betaFeedback

WorksPhysics and Astronomy

A Time-Dependent Ginzburg-Landau Framework for Sample-Specific Simulation of Superconductors for SRF Applications

Abstract

Modern superconducting radio frequency (SRF) applications demand precise control over material properties across multiple length scales - from microscopic composition, to mesoscopic defect structures, to macroscopic cavity geometry. We present a time-dependent Ginzburg-Landau (TDGL) framework that incorporates spatially varying parameters derived from experimental measurements and ab initio calculations, enabling realistic, sample-specific simulations. As a demonstration, we model Sn-deficient islands in Nb3_3Sn and calculate the field at which vortex nucleation first occurs for various defect configurations. These thresholds serve as a predictive tool for identifying defects likely to degrade SRF cavity performance. We then simulate the resulting dissipation and show how aggregate contributions from multiple small defects can reproduce trends consistent with high-field QQ-slope behavior observed experimentally. Our results offer a pathway for connecting microscopic defect properties to macroscopic SRF performance using a computationally efficient mesoscopic model.

Introduction

Superconducting Radio-Frequency (SRF) cavities are a crucial component of particle accelerators, as they utilize AC electromagnetic fields to accelerate beams of charged particles. Nb has been the industry standard for SRF applications for decades due to its high critical temperature (∼9\sim9K) relative to the other elemental superconductors. Within the past decade, the need for SRF cavities with capabilities beyond the limits of Nb cavity performance has led to the study of a variety of alternative SRF materials. Among these materials, Nb3_3Sn has emerged as a promising candidate; Nb3_3Sn boasts both a higher critical temperature (∼18\sim18K) and higher critical fields [1, 2]. One particular advantage of SRF cavities (as compared to traditional normal conducting RF cavities) is their high quality factors (QQ) [3]. A major benefit of Nb3_3Sn SRF cavities compared to their Nb counterparts is that they can maintain similar QQs (on the order of 101010^{10}) at higher temperatures (4.2K vs. 2K) [2], significantly reducing cryogenic costs. NbZr is another promising alternative SRF material which has seen recent attention [4, 5], with most existing NbZr samples exhibiting critical temperatures between 10 and 13K (the theoretical maximum is 17.7K), but this material has not yet been tested at cavity scale. The simulations in this paper focus on Nb3_3Sn, but the methods we will present can be generalized to any material of interest.

The oscillating electric fields used for acceleration in SRF cavities induce magnetic fields parallel to the cavity surface, so for large accelerating gradients, the critical magnetic fields of the cavity material are the fundamental limits of cavity performance. Type-II superconductors such as Nb and Nb3_3Sn have two critical fields, Hc1H_{c1} and Hc2H_{c2}.

For fields below Hc1H_{c1}, the Meissner state is stable and magnetic flux is expelled from the cavity. Between Hc1H_{c1} and Hc2H_{c2}, there is a mixed state in which superconducting vortices trap lines of magnetic flux, forming normal cores inside the otherwise superconducting state. Under an AC field, as is the case for SRF cavities, the vortices quickly move in and out of the cavity over the course of an AC cycle. This vortex motion leads to large amounts of dissipation [6]. For fields above Hc2H_{c2}, the mixed state becomes unstable and in this state SRF cavities will quench (i.e. go normal conducting). It is important to note that this is not the only mechanism for SRF cavity quench, the dissipation caused by moving vortices in the mixed state can cause heating in the cavity, which can also lead to quenching through a change in cavity temperature. As such, for SRF applications, it is important that the cavity remain within the Meissner state during operation.

While the Meissner state is no longer thermodynamically stable above Hc1H_{c1}, it can remain metastable up until the so-called superheating field, HshH_{sh} [7]. It is well known that many high-power SRF cavities operate in this metastable Meissner state [8]. As such, HshH_{sh} is the theoretical limiting field for operation of SRF cavities, since the dissipative vortices which are detrimental to SRF performance become unavoidable for fields above HshH_{sh}. HshH_{sh} has been studied for decades by condensed matter theorists. These studies have most commonly been within a Ginzburg-Landau (GL) framework [9, 10, 11, 12], but the superheating field has also been studied extensively utilizing the Eilenberger equations [13, 14, 15].

HshH_{sh} provides the maximum possible field (and therefore the maximum accelerating gradient) for SRF cavity operation, but local features of a material such as impurities or surface geometries can act as nucleation sites for vortices. This means that in practice, realistic material samples will be limited by what we will call the vortex penetration field, HvortH_{vort}, which is the lowest field at which the material nucleates vortices. This quantity can vary greatly between different samples and depends on a large variety of different effects, so estimation of HvortH_{\mathrm{vort}} for realistic sample materials remains a rich area of research. For the case of Nb3Sn\mathrm{Nb}_3\mathrm{Sn} in particular, theoretical HshH_{\mathrm{sh}} calculations suggest that Nb3Sn\mathrm{Nb}_3\mathrm{Sn} cavities could reach accelerating gradients as high as around 100 MV/m, yet the highest accelerating field achieved so far by a Nb3Sn\mathrm{Nb}_3\mathrm{Sn} SRF cavity is around 24 MV/m [16], with most other cavities reaching their quench field well below this. Additionally, many existing cavities exhibit a phenomenon in which QQ significantly degrades as the cavity approaches its quench field, a phenomenon dubbed “Q-slope” [17], also sometimes called high field Q-slope (HFQS) when it occurs primarily at higher fields near the quench field. These performance degradations are the result of material defects introduced during the Nb3Sn\mathrm{Nb}_3\mathrm{Sn} growth process, so understanding how different defects seen within samples affect things like HvortH_{\mathrm{vort}} or dissipation more generally is critical to developing better growth techniques.

The need to accurately model specific features such as material impurities within superconducting materials motivates us to develop a framework which will allow us to directly model the spatial variations of superconducting properties due to the material compositions observed in realistic sample materials. To do this, we use time-dependent Ginzburg-Landau (TDGL) theory. TDGL has already proven itself to be a powerful tool for mesoscopic-scale simulations relevant to SRF applications [18, 19, 20, 21]. Besides SRF simulations, TDGL has broad application such as in single photon detectors [22, 23], superconducting quantum interference devices (SQUIDs) [24, 25], weak links [26, 27, 28, 29, 30], or superconducting nanowires [31, 32, 33]. The methods we present in this paper are investigated with full 3D simulations. While prior work has studied vortex dynamics in 3D [34, 35, 36, 37, 38, 39, 40], the majority of TDGL research, especially in the context of SRF cavities, has been limited to 2D.

Whenever working with TDGL, it is important to acknowledge its limitations. TDGL is a mesoscopic-scale model which abstracts the microscopic details of superconductivity into quantities which can be used to describe things like vortex dynamics. Much of this abstraction is the direct result of restricting the quantitative validity to gapless superconductivity at temperatures near TcT_c. This means that outside of this fairly restrictive regime, the quantitative predictions of TDGL are not accurate in general. Despite this, there are three main reasons that TDGL is still has considerable value in a variety of studies, such as those referenced above. Firstly, it is well known that the solutions to the Ginzburg-Landau equations have a much wider range of quantitative applicability when under the dirty limit, and so properties of the TDGL equations which can be derived from steady-state dirty limit solutions, such as HshH_{\mathrm{sh}}, are still valuable quantitative outputs of the theory. Secondly, theories of superconductivity with larger ranges of quantitative accuracy (such as the Eilenberger equations [13, 14, 15] or other quasi-classical approaches) scale extremely poorly in terms of computational complexity when it comes to numerical simulations, whereas TDGL is at least feasible for larger mesoscopic-scale simulations. And finally, TDGL offers qualitative and semi-quantitative predictions that provide useful insight into phenomena which may be difficult or impossible to measure experimentally. For example, it can be used to compare how variations in the size and depth of stoichiometric defects relative to the superconductor’s surface affect SRF-relevant metrics such as vortex nucleation and energy dissipation—enabling prioritization of which defect characteristics are most critical to address when perfect control is not possible. In what follows, we apply this approach to a specific class of such defects relevant to Nb3Sn\mathrm{Nb}_3\mathrm{Sn}-coated SRF cavities.

To model sample-specific materials and investigate mechanisms behind Q-slope and other quenching phenomena in Nb3Sn\mathrm{Nb}_3\mathrm{Sn} SRF cavities, we draw on both experimental and theoretical studies that characterize the microscopic properties of Nb3Sn\mathrm{Nb}_3\mathrm{Sn}. There has been a large body of work experimentally characterizing SRF grade vapor-diffused Nb3Sn\mathrm{Nb}_3\mathrm{Sn} samples. The primary suspect for SRF performance degradation comes from defects or other imperfections in the Nb3Sn\mathrm{Nb}_3\mathrm{Sn} surface significantly lowering the barrier to flux penetration [41]. In particular, defects which have been studied are abnormally thin or patchy grains [42, 43, 44], Sn-segregated grain boundaries [19, 46] [45], and Sn-deficient regions [42, 43, 47, 48]. In addition to experimental characterizations, there have also been a variety of ab initio calculations for Nb3Sn\mathrm{Nb}_3\mathrm{Sn} using density functional theory. In addition to calculations of general properties of Nb3Sn\mathrm{Nb}_3\mathrm{Sn} [49], such as the electron and phonon density of states and Eliashberg spectral function, these quantities have also been estimated with respect to varying intrinsic strain [50] as well as normal resistivity [51]. Variations in the superconducting TcT_c as well as electron density of states have also been calculated with respect to varying tin concentration [52, 53], which applies to both Sn-segregation at grain boundaries and Sn-deficient regions. These Sn-deficient regions, which we will call Sn-deficient islands, are the primary material defect we will study in order to validate our methods.

Experimental characterizations can give data about the material compositions and physical structure of superconducting materials, and ab initio calculations provide detailed descriptions of the electronic/phononic structure and the resulting superconducting properties, both of these a microscopic scale. TDGL plays the role of modeling mesoscopic scale phenomena (such as vortex dynamics) which are difficult or even impossible to measure directly via experiment, and are too large to easily model with microscopic scale theories such as DFT. There have been a number of studies which have used TDGL to model material inhomogeneities [18, 19, 54, 55, 56, 57, 58], but these studies did not use the explicit dependencies of the TDGL parameters on microscopic material properties to inform their choice of parameters. The limitation to the approach used in these references is that it requires either looking through a large portion of the TDGL parameter space in order to find values which lead to expected predictions, or more commonly, picking somewhat arbitrary values, which limits confidence in the results. In this paper, we outline a new framework in TDGL theory which allows us to directly calculate the values of the TDGL parameters based on local properties of the superconductor. This framework enables modeling of realistic features of superconductor samples and supports estimation of critical fields and energy dissipation under dynamic electromagnetic conditions. Under our framework, TDGL serves as a bridge between experimental material characterizations and ab initio calculations of material-specific parameters, allowing us to further connect these microscopic characterizations with macroscopic SRF performance metrics in a sample-specific way. Because SRF cavity development is inherently multidisciplinary—bringing together accelerator physicists, materials scientists, and condensed matter theorists—a framework that integrates insights across these domains is particularly valuable. The method presented here enables such integration, offering a pathway to sample-specific predictions grounded in both microscopic characterization and macroscopic application.

This paper is organized as follows. In Section II, we present the TDGL equations and show how to calculate TDGL parameters from material properties. We then describe how spatial variations in these parameters are estimated by combining results from DFT calculations with experimental material characterizations. The section concludes with a discussion of how dissipation can be computed from TDGL solutions and how SRF cavity quality factors can be derived from this dissipation. In Section III, we apply our framework to estimate the critical fields associated with Sn-deficient islands of varying size, position, and stoichiometric composition. We then examine dissipation in the context of Nb SRF cavities and analyze a representative simulation in detail to probe the limits of TDGL. Finally, we demonstrate how combining critical field and dissipation results enables us to estimate high-field QQ-slope behavior for Nb3_3Sn SRF cavities. Section IV concludes the paper with a discussion of the justification for using TDGL despite its known limitations and the implications of our findings for future SRF cavity research.

Methods

The Time-Dependent Ginzburg-Landau Equations

Ginzburg-Landau (GL) theory is one of the oldest theories of superconductivity, and it remains relevant today owing to its relative simplicity and direct physical insights into the electrodynamic response of superconductors under static applied fields and currents [59]. The time-dependent Ginzburg-Landau (TDGL) equations were originally proposed by Schmid [60] in 1966 and Gor’kov and Eliashberg [61] derived them rigorously from BCS theory later in 1968. The TDGL equations (in Gaussian units) are given by:

Γ(∂ψ∂t+iesϕℏψ)+12ms(−iℏ∇−escA)2ψ+αψ+β∣ψ∣2ψ=0(1)\Gamma\left(\frac{\partial\psi}{\partial t}+\frac{i e_s\phi}{\hbar}\psi\right)+\frac{1}{2m_s}\left(-i\hbar\nabla-\frac{e_s}{c}\mathbf{A}\right)^2\psi+\alpha\psi+\beta|\psi|^2\psi=0 \tag*{(1)}
4πσnc(1c∂A∂t+∇ϕ)+∇×∇×A−2πiesℏmsc(ψ∗∇ψ−ψ∇ψ∗)+4πes2msc2∣ψ∣2A=0.(2)\frac{4\pi\sigma_n}{c}\left(\frac{1}{c}\frac{\partial\mathbf{A}}{\partial t}+\nabla\phi\right)+\nabla\times\nabla\times\mathbf{A}-\frac{2\pi i e_s\hbar}{m_sc}\left(\psi^*\nabla\psi-\psi\nabla\psi^*\right)+\frac{4\pi e_s^2}{m_sc^2}|\psi|^2\mathbf{A}=0. \tag*{(2)}

These equations are solved for the complex superconducting order parameter, ψ\psi, and the magnetic vector potential, A\mathbf{A}. The magnitude squared of ψ\psi is proportional to the density of superconducting electrons. The parameters α\alpha and β\beta are phenomenological, and were originally introduced as coefficients of the series expansion of the Ginzburg-Landau free energy. Additionally, ϕ\phi is the scalar potential; σn\sigma_n is the normal electron conductivity; Γ\Gamma is the phenomenological relaxation rate of ψ\psi. Furthermore, es=2ee_s=2e and ms=2mem_s=2m_e represent the total charge and total effective mass of a Cooper pair, respectively.

The TDGL equations are subject to boundary conditions

(iℏ∇ψ+escAψ)⋅n=0(3)\left(i\hbar\nabla\psi+\frac{e_s}{c}\mathbf{A}\psi\right)\cdot\mathbf{n}=0 \tag*{(3)}
(∇×A)×n=Ha×n(4)(\nabla\times\mathbf{A})\times\mathbf{n}=\mathbf{H}_a\times\mathbf{n} \tag*{(4)}
(∇ϕ+1c∂A∂t)⋅n=0,(5)\left(\nabla\phi+\frac{1}{c}\frac{\partial\mathbf{A}}{\partial t}\right)\cdot\mathbf{n}=0, \tag*{(5)}

where n\mathbf{n} is the outward normal vector to the boundary surface and Ha\mathbf{H}_a is the applied magnetic field. Eq. 3 ensures no current flows out of the superconducting domain, and noting that E=−∇ϕ−1c∂A∂tE=-\nabla\phi-\frac{1}{c}\frac{\partial\mathbf{A}}{\partial t}, Eqs. 4 and 5 are electromagnetic interface conditions with an applied magnetic field.

The parameters α\alpha, β\beta, and Γ\Gamma were originally introduced into the theory as phenomenological, temperature-dependent constants [62]. It is worth noting that α<0\alpha<0 corresponds to the superconducting state whereas α≥0\alpha\ge0 corresponds to the normal state; β\beta is strictly positive regardless of the system’s state. The TDGL equations can also be derived from microscopic theory using the time-dependent Gor’kov equations [61]. A useful consequence of this derivation is that it allows the TDGL parameters to be directly related to experimentally observable properties of the material in question. The material dependencies are given by Ref. 63:

α(ν(0),Tc,T)=−ν(0)(1−T2Tc21+T2Tc2)≈−ν(0)(1−TTc)(6)\begin{aligned} \alpha(\nu(0),T_c,T)=-\nu(0)\left(\frac{1-\frac{T^2}{T_c^2}}{1+\frac{T^2}{T_c^2}}\right) \\ &\approx-\nu(0)\left(1-\frac{T}{T_c}\right) \tag*{(6)} \end{aligned}
β(ν(0),Tc,T)=7ζ(3)ν(0)8π2Tc2(11+T2Tc2)2≈7ζ(3)ν(0)8π2Tc2(7)\begin{aligned} \beta(\nu(0),T_c,T)=\frac{7\zeta(3)\nu(0)}{8\pi^2T_c^2}\left(\frac{1}{1+\frac{T^2}{T_c^2}}\right)^2 \\ &\approx\frac{7\zeta(3)\nu(0)}{8\pi^2T_c^2} \tag*{(7)} \end{aligned}
Γ(ν(0),Tc)=ν(0)πℏ8Tc,(8)\Gamma(\nu(0),T_c)=\frac{\nu(0)\pi\hbar}{8T_c}, \tag*{(8)}

where ν(0)\nu(0) is the density of states at the Fermi-level, TcT_c is the critical temperature, TT is the temperature, and ζ(x)\zeta(x) is the Riemann zeta function. Additionally, the effective mass can also be expressed in terms of these same quantities in addition to the Fermi velocity, vfv_f, and electron mean free path, ℓ\ell:

ms=12ℏTcπν(0)vfℓ.(9)m_s=\frac{12\hbar T_c}{\pi\nu(0)v_f\ell}. \tag*{(9)}

Equation 9 gives the effective mass under the dirty limit. A major contribution of this paper is to introduce a framework for modeling sample-specific features of superconducting materials by connecting ab initio calculations of the material’s properties to experimental characterizations of the material.

from experimental characterizations. When converting these constant coefficient into a spatially-varying ones, it is important to consider how this may alter the TDGL equations due to potential alterations to the functional derivatives of the free energy. In the case of α\alpha, β\beta, and Γ\Gamma, they do not appear on terms of the free energy with spatial gradients. However, msm_s does appear in terms including gradients, and so when including spatially-varying effective mass, Equation 1 should be augmented with an additional term:

iℏ2∇1ms⋅(iℏ∇+ecA)ψ\frac{i\hbar}{2}\nabla\frac{1}{m_s}\cdot\left(i\hbar\nabla+\frac{e}{c}\mathbf{A}\right)\psi

Additionally, when solving the TDGL equations numerically, it is standard to normalize all the parameters of the model in order to obtain dimensionless quantities. In order to satisfy the above requirements, the steps are as follows: Pick a reference value for each of α\alpha, β\beta, Γ\Gamma, msm_s, and σn\sigma_n, these will most often just be the corresponding values for the bulk material. Label these reference values α0\alpha_0, β0\beta_0, Γ0\Gamma_0, m0m_0, and σn0\sigma_{n0}. Define dimensionless spatially varying functions, a(r)a(\mathbf{r}), b(r)b(\mathbf{r}), γ(r)\gamma(\mathbf{r}), μ(r)\mu(\mathbf{r}), and s(r)s(\mathbf{r}), relative to their reference values. Apply the following transformations to Eqs. 1 and 2:

α⟶α0a(r)(10)\alpha\longrightarrow\alpha_0a(\mathbf{r}) \tag*{(10)}
β⟶β0b(r)(11)\beta\longrightarrow\beta_0b(\mathbf{r}) \tag*{(11)}
Γ⟶Γ0γ(r)(12)\Gamma\longrightarrow\Gamma_0\gamma(\mathbf{r}) \tag*{(12)}
ms⟶m0μ(r)(13)m_s\longrightarrow m_0\mu(\mathbf{r}) \tag*{(13)}
σn⟶σn0s(r).(14)\sigma_n\longrightarrow\sigma_{n0}s(\mathbf{r}). \tag*{(14)}

Then proceed with standard nondimensionalization procedures for TDGL (see the Appendix for more details). The advantage of this approach is that the nondimensionalization procedures, when used on these transformed equations, leave behind only the dimensionless functions which capture the spatial variation of the sample material in natural units. The resulting equations are

γ(∂ψ∂t+iκ0ϕψ)+1μ(−iκ0∇−A)2ψ+1κ0∇1μ⋅(1κ0∇−iA)ψ−aψ+b∣ψ∣2ψ=0(15)\gamma\left(\frac{\partial\psi}{\partial t}+i\kappa_0\phi\psi\right)+\frac{1}{\mu}\left(\frac{-i}{\kappa_0}\nabla-\mathbf{A}\right)^2\psi+\frac{1}{\kappa_0}\nabla\frac{1}{\mu}\cdot\left(\frac{1}{\kappa_0}\nabla-i\mathbf{A}\right)\psi-a\psi+b|\psi|^2\psi=0 \tag*{(15)}
su0(∂A∂t+∇ϕ)+∇×∇×A+i2κ0μ(ψ∗∇ψ−ψ∇ψ∗)+1μ∣ψ∣2A=0,(16)\frac{s}{u_0}\left(\frac{\partial\mathbf{A}}{\partial t}+\nabla\phi\right)+\nabla\times\nabla\times\mathbf{A}+\frac{i}{2\kappa_0\mu}\left(\psi^*\nabla\psi-\psi\nabla\psi^*\right)+\frac{1}{\mu}|\psi|^2\mathbf{A}=0, \tag*{(16)}

where κ0=λ0ξ0\kappa_0=\frac{\lambda_0}{\xi_0} is the Ginzburg–Landau parameter of the reference material. The quantity λ0=m0c2β04πe2∣α0∣\lambda_0=\sqrt{\frac{m_0c^2\beta_0}{4\pi e^2|\alpha_0|}} is the penetration depth of the reference material, and ξ0=ℏ22m0∣α0∣\xi_0=\sqrt{\frac{\hbar^2}{2m_0|\alpha_0|}} is its coherence length. The parameter u0=τψ0τj0u_0=\frac{\tau_{\psi0}}{\tau_{j0}} is a ratio of characteristic time scales in the reference material, where τψ0=Γ0∣α0∣\tau_{\psi0}=\frac{\Gamma_0}{|\alpha_0|} is the charcharacteristic relaxation time of ψ\psi in the reference material and τj0=σn0m0β0e2∣α0∣\tau_{j0} = \frac{\sigma_{n0}m_0\beta_0}{e^2\lvert\alpha_0\rvert} is the characteristic relaxation time of the current. We have also inserted a minus in front of aa, which is just a convention to make positive values of aa correspond to the superconducting state (Note: this is the opposite of how α\alpha is usually interpreted within Ginzburg-Landau theory, however it is standard to make this change when performing nondimensionalization). The boundary conditions become:

(iκ0∇ψ+Aψ)⋅n=0(17)\left(\frac{i}{\kappa_0}\nabla\psi+ \mathbf{A}\psi\right)\cdot\mathbf{n}=0 \tag*{(17)}
(∇×A)×n=Ha×n(18)(\nabla\times\mathbf{A})\times\mathbf{n}=\mathbf{H}_a\times\mathbf{n} \tag*{(18)}
(∇ϕ+∂A∂t)⋅n=0.(19)\left(\nabla\phi+\frac{\partial\mathbf{A}}{\partial t}\right)\cdot\mathbf{n}=0. \tag*{(19)}

It should be noted that γ(r)\gamma(\mathbf{r}) and s(r)s(\mathbf{r}) allow the local characteristic time relaxations to vary, which only affects the dynamics of the solutions. Where the time dynamics are relevant these parameters cannot be ignored; however, in many cases, the primary interest of TDGL calculations is in determining the energetic stability and critical fields, such as the superheating field. In these cases, the time dynamics are not relevant and γ\gamma and ss can be safely set to arbitrary constant values, such as unity.

Determining Spatial Variation of TDGL Parameters

In the previous section, we have shown how to introduce spatial variation to the TDGL parameters. The process of calculating the values of a(r)a(\mathbf{r}), b(r)b(\mathbf{r}), γ(r)\gamma(\mathbf{r}), and μ(r)\mu(\mathbf{r}) is where we bring in experiment and ab initio theory. From Eqs. 6-8, we know these parameters mostly depend on well-defined microscopic quantities, namely ν(0)\nu(0), TcT_c, and vfv_f. These quantities can be calculated using density-functional theory (DFT). Local densities of states and Fermi velocities are straightforward to compute in DFT, providing local values for ν(0)\nu(0) and vfv_f. Superconducting quantities such as TcT_c are calculated by applying Eliashberg theory within a DFT framework and directly calculating electron-phonon coupling from first principles. Experiment can then give information about the compositions of sample materials, and DFT calculations can determine the ν(0)\nu(0), vfv_f, and TcT_c associated with these compositions. Using these values in addition to estimates of the electron mean free path (which can be derived from DFT or can come from experimental characterizations), a(r)a(\mathbf{r}), b(r)b(\mathbf{r}), γ(r)\gamma(\mathbf{r}), and μ(r)\mu(\mathbf{r}) are calculated from Eqs. 6-9, and the material geometries from the experimental results determine the spatial variation.

In this paper, we will demonstrate our framework on Sn-deficient islands. These defects have been studied extensively [42, 43, 47, 48] so it is straightforward to find estimates of the general size of these islands and their material compositions in the literature. DFT has also been used to calculate ν(0)\nu(0), vfv_f, and TcT_c for Nb-Sn systems of different concentrations. In this paper, we use the DFT values from Ref. 53 to construct a(r)a(\mathbf{r}), b(r)b(\mathbf{r}), γ(r)\gamma(\mathbf{r}), and μ(r)\mu(\mathbf{r}) for islands of different Sn concentration. This paper will primarily focus on the results of our TDGL calculations, for further information regarding the details and results from the DFT calculations used in this paper, we refer the interested reader to Refs. 52 and 53.

Simulation Geometry and Numerical Implementation

The simulations in this paper were solved using a finite element formulation of the TDGL equations proposed by Gao [64], under the temporal gauge, which sets the scalar potential ϕ=0\phi=0 (see Du [65] for a more detailed overview of gauge choices in TDGL). All computations were carried out using the open-source finite element software FEniCS [66].

The simulation domain consists of a rectangular cuboid with periodic boundary conditions applied in the xx- and yy-directions, as indicated by the yellow and light blue highlights in Figure 1. An external magnetic field of constant magnitude HaH_a is applied in the xx-direction on the red surface, while no field is applied on the opposing green surface. For simulations that include Sn-deficient islands, the island geometry is also shown in Figure 1. These defects are modeled as ellipsoids with equal radii in the xx- and yy-directions and a zz-radius equal to half of the xx-radius. The island’s distance from the surface, denoted by dd, is measured from the surface to the outer edge of the ellipsoid.

Schematic of Simulation Geometry

Figure 1. Schematic of Simulation Geometry. Like-colored surfaces (yellow and light blue outlines) have periodic boundary conditions. The red and green surfaces indicate the faces where external magnetic field is applied; in simulations, a field is applied to the red surface in the xx-direction (indicated by the red arrow), with no field applied on the green surface. The Sn-deficient island is shown as a dark blue ellipsoid, with dotted lines indicating its projections into the xyxy, xzxz, and yzyz planes. The distance dd is measured from the surface to the outer edge of the ellipsoid.

All meshes were generated using the open-source mesh generation software Gmsh [67], with cubic tetrahedral elements. For simulations involving Sn-deficient islands, the OpenCASCADE geometry kernel in Gmsh was used to ensure that the mesh conforms to the ellipsoidal shape of the defects.

Dissipation in TDGL

When simulating SRF materials, dissipation is often a physical quantity of interest. Under TDGL, a formula for dissipation can be derived by considering the time derivative of the free energy and the free energy current flux density. A more detailed derivation is found in Ref. 63, but we quote the final result here:

D=2Γ∣∂ψ∂t+iesϕψℏ∣2+σnE2.(20)D=2\Gamma\left|\frac{\partial\psi}{\partial t}+\frac{ie_s\phi\psi}{\hbar}\right|^2+\sigma_n E^2. \tag*{(20)}

This quantity is a power density, with the first term corresponding to the superconducting dissipation arising from the relaxation of the order parameter. The second term is the dissipation of normal currents which are largest near the surface where magnetic field can still appreciably penetrate.

It is worth considering how this expression for the dissipation density is related to existing theories of RF power loss and surface resistance. The first term in Eq. (20) is associated with the dissipation due to the rate of change of ψ\psi. This term is typically small, except in the vortex state where it becomes the dominant source of dissipation. A dissipation of this form is similar to that proposed by Tinkham [68], who suggested the vortex dissipation should be proportional to

(∂ψ∂t)2\left(\frac{\partial\psi}{\partial t}\right)^2

The additional term within the parenthesis in Eq. (20) is a result of the gauge invariance of TDGL.

The second term in Eq. (20) can be directly related to the phenomenological “two-fluid model,” which was first proposed by Gorter and Casimir [69] in 1934 and was applied by London [70] later that year to calculate the power loss of a superconductor. The two-fluid model approximates the electrons of the system as consisting of two non-interacting ‘fluids’: the superconducting electrons, in the form of cooper pairs which carry lossless supercurrent, and the normal electrons, which exist as thermally excited quasiparticles that produce typical dissipative currents. Under the two-fluid model, the normal fluid losses produce dissipation of the form [71, 72]

P∝σnE2,P \propto\sigma_n E^2,

which is identical to the second term of Eq. (20). For AC applied currents, the electric field is proportional to the frequency and magnetic field, E∝ωHE \propto\omega H, meaning that overall the power loss will be of the form

P∼ω2H2.(21)P \sim\omega^2 H^2. \tag*{(21)}

It is also well known that within RF cavities, the power loss is given by

P=12∮ARs∣H(r)∣2 dA.(22)P = \frac{1}{2}\oint_A R_s \lvert H(r)\rvert^2\,\mathrm{d}A. \tag*{(22)}

where RsR_s is the surface resistance of the cavity walls and AA is the surface area. Comparing Eqs. (21) and (22), we have that the second term of Eq. (20) directly leads to a surface resistance proportional to the square of the frequency, Rs∝ω2R_s \propto\omega^2. In the late 1950s, Mattis and Bardeen [73] as well as Abrikosov, Gor’Kov, and Khalatnikov [74] independently derived the now well-known “BCS resistance.” Under the low frequency and low field limit, the BCS resistance reduces to the form [75, 76]

Rbcs≃ω2Te−ΔkBT(23)R_{\mathrm{bcs}} \simeq\frac{\omega^2}{T}e^{-\frac{\Delta}{k_B T}} \tag*{(23)}

where Δ=1.76kBTc\Delta= 1.76k_B T_c is the superconducting energy gap [63]. We thus see that our calculated expression for the surface resistance has the same ω2\omega^2 frequency dependence of the BCS prediction in the low frequency and field limit, though the quantitative values may differ.

Given this connection to surface resistance, we now consider how our dissipation estimates can be used to calculate the cavity quality factor, QQ, a standard figure of merit in SRF cavity performance.

Estimating Cavity Quality Factor

While TDGL allows us to estimate dissipation, it is important to emphasize that both the dissipation calculations and the quality factor estimates that follow lie well outside the regime of quantitative validity for the theory. In particular, TDGL is strictly valid only near TcT_c in the gapless limit, and its predictions for dissipation under RF-like dynamic fields at low temperatures should be interpreted with caution. Despite this, we believe the calculations presented here remain qualitatively valuable. They provide a means of linking mesoscopic-scale simulations to macroscopic cavity performance metrics, and enable relative comparisons between different material configurations that may inform experimental priorities.

A particularly relevant quantity that can be estimated from the dissipation is the cavity quality factor, QQ, which is given by

Q=2πEΔE.(24)Q = \frac{2\pi E}{\Delta E}. \tag*{(24)}

Here, EE is the energy stored in the cavity, and ΔE\Delta E is the energy dissipated in the cavity walls during each RF cycle. It is common to express the quality factor in terms of the surface resistance as

Q=GRs,(25)Q = \frac{G}{R_s}, \tag*{(25)}

where RsR_s is the cavity surface resistance and GG is a geometric factor that depends only on quantities which are determined by the cavity geometry. For a typical 1.3 GHz 9-cell Nb TESLA cavity, G=270 ΩG = 270\ \Omega [77]. The surface resistance is given by

Rs=μ0ωλ3H~a2LxLy(Iψ+ωω~σnμ0λ2IA),(26)R_s = \frac{\mu_0\omega\lambda^3}{\widetilde{H}_a^2 L_x L_y}\left(I_\psi+ \frac{\omega}{\widetilde{\omega}}\sigma_n\mu_0\lambda^2 I_A\right), \tag*{(26)}

where μ0\mu_0 is the permeability of free space, ω\omega is the cavity frequency, λ\lambda is the penetration depth, H~a\widetilde{H}_a is the maximum applied magnetic field value in simulation units, LxL_x and LyL_y are the size of a simulation domain in the X and Y directions respectively, σn\sigma_n is the normal conductivity, and ω~\widetilde{\omega} is the frequency of the applied field in simulation units. IψI_\psi and IAI_A are integrals over the squared time derivatives of ψ\psi and A\mathbf{A}:

Iψ≡∫dt~∫dx~∫dy~∫dz~∣∂ψ~∂t~∣2(27)I_\psi\equiv\int d\widetilde{t}\int d\widetilde{x}\int d\widetilde{y}\int d\widetilde{z}\left|\frac{\partial\widetilde{\psi}}{\partial\widetilde{t}}\right|^2 \tag*{(27)}
IA=∫dt~∫dx~∫dy~∫dz~(∂A∂t~)2(28)I_A = \int d\widetilde{t}\int d\widetilde{x}\int d\widetilde{y}\int d\widetilde{z}\left(\frac{\partial\mathbf{A}}{\partial\widetilde{t}}\right)^2 \tag*{(28)}

where the tilde variables denote ones which are in simulation units. A much more detailed derivation of these equations can be found in the Appendix.

The σn\sigma_n appearing in Eq. (26) refers specifically to the conductivity of the normal quasiparticles. In general, calculating this quantity from first principles is a complex task that depends sensitively on microscopic material properties, and is largely orthogonal to the rest of the quality factor calculation. For this reason, we choose to treat σn\sigma_n as a free parameter in our model, which allows us to explore how dissipation and quality factor vary across a range of plausible conductivity values without relying on potentially oversimplified assumptions. For completeness, we outline a possible approximate derivation of σn\sigma_n based on the Drude model in the Appendix, though we emphasize that this is intended primarily as a qualitative reference.

In the derivation of Equation 26, it is assumed that the cavity surface is partitioned into small fractional areas, and the dissipated energy is calculated over one of these areas, and then multiplied by the total number of them. If some defect is present in the simulation domain, it would mean that a QQ calculation based purely on that value would implicitly assume that such a defect is uniformly distributed over the surface of the cavity. In Equation 26, the only thing that changes when simulating a different material or different defect is the value of the quantity in parenthesis, every other part of the process for calculating QQ remains the same. This means that we can calculate a more realistic cavity surface resistance by calculating RsR_s with Equation 26 for a few different simulations, and then taking a weighted average of these values. Let RiR_i be the surface resistance of the iith simulation, and let pip_i be the percentage of the fractional areas partitioning the cavity surface which are represented by this simulation. Then

Rtot=∑ipiRi,(29)R_{\mathrm{tot}} = \sum_i p_i R_i, \tag*{(29)}

where ∑ipi=1\sum_i p_i = 1. The simplest application for Equation 29 is to perform two simulations, one baseline simulation with no defects or material inhomogeneity, and then another simulation containing some defect of interest. Choosing a value pp to represent the percentage of fractional areas containing the defect and then following through with the rest of the quality factor calculation provides a simple way to estimate QQ for different average defect densities by simply changing the value of pp.

The value of QQ calculated from TDGL outputs will typically be underestimated at low field. This is because of the assumption of gapless superconductivity, which results in higher surface resistances than is predicted with the BCS surface resistance (Equation (23)). Despite this, our approach still often predicts quality factors within an order of magnitude of the experimental values. Additionally, the relative behavior of QQ at different applied fields, especially when averaging the impact of multiple kinds of defects as described above, qualitatively captures effects such as high field QQ-slope. Nonetheless, we emphasize that this quality factor calculation is best treated as a qualitative tool; for quantitatively accurate predictions, more rigorous superconductivity theories should be used.

Validation Study

Sn-deficient Islands in Nb3_3Sn

There has been a considerable amount of work both experimentally and theoretically understanding the various defects present in Nb3_3Sn [19, 41, 42, 43, 44, 46, 47, 48] [45]. For the purposes of this paper, we will focus on Sn-deficient islands, small regions within Nb3_3Sn grains that have lower than expected Sn concentrations. To demonstrate the use of our sample specific TDGL framework, we will estimate the potential impact to SRF performance by determining the vortex penetration field, HvortH_{\mathrm{vort}}. This field represents the lowest applied field at which a vortex nucleates and generalizes the concept of the superheating field to surfaces containing defects. In the limit of a uniform, flat surface, HvortH_{\mathrm{vort}} is equal to the superheating field.

In order to simulate Sn-deficient islands, we require estimates of the material properties within these regions. Such estimates are provided in Ref. 53, which reports values of TcT_c and ν(0)\nu(0) for Nb-Sn systems at 18.7, 20.8, and 23.4 at.% Sn. Our vfv_f estimates for these compositions were obtained via private communication [78]. We first choose a Sn concentration of 18.7 at.% for our islands. Using the corresponding values of TcT_c and ν(0)\nu(0), we construct a(r)a(r), b(r)b(r), and γ(r)\gamma(r); the values of these quantities in the bulk as well as in the island are reported in Table I. The simulation geometry is the same as depicted in Figure 1, with the domain a cube of side length 10λ10\lambda. The applied field for these simulations is held constant.

QuantityBulk ValueIsland Value
TcT_c (K)186
ν(0)\nu(0) (states/(eV*nm3^3))101.3340.53
vfv_f (10510^5 m/s)1.251.4
ℓ\ell (nm)5.52.25
aa (unitless)10.15
bb (unitless)11.8
γ\gamma (unitless)11.2
ss (unitless)11
μ\mu (unitless)11.79

Table I. A summary of the TDGL parameter and material values used in the results of this paper. All TDGL parameters were calculated from Equations 6-8 using the higher order temperature dependence and normalized with respect to the bulk value. The TcT_c and ν(0)\nu(0) values were taken from Ref. 53 assuming a Sn concentration of 18% for the island. The vfv_f values came from private correspondence [78], and the bulk value of ℓ\ell was taken from Ref. 79, we set the island value to half of the bulk value. The normal electron conductivity, ss, was assumed to remain constant throughout the simulation domain.

Figure 2 depicts vortex nucleation occurring at HvortH_{vort} for a particular island. We find that varying the distance of the island from the surface between 0.1 and 2 penetration depths (i.e. between ∼10\sim10-200 nm for Nb3_3Sn), we observe a reduction in HvortH_{vort} by as much as ∼60%\sim60\% below the bulk value of the superheating field for islands very near the surface, as shown in Figure 3. We did this for several different island sizes, and found that the severity of this drop in HvortH_{vort} increases with increasing island size. As such, we conclude that large islands within 200 nm of the cavity surface are a potential cause for concern when it comes to SRF performance, particularly ones within 1-2 penetration depths of the surface. Figure 4 shows how these effects change with respect to the Sn at.% inside each island. We see that as the Sn deficiency becomes weaker, so does the impact on vortex penetration, though it is worth noting that even an island with 23.4% Sn (only 1.6% off a perfect stoichiometric ratio) has a ∼30%\sim30\% reduction in HvortH_{vort} for islands near the surface. Currently, Nb3_3Sn-coated SRF cavities are only achieving at most ∼25%\sim25\% of their theoretical maximum accelerating fields. There are likely many factors which contribute to this outcome, but these results indicate that Sn-deficient

Example of a Sn-deficient Island Simulation

Figure 2. Example of a Sn-deficient Island Simulation. The domain is a cube with side length 10λ10\lambda, this particular simulation is for an island 0.5λ0.5\lambda from the surface, with a XX and YY radius of 2λ2\lambda and a ZZ radius of 1λ1\lambda. The applied field is Ha2Hc=0.38\frac{H_a}{\sqrt{2}H_c}=0.38, which is the value of HvortH_{vort} for this island size and distance from the surface. (a) depicts a volume plot of ∣ψ∣2|\psi|^2 over the whole domain during a vortex nucleation event. (b) is a 2D slice in the XZ plane at yλ=5\frac{y}{\lambda}=5. (c) is the ∣ψ∣2=0.1|\psi|^2=0.1 isosurface, which shows the shape of the vortex as well as the island that induced the nucleation.

Vortex Penetration Field Versus Distance from Surface for Different Island Size

Figure 3. Vortex Penetration Field Versus Distance from Surface for Different Island Size. We plot our calculated vortex penetration field estimates, varying the distance of the Sn deficient island from the superconductor surface. The field values are reported in units of 2Hc\sqrt{2}H_c and the distances are in units of penetration depths. We did this for 3 different islands, with x radii of 0.5, 1, and 2 penetration depths. The black dotted line denotes the superheating field of bulk Nb3_3Sn. Smaller volume islands have a diminishing impact on HvortH_{\mathrm{vort}}.

Vortex Penetration Field Versus Distance from Surface for Different Island Sn %

Figure 4. Vortex Penetration Field Versus Distance from Surface for Different Island Sn %. We plot our calculated vortex penetration field estimates, varying the distance of the Sn deficient island from the superconductor surface. The field values are reported in units of 2Hc\sqrt{2}H_c and the distances are in units of penetration depths. We did this for 3 different island Sn percentages, 18.7%, 20.8%, and 23.4%. The black dotted line denotes the superheating field of bulk Nb3_3Sn.

islands are among the defects which contribute to SRF performance degradations.

Sources of Dissipation in Nb SRF Cavities

In the previous section, we focused on steady-state properties of TDGL in the dirty limit, where the theory is known to have broader quantitative validity. We now extend this framework to examine the two dissipation terms that naturally arise in TDGL simulations. While these calculations fall outside the strict regime of TDGL’s quantitative accuracy, particularly at the low temperatures and high frequencies relevant for SRF cavities, they remain qualitatively informative. In particular, they allow us to interpret distinct contributions to RF dissipation in a way that parallels the conventional decomposition of surface resistance into BCS and residual components. More broadly, they provide a means of linking mesoscopic TDGL behavior to macroscopic cavity losses, and of understanding how steady-state properties, such as HvortH_{\mathrm{vort}} for individual defects, can collectively influence large-scale phenomena like the high-field Q-slope.

It is standard in SRF literature to decompose surface resistance into two components: the BCS resistance, which arises from thermally excited quasiparticles, and a more poorly understood residual resistance, which dominates at low temperatures and is often attributed to trapped flux, impurities, or surface defects. In our TDGL framework, these two types of dissipation emerge natu-

rally. Specifically, in Eq. (26), the term proportional to σn\sigma_n (involving IAI_A) captures dissipation due to normal current response, and is the TDGL analog of the BCS resistance. The second term (IψI_\psi), which arises from relaxation of the superconducting order parameter, reflects dissipation mechanisms not captured by traditional two-fluid or BCS models and can be viewed as a proxy for residual resistance in our simulations. Studying the behavior and interplay of these two terms as a function of applied field and conductivity provides insight into which mechanisms dominate under different conditions.

To establish a baseline for comparison, we simulate a uniform Nb domain subjected to a sinusoidal applied field with period 2000τψ2000\tau_\psi. This choice is motivated by the fact that, for Nb at 2 K, we estimate τψ≈3.7×10−13 s\tau_\psi\approx3.7 \times10^{-13}\,\mathrm{s} using Eqs. (6) and (8) and a value of ν(0)=90\nu(0) = 90 states/(eV·nm3^3) [78], corresponding to a driving frequency of approximately 1.3 GHz—typical of elliptical SRF cavities. The simulation domain is a cube of side length 20λ20\lambda. In these calculations, we treat the normal conductivity σn\sigma_n as a free parameter, and define the dimensionless ratio σn/σrt\sigma_n/\sigma_{rt}, where σrt=6.7×106\sigma_{rt} = 6.7 \times10^6 S/m is the room-temperature conductivity of Nb.

Figure 5 plots the two dissipation terms as functions of applied field for several values of σn/σrt\sigma_n/\sigma_{rt}. The IAI_A term scales linearly with σn\sigma_n, while IψI_\psi is independent of conductivity and increases more steeply with field. Although we treat σn\sigma_n as a free parameter in these simulations, we are nominally modeling Nb, for which a physically meaningful value of σn\sigma_n exists. Rigorously determining this value at cryogenic temperatures and GHz frequencies lies beyond the scope of this paper, but physical intuition provides useful bounds: at low field, the BCS-like (i.e., IAI_A) term should dominate in clean Nb cavities, while at high field, particularly once vortices nucleate, the residual-like (i.e., IψI_\psi) term should become dominant. These trends imply that the true conductivity is unlikely to differ from σrt\sigma_{rt} by more than an order of magnitude in either direction. This observation motivates our use of σn=σrt\sigma_n=\sigma_{rt} in the simulations presented in the following section.

Dissipation contributions from current response and order parameter dynamics

Figure 5. Dissipation Contributions from Current Response and Order Parameter Dynamics. The two dissipation terms from Equation 26 are plotted separately as functions of applied field, for several values of the normal conductivity σn\sigma_n, shown here as ratios to the room-temperature conductivity σrt\sigma_{rt} of Nb. The term proportional to σn\sigma_n and involving IAI_A represents dissipation from normal current response, while the IψI_\psi term reflects dissipation due to relaxation of the superconducting order parameter. As field increases, IψI_\psi grows more rapidly and eventually overtakes IAI_A. The crossover point depends on the value of σn/σrt\sigma_n/\sigma_{rt}: for small values (e.g., 0.1), IψI_\psi dominates even at low field, while for large values (e.g., 10), the current-induced dissipation remains dominant until vortex nucleation occurs near the superheating field HshH_{sh}. While σn\sigma_n is varied here as a free parameter, this analysis helps illustrate how the relative importance of the two dissipation mechanisms depends on conductivity and field strength within the TDGL framework.

While our primary emphasis is on the dissipation terms themselves, we can also use them to estimate quality factors via Eqs. (25) and (26). As shown in Figure 6, the shape of the resulting QQ curves depends on the choice of σn/σrt\sigma_n/\sigma_{rt}. For moderate and large values (e.g., 1 or 10), the quality factor remains relatively flat at low fields and drops sharply near the superheating field—consistent with expectations for clean Nb cavities. In contrast, for small conductivity values (e.g., 0.1), QQ decreases too rapidly with field, indicating unrealistic dissipation behavior. Although lower σn\sigma_n values shift the low-field QQ magnitude closer to the experimental range of 101010^{10}–101110^{11}, this comes at the cost of distorting the overall field dependence. Due

Nb SRF cavity quality factor versus applied field

Figure 6. Nb SRF Cavity Quality Factor vs. Applied Field. A plot of quality factor for a uniform Nb simulation versus the applied field (in units of 2Hc\sqrt{2}H_c) for several different values of the normal conductivity, given by σn=σn1⋅σn′\sigma_n=\sigma_{n1}\cdot\sigma'_n, where σn1\sigma_{n1} is the room temperature conductivity of Nb. The quality factor is calculated using Eqs. (25) and (26) with G=270 ΩG=270\ \Omega. The calculation for several different values of σn′\sigma'_n change the low field value, and the shape of the curves, but has much less of an impact near HshH_{sh}. The dotted black line indicates the superheating field of Nb. Because it is above HshH_{sh}, vortex nucleation occurred for H~a=1.0\tilde{H}_a=1.0 which is the reason for the sharp drop in QQ.

to the gapless nature of TDGL, we generally expect QQ to be underestimated. While we do vary σn\sigma_n in these calculations, it is not used as a fitting parameter in the traditional sense—namely, we do not select σn\sigma_n to match the low-field QQ magnitude from experiment, as we see that doing so would yield qualitatively inaccurate results. Instead, we explore a range of values to identify physically plausible regimes and to qualitatively interpret the resulting dissipation behavior.

These reference simulations validate the dissipation decomposition and demonstrate how our framework reproduces key qualitative features of SRF performance. In the following section, we apply the same methodology to simulations of Sn-deficient islands in Nb3_3Sn to investigate how localized stoichiometric defects affect the balance between current-induced and order parameter–related dissipation. Building on this, we introduce an approach for estimating cavity-level losses by linking local vortex nucleation behavior to global quality factor trends.

Dissipation and Quality Factor for Nb3_3Sn cavities with Sn-deficient Islands

To explore how localized stoichiometric defects alter RF dissipation, we now apply our framework to a representative simulation of a Sn-deficient island in Nb3_3Sn. This case illustrates how such inhomogeneities can shift the balance between current-induced and order parameter–related losses.

Dissipation Terms for Sn-deficient Island Simulations

Figure 7. Dissipation Terms for Sn-deficient Island Simulations. Dissipation terms IψI_{\psi} (solid lines) and IAI_A (dotted lines) from Eq. (26) are plotted as functions of applied field for simulations with (blue) and without (red) a Sn-deficient island. The ellipsoidal island has in-plane radii of 0.5λ0.5\lambda, a vertical radius of 0.25λ0.25\lambda, and its top edge located 0.5λ0.5\lambda beneath the surface. In the island case, vortices nucleate at Ha≈0.525H_a \approx0.525, causing both dissipation terms—especially IψI_{\psi}—to increase sharply by several orders of magnitude. In contrast, the no-island simulation exhibits smoother behavior with more modest growth, and IψI_{\psi} exceeds IAI_A by a smaller factor (typically 2–5×) at high field. These results highlight how local vortex penetration dramatically enhances both forms of dissipation, with IψI_{\psi} becoming the dominant contribution.

To evaluate dissipation in the presence of Sn-deficient islands, we simulate a time-dependent applied field with period 1000τψ01000\tau_{\psi0}, corresponding to a frequency of approximately 5GHz. This is higher than the ∼1.3GHz typically used in Nb3_3Sn cavities, but was chosen to reduce simulation time for these computationally intensive cases. Each simulation required roughly five weeks of continuous runtime on a computing cluster, making longer periods prohibitively expensive. While quantitative estimates of dissipation will differ at lower frequency, we expect the qualitative features of the results to remain representative.

To manage computational cost, we model a single island with radius 0.5λ0.5\lambda located 0.5λ0.5\lambda beneath the surface, using a reduced domain size of 5λ×2.5λ×2.5λ5\lambda\times2.5\lambda\times2.5\lambda oriented to align the applied field and potential vortex motion along the xx-axis.

Figure 7 shows the dissipation terms IψI_{\psi} and IAI_A for simulations with and without a Sn-deficient island. Below the vortex penetration field (Hvort≈0.525H_{vort} \approx0.525), both simulations yield nearly identical dissipation, indicating that embedded islands have minimal effect in the Meissner state. Once HaH_a exceeds HvortH_{vort}, however, the island simulation exhibits an abrupt and sustained increase in both dissipation terms—particularly in IψI_{\psi}, which becomes the dominant contribution and rises by several orders of magnitude. The dissipation values above HvortH_{vort} also appear somewhat noisy, which we attribute to numerical variations in vortex entry and dynamics. This binary behavior—where dissipation remains low until vortex entry—appears characteristic of embedded Sn-deficient regions in Nb3_3Sn. Whether other types of defects, such as surface-connected grain boundaries, show similar trends remains an open question. Exploring such cases is a clear next direction for future studies.

Building on this result, we develop an approximate method for estimating how a distribution of Sn-deficient islands could collectively affect the quality factor. The dissipation behavior observed in Figure 7 suggests that embedded islands have negligible impact below their respective vortex penetration fields, but trigger a sharp increase in dissipation once vortices begin to nucleate. This motivates a two-state approximation: for each island geometry, we assume dissipation follows the defect-free case up to HvortH_{vort}, and then transitions abruptly to constant, elevated values beyond that point. For the vortex state, we use fixed dissipation values of Iψ=2I_{\psi}=2 and ωξ0σnμ0λ2⋅IA=0.1\frac{\omega}{\xi_0} \sigma_n \mu_0 \lambda^2 \cdot I_A=0.1, based on the average values from the simulation in Figure 7.

Using this approximation, we construct dissipation profiles for each of the simulated island geometries shown in Figure 3, focusing on the cases with Sn concentration 18.7%. For each defect, the dissipation follows the defect-free simulation up to its corresponding vortex penetration field HvortH_{vort}, and then transitions to constant values of Iψ=2I_{\psi}=2 and ωξ0σnμ0λ2⋅IA=0.1\frac{\omega}{\xi_0} \sigma_n \mu_0 \lambda^2 \cdot I_A=0.1 once vortices nucleate. To estimate the aggregate dissipation and resulting quality factor, we assign weights to each defect based on an assumed distribution and compute a weighted average of the dissipation terms. For the red curve in Figure 8, we assume all three island sizes are equally likely and that the depth dd from the surface to the outer edge of the island follows an exponential distribution p(d)∝e−dp(d) \propto e^{-d}, normalized so the total probability sums to unity.

To evaluate the hypothetical benefit of eliminating large, near-surface defects, we also construct the blue curve in Figure 8 using a modified piecewise distribution. In this case, the probability p(d)p(d) is defined as

p(d)∝{0for r=0.5λ and d≤0.1λ0for r=λ or 2λ and d<λe−dotherwisep(d) \propto \begin{cases} 0 & \text{for } r=0.5\lambda\text{ and } d \le0.1\lambda\\ 0 & \text{for } r=\lambda\text{ or } 2\lambda\text{ and } d < \lambda\\ e^{-d} & \text{otherwise} \end{cases}

and the resulting weights are renormalized to ensure they sum to one. Comparing the red and blue curves illustrates how removing such high-impact defects can delay the onset of increased dissipation and improve the high-field behavior of the cavity.

While the estimated quality factors in Figure 8 provide valuable qualitative insight, they should not be interpreted quantitatively. The low-field QQ magnitude in our simulations is just under 10810^{8}, significantly lower than the ∼1010\sim10^{10} typically observed in Nb3_{3}Sn cavities. This discrepancy is primarily due to the higher simulation frequency (∼5\sim5 GHz versus 1.3 GHz in practice), though the quantitative limitations of TDGL—such as the assumption of gapless superconductivity and its reduced validity far below TcT_{c}—also contribute. Furthermore, we approximate the dissipation above HvortH_{vort} as constant, which neglects the increasing losses and possible thermal effects that would realistically occur as more vortices enter. In practice, such heating could trigger early cavity quench, while in our simulations the dissipation simply continues to increase with field without ever initiating a quench. Despite these simplifications, the structure of the estimated quality factors still captures important qualitative features.

Estimated quality factor curves based on weighted dissipation from different Sn-deficient islands

Figure 8. Estimated quality factor curves based on weighted dissipation from different Sn-deficient islands. The black curve shows the quality factor for a defect-free cavity. The red curve represents a hypothetical quality factor obtained by averaging dissipation profiles from multiple Sn-deficient island geometries, each transitioning to fixed dissipation values of Iψ=2I_{\psi}=2 and σnμ0λ2ξIA=0.1\frac{\sigma_{n}\mu_{0}\lambda^{2}}{\xi}I_{A}=0.1 above their respective HvortH_{vort}. The weighting assumes all island sizes are equally likely and that the depth dd from the surface to the island edge follows an exponential distribution, p(d)∝e−dp(d)\propto e^{-d}, normalized so the weights sum to one. The blue curve uses the same assumptions, except that p(d)=0p(d)=0 for the r=0.5λr=0.5\lambda island when d≤0.1λd\leq0.1\lambda, and for the r=λr=\lambda and r=2λr=2\lambda islands when d<λd<\lambda. All remaining weights are re-normalized so the distribution still sums to unity. Removing these large, near-surface defects in the model delays the onset of dissipation and improves high-field performance.

In particular, the onset of high-field QQ-slope in these plots arises directly from vortex nucleation, a steady-state property with broader validity in the dirty limit where TDGL is more appropriate. While the precise drop in QQ is uncertain, the onset of declining QQ in our

aggregated curves is governed by the lowest vortex penetration field (HvortH_{vort}) among the defect types with nonzero probability, highlighting how vortex entry from local defects could plausibly contribute to high-field QQ-slope in at least some Nb3_{3}Sn cavities. More broadly, our results demonstrate a mechanism in which HFQS can emerge from the collective behavior of many small defects. Each defect follows a binary dissipation pattern—remaining nearly inert below HvortH_{vort} and sharply increasing in loss once vortex nucleation occurs—and as the field increases, more defects transition into this high-loss state. The aggregated effect produces a decline in QQ that qualitatively resembles the high-field QQ-slope observed experimentally. While this is unlikely to be the sole mechanism underlying HFQS, it offers a plausible explanation for how distributed sub-surface defects may contribute to performance degradation in certain cavities, as we elaborate further in the conclusion.

Conclusion

In this paper, we have demonstrated a framework for incorporating sample-specific information into time-dependent Ginzburg–Landau (TDGL) simulations of superconducting radio-frequency (SRF) cavities. While TDGL is inherently suited for mesoscopic modeling, its application to non-uniform materials often involves arbitrary parameter choices. Our approach addresses this gap by linking TDGL parameters to well-defined microscopic material properties, allowing simulations to reflect characteristics obtained from density functional theory (DFT) or experimental measurements. This enables more realistic modeling of sample-specific features while preserving the computational tractability of TDGL on mesoscopic domains.

In addition to the sample-specific framework, we introduced a method for estimating dissipation and quality factor directly from TDGL simulations. While these estimates are subject to limitations—such as TDGL’s quantitative inaccuracy far below TcT_{c} and the computational necessity of simulating at higher frequencies—they offer a valuable qualitative connection between mesoscopic defect behavior and macroscopic performance degradation. By decomposing dissipation into physically meaningful components and linking vortex nucleation to high-field QQ-slope, our approach provides insight into how distributed material inhomogeneities may impact SRF cavity performance. This type of dissipation-based analysis remains relatively uncommon in the TDGL literature and may serve as a useful tool for future studies aiming to bridge simulation and experiment.

We applied our framework to model Sn-deficient islands in Nb3_{3}Sn and found that they can reduce the vortex penetration field by as much as 60% when located near the surface. This suggests that such sub-surface defects may play a role in limiting the achievable accelerating gradients in Nb3_{3}Sn cavities, particularly given that experimentally observed gradients remain well below the theoretical maximum. To explore this further, we computed the dissipation for representative island configurations and observed sharp increases in both dissipation terms above HvortH_{vort} due to vortex-induced losses. These effects are expected to be even more pronounced for larger or more exposed defects, and in future simulations that incorporate thermal feedback to model cavity quenching. Our method can also be readily extended to other defect types, such as grain boundaries, which are widely believed to impact cavity performance.

To illustrate how these mesoscopic effects might collectively influence macroscopic cavity behavior, we constructed hypothetical quality factor curves by assigning defect-specific HvortH_{vort} thresholds and aggregating their dissipation using simple assumptions about defect size and depth distributions. The resulting QQ curves qualitatively resemble the high-field QQ-slope observed in experimental measurements, offering a plausible explanation in which HFQS emerges from the collective activation of many small, embedded defects. This analysis highlights the potential of TDGL-based simulations to inform defect mitigation strategies and guide the development of higher-performance SRF materials.

A recent experimental study by Viklund et al. [80] found that applying centrifugal barrel polishing (CBP) to Nb3_3Sn SRF cavities decreased the overall quality factor, lowered the field at which QQ-slope began, and reduced the quench field. After an additional Sn vapor deposition, however, the cavity recovered its previous QQ performance and reached a higher quench field. Our results offer a potential explanation for these findings. CBP is known to smooth the cavity surface but also removes surface material, which may expose previously buried Sn-deficient islands. As shown in our simulations, bringing such defects closer to the surface significantly enhances their dissipation and lowers HvortH_{vort}, potentially triggering early vortex entry, high-field QQ-slope, and premature quenching. The follow-up Sn coating likely fills in these exposed islands, eliminating the source of additional dissipation. This healing process restores QQ and enables higher accelerating gradients, consistent with our finding that suppressing near-surface Sn-deficient regions improves both quality factor and vortex-related performance limits.

Whenever TDGL equations are used to model experimental systems, it is important to consider their limitations. The standard TDGL formalism is derived under the assumption of gapless superconductivity, since a gapped density of states introduces a singularity that precludes expansion in powers of the energy gap [81]. Additionally, the equations are only quantitatively valid near the superconducting critical temperature. The former limitation can be addressed by using a generalized TDGL formulation developed by Kramer and Watts-Tobin [82], which extends validity to gapped superconductors. However, several studies have observed conditions in SRF cavities that support the use of conventional TDGL. For example, Proslier et al. [83] reported a broadened density of states at the surface of Nb cavities due to oxide layers, resulting in gapless surface superconductivity. Further work by Gurevich and Kubo[76, 84] demonstrated that typical material compositions and SRF operating conditions often produce a broadened density of states and suppressed energy gap, reinforcing the relevance of TDGL for SRF applications. Nevertheless, such gapless behavior may come at a cost: broader density of states near the surface could contribute to increased dissipation and reduced quality factors.

The methods presented here for simulating realistic, sample-specific defects and estimating their impact on quality factor provide a new framework for linking microscopic material features to macroscopic SRF performance. While the underlying simulations are conceptually straightforward, their successful application depends on two critical inputs: accurate theoretical models of a material’s microscopic properties, and detailed experimental characterization of the defects present in real samples. For this reason, our method is best suited to research contexts where both inputs are available—either in well-studied materials systems or through interdisciplinary collaborations like those found in the NSF Center for Bright Beams, which helped motivate this work. SRF research is particularly well positioned for such integration. SRF cavities are inherently macroscopic devices whose performance depends sensitively on microscopic structure, yet direct experimental probing of such features is often infeasible. The approach we describe enables mesoscopic-scale simulations rooted in real material data, offering a valuable tool for designing and optimizing future generations of SRF cavities.

Acknowledgement

This work was supported by the US National Science Foundation under Award OIA-1549132, the Center for Bright Beams. We would like to thank Dr. Nathan Sitaraman for helpful conversations and estimates of the Fermi level density of states and Fermi velocities, and Dr. Michelle Kelley for helpful comments about early drafts of the methods section.

TDGL Non-dimensionalization

To nondimensionalize the TDGL equations, we start with Eqs. 1 and 2, and make the following coordinate transformations:

∇⟶1λ0∇~\nabla\longrightarrow\frac{1}{\lambda_0}\widetilde{\nabla}
∂∂t⟶1τψ0∂∂t~\frac{\partial}{\partial t} \longrightarrow\frac{1}{\tau_{\psi0}}\frac{\partial}{\partial\widetilde{t}}
A⟶2Hc0λ0A~\mathbf{A} \longrightarrow\sqrt{2}H_{c0}\lambda_0\widetilde{\mathbf{A}}
ψ⟶∣α0∣β0ψ~\psi\longrightarrow\sqrt{\frac{|\alpha_0|}{\beta_0}}\widetilde{\psi}
ϕ⟶ϕ0ϕ~.\phi\longrightarrow\phi_0\widetilde{\phi}.

If we substitute in Eqs. 10, 11, 12, and 14 for α\alpha, β\beta, Γ\Gamma, and σn\sigma_n respectively, we can then define the quantities:

λ0=msc2β04πes2∣α0∣\lambda_0 = \sqrt{\frac{m_sc^2\beta_0}{4\pi e_s^2|\alpha_0|}}
ξ0=ℏ22ms∣α0∣\xi_0 = \sqrt{\frac{\hbar^2}{2m_s|\alpha_0|}}
κ0=λ0ξ0\kappa_0 = \frac{\lambda_0}{\xi_0}
Hc0=4πα02β0H_{c0} = \sqrt{\frac{4\pi\alpha_0^2}{\beta_0}}
τψ0=Γ0∣α0∣\tau_{\psi0} = \frac{\Gamma_0}{|\alpha_0|}
τj0=σn0msβ0es2∣α0∣\tau_{j0} = \frac{\sigma_{n0}m_s\beta_0}{e_s^2|\alpha_0|}
u0=τψ0τj0u_0 = \frac{\tau_{\psi0}}{\tau_{j0}}
ϕ0=ℏκ0esτψ0.\phi_0 = \frac{\hbar\kappa_0}{e_s\tau_{\psi0}}.

Using these relations, the resulting equations under the above coordinate transformations simplify into Eqs. 15 and 16 (where we then drop the tildes).

Approximate Estimation of Normal-State Conductivity

We outline here a possible route to estimate σn\sigma_n using the Drude model [85], where the electrical conductivity is given by

σ=ne2τm,(30)\sigma= \frac{ne^2\tau}{m}, \tag*{(30)}

with ee and mm the electron charge and mass, nn the carrier density, and τ\tau the mean free collision time. For the normal quasiparticle density at low temperatures, we adopt the approximation from Ref. 86,

nn=8ne−Δ/kbT,(31)n_n = 8ne^{-\Delta/k_bT}, \tag*{(31)}

which leads to an expression for the normal-state conductivity in the Meissner regime:

σn=ne2τm(8e−Δ/kbT).(32)\sigma_n = \frac{ne^2\tau}{m}\left(8e^{-\Delta/k_bT}\right). \tag*{(32)}

We emphasize that this expression is approximate, and moreover that the energy gap Δ\Delta is inferred from the TDGL order parameter ψ\psi, which is known to overestimate Δ\Delta at low temperatures [87]. Given these limitations, we regard this estimation primarily as a qualitative reference and have proceed with treating σn\sigma_n as a free parameter in the main text.

Quality Factor Derivation

We start with the quality factor:

Q=2πEΔE.(33)Q = \frac{2\pi E}{\Delta E}. \tag*{(33)}

These quantities (working in SI units for this section) can be expressed as integrals:

E=12μ0∫VdVH2(34)E = \frac{1}{2}\mu_0\int_V dV\mathbf{H}^2 \tag*{(34)}
ΔE=∫0Tdt∫VsurfdVsurfD,(35)\Delta E = \int_0^T dt\int_{V_{\mathrm{surf}}}dV_{\mathrm{surf}}D, \tag*{(35)}

where VV is the cavity volume, TT is the RF period, and DD is given by Equation 20. VsurfV_{\mathrm{surf}} is the volume in the first few penetration depths of the cavity surface where essentially all of the dissipation occurs. TDGL simulation outputs are unit-free, so it is helpful to pull constants with units out of these integrals, leaving behind dimensionless functions which can be calculated from TDGL solutions. We start by expressing Equation A.2 in cylindrical coordinates:

E=12μ0∫r dr∫dϕ∫dzH2E = \frac{1}{2}\mu_0\int r\,dr\int d\phi\int dz\mathbf{H}^2

We then define some dimensionless quantities:

r~=rR(36)\widetilde{r} = \frac{r}{R} \tag*{(36)}
z~=zL(37)\widetilde{z} = \frac{z}{L} \tag*{(37)}
H~=HHa(38)\widetilde{\mathbf{H}} = \frac{\mathbf{H}}{H_a} \tag*{(38)}

Where RR is the maximum radius of the cavity, LL is the length of the cavity in the axial direction, and HaH_a is the maximum value of the applied field at the surface of the cavity during an RF period. These quantities allow the definition of a unit-less integral that only depends on the cavity geometry:

IH≡∫r~dr~∫dz~H~2(39)I_H \equiv\int\widetilde{r}d\widetilde{r}\int d\widetilde{z}\widetilde{\mathbf{H}}^2 \tag*{(39)}

Using these definitions with Equation A.2 and assuming that H\mathbf{H} has azimuthal symmetry results in

E=πμ0Ha2LR2IH.(40)E = \pi\mu_{0} H_{a}^{2} L R^{2} I_{H}. \tag*{(40)}

Turning to the dissipated energy integral, suppose all of the dissipation occurs within a distance dd below the cavity surface, where d≪Rd \ll R. This allows the cylindrical integral to be converted into cartesian coordinates, with the azimuthal direction becoming the new xx direction, the axial direction becoming the new yy direction, and the radial direction becoming the new zz direction. A diagram of these transformations is found in Figure 9. With these transformations, we have

Schematic of transformations for the quality factor calculation

Figure 9. Schematic of Transformations for the Quality Factor Calculation. The cylindrical geometry is shown on the left, with the cavity radius RR and length LL depicted, and the coordinate directions, r^\hat{r}, ϕ^\hat{\phi}, and z^\hat{z}. Under the transformation (on the right) the coordinates become cartesian, with the r^\hat{r} direction becoming the new z^\hat{z} direction, the ϕ^\hat{\phi} direction becomes the x^\hat{x} direction, and the old z^\hat{z} direction becomes the y^\hat{y} direction.

ΔE=∫0Tdt∫02πRdx∫0Ldy∫0ddzD.(41)\Delta E = \int_{0}^{T} dt \int_{0}^{2\pi R} dx \int_{0}^{L} dy \int_{0}^{d} dz D. \tag*{(41)}

When calculating this from simulation outputs, the integral is necessarily calculated over a small region of the overall cavity surface. Let LxL_{x} and LyL_{y} be the simulation domain size in the xx and yy directions respectively, and let NN be the total number of simulation areas needed to fully partition the cavity surface. Then the dissipation integral becomes

ΔE=N∫0Tdt∫0Lxdx∫0Lydy∫0ddzD,(42)\Delta E = N \int_{0}^{T} dt \int_{0}^{L_{x}} dx \int_{0}^{L_{y}} dy \int_{0}^{d} dz D, \tag*{(42)}

and NN can be approximated as

N=2πRLLxLy.(43)N = \frac{2\pi R L}{L_{x}L_{y}}. \tag*{(43)}

Continuing as before we again define dimensionless coordinates:

x~=xλ(44)\tilde{x} = \frac{x}{\lambda} \tag*{(44)}
y~=yλ(45)\tilde{y} = \frac{y}{\lambda} \tag*{(45)}
z~=zλ(46)\tilde{z} = \frac{z}{\lambda} \tag*{(46)}
t~=TsimTt(47)\tilde{t} = \frac{T_{\mathrm{sim}}}{T}t \tag*{(47)}

where λ\lambda is the penetration depth and TsimT_{\mathrm{sim}} is the period in units of simulation time. These convert the integral to

ΔE=λ3TTsimN∫0Tsimdt~∫0Lx/λdx~∫0Ly/λdy~∫0d/λdz~D.(48)\Delta E = \lambda^{3}\frac{T}{T_{\mathrm{sim}}}N \int_{0}^{T_{\mathrm{sim}}}d\tilde{t}\int_{0}^{L_{x}/\lambda}d\tilde{x}\int_{0}^{L_{y}/\lambda}d\tilde{y}\int_{0}^{d/\lambda}d\tilde{z}D. \tag*{(48)}

Additionally, under the temporal gauge (ϕ=0\phi= 0), Equation (20) can be expressed as

D=2μ0Hc2TsimT(∣∂ψ~∂t~∣2+σnμ0λ2TsimT(∂A~∂t~)2),(49)D = 2\mu_{0}H_{c}^{2}\frac{T_{\mathrm{sim}}}{T}\left(\left|\frac{\partial\tilde{\psi}}{\partial\tilde{t}}\right|^{2}+\sigma_{n}\mu_{0}\lambda^{2}\frac{T_{\mathrm{sim}}}{T}\left(\frac{\partial\tilde{\mathbf{A}}}{\partial\tilde{t}}\right)^{2}\right), \tag*{(49)}

where ψ~\tilde{\psi} and A~\tilde{\mathbf{A}} are the unit-free versions of the vector potential and order parameter that are solved for with Eqs. (15) and (16) (a derivation of Equation A.17 can be found in the next section of the Appendix). Finally, we define some more dimensionless integrals over the TDGL solutions:

Iψ≡∫0Tsimdt~∫0Lx/λdx~∫0Ly/λdy~∫0d/λdz~∣∂ψ~∂t~∣2(50)I_{\psi} \equiv\int_{0}^{T_{\mathrm{sim}}}d\tilde{t}\int_{0}^{L_{x}/\lambda}d\tilde{x}\int_{0}^{L_{y}/\lambda}d\tilde{y}\int_{0}^{d/\lambda}d\tilde{z}\left|\frac{\partial\tilde{\psi}}{\partial\tilde{t}}\right|^{2} \tag*{(50)}
IA≡∫0Tsimdt~∫0Lx/λdx~∫0Ly/λdy~∫0d/λdz~(∂A~∂t~)2(51)I_{A} \equiv\int_{0}^{T_{\mathrm{sim}}}d\tilde{t}\int_{0}^{L_{x}/\lambda}d\tilde{x}\int_{0}^{L_{y}/\lambda}d\tilde{y}\int_{0}^{d/\lambda}d\tilde{z}\left(\frac{\partial\tilde{\mathbf{A}}}{\partial\tilde{t}}\right)^{2} \tag*{(51)}

Combining everything and noting that ω=2πT\omega= \frac{2\pi}{T}, we get

ΔE=2μ0Hc2λ32πRLLxLy(Iψ+ωσnμ0λ2Tsim2πIA).(52)\Delta E = 2\mu_{0}H_{c}^{2}\lambda^{3}\frac{2\pi R L}{L_{x}L_{y}}\left(I_{\psi}+\omega\frac{\sigma_{n}\mu_{0}\lambda^{2}T_{\mathrm{sim}}}{2\pi}I_{A}\right). \tag*{(52)}

Now using Equations A.1, A.8, and A.20 we get an expression for the quality factor,

Q=H~a2RLxLyIH2λ3(Iψ+ωσnμ0λ2Tsim2πIA),(53)Q = \frac{\tilde{H}_{a}^{2}R L_{x}L_{y}I_{H}}{2\lambda^{3}\left(I_{\psi}+\omega\frac{\sigma_{n}\mu_{0}\lambda^{2}T_{\mathrm{sim}}}{2\pi}I_{A}\right)}, \tag*{(53)}

where H~a≡Ha2Hc\tilde{H}_{a} \equiv\frac{H_{a}}{\sqrt{2}H_{c}} is the applied field in simulation units. It is common to express the quality factor as

Q=GRs,(54)Q = \frac{G}{R_{s}}, \tag*{(54)}

where RsR_{s} is the cavity surface resistance and GG is a geometric factor that depends only on quantities which are determined by the cavity geometry. We can define these quantities under the framework we have presented as

G=12μ0ωRIH(55)G = \frac{1}{2}\mu_0\omega R I_H \tag*{(55)}
Rs=μ0ωλ3H~a2LxLy(Iψ+ωσnμ0λ2Tsim2πIA).(56)R_s = \frac{\mu_0\omega\lambda^3}{\widetilde{H}_a^2L_xL_y}\left(I_\psi+\omega\frac{\sigma_n\mu_0\lambda^2T_{\mathrm{sim}}}{2\pi}I_A\right). \tag*{(56)}

For a typical 1.3 GHz 9-cell Nb TESLA cavity, G=270G=270 Ω\Omega [77], so in practice we can just use this value or other known values of GG, and only calculate RsR_s from Equation A.24.

Nondimensionalizing the TDGL Dissipation

We start with Equation (20),

D=2Γ[∣∂ψ∂t+iesϕψℏ∣2]+σnE2D=2\Gamma\left[\left|\frac{\partial\psi}{\partial t}+\frac{ie_s\phi\psi}{\hbar}\right|^2\right]+\sigma_n\mathbf{E}^2

choosing the temporal gauge (ϕ=0\phi=0) we have

D=2Γ∣∂ψ∂t∣2+σn(∂A∂t)2.D=2\Gamma\left|\frac{\partial\psi}{\partial t}\right|^2+\sigma_n\left(\frac{\partial\mathbf{A}}{\partial t}\right)^2.

Next, we make the same coordinate transformations as from the previous section (and the same time transformation as from the methods section) and use the expressions for τψ\tau_\psi and HcH_c on the first term:

D=2ΓαβTsim2T2∣∂ψ~∂t~∣2+2σnHc2λ2Tsim2T2(∂A~∂t~)2=2τψα2βTsim2T2∣∂ψ~∂t~∣2+2σnHc2λ2Tsim2T2(∂A~∂t~)2=2Hc2Tsim4πT∣∂ψ~∂t~∣2+2σnHc2λ2Tsim2T2(∂A~∂t~)2,\begin{aligned} D=\frac{2\Gamma\alpha}{\beta}\frac{T_{\mathrm{sim}}^2}{T^2}\left|\frac{\partial\widetilde{\psi}}{\partial\widetilde{t}}\right|^2+2\sigma_nH_c^2\lambda^2\frac{T_{\mathrm{sim}}^2}{T^2}\left(\frac{\partial\widetilde{A}}{\partial\widetilde{t}}\right)^2 \\ &=\frac{2\tau_\psi\alpha^2}{\beta}\frac{T_{\mathrm{sim}}^2}{T^2}\left|\frac{\partial\widetilde{\psi}}{\partial\widetilde{t}}\right|^2+2\sigma_nH_c^2\lambda^2\frac{T_{\mathrm{sim}}^2}{T^2}\left(\frac{\partial\widetilde{A}}{\partial\widetilde{t}}\right)^2 \\ &=\frac{2H_c^2T_{\mathrm{sim}}}{4\pi T}\left|\frac{\partial\widetilde{\psi}}{\partial\widetilde{t}}\right|^2+2\sigma_nH_c^2\lambda^2\frac{T_{\mathrm{sim}}^2}{T^2}\left(\frac{\partial\widetilde{A}}{\partial\widetilde{t}}\right)^2, \end{aligned}

where in the last line we used the fact that T=τψTsimT=\tau_\psi T_{\mathrm{sim}}. Finally, this expression is in Gaussian units so we convert to SI units so that it is compatible with the other expressions in Section II E:

D=2τψμ0Hc2Tsim2T2∣∂ψ~∂t~∣2+2σnμ02Hc2λ2Tsim2T2(∂A~∂t~)2=2μ0Hc2TsimT(∣∂ψ~∂t~∣2+σnμ0λ2TsimT(∂A~∂t~)2).\begin{aligned} D=2\tau_\psi\mu_0H_c^2\frac{T_{\mathrm{sim}}^2}{T^2}\left|\frac{\partial\widetilde{\psi}}{\partial\widetilde{t}}\right|^2+2\sigma_n\mu_0^2H_c^2\lambda^2\frac{T_{\mathrm{sim}}^2}{T^2}\left(\frac{\partial\widetilde{A}}{\partial\widetilde{t}}\right)^2 \\ &=2\mu_0H_c^2\frac{T_{\mathrm{sim}}}{T}\left(\left|\frac{\partial\widetilde{\psi}}{\partial\widetilde{t}}\right|^2+\sigma_n\mu_0\lambda^2\frac{T_{\mathrm{sim}}}{T}\left(\frac{\partial\widetilde{A}}{\partial\widetilde{t}}\right)^2\right). \end{aligned}

References

  1. [1]S. Posen and M. Liepe, Advances in development of nb₃Sn superconducting radio-frequency cavities, Phys. Rev. ST Accel. Beams 17, 112001 (2014).DOI
  2. [2]S. Posen and D. L. Hall, nb₃Sn superconducting radiofrequency cavities: fabrication, results, properties, and prospects, Superconductor Science and Technology 30, 033004 (2017).DOI
  3. [3]A. Gurevich, Theory of rf superconductivity for resonant cavities, Superconductor Science and Technology 30, 034004 (2017).DOI
  4. [4]Z. Sun, T. Os̄eroff, Z. Baraissov, D. K. Dare, K. Howard, B. Francis, A. C. Hire, N. Sitaraman, T. A. Arias, M. K. Transtrum, R. Hennig, M. O. Thompson, D. A. Muller, and M. U. Liepe, ZrNb(Co) for superconducting thin film with high critical temperature in the theoretical limit, Advanced Electronic Materials 9, 2300151 (2023).DOI
  5. [5]N. S. Sitaraman, Z. Sun, B. L. Francis, A. C. Hire, T. Oseroff, Z. Baraissov, T. A. Arias, R. G. Hennig, M. U. Liepe, D. A. Muller, and M. K. Transtrum (Center for Bright Beams), Enhanced surface superconductivity of niobium by zirconium doping, Phys. Rev. Appl. 20, 014064 (2023).DOI
  6. [6]J. Bardeen and M. J. Stephen, Theory of the motion of vortices in superconductors, Phys. Rev. 140, A1197 (1965).DOI
  7. [7]D. B. Liarte, S. Posen, M. K. Transtrum, G. Catelani, M. Liepe, and J. P. Sethna, Theoretical estimates of maximum fields in superconducting resonant radio frequency cavities: stability theory, disorder, and laminates, Superconductor Science and Technology 30, 033002 (2017).DOI
  8. [8]S. Posen, N. Valles, and M. Liepe, Radio frequency magnetic field limits of nb and nb₃Sn, Phys. Rev. Lett. 115, 047001 (2015).pubmed.ncbi.nlm.nih.gov/26252705
  9. [9]L. Kramer, Stability limits of the meissner state and the mechanism of spontaneous vortex nucleation in superconductors, Phys. Rev. 170, 475 (1968).DOI
  10. [10]A. J. Dolgert, S. J. Di Bartolo, and A. T. Dorsey, Superheating fields of superconductors: Asymptotic analysis and numerical results, Phys. Rev. B 53, 5650 (1996).pubmed.ncbi.nlm.nih.gov/9984174
  11. [11]M. K. Transtrum, G. Catelani, and J. P. Sethna, Superheating field of superconductors within ginzburg-landau theory, Phys. Rev. B 83, 094505 (2011).DOI
  12. [12]D. B. Liarte, M. K. Transtrum, and J. P. Sethna, Ginzburg-landau theory of the superheating field anisotropy of layered superconductors, Phys. Rev. B 94, 144504 (2016).DOI
  13. [13]G. Catelani and J. P. Sethna, Temperature dependence of the superheating field for superconductors in the high-κ london limit, Phys. Rev. B 78, 10.1103/PhysRevB.78.224509 (2008), cited by: 53; All Open Access, Green Open Access.DOI
  14. [14]F. P.-J. Lin and A. Gurevich, Effect of impurities on the superheating field of type-ii superconductors, Phys. Rev. B 85, 10.1103/PhysRevB.85.054513 (2012), cited by: 40; All Open Access, Green Open Access.
  15. [15]T. Kubo, Superheating fields of semi-infinite superconductors and layered superconductors in the diffusive limit: structural optimization based on the microscopic theory, Superconductor Science and Technology 34, 045006 (2021).DOI
  16. [16]S. Posen, J. Lee, D. Seidman, A. Romanenko, B. Tennis, O. Melnychuk, and D. Sergatskov, Advances in nb3Sn superconducting radiofrequency cavities towards first practical accelerator applications, Superconductor Science and Technology 34, 10.1088/1361-6668/abc7f7 (2021), cited by: 19; All Open Access, Green Open Access, Hybrid Gold Open Access.DOI
  17. [17]G. Müller, H. Piel, J. Pouryamout, P. Boccard, and P. Kneisel, Proceedings of the workshop on thin film coating methods for superconducting accelerating cavities (2000).
  18. [18]A. R. Pack, J. Carlson, S. Wadsworth, and M. K. Transtrum, Vortex nucleation in superconductors within time-dependent ginzburg-landau theory in two and three dimensions: Role of surface defects and material inhomogeneities, Phys. Rev. B 101, 144504 (2020).DOI
  19. [19]J. Carlson, A. Pack, M. K. Transtrum, J. Lee, D. N. Seidman, D. B. Liarte, N. S. Sitaraman, A. Senanian, M. M. Kelley, J. P. Sethna, T. Arias, and S. Posen, Analysis of magnetic vortex dissipation in sn-segregated boundaries in nb3Sn superconducting rf cavities, Phys. Rev. B 103, 024516 (2021).DOI
  20. [20]B. Oripov and S. M. Anlage, Time-dependent ginzburg-landau treatment of rf magnetic vortices in superconductors: Vortex semiloops in a spatially nonuniform magnetic field, Physical Review E 101, 10.1103/PhysRevE.101.033306 (2020), cited by: 13; All Open Access, Bronze Open Access, Green Open Access.DOI
  21. [21]B. Oripov, T. Bieler, G. Ciovati, S. Calatroni, P. Dhakal, T. Junginger, O. B. Malyshev, G. Terenziani, A.-M. Valente-Feliciano, R. Valizadeh, S. Wilde, and S. M. Anlage, High-frequency nonlinear response of superconducting cavity-grade Nb surfaces, Phys. Rev. Appl. 11, 064030 (2019).
  22. [22]J. Allmaras, A. Kozorezov, B. Korzh, K. Berggren, and M. Shaw, Intrinsic timing jitter and latency in superconducting nanowire single-photon detectors, Phys. Rev. Appl. 11, 034062 (2019).
  23. [23]M. Jönsson, R. Vedin, S. Gyger, J. A. Sutton, S. Steinhauer, V. Zwiller, M. Wallin, and J. Lidmar, Current crowding in nanoscale superconductors within the ginzburg-landau model, Phys. Rev. Appl. 17, 064046 (2022).arxiv.org/abs/2112.05443
  24. [24]L. Bishop-Van Horn, E. Mueller, and K. A. Moler, Vortex dynamics induced by scanning squid susceptometry, Phys. Rev. B 107, 224509 (2023).
  25. [25]L. Bishop-Van Horn, pytdgl: Time-dependent ginzburg-landau in python, Computer Physics Communications 291, 108799 (2023).
  26. [26]T. J. Rieger, D. J. Scalapino, and J. E. Mercereau, Charge conservation and chemical potentials in time-dependent ginzburg-landau theory, Phys. Rev. Lett. 27, 1787 (1971).DOI
  27. [27]T. J. Rieger, D. J. Scalapino, and J. E. Mercereau, Time-dependent superconductivity and quantum dissipation, Phys. Rev. B 6, 1734 (1972).
  28. [28]S. Frota-Pessôa, J. A. Blackburn, and B. B. Schwartz, Short weak link with distinct chemical potentials at the boundary, Phys. Rev. B 20, 993 (1979).
  29. [29]G. R. Berdiyorov, X. H. Chao, F. M. Peeters, H. B. Wang, V. V. Moshchalkov, and B. Y. Zhu, Magnetoresistance oscillations in superconducting strips: A ginzburg-landau study, Phys. Rev. B 86, 224504 (2012).DOI
  30. [30]G. Kimmel, A. Glatz, and I. S. Aranson, Phase slips in superconducting weak links, Phys. Rev. B 95, 014518 (2017).DOI
  31. [31]L. Peng, Y. Hu, Z. Li, K. Deng, Y. Zhu, L. Xu, and Y. Zhou, Finite element treatment of vortex states in 3d mesoscopic cylindrical superconductors in a tilted magnetic field, Acta Physica Polonica A 133, 152 – 156 (2018), cited by: 2; All Open Access, Bronze Open Access.DOI
  32. [32]K. Arutyunov, D. Golubev, and A. Zaikin, Superconductivity in one dimension, Physics Reports 464, 1 (2008).
  33. [33]S. Michotte, S. Mátéfi-Templfi, L. Piraux, D. Y. Vodolazov, and F. M. Peeters, Condition for the occurrence of phase slip centers in superconducting nanowires under applied current or voltage, Phys. Rev. B 69, 094512 (2004).DOI
  34. [34]B. Petukhov and V. Chechetkin, Rate of penetration of a magnetic flux into type-ii superconductors, Zh. Eksp. Teor. Fiz 65, 1653 (1973).
  35. [35]L. Burchakov, Magnetic relaxation over the bean-livingston surface barrier, Physical Review B 47, 8056 (1993).
  36. [36]D. Feinberg, Vortex lines in layered superconductors. i. from 3d to 2d behaviour, Journal de Physique III 4, 169 (1994).DOI
  37. [37]H. Guo, C. L. Phillips, T. Peterka, D. Karpyev, and A. Glatz, Extracting, tracking, and visualizing magnetic flux vortices in 3d complex-valued superconductor simulation data, IEEE Transactions on Visualization and Computer Graphics 22, 827 (2015).DOI
  38. [38]I. Sadovskyy, A. Koshelev, A. Glatz, V. Ortalan, M. Rupich, and M. Leroux, Simulation of the vortex dynamics in a real pinning landscape of yba 2 cu 3 o 7-δ coated conductors, Physical Review Applied 5, 014011 (2016).
  39. [39]H. Guo, T. Peterka, and A. Glatz, In situ magnetic flux vortex visualization in time-dependent ginzburg-landau superconductor simulations, in 2017 IEEE Pacific Visualization Symposium (PacificVis) (IEEE, 2017) pp. 71–80.DOI
  40. [40]E. Zhakina, L. Turnbull, W. Xu, M. König, P. Simon, W. Carrillo-Cabrera, A. Fernandez-Pacheco, U. Vool, D. Suess, C. Abert, V. Fomin, and C. Donnelly, Vortex motion in reconfigurable three-dimensional superconducting nanoarchitectures, arXiv preprint arXiv:2404.12151 (2024).arxiv.org/abs/2404.12151
  41. [41]S. Posen, N. Valles, and M. Liepe, Radio frequency magnetic field limits of nb and nb3Sn, Phys. Rev. Lett. 115, 047001 (2015).pubmed.ncbi.nlm.nih.gov/26252705
  42. [42]Y. Trenikhina, S. Posen, A. Romanenko, M. Sardela, J.-M. Zuo, D. L. Hall, and M. Liepe, Performance-defining properties of nb3Sn coating in srf cavities, Superconductor Science and Technology 31, 015004 (2017).DOI
  43. [43]J. Lee, S. Posen, Z. Mao, Y. Trenikhina, K. He, D. L. Hall, M. Liepe, and D. N. Seidman, Atomic-scale analyses of nb3Sn on nb prepared by vapor diffusion for superconducting radiofrequency cavity applications: a correlative study, Superconductor Science and Technology 32, 024001 (2018).DOI
  44. [44]U. Pudasaini, G. V. Eremeev, J. W. Angle, J. Tuggle, C. E. Reece, and M. J. Kelley, Growth of Nb3Sn coating in tin vapor-diffusion process, Journal of Vacuum Science & Technology A 37, 051509 (2019), https://pubs.aip.org/avs/jva/article- pdf/doi/10.1116/1.5113597/13962261/051509_1_online.pdf. [45] J. Lee, Z. Mao, K. He, Z. H. Sung, T. Spina, S.-I. Baik, D. L. Hall, M. Liepe, D. N. Seidman, and S. Posen, Grain- boundary structure and segregation in nb3sn coatings on nb for high-performance superconducting radiofrequency cavity applications, Acta Materialia 188, 155 (2020).
  45. [46]S.-H. Oh, D. Seol, Y.-J. Jeong, S.-H. Na, J. Kim, W.-S. Ko, J. B. Jeon, and B.-J. Lee, Diffusion in a15 nb3sn: An atomistic study, Acta Materialia 234, 118050 (2022).DOI
  46. [47]C. Becker, S. Posen, N. Groll, R. Cook, C. M. Schleputz, D. L. Hall, M. Liepe, M. Pellin, J. Za- sadzinski, and T. Proslier, Analysis of Nb3Sn sur- face layers for superconducting radio frequency cavity applications, Applied Physics Letters 106, 082602 (2015), https://pubs.aip.org/aip/apl/article- pdf/doi/10.1063/1.4913617/13593886/082602_1_online.pdf.
  47. [48]E. Wiklund, J. Lee, D. Seidman, and S. Posen, Three-dimensional reconstruction of nb3sn films by focused ion beam cross sectional microscopy, IEEE Transactions on Applied Superconductivity 33, 10.1109/TASC.2023.3257819 (2023), cited by: 0.
  48. [49]J. K. Freericks, A. Y. Liu, A. Quandt, and J. Geerk, Nonconstant electronic density of states tunneling inver- sion for a15 superconductors: nb3Sn, Phys. Rev. B 65, 224510 (2002).
  49. [50]W. Markiewicz, Elastic stiffness model for the critical temperature tc of nb3sn including strain dependence, Cryogenics 44, 767 (2004).
  50. [51]M. G. T. Mentink, M. M. J. Dhalle, D. R. Dieterich, A. Godeke, F. Hellman, and H. H. J. ten Kate, The ef- fects of disorder on the normal state and superconducting properties of nb3sn, Superconductor Science and Tech- nology 30, 025006 (2016).
  51. [52]M. M. Kelley, N. Sitaraman, and T. A. Arias, Ab ini- tio theory of the impact of grain boundaries and substitu- tional defects on superconducting nb3Sn, Superconductor Science and Technology 34, 015015 (2020).
  52. [53]N. S. Sitaraman, M. M. Kelley, R. D. Porter, M. U. Liepe, T. A. Arias, J. Carlson, A. R. Pack, M. K. Transtrum, and R. Sundararaman, Effect of the density of states at the fermi level on defect free energies and superconduc- tivity: A case study of nb3Sn, Phys. Rev. B 103, 115106 (2021).
  53. [54]A. E. Koshelev, I. A. Sadovskyy, C. L. Phillips, and A. Glatz, Optimization of vortex pinning by nanopar- ticles using simulations of the time-dependent ginzburg- landau model, Phys. Rev. B 93, 060508 (2016).
  54. [55]I. Sadovskyy, A. Koshelev, C. Phillips, D. Karpeyev, and A. Glatz, Stable large-scale solver for ginzburg–landau equations for superconductors, Journal of Computational Physics 294, 639 (2015).
  55. [56]M. P. Sørensen, N. F. Pedersen, and M. Ögren, The dynamics of magnetic vortices in type ii superconduc- tors with pinning sites studied by the time dependent ginzburg–landau model, Physica C: Superconductivity and its Applications 533, 40 (2017).
  56. [57]S. J. Chapman, C. M. Elliott, A. K. Head, S. D. How- ison, F. M. Leslie, J. R. Ockendon, J. Deang, Q. Du, M. Gunzburger, and J. Peterson, Vortices in supercon- ductors: modelling and computer simulations, Philosoph- ical Transactions of the Royal Society of London. Se- ries A: Mathematical, Physical and Engineering Sciences 355, 1957 (1997).
  57. [58]A. Al Luhaibi, A. Glatz, and J. B. Ketterson, Driven re- sponses of periodically patterned superconducting films, Phys. Rev. B 106, 224516 (2022).
  58. [59]V. L. Ginzburg and L. D. Landau, On the theory of super- conductivity, in On Superconductivity and Superfluidity: A Scientific Autobiography (Springer Berlin Heidelberg, Berlin, Heidelberg, 2009) pp. 113–137.
  59. [60]A. Schmid, A time dependent ginzburg-landau equation and its application to the problem of resistivity in the mixed state, Physik der kondensierten Materie 5, 302 (1966).
  60. [61]L. Gor’kov and G. Eliashberg, Generalization of ginzburg-landau equations for non-stationary problems in the case of alloys with paramagnetic impurities, Sov. Phys. JETP 27, 328 (1968).
  61. [62]Q. Du, M. D. Gunzburger, and J. S. Peterson, Analy- sis and approximation of the ginzburg–landau model of superconductivity, SIAM Review 34, 54–81 (1992).
  62. [63]N. Kopnin, Theory of nonequilibrium superconductivity, Vol. 110 (Oxford University Press, 2001).DOI
  63. [64]H. Gao and W. Aun, A new mixed formula- tion and efficient numerical solution of ginzburg– landau equations under the temporal gauge, SIAM Journal on Scientific Computing 38, A1339 (2016), https://doi.org/10.1137/15M1022744.
  64. [65]Q. Du, Global existence and uniqueness of solutions of the time-dependent ginzburg–landau model for supercon- ductivity, Applicable Analysis 53, 1 (1994).
  65. [66]M. S. Alnæs, J. Blechtа, J. Hake, A. Johansson, B. Kehlet, A. Logg, C. Richardson, J. Ring, M. E. Rognes, and G. N. Wells, The fenics project version 1.5, Archive of Numerical Software 3, 9 (2015).
  66. [67]C. Geuzaine and J.-F. Remacle, Gmsh: A 3-d finite element mesh generator with built-in pre-and post- processing facilities, International journal for numerical methods in engineering 79, 1309 (2009).
  67. [68]M. Tinkham, Viscous flow of flux in type-ii superconduc- tors, Phys. Rev. Lett. 13, 804 (1964).
  68. [69]C. J. Gorter and H. Casimir, On superconductivity i, Physica 1, 306 (1934).
  69. [70]H. London, Production of heat in superconductors by alternating currents, Nature 133, 497 (1934).
  70. [71]J. Halbritter, On surface resistance of superconductors, Zeitschrift für Physik 266, 209 (1974).
  71. [72]J. Turneаure, J. Halbritter, and H. Schwettman, The sur- face impedance of superconductors and normal conduc- tors: The mattis-bardeen theory, Journal of Supercon- ductivity 4, 341 (1991).
  72. [73]D. C. Mattis and J. Bardeen, Theory of the anomalous skin effect in normal and superconducting metals, Phys- ical Review 111, 412 (1958).
  73. [74]A. A. Abrikosov, L. Gor’Kov, and I. I. Khalatnikov, A super- conductor in a high frequency field, Sov. Phys. JETP 35, 182 (1959).
  74. [75]A. Gurevich, Theory of rf superconductivity for reso- nant cavities, Superconductor Science and Technology 30, 034004 (2017).
  75. [76]A. Gurevich and T. Kubo, Surface impedance and op- timum surface resistance of a superconductor with an imperfect surface, Phys. Rev. B 96, 184515 (2017).
  76. [77]B. Aune, R. Bandelmann, D. Bloess, B. Bonin, A. Bosotti, M. Champain, C. Crawford, G. Deppe, B. Dwersteg, D. A. Edwards, H. T. Edwards, M. Ferrario, M. Fouaidy, P.-D. Gall, A. Gamp, A. Gössel, J. Graber, D. Hubert, M. Hüning, M. Juillard, T. Junquera, H. Kaiser, G. Kreps, M. Kuchnir, R. Lange, M. Leenen, M. Liepe, L. Lilje, A. Matheisen, W.-D. Möller, A. Mosnier, H. Padamsee, C. Pagani, M. Pekeler, H.-B. Peters, O. Peters, D. Proch, K. Rehlich, D. Reschke, H. Safa, T. Schlicher, P. Schmüser, J. Sekutowicz, S. Simrock, W. Singer, M. Tigner, D. Trines, K. Twarowski, G. Weichert, J. Weisend, J. Wojtkiewicz, S. Wolff, and K. Zapfe, Superconducting tesla cavities, Phys. Rev. ST Accel. Beams 3, 092001 (2000).
  77. [78]N. Sitaraman, Private correspondence (2025).
  78. [80]E. Viklund, D. N. Seidman, D. Burk, and S. Posen, Improving nb3sn cavity performance using centrifugal barrel polishing (2023), arXiv:2305.10226 [physics.acc-ph].
  79. [81]M. Tinkham, Introduction to Superconductivity (Courier Corporation, 2004).
  80. [82]L. Kramer and R. J. Watts-Tobin, Theory of dissipative current-carrying states in superconducting filaments, Phys. Rev. Lett. 40, 1041 (1978).DOI
  81. [83]T. Proslier, J. F. Zasadzinski, L. Cooley, C. Antoine, J. Moore, J. Norem, M. Pellin, and K. E. Gray, Tunneling study of cavity grade nb: Possible magnetic scattering at the surface, Applied Physics Letters 92, 212505 (2008), https://doi.org/10.1063/1.2913764.
  82. [84]T. Kubo and A. Gurevich, Field-dependent nonlinear surface resistance and its optimization by surface nanostructuring in superconductors, Phys. Rev. B 100, 064522 (2019).DOI
  83. [85]N. W. Ashcroft and N. D. Mermin, Solid State Physics (Holt, Rinehart and Winston, New York, 1976).
  84. [87]K. Maki, On persistent currents in a superconducting alloy. ii, Progress of Theoretical Physics 29, 333 (1963).DOI

Paper details

vlnc:W_8bx43ve6
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents