Phys. Rev. B 114, L080201 (2026)
21 September 2026
First-order topology
Chiral Majorana edge modes
Class D in 2D
Second-order topology
Majorana corner modes
Particle–hole and C₄ symmetry in 2D
All clean higher-order phases have been classified,
but how do we identify the ones that are not?
Local environments fluctuate. Rotating a neighborhood need not map its sites onto one another.
Angular correlations recur around many sites within a single sample.
Diffraction reveals the order: sharp angular peaks on a ring, without a periodic lattice.
Aperiodic, with long-range quasiperiodic order and sharp diffraction peaks.
Liquid-like translational disorder, with sharp rotational order on a ring in diffraction.
S_{\mathrm{gyro}}(\mathbf k)\;\longleftrightarrow\;g_{\mathrm{quasi}}(\mathbf r)
Gyromorph diffraction ↔︎ quasicrystal correlations
g_{\mathrm{gyro}}(\mathbf r)\;\longleftrightarrow\;S_{\mathrm{quasi}}(\mathbf k)
Gyromorph correlations ↔︎ quasicrystal diffraction
14-fold examples.
Casiulis et al., Phys. Rev. Lett. 135, 196101 (2025)
Two coupled superconducting copies
\mathcal H=\sum_j\mathbf c_j^\dagger\epsilon_j\mathbf c_j +\sum_{j,k}f(r_{jk})\mathbf c_j^\dagger T_{jk}\mathbf c_k
\epsilon_j=\mu\,\sigma_z\otimes\tau_z
\begin{aligned} T_{jk}&=\frac{t}{2}\sigma_z\otimes\bigl[\tau_z-i(\tau_x\cos\theta_{jk}+\tau_y\sin\theta_{jk})\bigr]+\frac V2\cos(4\theta_{jk})\,\sigma_y\otimes\tau_0,\\ f(r)&=e^{1-r/R}\Theta(R-r). \end{aligned}
Symmetries: particle–hole and average local C_8
Are these corner modes protected?
Project the real-space Green’s function onto plane waves.
G_{\mathrm{eff}}(\mathbf k)_{ab}=\langle\mathbf k,a|(\mathcal H+i\eta)^{-1}|\mathbf k,b\rangle
H_{\mathrm{eff}}=\frac12\left[G_{\mathrm{eff}}^{-1}+(G_{\mathrm{eff}}^{-1})^\dagger\right]
Find momenta where the rotation becomes a good label:
\|[H_{\mathrm{eff}}(\mathbf k),C_8]\|\approx0.
At Γ and A, symmetry eigenvalues invert.
Caveat: C_8 is only approximate and H_{\mathrm{eff}} gets less accurate as k increases.
(H-E)\left(\Psi^{\mathrm{in}}\alpha^{\mathrm{in}}+\Psi^{\mathrm{out}}S(E)\alpha^{\mathrm{in}}+\Psi^{\mathrm{loc}}\alpha^{\mathrm{in}}\right)=0
S=\begin{pmatrix}r_{LL}&0\\0&r_{RR}\end{pmatrix},\qquad rr^\dagger=1
Recipe
Construct an exactly C_8-symmetric annulus from a ⅛ gyromorph sector.
C_8:\quad r(\phi)=Q_{C_8}\,r(\phi)\,V_{C_8}^{\dagger}
\mathcal P:\quad r(\phi)=r^*(-\phi)
At \phi=\pi, the +1 rotation block r' is real in \mathcal{P}-basis
\nu=\det r'(\pi) \in \{\pm1\}.
Caveat: repeating one ⅛ sector imposes exact C_8 symmetry.
Combine the Hamiltonian, position and chiral symmetry:
\widetilde L_{\mathbf x}= \left[i\mathcal H+\kappa\,\hat{\vphantom{X}e}_d\cdot(\mathbf X-\mathbf x)\right]\mathcal C.
Its half-signature counts local chiral charge:
\zeta_{\mathbf x}=\frac12\operatorname{sig}\widetilde L_{\mathbf x} =\frac12(N_+-N_-).
Caveat: Spectral localizers did not exist for spatially symmetric systems.
Conclusions: