneously developed using a developer solution, wherein the exposed portion of the third polymer layer is less soluble in the developer solution than the exposed portion of the second polymer layer.

8. The method of claim 6, wherein before the third polymer layer is formed on the second polymer layer, the entire second polymer layer is flood exposed with deep ultraviolet light.

9. The method of claim 6, wherein the third polymer layer and the second polymer layer are exposed simultaneously.

10. The method according to claim 9, wherein: After developing the third polymer layer, the second polymer layer is exposed a second time. 11 . The method of claim 1 , wherein dry etching the first polymer layer comprises reactive ion etching the first polymer layer.

12. The method of claim 1, wherein the first polymer layer recedes laterally as a result of the dry etching.

13. The method of claim 12, comprising controlling a degree of recessing of the first polymer layer during the dry etching.

14. The method of claim 1, wherein dry etching the metal layer comprises ion milling the metal layer.

15. The method of claim 1, wherein dry etching the metal layer comprises performing a reactive ion etch of the metal layer.

16. The method of claim 1, wherein the first polymer layer has a thickness between 10 nm and 100 nm.

17. The method of claim 1, wherein the metal layer comprises a superconductor.

18. The method of claim 1, wherein the device comprises a Josephson junction, a superconducting quantum interference device (SQUID), or a qubit.