🔬 materials science

NaFeP: first pairing prediction for an unmade 111 iron phosphide, and the pnictogen-height rule inside one compound

This paper applies a physics-based prediction framework to the unsynthesized iron phosphide NaFeP, revealing that while structural rules suggest nodal superconductivity, the electronic structure predicts an s± state with a near-nodal inner hole sheet, thereby demonstrating the height-dependent transition between nodal and nodeless pairing within a single compound.

Original authors: Reinaldo Inácio

Published 2026-09-16

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Reinaldo Inácio

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Superconductivity is a state where electricity flows without any resistance, a phenomenon that could revolutionize how we transmit power and build machines. For decades, scientists have searched for new materials that can carry this current, but the path is littered with candidates that look promising on paper but fail in the lab. The challenge is not just finding a material that works, but finding one that can be drawn into a wire. Some superconductors are fragile; their ability to carry current collapses if the material contains tiny imperfections or grain boundaries, forcing engineers to use complex, expensive manufacturing techniques. Others are robust, tolerating these imperfections and allowing for simple, round wires. The key difference often lies in the invisible structure of the electrons within the material, specifically how they pair up. If the pairing has a certain symmetry, the material is wire-friendly; if it has another, it is not. The question facing the field is whether we can predict this behavior before we ever melt a single atom in a furnace.

A researcher has taken a significant step toward answering this by applying a new kind of digital judge to a compound that has never been made. They focused on a theoretical material called sodium iron phosphide, a substance that sits in a family of iron-based compounds known for superconductivity but has never been synthesized as a bulk solid. Using a sophisticated computer pipeline, they screened dozens of potential candidates and found that this specific one, NaFeP, was the only survivor of a rigorous stability check. It appeared to be stable, unreported in any database, and chemically distinct because it uses phosphorus instead of the more toxic arsenic found in many similar compounds. The researcher then ran their digital judge on this unmade material to predict its electronic behavior, aiming to determine if it could be a viable superconductor and, crucially, if it could be made into a wire.

The computer simulation revealed that the material would likely be a superconductor with a specific type of electron pairing that is generally tolerant of the grain boundaries found in wires. The simulation showed that the electrons would pair up in a way that allows current to flow smoothly across the material's internal boundaries, a property that makes it a strong candidate for practical wire manufacturing. However, the prediction came with a subtle but important caveat. While the initial, sealed verdict suggested the material was perfectly robust, a deeper, more detailed analysis introduced a nuance. The researcher found that the gap in the electron energy levels, which protects the superconducting state, might be extremely thin on one specific part of the material's structure. It is not broken, but it is dangerously close to the edge where it could fail. This near-miss suggests that while the material is likely a superconductor, its ability to carry high currents might be more fragile than the first, simpler model predicted.

The study also explored how the physical shape of the atoms within the material dictates these electronic properties. By virtually stretching and compressing the distance between the phosphorus atoms and the iron layers, the researcher observed a clear pattern. As the distance changed, the material's electronic behavior shifted, confirming a known rule in the field that the height of the phosphorus atom above the iron plane controls whether the superconducting gap is robust or fragile. In the case of sodium iron phosphide, the calculated height placed it on the side of the rule where the gap is expected to be thin or nearly broken, aligning the computer's detailed reading with established physical laws. This internal consistency gave the researcher confidence that their tool was working correctly, even as it delivered a more cautious final verdict.

The researcher is careful to state that this is a prediction, not a discovery. No one has yet built this material, and no one has measured its temperature or current-carrying capacity. The entire exercise was a simulation, a test of a new method for filtering candidates before they are synthesized. The researcher has provided a clear, falsifiable target for experimentalists: they have predicted the crystal structure, the magnetic state, and the specific nature of the superconducting gap. They have also outlined exactly what measurements would prove them right or wrong. If a lab can synthesize this sodium iron phosphide and find that it is a superconductor with a thin gap on its inner electron layer, the prediction is confirmed. If the material is not superconducting, or if the gap is robust everywhere, the prediction is refuted. The work stands as a bridge between the digital and the physical, offering a specific, testable hypothesis for a material that does not yet exist, with the potential to guide the search for the next generation of practical superconducting wires.

Technical Summary: NaFeP – First Pairing Prediction for an Unmade 111 Iron Phosphide

Problem and Context
This paper addresses the bottleneck in superconductor discovery: while computational generation of crystal structures is abundant, the field lacks a reliable "judge" to determine if a candidate can actually pair, with what symmetry, and whether it is viable for wire fabrication. Previous work (arXiv:2609.10614) established that generative models largely rediscover known physics and that "judgment"—classifying gap symmetry from structure—is the critical filter. This study applies that judgment pipeline to a single, genuinely unreported candidate: NaFeP (an 111-type iron phosphide). The goal is not to synthesize a new material but to provide a specific, falsifiable prediction for a compound that has no existing experimental bulk phase, band structure, or pairing calculation in the literature.

Methodology
The study employs a refined "judge" pipeline, building on the framework established in the companion paper [1]. The process involves:

Candidate Selection: A zero-cost screen of 58 iron-pnictide compositions absent from the Materials Project. Using literature searches and stability checks (CHGNet and high-throughput databases OQMD/Alexandria), 57 were eliminated. NaFeP survived as the only compound with no prior experimental report, no band-structure calculation, and a hull position indicating stability ( $e_{hull} \approx 0$ ).
Structural and Magnetic Analysis: NaFeP was relaxed using DFT (PBE). Four magnetic configurations (paramagnetic, ferromagnetic, Néel, and stripe) were tested. The paramagnetic state was found to be the ground state, with magnetic orders collapsing or yielding negligible moments ( $\sim 0.4 \mu_B$ /Fe, 0.4 meV/Fe below non-magnetic).
Electronic Structure and Wannier Modeling: Two Wannier tight-binding bases were constructed:
- d-only: Fe-3d orbitals only (10 functions), matching the basis of the positive control BaFe2As2.
- d+p: Fe-3d + P-3p orbitals (16 functions), explicitly including the ligand.
  Both models were validated against DFT bands and filling gates.
Pairing Calculation: The full-rank bare susceptibility ( $\chi_0$ ) was computed on a $24^3$ mesh. The linearized gap equation was solved within the Random Phase Approximation (RPA) using a Kanamori interaction ( $J/U = 0.15$ ) on the d-blocks.
Refinements: Two key refinements were introduced since the companion paper:
- Stoner Gate: A threshold ( $U_{crit} > 2.5$ eV) to classify materials as "null/negative" (far from spin-fluctuation instability) before symmetry analysis.
- Frame-Independent Symmetry Reader: A classifier based on the physical sign structure across Fermi sheets and node ratios, rather than standard symmetry projection strings which are sensitive to coordinate rotation.

Key Results

Gap Symmetry: The pipeline predicts NaFeP to be an $s_{\pm}$ superconductor (sign reversal between hole cylinders at $\Gamma$ and electron sheets at the zone corner). The nesting vector is identified as $(\frac{1}{2}, \frac{1}{2}, \frac{1}{2})$ .
The Node Debate (d-only vs. d+p):
- The d-only basis (sealed verdict) predicts a nodeless gap, with the weakest sheet (inner hole cylinder) having a minimum-to-maximum gap ratio of 0.15–0.18.
- The d+p basis (official prediction) predicts a near-nodal gap. On the inner hole cylinder, the ratio drops to 0.07–0.10, sitting at the reader's node threshold. This basis explicitly captures the phosphorus weight on the inner hole sheet, which the d-only model misses.
Pnictogen Height Rule: The phosphorus height ( $h_{Pn}$ $h_{P n}$ ) in NaFeP is $\approx 1.16$ $\approx 1.16$ Å (PBE-relaxed), well below the empirical $\approx 1.33$ $\approx 1.33$ Å threshold separating nodal (e.g., LiFeP) from nodeless (e.g., LiFeAs) iron pnictides.
- A scan of $h_{Pn}$ within the compound reproduces the height rule: as $h_{Pn}$ increases, a third hole pocket ( $d_{xy}$ ) appears, the nesting vector switches to $(\frac{1}{2}, \frac{1}{2}, 0)$ , and the system approaches a Stoner instability.
- The d+p reading aligns with the empirical rule, placing NaFeP on the "nodal" side of the height axis, specifically with a near-nodal inner hole sheet.
Stoner Stability: $U_{crit}$ is 1.68 eV (d-only) and 2.37 eV (d+p). Both values are below the 2.5 eV Stoner gate, confirming the material is a moderately correlated system capable of spin-fluctuation pairing, but distinct from conventional metals like Nb ( $U_{crit} = 3.65$ eV).

Manufacturability Profile
Conditional on the nodeless reading, NaFeP is projected as a Class B wire candidate (grain-boundary tolerant, round wire viable) with:

Anisotropy ( $\gamma$ ): 4.2 (within the iron pnictide range of 1–5, far more isotropic than cuprates).
Coherence Length ( $\xi_0$ ): $\approx 15$ nm (assuming $T_c = 20$ K).
Arsenic-free: A potential advantage for toxicity and processing.
Caveat: The near-nodal reading (d+p basis) erodes the grain-boundary argument, as a gap that nearly closes on one sheet is more sensitive to scattering than a fully open one.

Significance and Claims
The paper claims to provide the first falsifiable gap-symmetry prediction for NaFeP, a compound with no prior bulk synthesis or electronic structure data.

Specific Prediction: NaFeP is an $s_{\pm}$ superconductor with a near-nodal inner hole sheet (d+p basis) or a nodeless gap (d-only basis). The author adopts the d+p reading as the official prediction, acknowledging the instrument's disagreement between bases.
Methodological Proof: It demonstrates the application of the "judge" to a single, unmade candidate, moving beyond the "generation" phase to "judgment."
Internal Validation: The study validates the "pnictogen-height rule" internally by scanning the phosphorus height within NaFeP, showing the instrument reproduces the known transition from nodeless to nodal behavior as the height increases.
Modesty: The paper explicitly states no $T_c$ is asserted. The prediction is limited to symmetry class and manufacturability potential. The "nodeless" vs. "near-nodal" distinction is reported as an instrument disagreement rather than resolved by choice, inviting experimental measurement (penetration depth, thermal conductivity) to decide the verdict.
Cost: The entire characterization, including the height scan, cost approximately US$ 25 in cloud compute.

The paper concludes by inviting experimental collaboration to synthesize NaFeP and measure the gap on its inner hole sheet, framing the prediction as a synthesis target with a clearly stated, pre-registered measurement to refute or confirm the instrument's reading.

More like this