Vacancy-free cubic superconducting NbN enabled by quantum anharmonicity

This study demonstrates that quantum anharmonic effects stabilize a previously unknown, vacancy-free cubic phase of stoichiometric NbN with a superconducting transition temperature of 20 K, challenging the long-held belief that vacancies are essential for its structural stability.

The Gist

Imagine you have a team of dancers (atoms) trying to form a perfect, rigid square formation on a dance floor. For years, scientists believed that for Niobium Nitride (NbN)—a material famous for its superconducting abilities (conducting electricity with zero resistance)—this perfect square formation was impossible to hold together.

The old story went like this: To keep the dancers from tripping over each other and collapsing the formation, you had to have some empty spots on the floor (vacancies). You needed to remove a few dancers to make the square stable. If you tried to fill every spot perfectly (a 1:1 ratio), the formation would wobble and fall apart.

The New Discovery: The "Quantum Wiggle"

This paper tells a different story. The researchers found that if you stop looking at the dancers as stiff, frozen statues and realize they are actually quantum particles, the whole picture changes.

In the quantum world, atoms aren't still; they are constantly jittering and vibrating, even at absolute zero. This is called "zero-point motion." Furthermore, the forces holding them together aren't like a simple spring that pulls back evenly; they are "anharmonic," meaning the spring gets weird and stretchy when pulled hard.

The authors used super-computers to simulate these "quantum wiggles" and "stretchy springs." They discovered that when the atoms are allowed to dance with these quantum movements, they don't need empty spots to stay stable. Instead, they naturally shift into a new, slightly distorted shape that is actually more stable than the old, perfect square.

The Metaphor: The Jiggling Jello

Think of the old "perfect square" structure as a block of Jello that is too stiff to stand up; it collapses. Scientists used to think you had to poke holes in the Jello (vacancies) to make it hold its shape.

This paper shows that if you let the Jello jiggle (quantum anharmonicity), it doesn't collapse. Instead, the jiggle causes the Jello to settle into a slightly squashed, wobbly shape that is actually stronger and more comfortable than the rigid block. This new shape is the "vacancy-free" cubic phase the authors found.

What They Found

1. A New Shape: They identified a specific, previously unknown arrangement of atoms (with a space group called $P\bar{4}3m$). It's like the dancers found a new, slightly off-center formation that works better than the perfect square.
2. It's More Stable: This new, wobbly shape is energetically happier (lower in energy) than the old "perfect" square shape, even without any missing dancers.
3. Superconducting Performance: They calculated how well this new shape conducts electricity without resistance. They found it works at a temperature of 20 Kelvin. This matches very closely with what experiments see in real-world samples that are almost perfect (near-stoichiometric).
4. Why the Old Math Failed: Previous computer models assumed the atoms were stiff springs (harmonic). Those models said the perfect square was unstable. When the researchers added the "quantum wiggle" (anharmonicity), the math finally agreed with reality: the perfect square can exist, but it just needs to be slightly distorted to stay standing.

The Bottom Line

For a long time, scientists thought you needed defects (missing atoms) to make cubic Niobium Nitride work. This paper argues that you don't. The "defects" we see in experiments might just be the result of us not understanding the quantum dance moves of the atoms. If we can synthesize this perfect, vacancy-free material, it might actually perform even better as a superconductor than we currently think.

The paper suggests that instead of trying to fix the material by adding or removing atoms, we might just need to let the atoms do their natural quantum dance to find their most stable, high-performing shape.

Technical Summary

Technical Summary: Vacancy-free cubic superconducting NbN enabled by quantum anharmonicity

Problem Statement
Niobium nitride (NbN) is a technologically critical material known for its high critical magnetic field and superconducting transition temperature ($T_c$), reaching up to 17 K in its face-centered cubic (fcc) $\delta$-phase ($Fm\bar{3}m$). However, the ideal 1:1 stoichiometric $\delta$-NbN phase is theoretically predicted to be dynamically unstable within the harmonic approximation, exhibiting imaginary phonon modes across large portions of the Brillouin zone. Consequently, experimental observations have consistently indicated that nitrogen vacancies are necessary to stabilize the cubic structure. These vacancies induce structural distortions that stabilize the phase but also alter its electronic properties. The challenge lies in determining whether a stable, vacancy-free, 1:1 stoichiometric cubic phase exists, and if so, whether its superconducting properties differ significantly from the vacancy-stabilized $\delta$-phase. Traditional computational approaches struggle with this problem because modeling disorder and vacancies requires prohibitively large supercells, while standard harmonic approximations fail to capture the true stability of the ideal structure.

Methodology
The authors employed a multi-faceted computational approach combining state-of-the-art first-principles calculations with machine learning to overcome the limitations of standard density functional theory (DFT) and harmonic approximations.

1. Quantum Anharmonicity: To account for quantum ionic motion (zero-point motion) and the non-parabolic nature of the ionic potential, the study utilized the Stochastic Self-Consistent Harmonic Approximation (SSCHA). This method generates an effective harmonic description that explicitly includes anharmonic effects.
2. Machine-Learned Potentials: Given the computational cost of SSCHA and Molecular Dynamics (MD) simulations (requiring hundreds of thousands of DFT calculations), the authors trained high-fidelity machine-learned interatomic potentials (specifically Moment Tensor Potentials, MTP, and Ephemeral Data-Derived Potentials, EDDPs). These potentials retained DFT-level accuracy for energies and forces, enabling a computational speed-up of approximately $10^4$–$10^5$.
3. Complementary Validation: The stability of the identified phases was cross-verified using three distinct methods:
   • SSCHA: To relax the structure and calculate anharmonic phonon dispersions.
   • Molecular Dynamics Spectral Energy Density (MD-SED): To compute vibrational properties directly from equilibrium MD simulations, capturing anharmonicity to all orders.
   • Ab Initio Random Structure Search (AIRSS): To independently search for low-energy structures without bias, using 'shaken' supercells and symmetry-constrained searches.
4. Superconducting Properties: Electron-phonon coupling and superconducting transition temperatures were calculated using the isotropic Migdal-Eliashberg (ME) formalism (via the IsoME code) and anisotropic ME calculations (via EPW). The screened Coulomb interaction ($\mu^*$) was evaluated using the GW approximation.

Key Contributions and Results

• Discovery of a New Stable Phase: The study identifies a previously unreported, dynamically stable cubic phase of NbN with space group $P\bar{4}3m$. This structure emerges when the ideal 1:1 stoichiometric $\delta$-phase is allowed to fully relax under quantum anharmonic effects.
• Thermodynamic Stability: The $P\bar{4}3m$ phase is found to be 65 meV/atom lower in free energy than the ideal $Fm\bar{3}m$ $\delta$-phase. While still higher in energy than the hexagonal $\epsilon$-NbN phase (the predicted ground state at $x=1$), the $P\bar{4}3m$ phase is thermodynamically more favorable than the conventional cubic $\delta$-phase.
• Mechanism of Stabilization: The stabilization of the $P\bar{4}3m$ phase is driven primarily by quantum ionic motion (zero-point fluctuations) rather than higher-order anharmonic effects. The structural distortion lifts the high degeneracy of the $Fm\bar{3}m$ phase, particularly splitting bands at the X and $\Gamma$ points.
• Electronic Structure and Coupling: The lattice distortion in the $P\bar{4}3m$ phase reduces the electronic density of states (DOS) at the Fermi level compared to the ideal $\delta$-phase. This reduction leads to a markedly weakened electron-phonon coupling constant ($\lambda$) compared to calculations that artificially stabilize the $\delta$-phase using large electronic smearing.
• Superconducting Transition Temperature ($T_c$):
  • Using the SSCHA phonon dispersion for the $P\bar{4}3m$ phase, the calculated $T_c$ is 20 K.
  • This value aligns closely with experimentally reported $T_c$ values (~16 K) for near-stoichiometric bulk NbN.
  • In contrast, calculations that artificially stabilize the $Fm\bar{3}m$ phase via large electronic smearing overestimate $T_c$ to 27 K.
• Specificity to NbN: The authors note that similar SSCHA relaxation procedures applied to the related compounds TiN and NbC did not result in a symmetry-lowering phase transition, suggesting this phenomenon is specific to the quantum anharmonic landscape of NbN.

Significance and Claims
The paper challenges the long-held assumption that nitrogen vacancies are strictly necessary to stabilize the cubic phase of NbN. Instead, it proposes that quantum anharmonicity can stabilize a distinct, vacancy-free, distorted cubic phase ($P\bar{4}3m$) at the ideal 1:1 stoichiometry.

The authors claim that this discovery offers a new perspective on the synthesis of NbN. While previous efforts focused on tuning vacancy concentrations to tailor material properties, the findings suggest that synthesizing the ideal 1:1 stoichiometric cubic phase (specifically the $P\bar{4}3m$ structure) could provide a more direct route to achieving enhanced superconducting performance. The calculated $T_c$ of 20 K for the vacancy-free phase suggests that achieving perfect stoichiometry could potentially lead to even higher critical temperatures than currently observed in vacancy-stabilized samples.

The authors conclude by encouraging further experimental investigations, particularly using high-resolution diffraction techniques on single crystals or well-controlled powders, to validate the presence of the $P\bar{4}3m$ phase and assess its potential for superconducting applications. They also suggest that future studies should explore the impact of strain and chemical substitution on the stability of this phase and investigate the role of quantum anharmonicity in other superconducting materials.