Calculations predict a fully gapped p-wave triplet and chiral Majorana edge modes at a V2Se2O/conventional s-wave superconductor interface

At experimentally attainable hole doping of 2×10^13 cm^-2, monolayer V2Se2O has Néel-ordered vanadium moments of ±1.7µB, a maximal altermagnetic spin splitting of about 1 eV, and a Fermi surface of four disconnected hole pockets that are fully spin polarized and do not cross the Brillouin zone diagonals. Earlier proximity models for altermagnets used a minimal Fermi-surface geometry in which the induced state was generically nodal; this work replaces that abstract geometry with the ab-initio Fermi-surface topology of a specific material, V2Se2O. Based on full-potential local-orbital (FPLO) calculations with a 12×12×1 k-point mesh, the PBE functional, U=5.1 eV and J=0.8 eV for V d states, and doping simulated by the virtual crystal approximation; the Fermi surface is shown both from ab initio and from a fitted four-band effective model.

Coupling this altermagnetic layer to a conventional s-wave superconductor makes the momentum-dependent exchange splitting strongly favor equal-spin pairing and converts the proximity-induced order parameter into a fully gapped p-wave state, with Bogoliubov–de Gennes bands carrying a nonzero first Chern number over extended regions of the interface parameter space. The paper states this realizes the fully gapped descendant anticipated but not obtained in the earlier proximity model, and that the mechanism does not rely on fine-tuned spin-orbit coupling or magnetic exchange. The 12×12 heterostructure Hamiltonian is diagonalized for Δ0=0.2, λR=0.2, γ=0.9, η=1 on a 51×51 momentum mesh, yielding a topological phase diagram in the interface tunneling spin polarization p and hybridization vM; extended ch1=±1 regions survive at η=0.5, while no ch1=±2 phase is found.

In a 100-layer slab at vM=0.1 and p=0.35 (ch1=−1), the boundary spectral function exhibits one chiral Majorana edge mode crossing the superconducting gap, whereas the ch1=0 regime at vM=0 and p=0 shows two counterpropagating Majorana modes that are not protected by the total Chern number. The work ties bulk topological invariants directly to boundary spectra and distinguishes Chern-protected chiral modes from counterpropagating modes that may hybridize or merge with the bulk spectrum. The slab calculation uses the five outermost layers along a1 with periodic boundary conditions along a2; the ch1=0 example sets γ, µ, Δ0 and λR to unity for visual clarity, giving a gap comparable to the chiral case.

Self-consistent layer-resolved singlet gaps in a finite stack show that the inverse proximity effect strongly suppresses the first superconducting layer adjacent to the magnetic interface, yet the gap remains finite for all parameters considered and recovers its bulk value within a few layers, with the healing length growing from about one superconducting layer at t⊥/t=0.25 to about four layers at t⊥/t=1. The paper incorporates the inverse proximity effect and separates magnetic pair breaking from the generic loss caused by tunneling, noting that the spin-splitting parameter η has little effect beyond the first superconducting layer. Calculations use N=100 superconducting layers, T=0.01, a local attractive interaction G=2.82 per layer, and fixed-point iteration of the layer-resolved gap until convergence; reducing the tunneling scale from γ=0.9 to 0.4 leaves the bulk order parameter and qualitative t⊥ dependence essentially unchanged while substantially reducing suppression in the first few layers.

The result targets interfaces between monolayer V2Se2O and a conventional s-wave superconductor, and applies to altermagnetic thin films whose Fermi surface consists of disconnected, fully spin-polarized pockets with finite interfacial Rashba coupling. The design rule offered is that the superconducting film should be thicker than several healing lengths so a robust parent condensate survives away from the interface, while enough hybridization is retained to induce the topological gap; p and vM are described as tunable through interface engineering such as surface termination, stacking registry or twist angle, interlayer separation, strain, electrostatic gating, and vertical pressure. The paper also proposes experimentally testable signatures: tunneling spectroscopy should detect a hard induced gap in the bulk, edge-sensitive scanning tunneling microscopy should reveal in-gap spectral weight confined to step edges or patterned boundaries, thermal transport provides a charge-neutral probe of the chiral boundary channel, Josephson interferometry can test the odd-parity character of the induced order, and advanced scanning magnetometry can detect magnetic fields generated by currents associated with the protected edge modes.

This is a theoretical work: the conclusions rest on an effective model parameterized by ab-initio input and on self-consistent mean-field Bogoliubov–de Gennes calculations, with no experimental realization or measurement yet. The microscopic values of the interface parameters p and vM are not fixed by bulk altermagnetic symmetry but depend on the specific termination, strain, reconstruction, or moiré structure, so where a real device falls in the topological regions of the phase diagram remains to be determined experimentally. The counterpropagating Majorana modes in the ch1=0 regime are not protected by the total Chern number and may hybridize or merge with the bulk spectrum when the symmetry relating the two spin sectors is broken. The healing length ξh is presented as a useful estimate of the minimum superconducting thickness rather than a sharp critical thickness. Data and code are available from the corresponding author upon reasonable request, so independent reproduction currently depends on the model and parameters given in the text.