Modelling Realistic Multi-layer devices for superconducting quantum electronic circuits

This paper presents a flexible and accurate numerical model for 3D multilayer superconducting devices that validates its ability to enhance qubit anharmonicity and study proximity effects by calculating critical currents and energy gaps without approximating physical layouts or limiting constituent materials.

The Gist

Imagine you are trying to build a tiny, ultra-fast electronic switch using superconducting materials (metals that conduct electricity with zero resistance when cold). These switches, called Josephson junctions, are the heart of quantum computers.

For a long time, scientists built these switches using a "sandwich" method: two metal layers separated by a thin, insulating oxide layer (like a piece of bread with a layer of jelly in the middle). However, that "jelly" (the oxide) can be messy. It creates unwanted noise, loses energy, and makes it hard to predict exactly how the switch will behave.

The New Approach: The "Bridge"
The researchers in this paper propose a different design. Instead of a sandwich with jelly, they build a nanobridge. Imagine two islands (the metal electrodes) connected by a tiny, narrow bridge made of metal. There is no insulating jelly in the middle; the metals touch directly. This removes the messy oxide layer, making the connection cleaner and more precise.

The Problem: It's Hard to Predict
While the bridge idea sounds great, it's incredibly difficult to predict exactly how electricity will flow through these tiny, 3D structures, especially when they have different shapes (like rounded corners instead of sharp squares) or are made of multiple layers of different metals. Existing computer models were too simple; they either ignored the 3D shape or assumed the materials were perfect, leading to inaccurate designs.

The Solution: A "Digital Twin" Simulator
The team created a new, highly detailed computer model (a "digital twin") that simulates these 3D multilayer devices exactly as they are built in real life.

• No Shortcuts: Unlike older models, this one doesn't pretend the bridge is a perfect rectangle or ignore the different materials. It accounts for rounded edges (which happen naturally when you carve these tiny bridges) and layers of different metals.
• The Physics: It uses complex math (called Usadel equations) to track how electrons move and how the "superconducting energy gap" (the energy needed to break the superconducting state) changes across the device.

Key Discoveries: Why Shape and Layers Matter
By running their new simulator, the team found some surprising and useful things:

1. Rounded Edges Change the Flow: When the edges of the bridge are rounded (like a real bridge) rather than sharp (like a digital drawing), the maximum current the bridge can carry drops slightly. This is because the rounded shape weakens the connection between the two sides, making the device behave more like a theoretical "ideal" model.
2. The "Variable Thickness" Trick: They tested a design where the bridge gets thinner in the middle (like a dumbbell). They found that this shape creates a more stable and predictable flow of electricity compared to a flat, uniform bridge. This is crucial for qubits (the basic units of quantum computers) because it helps them stay "tuned" to the right frequency, making them more reliable.
3. The "Proximity Effect" (The Contagion): When they placed a normal metal on top of a superconductor (a technique called "encapsulation" to protect the surface), they saw a "contagion" effect. The superconducting power of the metal "leaked" into the normal metal, but in doing so, the superconductor's own power (the energy gap) got weaker.
   • The Analogy: Imagine a group of people holding hands tightly (superconducting). If you add a few people who don't hold hands well (normal metal) to the chain, the whole group has to loosen their grip to accommodate them. The researchers' model helps calculate exactly how much the grip loosens so engineers can choose the right materials to keep the quantum computer stable.

Why This Matters
The paper doesn't promise a new quantum computer tomorrow. Instead, it provides a better blueprint tool.

• It allows engineers to design these tiny bridges with much higher confidence.
• It shows that using multilayer films (stacking different materials) gives them better control over the device's performance.
• It proves that their new simulation matches real-world experiments better than previous models, especially when they account for the fact that the materials might be slightly different than originally thought (like the "coherence length" being larger than expected).

In short, the researchers built a more accurate "GPS" for designing the tiny bridges that power the next generation of quantum computers, helping engineers avoid dead ends and build more reliable machines.

Technical Summary

Technical Summary: Modelling Realistic Multi-layer Devices for Superconducting Quantum Electronic Circuits

Problem Statement
Superconducting quantum technologies, including qubits and Single Flux Quantum (SFQ) circuits, rely heavily on Josephson junctions and coplanar waveguides (CPW). While traditional tunnel junctions are common, they suffer from performance degradation due to oxide layers that introduce localized states, energy loss, and reduced coherence. Nanobridge junctions offer a promising alternative by eliminating the oxide layer, yet existing numerical models for these devices are limited. Previous approaches were restricted to either 3D simulations lacking quantum technology applications or 2D designs that failed to account for realistic geometric features, such as rounded edges resulting from etching processes. Furthermore, there is a need for accurate modeling of multilayer structures to study the proximity effect and optimize device performance, particularly regarding critical currents, current-phase relationships (CPR), and energy gaps.

Methodology
The authors developed a numerical model specifically designed for 3D multilayer devices, focusing on nanobridge junctions and CPWs. Unlike prior models, this approach does not approximate the physical layout or limit the number of constituent materials. The core of the methodology involves solving the Usadel equations under the assumption of negligible magnetic effects and the "dirty limit" condition ($l \ll \xi$).

Key methodological steps include:

• Governing Equations: The model utilizes the Usadel equations to describe current transport, parametrizing the Green's function using the $\Phi$-parametrization.
• Numerical Implementation: To handle the non-linear and non-differentiable nature of the equations (due to the $\|\Phi\|^2$ term), the authors separated the real and imaginary parts of the equations. These were reformulated into a set of coupled weak-form equations and solved using the sfepy finite element library with custom non-linear functions.
• Boundary Conditions: The model incorporates self-consistent equations for the pair potential and specific boundary conditions between different materials (Superconductor-Superconductor' or Superconductor-Normal metal), accounting for suppression parameters ($\gamma$) and boundary resistance ($\gamma_B$).
• Device Geometry: Simulations were performed on various layouts, including planar junctions and Variable-Thickness Nanobridge (VTB) junctions, with both rectangular and realistically rounded cross-sections.
• Qubit Analysis: The calculated CPRs were integrated into custom Hamiltonians for Transmon and Fluxonium qubits (using scqubits and qiskit metal) to determine energy levels and anharmonicity.

Key Contributions and Results

1. Validation Against Experimental Data: The model was validated by comparing simulation results with published experimental data on aluminum VTB and planar nanobridges. The simulated critical currents showed good agreement with experimental values, particularly when an adjusted coherence length (110 nm vs. the reported 40 nm) was used, suggesting that room-temperature resistivity calculations may underestimate the coherence length.
2. Impact of Geometry: The study demonstrated that realistically rounded edges reduce the critical current in junctions, even after normalizing for the reduced cross-sectional area. For planar junctions, rounded edges slightly shift the CPR maximum, bringing the behavior closer to the ideal one-dimensional KO-1 model but with lower values and increased skewness due to energy gap reduction in the bridge area.
3. Advantages of Multilayer (VTB) Designs: Variable-Thickness Nanobridge (VTB) junctions produced less skewed CPRs compared to planar junctions. This reduction in skewness is attributed to the electrodes acting as a robust phase reservoir, concentrating the phase change across the bridge. Consequently, VTB structures significantly improve qubit anharmonicity compared to monolayer junctions.
4. Proximity Effect and Encapsulation: The model successfully analyzed the proximity effect in coated multilayer CPWs. It was found that the suppression parameter $\gamma$ (dependent on resistivity and coherence length) critically influences the superconducting gap. An increase in $\gamma$ leads to a reduction in the energy gap ($\Delta$) within the superconducting layer, which increases quasiparticle density and susceptibility to environmental noise.
5. Qubit Performance: By applying the calculated CPRs to qubit Hamiltonians, the authors showed that while tunnel junctions (sinusoidal CPR) offer high anharmonicity, the skewed CPRs of simple planar nanobridges reduce this benefit. However, multilayer designs recover high anharmonicity levels, offering a pathway to better qubit performance without the coherence issues associated with oxide-based tunnel junctions.

Significance
The paper claims that this numerical model provides a more accurate and flexible design tool for superconducting quantum devices by avoiding physical approximations and accommodating complex multilayer geometries. The primary significance lies in demonstrating that multilayer films offer substantial control over critical currents, CPRs, and energy gaps. Specifically, the ability to engineer VTB structures allows for improved qubit anharmonicity, addressing a key performance metric. Additionally, the model's capability to simulate the proximity effect in encapsulated structures aids in the design of surface passivation techniques, which are crucial for mitigating surface oxide losses and enhancing qubit coherence times. This work aims to interface directly with standard design processes, facilitating the development of more reliable and scalable superconducting quantum technologies.