QAHI/QSHI/QAHI Junction Overview

• QAHI/QSHI/QAHI junction is a heterostructure combining quantum anomalous and spin Hall insulators, where edge modes are converted and filtered across interfaces.
• The device uses a buckled-honeycomb lattice with staggered antiferromagnetic and perpendicular electric fields to tune between QSHI, QAHI, and band-insulating phases.
• Robust quantized magnetoresistance and perfect chiral channel transmission are achieved through electrically switchable spin polarization and topological protection.

A QAHI/QSHI/QAHI junction is a heterostructure in which two quantum anomalous Hall insulator (QAHI) regions flank a central quantum spin Hall insulator (QSHI) region, so that transport is governed by the conversion, filtering, and matching of topological edge modes across interfaces. In the antiferromagnetic buckled-honeycomb realization proposed for silicene-, germanene-, and stanene-type nanoribbons, the junction is controlled by a staggered antiferromagnetic exchange field $M_{AF}$ and a perpendicular electric field $E_z$, which tune the bulk Dirac masses and thereby switch the system among QSHI, QAHI, and band-insulator (BI) phases. The resulting magnetoresistance is topologically protected by the Chern number, electrically switchable, and robust against finite-size variation, smooth boundaries, and substantial disorder (Lu et al., 4 Sep 2025).

1. Buckled-honeycomb realization and microscopic model

The canonical realization uses a buckled honeycomb lattice nanoribbon in which three ingredients are essential: nearest-neighbor hopping $t$, intrinsic spin–orbit coupling $\lambda$, and sublattice-staggered terms induced respectively by an antiferromagnetic exchange field and by the buckled structure under a perpendicular electric field. The tight-binding Hamiltonian is

$$\begin{aligned} H &= -t \sum_{\langle i,j \rangle, \alpha} c_{i \alpha}^{\dagger} c_{j \alpha} + i \frac{\lambda}{3\sqrt{3}} \sum_{\langle\langle i,j \rangle\rangle, \alpha, \beta} v_{ij}\, c_{i \alpha}^{\dagger} (\sigma_z)_{\alpha\beta} c_{j \beta} \ &\quad + M_{AF} \sum_{i,\alpha} \xi_i\, c_{i \alpha}^{\dagger} (\sigma_z)_{\alpha\alpha} c_{i \alpha} - l E_z \sum_{i,\alpha} \xi_i\, c_{i \alpha}^{\dagger} c_{i \alpha}. \end{aligned}$$

Here $\xi_i=+1$ on sublattice A and $-1$ on sublattice B, so the AF exchange is staggered and breaks time-reversal symmetry without net magnetization, while the electric field produces a staggered potential through the buckling height $l$. The reported values are $l=0.035\ \mathrm{e\AA}$ for silicene, $0.046\ \mathrm{e\AA}$ for germanene, and $E_z$0 for stanene. No Rashba term is included; the physics is governed by intrinsic SOC and staggered mass terms (Lu et al., 4 Sep 2025).

Near the $E_z$1 and $E_z$2 valleys, the low-energy continuum Hamiltonian is

$E_z$3

with $E_z$4 for valley, $E_z$5 for spin, and $E_z$6 acting in sublattice space. The bulk dispersion is

$E_z$7

so the topological character is controlled by the mass term

$E_z$8

This formulation makes the junction fundamentally a mass-domain device: the left, middle, and right segments differ by the sign structure of $E_z$9, and the interfaces inherit their transport properties from that mass inversion pattern.

2. Phase structure and topological classification

The topological phase diagram in the $t$0 plane is determined by the Chern number,

$t$1

with quantized Hall conductivity $t$2. Phase boundaries occur when one valley–spin mass vanishes,

$t$3

The resulting classification is compactly summarized as follows (Lu et al., 4 Sep 2025).

Phase	Condition	Edge/transport character
QSHI	$t$4	Helical edge states; $t$5; $t$6 conductance
BI	$t$7	No protected edge states; $t$8
QAHI	$t$9	Single chiral edge channel; $\lambda$0

In the QSHI regime, the spin-resolved valley Chern numbers satisfy

$\lambda$1

which yields counterpropagating helical edge states. In the BI regime, all masses keep the same sign across valleys and spins, and no protected edge states appear. In the QAHI regime, one spin sector undergoes band inversion and produces $\lambda$2. The first and third quadrants of the phase diagram, where the signs of $\lambda$3 and $\lambda$4 are aligned, correspond to a spin-up-polarized QAHE with $\lambda$5; the second and fourth quadrants correspond to a spin-down-polarized QAHE with $\lambda$6 (Lu et al., 4 Sep 2025).

A central quantitative feature is the QAHE bulk gap,

$\lambda$7

which increases with $\lambda$8 and $\lambda$9. This growth directly enlarges the parameter and energy windows in which the junction exhibits quantized magnetoresistance.

3. Edge states, spin polarization, and electrical switching

The transport distinction between QSHI and QAHI segments is encoded in their edge-state content. In the QSHI phase, spin-up and spin-down counterpropagate, giving a two-channel conductance $$\begin{aligned} H &= -t \sum_{\langle i,j \rangle, \alpha} c_{i \alpha}^{\dagger} c_{j \alpha} + i \frac{\lambda}{3\sqrt{3}} \sum_{\langle\langle i,j \rangle\rangle, \alpha, \beta} v_{ij}\, c_{i \alpha}^{\dagger} (\sigma_z)_{\alpha\beta} c_{j \beta} \ &\quad + M_{AF} \sum_{i,\alpha} \xi_i\, c_{i \alpha}^{\dagger} (\sigma_z)_{\alpha\alpha} c_{i \alpha} - l E_z \sum_{i,\alpha} \xi_i\, c_{i \alpha}^{\dagger} c_{i \alpha}. \end{aligned}$$0. In the QAHI phase, a single chiral channel appears, and its spin is fully polarized. The spin orientation is determined by the sign structure of $$\begin{aligned} H &= -t \sum_{\langle i,j \rangle, \alpha} c_{i \alpha}^{\dagger} c_{j \alpha} + i \frac{\lambda}{3\sqrt{3}} \sum_{\langle\langle i,j \rangle\rangle, \alpha, \beta} v_{ij}\, c_{i \alpha}^{\dagger} (\sigma_z)_{\alpha\beta} c_{j \beta} \ &\quad + M_{AF} \sum_{i,\alpha} \xi_i\, c_{i \alpha}^{\dagger} (\sigma_z)_{\alpha\alpha} c_{i \alpha} - l E_z \sum_{i,\alpha} \xi_i\, c_{i \alpha}^{\dagger} c_{i \alpha}. \end{aligned}$$1, so changing $$\begin{aligned} H &= -t \sum_{\langle i,j \rangle, \alpha} c_{i \alpha}^{\dagger} c_{j \alpha} + i \frac{\lambda}{3\sqrt{3}} \sum_{\langle\langle i,j \rangle\rangle, \alpha, \beta} v_{ij}\, c_{i \alpha}^{\dagger} (\sigma_z)_{\alpha\beta} c_{j \beta} \ &\quad + M_{AF} \sum_{i,\alpha} \xi_i\, c_{i \alpha}^{\dagger} (\sigma_z)_{\alpha\alpha} c_{i \alpha} - l E_z \sum_{i,\alpha} \xi_i\, c_{i \alpha}^{\dagger} c_{i \alpha}. \end{aligned}$$2 or $$\begin{aligned} H &= -t \sum_{\langle i,j \rangle, \alpha} c_{i \alpha}^{\dagger} c_{j \alpha} + i \frac{\lambda}{3\sqrt{3}} \sum_{\langle\langle i,j \rangle\rangle, \alpha, \beta} v_{ij}\, c_{i \alpha}^{\dagger} (\sigma_z)_{\alpha\beta} c_{j \beta} \ &\quad + M_{AF} \sum_{i,\alpha} \xi_i\, c_{i \alpha}^{\dagger} (\sigma_z)_{\alpha\alpha} c_{i \alpha} - l E_z \sum_{i,\alpha} \xi_i\, c_{i \alpha}^{\dagger} c_{i \alpha}. \end{aligned}$$3 flips the spin polarization of the chiral edge state and reverses the sign of the Chern number (Lu et al., 4 Sep 2025).

The transport spin polarization is defined as

$$\begin{aligned} H &= -t \sum_{\langle i,j \rangle, \alpha} c_{i \alpha}^{\dagger} c_{j \alpha} + i \frac{\lambda}{3\sqrt{3}} \sum_{\langle\langle i,j \rangle\rangle, \alpha, \beta} v_{ij}\, c_{i \alpha}^{\dagger} (\sigma_z)_{\alpha\beta} c_{j \beta} \ &\quad + M_{AF} \sum_{i,\alpha} \xi_i\, c_{i \alpha}^{\dagger} (\sigma_z)_{\alpha\alpha} c_{i \alpha} - l E_z \sum_{i,\alpha} \xi_i\, c_{i \alpha}^{\dagger} c_{i \alpha}. \end{aligned}$$4

In a spin-filter geometry where $$\begin{aligned} H &= -t \sum_{\langle i,j \rangle, \alpha} c_{i \alpha}^{\dagger} c_{j \alpha} + i \frac{\lambda}{3\sqrt{3}} \sum_{\langle\langle i,j \rangle\rangle, \alpha, \beta} v_{ij}\, c_{i \alpha}^{\dagger} (\sigma_z)_{\alpha\beta} c_{j \beta} \ &\quad + M_{AF} \sum_{i,\alpha} \xi_i\, c_{i \alpha}^{\dagger} (\sigma_z)_{\alpha\alpha} c_{i \alpha} - l E_z \sum_{i,\alpha} \xi_i\, c_{i \alpha}^{\dagger} c_{i \alpha}. \end{aligned}$$5 and $$\begin{aligned} H &= -t \sum_{\langle i,j \rangle, \alpha} c_{i \alpha}^{\dagger} c_{j \alpha} + i \frac{\lambda}{3\sqrt{3}} \sum_{\langle\langle i,j \rangle\rangle, \alpha, \beta} v_{ij}\, c_{i \alpha}^{\dagger} (\sigma_z)_{\alpha\beta} c_{j \beta} \ &\quad + M_{AF} \sum_{i,\alpha} \xi_i\, c_{i \alpha}^{\dagger} (\sigma_z)_{\alpha\alpha} c_{i \alpha} - l E_z \sum_{i,\alpha} \xi_i\, c_{i \alpha}^{\dagger} c_{i \alpha}. \end{aligned}$$6 act only in a central QAHI region, the choice $$\begin{aligned} H &= -t \sum_{\langle i,j \rangle, \alpha} c_{i \alpha}^{\dagger} c_{j \alpha} + i \frac{\lambda}{3\sqrt{3}} \sum_{\langle\langle i,j \rangle\rangle, \alpha, \beta} v_{ij}\, c_{i \alpha}^{\dagger} (\sigma_z)_{\alpha\beta} c_{j \beta} \ &\quad + M_{AF} \sum_{i,\alpha} \xi_i\, c_{i \alpha}^{\dagger} (\sigma_z)_{\alpha\alpha} c_{i \alpha} - l E_z \sum_{i,\alpha} \xi_i\, c_{i \alpha}^{\dagger} c_{i \alpha}. \end{aligned}$$7 produces, for $$\begin{aligned} H &= -t \sum_{\langle i,j \rangle, \alpha} c_{i \alpha}^{\dagger} c_{j \alpha} + i \frac{\lambda}{3\sqrt{3}} \sum_{\langle\langle i,j \rangle\rangle, \alpha, \beta} v_{ij}\, c_{i \alpha}^{\dagger} (\sigma_z)_{\alpha\beta} c_{j \beta} \ &\quad + M_{AF} \sum_{i,\alpha} \xi_i\, c_{i \alpha}^{\dagger} (\sigma_z)_{\alpha\alpha} c_{i \alpha} - l E_z \sum_{i,\alpha} \xi_i\, c_{i \alpha}^{\dagger} c_{i \alpha}. \end{aligned}$$8, a plateau with

$$\begin{aligned} H &= -t \sum_{\langle i,j \rangle, \alpha} c_{i \alpha}^{\dagger} c_{j \alpha} + i \frac{\lambda}{3\sqrt{3}} \sum_{\langle\langle i,j \rangle\rangle, \alpha, \beta} v_{ij}\, c_{i \alpha}^{\dagger} (\sigma_z)_{\alpha\beta} c_{j \beta} \ &\quad + M_{AF} \sum_{i,\alpha} \xi_i\, c_{i \alpha}^{\dagger} (\sigma_z)_{\alpha\alpha} c_{i \alpha} - l E_z \sum_{i,\alpha} \xi_i\, c_{i \alpha}^{\dagger} c_{i \alpha}. \end{aligned}$$9

Reversing $\xi_i=+1$0 or $\xi_i=+1$1 flips the plateau to

$\xi_i=+1$2

This establishes that the chiral edge states are fully spin-polarized along $\xi_i=+1$3, and that their orientation is both electrically and magnetically controllable. In the junction context, this is the decisive mechanism: the left and right QAHI segments can be tuned to have either matching or opposite spin-polarized chiral channels, which in turn enforces either perfect transmission or complete blocking across the intervening QSHI region.

4. Junction construction, transport formalism, and selection rules

The proposed device is a nanoribbon of width $\xi_i=+1$4, extended along $\xi_i=+1$5, with three segments of equal length $\xi_i=+1$6 unit cells. The left and right segments are QAHI regions with applied fields, and the central segment is a QSHI region with fields turned off. The fields can be configured in two ways:

• Parallel: $\xi_i=+1$7, with the same $\xi_i=+1$8 on both sides.
• Antiparallel: $\xi_i=+1$9, or equivalently by flipping $-1$0 on one side.

Two-terminal conductance is computed with nonequilibrium Green’s functions and the Landauer–Büttiker formula,

$-1$1

where

$-1$2

In the QAHE energy window $-1$3, the junction exhibits sharply different selection rules in the two configurations. In the parallel case, the left and right QAHI regions host the same spin-polarized chiral edge state with identical Chern number, so a single perfectly transmitting channel persists and

$-1$4

In the antiparallel case, the right QAHI chiral channel has the opposite spin polarization, so the QAHI–QSHI interfaces impose a spin mismatch and forbid transmission of the chiral edge state across the full device: $-1$5

The stated origin of these rules is that the QAHI–QSHI interface couples only to the spin/valley sector with matching mass inversion; the antiparallel configuration lacks such a matching interface mode across the junction (Lu et al., 4 Sep 2025).

5. Magnetoresistance, quantization, and robustness

The junction’s magnetoresistance is defined by

$-1$6

For the representative choice $-1$7, $-1$8, and $-1$9, the QAHE regime $l$0 yields the quantized plateau

$l$1

As $l$2 and $l$3 increase, the bulk QAHE gap grows, and the MR plateau broadens in energy. In parameter-space maps, the QAHI regions yield $l$4, $l$5, and $l$6; the QSHI regions yield $l$7 and $l$8; the BI regions yield $l$9 and $l=0.035\ \mathrm{e\AA}$0. For fixed energy, such as $l=0.035\ \mathrm{e\AA}$1, MR as a function of $l=0.035\ \mathrm{e\AA}$2 exhibits wide plateaus centered in the QAHE intervals, with width about $l=0.035\ \mathrm{e\AA}$3 when $l=0.035\ \mathrm{e\AA}$4 (Lu et al., 4 Sep 2025).

The robustness analysis is unusually explicit. Varying $l=0.035\ \mathrm{e\AA}$5, $l=0.035\ \mathrm{e\AA}$6, $l=0.035\ \mathrm{e\AA}$7, or $l=0.035\ \mathrm{e\AA}$8 leaves the quantized MR plateau intact. Reducing $l=0.035\ \mathrm{e\AA}$9 and $0.046\ \mathrm{e\AA}$0 slightly widens the plateau because of tunneling, whereas changing the QSHI length $0.046\ \mathrm{e\AA}$1 does not affect MR. Increasing $0.046\ \mathrm{e\AA}$2 adds bands and shifts edge dispersions slightly, modestly broadening the QAHE windows without destroying the plateaus. For $0.046\ \mathrm{e\AA}$3, the plateau position is approximately the QAHE interval

$0.046\ \mathrm{e\AA}$4

Disorder is modeled as Anderson-type on-site randomness in the central scattering region, uniformly distributed in $0.046\ \mathrm{e\AA}$5, and averaged over up to 3600 configurations. For a region of total length $0.046\ \mathrm{e\AA}$6 and width $0.046\ \mathrm{e\AA}$7, the chiral-edge-state plateau $0.046\ \mathrm{e\AA}$8 remains robust up to large disorder strengths, $0.046\ \mathrm{e\AA}$9, while $E_z$00 remains zero inside the gap. The MR plateau can even widen for weak to moderate disorder because $E_z$01 preserves its plateau across broader parameter ranges while $E_z$02 stays zero. Only at extremely strong disorder, for example $E_z$03, does the MR plateau shrink and negative MR emerge because bulk states begin to contribute. The protection mechanism is the absence of backscattering for chiral edge states together with topological mismatch at the QAHI–QSHI interfaces in the antiparallel case (Lu et al., 4 Sep 2025).

A related practical result concerns interface smoothness. Spatially varying $E_z$04 and $E_z$05 are modeled by error-function profiles over $E_z$06 unit cells per interface. The QAHE-regime MR remains perfect for $E_z$07 regardless of boundary smoothness; only bulk-state transport outside the gap is modified. This directly attributes the device response to the segment topology rather than to sharp-interface idealization.

6. Alternative platforms, conceptual variants, and related junction physics

The buckled-honeycomb antiferromagnetic junction is not the only framework in which a QAHI/QSHI/QAHI sequence can be engineered. A distinct route is provided by Cr-doped $E_z$08 thin films under high-frequency circularly polarized light, where the interplay between the static exchange field $E_z$09 and the light-induced term produces a phase diagram containing a normal insulator, a time-reversal-symmetry-broken QSHI, and two QAHI phases with opposite Chern numbers. In that setting, a spatial sequence $E_z$10 can be designed by modulating light amplitude, helicity, or film thickness; the interfaces then host chiral domain-wall channels fixed by the Chern-number change $E_z$11 (Qin et al., 2021).

A further extension arises in magnetically doped topological-insulator thin films with correlated quasi-periodic disorder. That work directly studies QSHI leads attached to a magnetic central region and identifies QAHI, QSHI, and quantum spin Chern insulator phases via conductance quantization and self-consistent Born approximation. Its discussion of a QAHI/QSHI/QAHI junction is explicitly an adaptation rather than a direct simulation: it argues that the two-terminal conductance should remain limited by the single chiral channel of the QAHI leads, so that an idealized QAHI/QSHI/QAHI geometry would have $E_z$12, with strong sensitivity to disorder orientation in the central region and with transverse quasi-periodicity offering the greatest robustness (Okugawa et al., 2022). This suggests that the precise conductance quantization depends on the lead-channel structure of the chosen platform, even when the interface physics remains topological.

Related superconducting hybrids reveal a different aspect of QAHI/QSHI/QAHI phenomenology. In QAHI Josephson junctions, chiral edge supercurrents produce a $E_z$13-periodic interference pattern, whereas bulk carriers or magnetic domains generate asymmetric Fraunhofer or Fraunhofer-like patterns. The discussion of QAHI/QSHI/QAHI hybrids in that context is again presented as an implication rather than a direct calculation: the expectation is a superposition of chiral $E_z$14-periodic responses from QAHI segments and helical or bulk-like contributions from the QSHI region, with asymmetry inherited from broken time-reversal symmetry in the QAHI leads (Qi et al., 2023). A plausible implication is that the same interface mode-conversion physics that governs normal-state magnetoresistance should also shape phase-sensitive superconducting transport.

Across these variants, the common structure is a junction between regions whose Chern and spin-resolved mass patterns differ. In the antiferromagnetic buckled-honeycomb proposal, that structure yields an electrically switchable MR element with $E_z$15 on wide plateaus, without an external magnetic field and with robustness set by topological invariance. The principal limitations identified are the requirement to remain in the QAHE regime, the sensitivity of small negative MR near phase boundaries to energy and disorder, and the eventual loss of protection once disorder is strong enough to close the bulk gap (Lu et al., 4 Sep 2025).