Ab-initio superfluid weight and superconducting penetration depth

arXiv:2603.10955 · cond-mat.supr-con, cond-mat.mtrl-sci · Submitted 2026-03-11 · Read on arXiv

Listen

Transcript

Introduction to the show: ident: Quantum Radio. Generated commentary on the latest quantum physics and condensed matter papers.

Kai: I'm Kai, and with me are Mira and Lev, guest researcher.

Mira: Today's paper: "Ab-initio superfluid weight and superconducting penetration depth".

Kai: This paper develops a computationally efficient framework to calculate zero-temperature, mean-field superfluid weight from density functional theory (DFT) data,

Mira: First, who's behind it and why it matters.

Title and authors: Kai: So, we're diving into this paper now titled "Ab-initio superfluid weight and superconducting penetration depth." It looks like they’re tackling how we can calculate something fundamental for superconductivity right from Density Functional Theory data.

Mira: That’s right, Kai; the title itself tells us they are connecting the microscopic electronic structure calculations to macroscopic experimental observables. They're focusing on the superfluid weight, which is a key descriptor for understanding how these materials behave in experiments.

Lev: From my side, I'm wondering if this approach is computationally feasible for real hardware, especially given the complexity of incorporating quantum geometry into standard DFT outputs. We need to make sure we aren't just generating massive data sets that are impossible to process reliably on actual quantum systems.

Kai: Exactly; the authors are proposing a new strategy to handle the integration over the Brillouin zone, moving away from traditional finite-difference derivatives toward a Nadaraya–Watson kernel regression approach. This is intended to make it much more computationally efficient for high-throughput screening of materials.

Mira: The paper explains that they're separating the total superfluid weight into two parts: a conventional component based on band curvature and another geometric component stemming from the quantum geometry of the Bloch wavefunctions themselves. This separation helps them understand which physical mechanism might be driving superconductivity in different material classes.

Lev: And if I can get that separation, I can start thinking about error correction protocols for when we try to implement these calculations on actual qubits; knowing whether we're dealing with a dominant conventional or geometric term could drastically affect the required fidelity of the measurement.

Kai: Right, so they use standard DFT codes to get those matrix elements for the geometric term, labeled as Mk,bmn, and then they use kernel regression to define an energy-resolved quantity Bµν(ϵ) by taking a ratio with the density of states. This lets them calculate the conventional superfluid weight as Dµν conv = Z ∫ A(ϵ)Bµν(ϵ)ρ(ϵ) dϵ.

Mira: That integral formulation is what makes it tractable; instead of needing perfect derivatives across a grid, they can evaluate it even when their k-mesh isn't perfectly fine, as long as the prefactor A(epsilon) is analytically known. They’ve shown this allows for controlled evaluation even on relatively coarse k-meshes.

Lev: Coarse meshes are always a concern when you move from theory to experimental realization; if we were running this on an actual superconducting circuit simulation, I'd be worried about how much noise in the input DFT data would propagate through that kernel regression step.

Title and authors: Kai: They do benchmark MgB2 and Al for this displacement of zero point zero zero one Å, showing that the resulting derivatives were well converged with respect to that displacement, which gives some confidence in their numerical setup. They also noted that the geometric contribution is typically three to four orders of magnitude smaller than the conventional one in many materials studied.

Mira: That observation is interesting because it sets up a clear physical dichotomy: you expect the conventional term to dominate in wide-band materials where pairing gaps are smaller than the bandwidth, but you anticipate that flatness or gapped structures might cause the geometric term to take over and become more significant.

Lev: If the geometric contribution does dominate, that means we might need entirely different theoretical models for those unconventional systems because mean-field pairing temperature just won't tell us the whole story anymore.

Kai: That’s a big implication; they argue that if superconductivity in unconventional materials is limited by phase coherence rather than just the mean-field pairing temperature, then the superfluid weight becomes a more crucial descriptor than the pairing gap itself for predicting high-temperature superconductivity.

Mira: Exactly, and this paper directly supports that idea by proposing it as a crucial limiting factor for high-Tc materials because it quantifies phase coherence. This moves us beyond just looking at how strong the pairing interaction is to understanding how well the Cooper pairs can actually establish order across the entire system.

Lev: For error correction researchers, if we can reliably calculate this weight, it gives us a physical target that’s independent of some of the more complex noise models we have to invent for real hardware implementation. It provides a tangible metric for success in characterizing those phase coherence issues.

Kai: So, as they move toward large-scale screening, the framework is presented as computationally accessible because it only requires standard DFT outputs and can be integrated into existing high-throughput workflows using packages like Quantum ESPRESSO.

Mira: The authors are pushing this paper toward a future where we can screen thousands of materials based on these physically meaningful descriptors, allowing us to filter candidates based on target properties like achieving a specific magnetic penetration depth, as they showed with the examples of Al and Nb.

Lev: If you can screen materials based on a property that relates so directly to experimental observables like the London penetration depth, it provides a much better filter than relying solely on complex pairing function calculations for initial material selection.

Title and authors: Kai: This paper lays a foundation for using superfluid properties to guide the search for conventional superconductors by targeting specific penetration depths, while also opening up avenues to explore unconventional superconductivity where phase coherence might be the real bottleneck.

Mira: The structure of the paper clearly shows they are trying to establish a clear path from first-principles calculations directly into experimentally verifiable parameters. This bridge between DFT and experimental observables is what makes this work so valuable for condensed matter theory.

Lev: From a hardware standpoint, if the theoretical predictions for penetration depth align well with our measurements on actual thin films, it validates the entire computational chain we'd need to build for actual experimental testing of these materials.

Kai: So, to wrap up this discussion on "Ab-initio superfluid weight and superconducting penetration depth," this paper gives us a powerful way to calculate a physically meaningful descriptor from DFT data that ties directly into key experimental measurements like magnetic penetration depth.

Mira: Indeed, it suggests that the total superfluid weight, with its conventional and geometric components, offers a deeper understanding of what limits superconductivity in different material architectures than just the pairing gap alone.

Lev: And for those of us working on error correction, having this framework means we have a more robust theoretical anchor when we start thinking about how to manage phase coherence in complex superconducting systems.

Kai: It's exciting because it shows how these fundamental calculations can be structured to be computationally tractable enough for real-world screening applications, which is what the authors really want to achieve with this work.

Mira: I think the most significant implication is shifting our focus from just finding materials that have a high pairing gap to finding materials that possess the correct phase coherence structure dictated by their band geometry.

Lev: That distinction is important because it suggests we need to look beyond simple mean-field approximations when predicting behavior in these more complex, unconventional regimes.

Kai: So, this paper provides a clear roadmap for using superfluid weight as a filter for material discovery and a tool to probe the physics of phase coherence in high-temperature superconductors.

Mira: And looking ahead, the future work they mention—incorporating multiband non-uniform pairing or extending calculations to finite temperatures—will be key to fully unlocking this predictive power.

Lev: I'd like to see those finite temperature extensions because real materials operate at temperatures where thermal fluctuations and phase coherence effects become much more pronounced.

Kai: Well, that’s our time on "Ab-initio superfluid weight and superconducting penetration depth"; it’s a solid piece of work that connects deep theory to practical screening tools.

The paper's summary: Kai: So, to summarize what we just read, this paper is essentially taking those complex Density Functional Theory calculations and distilling them down into something that relates directly to how superconductors actually behave in experiments, specifically looking at the superfluid weight as a key property.

Mira: That’s right, Kai; the authors are proposing a way to calculate this weight from just standard DFT data, aiming to give us a physically meaningful number we can compare with things like magnetic penetration depth. They break it down into two parts: one that comes from how bands curve and another one coming from the geometry of the electronic wavefunctions.

Lev: From my side, I’m thinking about what this means for experimental verification; if they can provide a numerical value for something like penetration depth directly from first principles, that would be a huge step in validating our experimental setup before we even start fabricating things on real hardware.

Kai: Exactly; it’s about building that bridge between the theoretical calculation and the physical measurement. They use this weight as a descriptor because it tells us about phase coherence, which is something we often struggle to measure directly in complex systems like high-Tc materials.

Mira: And they point out that this weight isn't just one number; it shows a competition between different physical effects; the conventional part, driven by band curvature, tends to dominate in materials with broad bands, while the geometric part becomes more important when those conventional contributions are suppressed.

Lev: That distinction is vital for error correction research because if we're trying to model phase coherence in a system meant for quantum computation, knowing which term dominates tells us where our theoretical approximations are most likely to fail.

Kai: Right; it’s like they’re giving us a diagnostic tool for different classes of superconductors; you use the weight to quickly figure out if you’re dealing with a conventional system or something more exotic where phase coherence is the main limiting factor.

Mira: The authors really highlight how this could be used for high-throughput screening, suggesting that we can filter thousands of materials based on these calculated properties, which is a big deal for discovering new candidates in condensed matter physics.

Lev: If the AI system can automate this filtering based on penetration depth targets, it means we don't have to spend as much time running expensive experiments just to narrow down the initial material pool.

Kai: So, they’re essentially proposing a pipeline where DFT feeds into a screening tool that guides experimentalists toward materials that are most likely to show strong superconducting behavior based on this weight calculation.

Mira: And while they show promise for conventional superconductors like Al and Nb, the paper also points out that the geometric contribution is often very small compared to the conventional term unless you’re dealing with specific band structures, so we need to be careful not to over-interpret it in every case.

Lev: If we look at this through a quantum error-correction lens, quantifying phase coherence via this weight gives us a way to characterize the system's inherent susceptibility to decoherence before we even try building a qubit on top of it.

Kai: It’s clear that the next step is taking these theoretical results and seeing if they hold up when we look at those actual experimental data sets for things like penetration depth, which is what the paper validates with its benchmark results.

Mira: And while the framework itself is accessible using standard DFT codes, the authors do flag a limitation: they haven't fully incorporated multiband non-uniform pairing yet, so it’s currently focused on simpler band structures.

Lev: That limitation makes sense; when we move to those more complex systems you mentioned, like underdoped cuprates, you need a much more robust way to handle the coupling between different bands that this current kernel regression might not capture perfectly.

Kai: So the big takeaway is that this work provides a practical method for using quantum geometry as a descriptor for superconducting properties, setting the stage for better material discovery and deeper understanding of phase coherence limits.

The paper's improvements: Tom: So, to recap, this paper lays out a computational framework using density functional theory data to calculate the superfluid weight from first principles and then proposes a specific strategy—using kernel regression—to integrate those results across the entire material structure efficiently.

Kai: That's right; they're essentially showing us how to take complex electronic structure information and turn it into a number that directly relates to an experimental measurement like the magnetic penetration depth. It’s about making sure our theoretical tools actually map onto what we can build and measure in the lab.

Mira: I see, the improvements section focuses on how they’re refining that methodology, moving beyond just the initial calculation to make it more robust for real-world application. They are specifically focusing on rewriting those conventional contributions and defining energy-resolved quantities using a kernel regression approach instead of just relying on traditional finite-difference derivatives.

Lev: That’s interesting; from an error correction standpoint, if you use a kernel regression approach, you're essentially introducing a continuous approximation to the underlying data, which might simplify the required noise model for simulating that calculation on real quantum hardware.

Kai: Exactly; it makes the integration process much smoother and less prone to numerical instability when we try to implement this in simulation environments. They’re making it easier for us to get a reliable number from a relatively coarse k-mesh, which is crucial when we're dealing with large material systems.

Mira: The implication here is that they’re improving the computational accessibility; it means we can use this framework on much larger and more diverse material databases than previously possible, accelerating the screening process significantly.

Lev: If the AI system can reliably calculate this weight across a massive database, it provides us with a much richer set of data points to test our models for predicting phase coherence in complex systems.

Kai: Right; it’s about making sure that when we finally get to experimental validation—when we’re actually cooling down superconducting circuits and measuring those penetration depths—we have a solid theoretical prediction to compare against.

Mira: The authors are also looking ahead by suggesting incorporating multiband non-uniform pairing, which addresses a major assumption they made in their initial work by accounting for more complex electronic structures found in many real materials.

Lev: That extension is important because it tackles the complexity of how different bands interact, which is exactly what we need to understand when predicting behavior in unconventional superconductors where those band interactions are crucial.

Kai: So, they’re not just stopping at a single material type; they’re building a pathway to handle the more intricate physics found in high-temperature materials by adding these more advanced theoretical layers.

Mira: And looking ahead, combining this superfluid weight calculation with other machine learning descriptors is another key suggestion, which points toward using AI to find even more subtle correlations that traditional theories might miss.

Lev: That makes sense; using AI to find those subtle correlations could help us identify the specific geometric or conventional contributions that are most predictive of phase coherence limitations in those hard-to-character materials.

Kai: So, the overall goal of these improvements is to move this from a single calculation for one material into a general, scalable toolkit for exploring the physics across many different candidates.

Conclusion: Kai: So, to wrap up our discussion on "Ab-initio superfluid weight and superconducting penetration depth," this paper provides a solid framework for using Density Functional Theory to calculate a fundamental property of superconductors—the total superfluid weight—and then linking that number directly to experimental observables like the magnetic penetration depth.

Mira: It really does establish a way to use quantum geometry information, which is often hidden in band structure calculations, as a direct probe for phase coherence, which is what limits superconductivity in many high-Tc systems.

Lev: And from my end, if this framework can be made computationally accessible through kernel regression, it means we have a more reliable theoretical anchor to test against the noisy measurements we get from real quantum devices.

Kai: Exactly; it's about giving experimentalists a way to filter materials based on these calculated properties before they even start the costly fabrication process.

Mira: The implication is that we can move beyond just looking at how strong the pairing interaction is and start considering the structural and geometric factors that govern how well those pairs actually establish order across a material.

Lev: For error correction, having this descriptor means we aren't guessing about phase coherence; we have a quantifiable metric to see what's holding those superconducting states together.

Kai: It’s exciting because it shows how deep theoretical calculations can translate into practical tools for real-world material discovery and experimental validation.

Mira: We should definitely keep an eye on the future work they mention, especially regarding multiband pairing, because that will be necessary to fully capture the physics in many of the materials we're interested in.

Lev: I agree; those extensions are necessary to bridge the gap between these simple models and the complex reality of high-temperature superconductors.

Kai: Well, it’s been a fascinating look at how DFT can become a powerful screening tool for finding new superconducting candidates based on their inherent quantum geometry.

Mira: Indeed, it's a significant step in connecting microscopic theory to macroscopic experimental realities through this paper on "Ab-initio superfluid weight and superconducting penetration depth."

Lev: We’re ready to take these theoretical predictions and see what they mean for building more resilient superconducting systems.

Department of Applied Physics, Aalto University School of Science · Research Center Future Energy Materials and Systems of the University Alliance Ruhr and Interdisciplinary Centre for Advanced Materials Simulation, Ruhr University Bochum, Center for Materials Theory, Department of Physics and Astronomy, Rutgers University

cond-mat.supr-con, cond-mat.mtrl-sci

Submitted: 2026-03-11

Updated: 2026-03-11

Comments: 12 pages, 3 figures. Submitted to Physical Review B as a regular article on the 2nd of March 2026

DOI: 10.1103/2xrg-6fy6

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Importance score: 83/100

The gist: This paper develops a computationally efficient framework to calculate zero-temperature, mean-field superfluid weight from density functional theory (DFT) data, aiming to provide physically

Key concepts

Superfluid Weight: This is a measure derived from the current-current response function that quantifies the total superfluid density in a material. It is crucial because it directly determines key experimental properties, such as how strongly a material repels magnetic fields inside it, making it an ideal descriptor for superconductors.
Conventional Contribution: This part of the superfluid weight arises from the curvature of electronic bands near the Fermi level. It is calculated using standard DFT outputs and is expected to be dominant in materials with wide energy bands that cross the Fermi level, reflecting standard band structure effects.
Geometric Contribution: This term originates from the quantum geometry of Bloch wavefunctions. It represents a different physical effect and is often much smaller than the conventional term. It becomes more significant in systems where the conventional contribution is weak, such as flat-band materials or those with band gaps.
Nadaraya–Watson Kernel Regression: Instead of using complex finite-difference derivatives to calculate integrals over the Brillouin zone, this strategy uses kernel regression. This technique reformulates the problem by treating the integration as a continuous variable, allowing for efficient calculation even on relatively coarse electronic meshes used in high-throughput screening.

Terminology

Benchmark — 8× in this paper · explained in 2 episodes
DFT Calculations — 3× in this paper · explained in 2 episodes
Accuracy — 2× in this paper · explained in 1 episode
Coherence Length — 2× in this paper · explained in 2 episodes
Density Functional Theory (DFT) — 2× in this paper · explained in 2 episodes
Anisotropy — 1× in this paper · explained in 3 episodes
Berry Curvature — 1× in this paper · explained in 3 episodes
Bogoliubov-de Gennes — 1× in this paper
Fidelity — 1× in this paper · explained in 2 episodes
Hubbard Model — 1× in this paper · explained in 3 episodes
Metastability — 1× in this paper · explained in 3 episodes
Phase Transitions — 1× in this paper · explained in 2 episodes
Regularization — 1× in this paper · explained in 3 episodes
Robustness — 1× in this paper · explained in 5 episodes
Verification — 1× in this paper · explained in 2 episodes

Summary

This paper develops a computationally efficient framework to calculate zero-temperature, mean-field superfluid weight from density functional theory (DFT) data, aiming to provide physically meaningful descriptors for high-throughput screening of superconducting materials. The superfluid weight is highlighted as an ideal descriptor because it determines key experimental observables like the magnetic penetration depth and the Berezinskii-Kosterlitz-Thouless transition temperature in two-dimensional systems, and it is proposed as a crucial limiting factor for superconductivity in unconventional high-Tc materials by quantifying phase coherence.

Theoretical Foundation of Superfluid Weight Calculation

The superfluid weight is defined as the static, long-wavelength limit of the current-current response function, expressed as:

The total superfluid weight being separated into a conventional and a geometric contribution:
Conventional contribution:
Geometric contribution:

The conventional term is derived from band curvature and is given by Equation (6), while the geometric term arises from the quantum geometry of Bloch wavefunctions in Equation (7). The matrix elements required for the geometric term, denoted as Mk,bmn, are obtainable from standard DFT codes.

Computational Strategy for High-Throughput Screening

The paper addresses the challenge of integrating these quantities over the entire Brillouin zone by proposing an alternative strategy based on a Nadaraya–Watson kernel regression instead of traditional finite-difference derivatives. This approach reformulates the Brillouin-zone integral in an energy representation, treating it as a continuous variable where the prefactor depends only on energy and is known analytically.

Key aspects of this strategy include:

(10) Rewriting the conventional contribution by separating the analytic prefactor from the derivatives to be interpolated.

(13) Defining an energy-resolved quantity Bµν(ϵ) through kernel regression as a ratio between the energy-resolved density of Bµν and the density of states.

(15) Expressing the conventional superfluid weight as Dµν conv = Z ∫ A(ϵ)Bµν(ϵ)ρ(ϵ) dϵ, which allows for controlled evaluation even on relatively coarse k meshes.

Validation and Benchmark Results

The framework is validated by calculating London penetration depths for several conventional superconductors, including Al, Pb, Nb, MgB2, LuRu3B2, and YRu3B2. The penetration depth is extracted from the superfluid weight using Equations (16) and (17), relating it to the magnetic field decay inside the superconductor.

The results show:

(Table I) A comparison between calculated and experimental penetration depths for Pb, Al, Nb, MgB2, LuRu3B2, and YRu3B2.

(A. Single-element superconductors) For Aluminium (Al), the superfluid weight calculations yield a London penetration depth of 13 nm compared to 16 nm experimentally.

(Niobium (Nb)) Accounting for strong coupling via the renormalization λL → λL/√Z with Z ≈ 2.1 leads to an experimental range of 16–19 nm, which is very close to our calculated penetration depth of 15 nm.

Comparison of Conventional and Geometric Contributions

For the materials studied, the geometric contribution is typically three to four orders of magnitude smaller than the conventional contribution. The conventional term is expected to dominate in wide-band materials with dispersive bands crossing the Fermi level. Conversely, the geometric term is expected to dominate in systems where the conventional contribution is suppressed, such as in flat-band systems or gapped band structures.

Furthermore, Figure 3 illustrates that while the conventional term shows a weak dependence on ∆, the geometric contribution exhibits a clear dependence on the gap and increases with ∆, which is consistent with known limiting cases like the isolated flat-band limit where it scales linearly with the superconducting gap.

Conclusion and Future Directions

The developed framework provides a foundation for large-scale screening of superconducting candidates guided by their superfluid properties. The computational accessibility, requiring only standard DFT outputs, enables straightforward integration into existing high-throughput workflows using packages like Quantum ESPRESSO. The potential applications include screening for conventional superconductors based on desired penetration depths and exploring the role of the superfluid weight as a predictor for transition temperature in unconventional superconductors where phase coherence may be limiting. Future extensions include incorporating multiband non-uniform pairing, extending calculations to finite temperatures, and combining superfluid weight calculations with other machine learning descriptors.

Improvements for AI systems

Here are the specific improvements that an AI system, leveraging this scientific paper, could make:

Enhanced Superconductor Discovery Screening:
Predictive Modeling for Unconventional Superconductivity:
Quantum Geometry Property Mapping:

Enhancement in Superconductor Discovery Screening:

An improved AI system can perform high-throughput, physics-informed screening of vast chemical and structural databases (e.g., Materials Project, OQMD) to rapidly identify promising superconducting candidates based on their predicted zero-temperature superfluid weight. Specifically:

It can calculate the conventional and geometric contributions to the superfluid weight for thousands of materials using a computationally efficient framework derived from DFT data.
It can filter candidates based on target properties, such as achieving a specific magnetic penetration depth (e.g., selecting materials with a desired London penetration depth, as demonstrated in Table I).
It can prioritize materials exhibiting strong conventional contributions over geometric ones, guiding the search toward wide-band superconductors while flagging potential unconventional candidates where the geometric term might dominate.

Predictive Modeling for Unconventional Superconductivity:

The AI system can move beyond traditional pairing gap analysis to predict the limiting factors of superconductivity in complex systems (like underdoped cuprates). Specifically:

It can quantify the phase coherence contribution to the superconducting state by calculating the total superfluid weight, which is hypothesized to be more relevant than the mean-field pairing temperature.
It can distinguish between conventional and unconventional superconducting orders by analyzing whether the conventional contribution dominates (as seen in Pb and Nb) or if a significant geometric contribution exists, providing a physical descriptor to categorize material types.
It can predict behavior in reduced dimensions (2D materials) by directly relating the calculated superfluid weight to the Berezinskii-Kosterlitz-Thouless transition temperature, aiding in the design of atomically thin superconducting devices.

Quantum Geometry Property Mapping:

The AI system can map and utilize quantum geometric descriptors that are otherwise computationally inaccessible or difficult to calculate directly. Specifically:

It can extract the geometric contribution to the superfluid weight, which is directly related to the quantum metric of Bloch wavefunctions, providing a proxy for fundamental properties like electron-phonon coupling.
It can correlate these calculated geometric terms with other measurable physical responses (e.g., optical responses), allowing for the discovery of materials where quantum geometry plays a crucial role in superconductivity.
It can serve as a computational tool to explore how the flatness or dispersion characteristics of electronic bands influence superconducting behavior, which is critical for understanding phenomena like unconventional pairing mechanisms limited by phase coherence.