betaFeedback

WorksPhysics and Astronomy

Collective Hall quantization in 3D Chern vector systems

Abstract

Understanding the three-dimensional (3D) quantum Hall effect, characterized by a Chern vector rather than a single Chern number, has been a central pursuit in topological physics. While quantized Hall transport is established for the special case with one nonzero Chern component, recent realizations of more general Chern-vector states raise a fundamental question: how does such topology protect quantized transport? Here, we predict a distinct 3D quantized Hall effect protected by C=(0,m,n)\mathbf{C}=(0,m,n). For samples with dimensions LyL_y and LzL_z, global topology enforces collective quantization of mLy+nLzmL_y+nL_z, despite the fragility of individual edge modes to disorder. This reveals a novel transport regime in which robust quantization emerges as a holistic property of a unified chiral network,rather than of its constituent modes. Crucially, this collective response manifests as a quantized Hall conductance Gxy=±(mLy+nLz)e2/hG_{xy}=\pm(mL_y+nL_z)e^2/h, with a distinctive dependence on both Chern-vector components and sample dimensions. We propose candidate materials to facilitate experimental observation.

Introduction

The two-dimensional (2D) quantized Hall effect (QHE) [5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20] [1, 2, 3, 4] has long been a cornerstone of topological physics, establishing a profound connection between the Chern number CxyC_{xy}, chiral edge modes, and the precisely quantized Hall conductance Gxy=Cxye2/hG_{xy}=C_{xy}e^2/h [5, 6, 7], as shown in Figure 1(A). Recently, interest has expanded to three-dimensional (3D) systems where various Hall responses have been proposed and experimentally verified [15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30]. Generalizing the scalar Chern number to 3D systems yields the Chern vector C=(Cyz,Cxz,Cxy)\mathbf{C}=(C_{yz},C_{xz},C_{xy}) [31, 32, 33, 34], which governs both the spatial configuration and transport of chiral boundary states. For example, when C=(0,0,n)\mathbf{C}=(0,0,n), as shown in Figure 1(B), each of the LzL_z structural periods along the zz-direction contributes a Chern number nn, resulting in a quantized Hall conductance Gxy=nLze2/hG_{xy}=nL_ze^2/h. While this framework has successfully underpinned research on tunable quantum anomalous Hall effects and led to significant experimental milestones [15, 16, 17, 18], Chern vectors with multiple nonzero components—which host far more intricate chiral boundary states—remain largely unexplored in quantum transport.

Figure 1. Chiral modes, front surface state spectra, conductances and Hall effect from Chern-number insulators to Chern-vector insulators. (A) 2D samples with Cxy=nC_{xy}=n; (B) 3D samples with a single nonzero Chern component C=(0,0,n)\mathbf{C}=(0,0,n); (C) 3D samples with multiple nonzero Chern components C=(0,m,n)\mathbf{C}=(0,m,n), with td=0t_d=0 for the schematic spectra. The sample sizes are LαL_\alpha with α∈x,y,z\alpha\in x,y,z. GαG_\alpha is the two-terminal conductance along α\alpha direction. To simplify the analysis, we set m=n=1m=n=1 and e2/h=1e^2/h=1 by default.

Very recently, 3D topological states characterized by Chern vectors C=(0,m,n)\mathbf{C}=(0,m,n) and their exotic surface states have been experimentally realized [31, 32, 34]. However, existing studies have primarily focused on their static band dispersions. How quantum transport is fundamentally governed by a multi-component Chern vector remains an unresolved and pressing challenge. Crucially, a Chern vector of the form C=(0,m,n)\mathbf{C}=(0,m,n) manifests as an interconnected chiral network [Fig. 1(C)], where complex loops and cross-surface links induce substantial scattering between opposite surfaces. This behavior fundamentally departs from that of conventional 2D Chern insulators [8] or simple 3D stacks with C=(0,0,Cxy)\mathbf{C}=(0,0,C_{xy}) [15]. Under such inter-surface scattering, the individual chiral modes—though nominally protected by Chern numbers—become “fragile”: they no longer sustain the backscattering-free transport that characterizes the 2D paradigm [35]. Consequently, standard frameworks, such as the layer-based TKNN formula [36, 37], fail to capture the transport physics, raising a fundamental question: Can any quantized transport signature survive when local edge-state protection is compromised by global network complexity?

Here, we demonstrate that a 3D quantized Hall effect is uniquely protected by the Chern vector C=(0,m,n)\mathbf{C}=(0,m,n) through an emergent mechanism of collective quantization. We reveal that although the transmission of individual edge states deviates from quantization due to inter-surface scattering, the total transmission coefficient of the unified chiral network remains robustly quantized at (mLy+nLz)(mL_y+nL_z). This collective immunity transcends the local fragility of constituent modes, revealing a shift from individual to holistic topological protection. Applying the Landauer-Büttiker formalism [38], we show that this topological protection underpins distinct multi-terminal responses measurable in standard Hall-bar setups, leading to four quantized Hall conductances: Gxy=Gxz=−Gyx=−Gzx=(mLy+nLz)e2/hG_{xy}=G_{xz}=-G_{yx}=-G_{zx}=(mL_y+nL_z)e^2/h. These conductances exhibit a distinctive scaling with the sample dimensions (Ly,Lz)(L_y,L_z), directly reflecting the intrinsic 3D nature of the effect. Finally, we propose realistic solid-state candidate materials to facilitate experimental realization. Our work bridges the gap between formal 3D topological invariants and measurable quantized transport.

Model

Our starting point is a minimal tight-binding model composed of two coupled 3D Chern-insulator sectors [see Appendix and Supplementary Material (SM) Note 0 [39] for details],

H=(Hxyz00Hxzy)+∑n(0TdTd0)cn†cn.(1)H=\begin{pmatrix}\mathcal{H}_{xyz} & 0 \\ 0 & \mathcal{H}_{xzy}\end{pmatrix}+\sum_n\begin{pmatrix}0 & T_d \\ T_d & 0\end{pmatrix}c_n^\dagger c_n. \tag*{(1)}

Here, Hxyz\mathcal{H}_{xyz} describes a 3D Chern insulator characterized by the Chern vector Cxyz=(Cyz,Cxz,Cxy)=(0,0,1)\mathbf{C}_{xyz}=(C_{yz},C_{xz},C_{xy})=(0,0,1), [7, 8, 9, 10, 11, 12, 13, 14, 15, 16] whereas Hxzy\mathcal{H}_{xzy} is obtained by exchanging the roles of the yy and zz directions and is characterized by Cxzy=(0,1,0)\mathbf{C}_{xzy}=(0,1,0). The explicit real-space and momentum-space forms of Hxyz\mathcal{H}_{xyz} and Hxzy\mathcal{H}_{xzy} are given in the Appendix.

According to the bulk-boundary correspondence, Hxyz\mathcal{H}_{xyz} gives rise to red chiral modes associated with the xx-yy boundaries, while Hxzy\mathcal{H}_{xzy} produces blue chiral modes associated with the xx-zz boundaries, as illustrated in Figure 2(A). In the decoupled limit, their combination therefore realizes the composite Chern vector C=(0,1,1)\mathbf{C}=(0,1,1).

Figure 2. Chiral surface states and quantized transport in a Chern-vector insulator. (A) Schematic illustration of the chiral surface states in a system with Chern vector C=(0,1,1)\mathbf{C}=(0,1,1). The sample details are indicated in the figure. (B) Half of the sample in (A), where solid (dashed) lines denote the visible (hidden) chiral modes. (C) Enlarged views of a typical chiral intersection in (A) for different inter-mode coupling strengths tdt_d. (D),(E) Chiral network transport for various lead configurations in (B). (F),(G) Energy spectra EE versus kyk_y with and without inter-mode coupling tdt_d, respectively. (H) Typical spatial distributions of the eigenstates, ρ=∣ψn∣2\rho=\lvert\psi_n\rvert^2, for the blue and red modes indicated in (F). (I) Two-terminal conductance Gγ(αβ)G_\gamma^{(\alpha\beta)} as a function of Fermi energy EE. We define Gγ(αβ)G_\gamma^{(\alpha\beta)} as the conductance of chiral modes localized at the α\alpha-β\beta boundaries, where the current flows along the γ\gamma axis, for α,β,γ∈{x,y,z}\alpha,\beta,\gamma\in\{x,y,z\}. (J) Gγ(αβ)G_\gamma^{(\alpha\beta)} as a function of disorder strength WW at Fermi energy E=−0.32tE=-0.32t. (K) shows the variance of the data in (J). The parameters are Lx=Ly=Lz=10L_x=L_y=L_z=10 and N=15N=15.

The two sectors are coupled through the hopping matrix Td=td(I2×2+τx)T_d=t_d(I_{2\times2}+\tau_x), where tdt_d denotes the inter-sector coupling strength. A finite tdt_d hybridizes the intersecting chiral surface modes and modifies their connectivity without closing the bulk gap in the parameter regime considered here [see Appendix and SM Note 1].

To examine the robustness of the transport properties, Anderson disorder is included through Hw=∑nϵncn†cnH_w=\sum_n\epsilon_n c_n^\dagger c_n, where ϵn\epsilon_n is uniformly distributed within [−W/2,W/2][-W/2,W/2], with WW denoting the disorder strength [40, 41, 42].

Chiral network configuration and terminal responses

We first investigate the influence of coupling tdt_d on the chiral mode configurations. As shown in Figure 2(A) and (C), the surface networks exhibit coupling-dependent scattering between red and blue chiral channels with probability P(td)P(t_d), spanning three limits: (i) the decoupling limit where Ptd=0=0P_{t_d=0}=0; (ii) the moderate coupling regime with Ptd≠0≠0P_{t_d\ne0}\ne0; and (iii) the strong coupling limit where Ptd→∞=1P_{t_d\to\infty}=1, which reconfigures the network geometry.

Crucially, the bulk Chern vector C=(0,1,1)\mathbf{C}=(0,1,1) remains invariant under varying tdt_d [see SM. Note 1]. Despite these structural variations (a feature absent in the C=(0,0,1)\mathbf{C}=(0,0,1) case), all configurations here obey a universal unidirectional scattering rule: tunneling occurs strictly from A/B to C/D [Fig. 2(C)]. As this rule is consistently captured by configuration (ii), this regime provides a unified framework to describe transport across all coupling strengths, as governed by C=(0,1,1)\mathbf{C}=(0,1,1).

To characterize the topology, we examine the two-terminal conductance Gγ(αβ)G_\gamma^{(\alpha\beta)}. Here, we introduce the notation Gγ(αβ)G_\gamma^{(\alpha\beta)} to denote the two terminal conductance in the γ\gamma-direction, originating from the chiral modes located on the α\alpha-β\beta boundaries. For example, in the conventional case [Fig. 1(B)], the conductance is topologically protected and determined solely by the planar Chern number with Gy(xy)=CxyLzG_y^{(xy)}=C_{xy}L_z.

However, for a general Chern vector C=(0,Cxz,Cxy)\mathbf{C}=(0,C_{xz},C_{xy}), the two-terminal responses lose this topological protection. Although the components of the Chern vector remain invariant as per the TKNN formula, the non-zero tdt_d induces scattering within the chiral network, fundamentally altering the transport. In the yy direction, the red modes exhibit traveling features that dominate the conductance, whereas the blue modes are nearly localized in flat bands [Fig. 2(F)]. However, scattering between these modes partially gaps out the spectrum [Fig. 2(G)], causing Gy(xy)G_y^{(xy)} to deviate from the quantized value CxyLzC_{xy}L_z. For instance, in both clean [Fig. 2(I)] and disordered [Fig. 2(J)] cases, Gy(xy)G_y^{(xy)} falls below the expected value CxyLz=10C_{xy}L_z=10. Intuitively, this reduction, expressed as Gy(xy)=[Cxy−P1(td)]LzG_y^{(xy)}=[C_{xy}-P_1(t_d)]L_z, originates from backscattering induced by the coupling between chiral modes on opposite surfaces [Fig. 2(D)].

In the xx-direction, Gx(xz)G_x^{(xz)} similarly deviates from the quantized value CxzLyC_{xz}L_y, but notably shifts toward a higher value in both clean and disordered cases [Figs. 2(I)–(J)]. This enhancement, described by Gx(xz)=CxzLy+P2(td)LzG_x^{(xz)}=C_{xz}L_y+P_2(t_d)L_z, stems from additional transport channels facilitated by inter-mode scattering [Fig. 2(E)]. Because these typical two-terminal responses Gy(xy)G_y^{(xy)} and Gx(xz)G_x^{(xz)} lack topological protection, they cannot directly reflect the intrinsic (0,1,1)(0,1,1) topology. This discrepancy necessitates an alternative mechanism to explain the robust, quantized transport observed in the system.

Remarkably, the scattering processes in Figure 2(B) that decrease Gy(xy)G_y^{(xy)} [Fig. 2(D)] and increase Gx(xz)G_x^{(xz)} [Fig. 2(E)] are physically identical, inherently ensuring charge conservation [See SM. Note 3 for details]. This suggests that the summation Gx(xz)+Gy(xy)G_x^{(xz)}+G_y^{(xy)} remains topologically protected provided P1(td)=P2(td)P_1(t_d)=P_2(t_d), as the scattering-induced deviations exactly cancel each other out:

Gx(xz)+Gy(xy)=[CxzLy+P2(td)Lz]+[CxyLz−P1(td)Lz]=CxzLy+CxyLz.(2)\begin{aligned} G_x^{(xz)}+G_y^{(xy)}&=[C_{xz}L_y+P_2(t_d)L_z]\\ &+[C_{xy}L_z-P_1(t_d)L_z]\\ &=C_{xz}L_y+C_{xy}L_z. \tag*{(2)} \end{aligned}

This quantized sum defines a total Chern vector Ct=(0,CxzLy,CxyLz)\mathbf{C}_t=(0,C_{xz}L_y,C_{xy}L_z), serving as a robust measure of the aggregate transport channels within a typical sample.

To verify this conclusion, Figure 2(I) shows numerical results with Ct=(0,10,10)\mathbf{C}_t=(0,10,10) demonstrating that Gx(xz)+Gy(xy)G_x^{(xz)}+G_y^{(xy)} is quantized at CxzLy+CxyLz=20C_{xz}L_y+C_{xy}L_z=20 within the shaded energy interval [43]. Furthermore, the robustness of this topological quantity is confirmed in Figure 2(J) and 2(K). Gx(xz)+Gy(xy)G_x^{(xz)}+G_y^{(xy)} remains quantized with vanishing fluctuations for significant disorder strengths and varying tdt_d [refer to SM. Note 4]. This stands in sharp contrast to the fragile nature of the individual components Gx(xz)G_x^{(xz)} and Gy(xy)G_y^{(xy)}, which fluctuate considerably.

Correspondingly, the total xx-direction conductance Gx=Gx(xz)+Gx(xy)G_x=G_x^{(xz)}+G_x^{(xy)} emerges as a similarly robust topological response.

Significantly, we find that the quantized sum (CxzLy+CxyLz)(C_{xz}L_y+C_{xy}L_z) is equivalent to the global Hall conductance, which is directly measurable in a standard Hall-bar geometry. To demonstrate this, we model the quantum transport of a sample with C=(0,1,1)\mathbf{C}=(0,1,1) using the Landauer-Büttiker formalism. For a multi-terminal device, the relationship between current and voltage is governed by [38]:

Ii=e2h∑j≠iTij(Vi−Vj).(3)I_i=\frac{e^2}{h}\sum_{j\ne i}T_{ij}(V_i-V_j). \tag*{(3)}

ViV_i represents the voltage at the ii-th lead, IiI_i denotes the current flowing out of the ii-th lead into the sample, and TijT_{ij} is the transmission coefficient from the jj-th to the ii-th lead.

We first investigate transport in a Hall bar oriented along the yy-direction, characterized by the total Chern vector Ct=(0,CxzLy,CxyLz)\mathbf{C}_t=(0,C_{xz}L_y,C_{xy}L_z). For the standard six-terminal geometry shown in Figure 3(A), current is injected between leads 1 and 4, while leads 2, 3, 5, and 6 serve as voltage probes. The interconnected chiral modes yield the following non-zero transmission coefficients:

Figure 3. Hall bar device measurement of quantized Hall conductance. (A) Schematic plot of Hall bar with current along the yy-direction. (B) and (C) the Hall conductances versus Fermi energy EE and CtC_t for the setup in (A). The total Chern vector Ct=(0,CxzLy,CxyLz)C_t=(0,C_{xz}L_y,C_{xy}L_z) is marked in the figure with td=0.1tt_d=0.1t. The solid lines plot (CxzLy+CxyLz)(C_{xz}L_y+C_{xy}L_z) versus LL. (D)-(F) are the same as those in (A)-(C), excepting currents along the xx-direction. We set E=−0.2tE=-0.2t for (C) and (F).

Tj,j+1=Gy(xy),T61=Gy(xy),T26/35=Gx(xz),T62/53=Gx(xz)(4)\begin{aligned} T_{j,j+1}&=G_y^{(xy)},\quad T_{61}=G_y^{(xy)},\\ T_{26/35}&=G_x^{(xz)},\quad T_{62/53}=G_x^{(xz)} \tag*{(4)} \end{aligned}

where j∈[1,2,3,4,5]j\in[1,2,3,4,5]. By solving the (3) yields the quantized Hall conductance between leads 2 and 6 [refer to SM. Note 5 for details]

Gyx=Iy(V2−V6)=−(CxzLy+CxyLz),(5)\begin{aligned} G_{yx}&=\frac{I_y}{(V_2-V_6)}\\ &=-(C_{xz}L_y+C_{xy}L_z), \tag*{(5)} \end{aligned}

where IyI_y denotes the current along the yy-direction.

Due to the 3D nature of the Chern vector, the Hall-bar response can also be measured along the xx-direction [Fig. 3(D)]. In this orientation, the Hall conductance remains quantized as [see SM. Note 5 for details]:

Gxy=Ix(V2−V6)=CxzLy+CxyLz.(6)\begin{aligned} G_{xy}&=\frac{I_x}{(V_2-V_6)}\\ &=C_{xz}L_y+C_{xy}L_z. \tag*{(6)} \end{aligned}

The validity of Eq. (5) and Eq. (6) is verified by numerical calculations using the non-equilibrium Green’s function technique [38]. As shown in Figure 3(B), the Hall conductance GyxG_{yx} aligns with the quantized value (CxzLy+CxyLz)(C_{xz}L_y+C_{xy}L_z) across varying sample sizes. A similar agreement is observed for GxyG_{xy} in Figure 3(E). The robustness of these Hall responses against disorder is further confirmed in SM. Note 6.

Unlike the single Chern number dependence of Gxy=nLzG_{xy}=nL_z, typically observed for Ct=(0,0,nLz)\mathbf{C}_t=(0,0,nL_z) [Fig. 1(B)], the 3D quantized Hall conductance of Ct=(0,mLy,nLz)\mathbf{C}_t=(0,mL_y,nL_z) exhibits double Chern number components dependence and can be observed on different directions. Specifically, both GyxG_{yx} and GxyG_{xy} exhibit a joint dependence on Chern numbers mLymL_y and nLznL_z [Figs. 3(C) and (F)], serving as a distinctive experimental fingerprint of the collective 3D quantized Hall effect.

Intrinsic network topology vs. principal axis rotation

To elucidate the topological essence of the general Chern vector C=(0,m,n)\mathbf{C}=(0,m,n), it is instructive to contrast our framework with the conventional (0,0,q)(0,0,q) case obtained via coordinate rotation. While a Chern vector is mathematically reducible to a principal axis, such a mapping obscures the rich surface physics emerging from the internal coupling within the chiral network.

As illustrated in Figure 4, the global connectivity of the chiral network persists even when the system is cut along high-index surfaces (Figure 4(A)), maintaining a complex topology fundamentally distinct from the decoupled stacking limit (Figure 4(C)). This inequivalence is further evidenced by the surface spectra: in the general regime, unique surface states emerge on both top and bottom surfaces (Figure 4(E)), a spectral hallmark of the 3D chiral network. In the strong-coupling limit, however, these features vanish as the system collapses into a simple stacking of 2D QAH layers (Figure 4(F)). Generally, the transport phenomena and quantized Hall responses observed in this work originate from the intrinsic network topology, providing a theoretical breadth that encompasses but significantly extends beyond the traditional rotation-and-stacking schemes.

Figure 4. Intrinsic 3D chiral network topology beyond stacked 2D QAH physics. Schematic of surface states for a 3D Chern insulator with a general Chern vector C=(0,m,n)\mathbf{C}=(0,m,n) in the general case (A) and the strong-coupling limit (C). (B) and (D) show surface configurations after coordinate rotation and cutting along high-index surfaces. In the general regime (B), the chiral network persists, maintaining a complex connectivity that cannot be reduced to simple 2D physics. In contrast, the strong-coupling limit (D) yields a configuration equivalent to a stacking of 2D QAH insulators with C=(0,0,q)\mathbf{C}=(0,0,q). (E) and (F) give their typical surface spectra.

In general, the standard stacked-layer picture is recovered only in the singular strong-coupling limit, where the system’s intrinsic complexity effectively decouples. However, this limit represents a constrained regime that is not representative of the general physics. In the broader regime of intermediate coupling, the system manifests a 3D collective state whose topology is governed by global network connectivity. This global network architecture accounts for the full variety of chiral mode configurations, that remains fundamentally inaccessible within the spatial rotation of the (0,0,q)(0,0,q) limit. Consequently, our generalized framework establishes a comprehensive theoretical platform, while the rotated (0,0,q)(0,0,q) case represents a limiting simplification rather than a full realization of the general physics.

Candidate systems and materials

The universality of this network-based framework allows for its realization across diverse topological platforms. Because its transport properties stem from intrinsic network topology rather than specific structural orientation, the framework transcends the constraints of

geometric rotation models. To demonstrate this versatility, we consider a representative hybrid system where 3D Chern insulator bands coexists with metallic bands-a configuration prevalent in many topological materials. This system, governed by the non-local connectivity of the network, is described by the Hamiltonian:

H=(Hmetal(B)00Hxzy(B))+∑n(0tdtd0)cn†cn(7)H = \begin{pmatrix} \mathcal{H}_{\mathrm{metal}}(\mathbf{B}) & 0 \\ 0 & \mathcal{H}_{xzy}(\mathbf{B}) \end{pmatrix} +\sum_n \begin{pmatrix} 0 & t_d \\ t_d & 0 \end{pmatrix} c_n^\dagger c_n \tag*{(7)}

Hxzy(B)\mathcal{H}_{xzy}(\mathbf{B}) contributes a robust, field-insensitive background Chern vector CQAHC_{\mathrm{QAH}}, while the metallic component Hmetal(B)\mathcal{H}_{\mathrm{metal}}(\mathbf{B}) provides the necessary degrees of freedom for Landau quantization under an applied magnetic field

B. Here, the metallic bands are described by

Hmetal(B)=∑nϵ0cn†cn+txeiϕycn+δx†cn+tycn+δy†cn+tzcn+δz†cn+h.c.\mathcal{H}_{\mathrm{metal}}(\mathbf{B})=\sum_n \epsilon_0 c_n^\dagger c_n+t_x e^{i\phi_y}c_{n+\delta_x}^\dagger c_n+t_y c_{n+\delta_y}^\dagger c_n+t_z c_{n+\delta_z}^\dagger c_n+\mathrm{h.c.}

and Hxzy(B)\mathcal{H}_{xzy}(\mathbf{B}) is identical to Eq. (1), with the inclusion of an applied magnetic field B\mathbf{B}. The magnetic field B=(0,0,B)\mathbf{B}=(0,0,B) is incorporated by introducing an extra flux ϕ=a2B/ϕ0\phi=a^2B/\phi_0 through the unit lattice area δx×δy=a2\delta_x\times\delta_y=a^2 [38]. As schematically illustrated in Figure 5(A)–(C), the coexistence of these bands enables the synthesis of a composite Chern vector C=(0,m,n)\mathbf{C}=(0,m,n). While the open Fermi

Figure 5. Quantized Hall response in Chern-vector-protected 3D quantum Hall insulators. (A) Schematic band structures for a 3D QAH and a 3D metallic phase. The pink planes indicate the constant-energy surfaces. (B) Corresponding Fermi surfaces for the spectra shown in (A). (C) Surface states associated with (A) under an applied magnetic field. (D),(F) Numerically calculated band spectra at magnetic flux ϕ=0.25π\phi= 0.25\pi. (E) and (G) are Hall conductances of (D) and (F), respectively. The sample sizes are Lx=Ly=Lz=10L_x=L_y=L_z=10. The other parameters are fixed at tx=ty=tt_x=t_y=t, ϵ0=−2t\epsilon_0=-2t, tz=0.1tt_z=0.1t, and td=0.1tt_d=0.1t.

surfaces of the 3D QAH surface states [Fig. 5(B)] are immune to Landau quantization [19, 20], the closed Fermi surface of the metallic component facilitates the 3D quantum Hall effect (QHE) under a field B∥z\mathbf{B}\parallel z [18]. This allows the system to reach target states with different Chern vector magnitudes.

Numerical spectra and transport calculations in Figure 5(D)–(G) confirm that the Hall conductances GxyG_{xy} and GyxG_{yx} remain quantized according to the predicted formulas, even in the presence of inter-band coupling tdt_d. For instance, the system hosts stable C=(0,1,1)\mathbf{C}=(0,1,1) and (0,1,2)(0,1,2) phases at distinct energy windows, demonstrating the existence of the collective transport.

From a material perspective, the requirement for multiple coexisting orbitals is intrinsically satisfied in many candidates. Based on topological material databases, compounds such as RbSb, CsSb, RbAs [44], and BiInN/Al2_2O3_3 [45] are identified as promising candidates, as they naturally exhibit metallic bands nested within a nontrivial QAH gap. Through strategic heterostructure engineering or stacking, the proposed model offers a highly feasible route for experimental observation of controllable 3D Chern vectors.

the quantized value (mLy+nLz)(mL_y+nL_z). This summation relation manifests as observable quantized Hall conductances ±(mLy+nLz)\pm(mL_y+nL_z). A hallmark of these responses is their distinctive double Chern number components dependence, which serves as a unique experimental fingerprint. To facilitate experimental validation, we propose a class of candidate condensed-matter systems.

Discussion and conclusion

Our work identifies a distinct quantized transport phenomenon that originates from the inherent chiral network manifested by the Chern vector C=(0,m,n)\mathbf{C}=(0,m,n). This network-based topology dictates unique collective properties and global connectivity. Specifically, we demonstrate that the sum of transmission coefficients along both the xx- and yy-directions, rather than the individual ones, is protected by the Chern vector, preserving

Beyond condensed-matter materials, the chiral surface states associated with C=(0,m,n)\mathbf{C}=(0,m,n) have been observed in photonic and acoustic crystals [31, 32, 34]. Although directly measuring Hall conductances in these platforms remains challenging, the characteristic chiral network scattering [Figs. 2(D) and 2(E)] can be validated by tracking wave-packet propagation. These distinct trajectories underpin the microscopic origin of the collective Hall conductances in Eqs. (5) and (6).

In conclusion, our study advances the understanding of quantized transport in 3D topological systems. Notably, exploring the transport properties of Chern vectors with three non-zero components, C=(Cyz,Czx,Cxy)\mathbf{C}=(C_{yz},C_{zx},C_{xy}), represents an exciting avenue for future investigation, potentially unveiling even richer physics.

Acknowledgments

We appreciate the valuable discussions with Hong Yao, Li-Yuan Zhang, Ming Lu, Ming Gong, Qiang Wei and Rui-Chun Xiao.

This work was supported by the National Basic Research Program of China (Grants No. 2024YFA1409003 and No. 2022YFA1403700), National Natural Science Foundation of China (Grants No. 12350401 and No. 12204044).

Data availability

All essential data are available in the paper. Additional data are given in the supplementary file.

Supplementary data

Further supporting data, including the codes for numerical calculations can be accessed via the GitHub website (https://github.com/jianghuaphy-arch/chernvector).

Lattice model and Chern-vector characterization

Here we provide the explicit lattice Hamiltonians and the calculation of the corresponding Chern vector C=(Cyz,Cxz,Cxy)\mathbf{C}=(C_{yz},C_{xz},C_{xy}) for the model introduced in Eq. (1).

The first sector, Hxyz\mathcal{H}_{xyz}, is written in real space as

Hxyz=∑nT0cn†cn+∑n(Txcn+δx†cn+Tycn+δy†cn+Tzcn+δz†cn)+h.c.(8)\mathcal{H}_{xyz}=\sum_n T_0c_n^\dagger c_n+\sum_n\left(T_xc_{n+\delta_x}^\dagger c_n+T_yc_{n+\delta_y}^\dagger c_n+T_zc_{n+\delta_z}^\dagger c_n\right)+\mathrm{h.c.} \tag*{(8)}

where

T0=tτz,Tx=M2τz−iA2τxT_0=t\tau_z,\qquad T_x=\frac{M}{2}\tau_z-\frac{iA}{2}\tau_x
Ty=M2τz−iA2τy,Tz=t1τz.T_y=\frac{M}{2}\tau_z-\frac{iA}{2}\tau_y,\qquad T_z=t_1\tau_z.

Here, τx,y,z\tau_{x,y,z} are Pauli matrices acting in the orbital space, and δα\delta_\alpha denotes the nearest-neighbor displacement along the α\alpha direction.

Fourier transforming the real-space Hamiltonian gives

Hxyz(k)=tτz+Mτzcos⁡kx+Aτxsin⁡kx+Mτzcos⁡ky+Aτysin⁡ky+2t1τzcos⁡kz.(9)\mathcal{H}_{xyz}(\mathbf{k})=t\tau_z+M\tau_z\cos k_x+A\tau_x\sin k_x+M\tau_z\cos k_y+A\tau_y\sin k_y+2t_1\tau_z\cos k_z. \tag*{(9)}

The second sector, Hxzy\mathcal{H}_{xzy}, is obtained from Hxyz\mathcal{H}_{xyz} by interchanging the roles of the yy and zz directions. Its momentum-space Hamiltonian is

Hxzy(k)=tτz+Mτzcos⁡kx+Aτxsin⁡kx+2t2τzcos⁡ky+Mτzcos⁡kz+Aτysin⁡kz.(10)\mathcal{H}_{xzy}(\mathbf{k})=t\tau_z+M\tau_z\cos k_x+A\tau_x\sin k_x+2t_2\tau_z\cos k_y+M\tau_z\cos k_z+A\tau_y\sin k_z. \tag*{(10)}

The two sectors are coupled through Td=td(I2×2+τx)T_d=t_d(I_{2\times2}+\tau_x), as introduced in the main text. For the numerical calculations, we take A=−tA=-t, M=2tM=2t, t1=−0.3tt_1=-0.3t, and t2=−0.1tt_2=-0.1t, with t=1t=1 taken as the energy unit. For these parameters, the bulk gap remains open for td<0.5tt_d<0.5t; the evolution of the bulk spectrum with tdt_d is shown in SM Note 1.

To characterize the topology, we calculate the Chern number Cαβ(kγ)C_{\alpha\beta}(k_\gamma) of each two-dimensional momentum slice at fixed kγk_\gamma using the TKNN formula [6]:

Cαβ(kγ)=−1π∬BZ∑Ec,EvIm⁡[⟨uc∣∂kαH(k)∣uv⟩×⟨uv∣∂kβH(k)∣uc⟩](Ec−Ev)2 dkα dkβ,(11)C_{\alpha\beta}(k_\gamma)=-\frac{1}{\pi}\iint_{\mathrm{BZ}}\sum_{E_c,E_v}\frac{\operatorname{Im}\left[\langle u_c|\partial_{k_\alpha}\mathcal{H}(\mathbf{k})|u_v\rangle\times\langle u_v|\partial_{k_\beta}\mathcal{H}(\mathbf{k})|u_c\rangle\right]}{(E_c-E_v)^2}\,dk_\alpha\,dk_\beta, \tag*{(11)}

where α,β,γ∈{x,y,z}\alpha,\beta,\gamma\in\{x,y,z\}, and ∣uc⟩|u_c\rangle (∣uv⟩|u_v\rangle) denotes an eigenstate in the conduction (valence) band with energy Ec>0E_c>0 (Ev<0E_v<0).

The system remains gapped throughout the Brillouin zone, and the corresponding slice Chern numbers are independent of the transverse momentum for all kx,y,z∈[−π,π]k_{x,y,z}\in[-\pi,\pi]. The resulting Chern vector is

C=(Cyz,Cxz,Cxy)=(0,1,1).\mathbf{C}=(C_{yz},C_{xz},C_{xy})=(0,1,1).

This Chern vector remains unchanged for finite inter-sector coupling tdt_d as long as the bulk gap remains open. At td≈0.5tt_d\approx0.5t, the bulk gap closes, signaling a topological phase transition [see SM Note 1].

References

  1. [5]R. B. Laughlin, Quantized Hall conductivity in two dimensions, Phys. Rev. B 23, 5632 (1981).
  2. [6]D. J. Thouless, M. Kohmoto, M. Nightingale, and M. den Nijs, Quantized Hall conductance in a two-dimensional periodic potential, Phys. Rev. Lett. 49, 405 (1982).DOI
  3. [7]B. I. Halperin, Quantized Hall conductance, current-carrying edge states, and the existence of extended states in a two-dimensional disordered potential, Phys. Rev. B 25, 2185 (1982).
  4. [8]C. Z. Chang et al., Experimental observation of the quantum anomalous Hall effect in a magnetic topological insulator, Science 340, 167 (2013).DOI
  5. [9]F. D. M. Haldane, Model for a quantum Hall effect without Landau levels: Condensed-matter realization of the parity anomaly, Phys. Rev. Lett. 61, 2015 (1988).pubmed.ncbi.nlm.nih.gov/10038961
  6. [10]E. Prodan, Disordered topological insulators: A non-commutative geometry perspective, J. Phys. A: Math. Theor. 44, 113001 (2011).
  7. [11]Y. Gong et al., Experimental realization of an intrinsic magnetic topological insulator, Chin. Phys. Lett. 36, 076801 (2019).DOI
  8. [12]H. Park et al., Observation of fractionally quantized anomalous Hall effect, Nature 622, 74 (2023).DOI
  9. [13]Z. G. Lu et al., Fractional quantum anomalous Hall effect in multilayer graphene, Nature 626, 759 (2024).DOI
  10. [14]F. Xu, Z. Sun, T. Jia, C. Liu, C. Xu, C. S. Li, et al., Observation of integer and fractional quantum anomalous Hall effects in twisted bilayer MoTe2, Phys. Rev. X 13, 031037 (2023).DOI
  11. [15]Y. F. Zhao et al., Tuning the Chern number in quantum anomalous Hall insulators, Nature 588, 419 (2020).DOI
  12. [16]Y. F. Zhao, R. X. Zhang, L. J. Zhou, R. B. Mei, Z. J. Yan, M. H. W. Chan, C. X. Liu, and C. Z. Chang, Zero magnetic field plateau phase transition in higher Chern number quantum anomalous Hall insulators, Phys. Rev. Lett. 128, 216801 (2022).DOI
  13. [17]M. M. Otrokov et al., Unique thickness-dependent properties of the van der Waals interlayer antiferromagnet MnBi2Te4 films, Phys. Rev. Lett. 122, 107202 (2019).DOI
  14. [18]F. Tang et al., Three-dimensional quantum Hall effect and metal-insulator transition in ZrTe5, Nature 569, 537 (2019).DOI
  15. [19]C. M. Wang, H. P. Sun, H. Z. Lu, and X. C. Xie, 3D quantum Hall effect of Fermi arcs in topological semimetals, Phys. Rev. Lett. 119, 136806 (2017).DOI
  16. [20]H. L. Li, H. W. Liu, H. J. Jiang, and X. C. Xie, 3D quantum Hall effect and a global picture of edge states in Weyl semimetals, Phys. Rev. Lett. 125, 036602 (2020).DOI
  17. [21]C. Zhang et al., Quantum Hall effect based on Weyl orbits in Cd3As2, Nature 565, 331 (2019).DOI
  18. [22]C. Zhang et al., Evolution of Weyl orbit and quantum Hall effect in Dirac semimetal Cd3As2, Nat. Commun. 8, 1272 (2017).
  19. [23]T. Schumann, L. Galletti, D. A. Kealhofer, H. Kim, M. Goyal, and S. Stemmer, Observation of the quantum Hall effect in confined films of the three-dimensional Dirac semimetal Cd3As2, Phys. Rev. Lett. 120, 016801 (2018).DOI
  20. [24]L. Li, J. Cao, C. Cui, Z. M. Yu, and Y. Yao, Planar Hall effect in topological Weyl and nodal-line semimetals, Phys. Rev. B 108, 085120 (2023).DOI
  21. [25]Z. Z. Du, H. Z. Lu, and X. C. Xie, Nonlinear Hall effects, Nat. Rev. Phys. 3, 744 (2021).DOI
  22. [26]J. X. Yin et al., Quantum-limit Chern topological magnetism in TbMn6Sn6, Nature 583, 533 (2020).DOI
  23. [27]N. Shumiya et al., Evidence of a room-temperature quantum spin Hall edge state in a higher-order topological insulator, Nat. Mater. 21, 1111 (2022).DOI
  24. [28]M. S. Hossain et al., Quantum transport response of topological hinge modes, Nat. Phys. 20, 776 (2024).DOI
  25. [29]Z. Wu, M. Cheng, Z. Wang, S. Xu, J. Chen, Y. Meng, X. Xi, X. Sun, P. P. Shum, H. Lu, and Z. Gao, Three-dimensional photonic quantum hall effect of fermi arcs, Natl. Sci. Rev. 13, nwag251 (2026).DOI
  26. [30]L. Ye, Q. Wang, Z. Fu, H. He, X. Huang, M. Ke, J. Lu, W. Deng, and Z. Liu, Hinge modes of surface arcs in a synthetic weyl phononic crystal, Phys. Rev. Lett. 133, 126602 (2024).DOI
  27. [31]G. G. Liu et al., Topological Chern vectors in three-dimensional photonic crystals, Nature 609, 925 (2022).DOI
  28. [32]L. Yang et al., Three-dimensional acoustic Chern insulators with arbitrary Chern vectors, Phys. Rev. Research 7, 023280 (2025).
  29. [33]F. Ma, J. Feng, F. Li, Y. Wu, and D. Zhou, Floquet Chern vector topological insulators in three dimensions, Phys. Rev. B 111, 134301 (2025).arxiv.org/abs/2412.00619
  30. [34]X. Xi et al., Photonic arbitrary Chern vectors, Sci. Adv. 12, eaef3784 (2026).DOI
  31. [35]M. Büttiker, Edge-state physics without magnetic fields, Science 325, 278 (2009).DOI
  32. [36]H. L. Störmer, J. P. Eisenstein, A. C. Gossard, W. Wiegmann, and K. Baldwin, Quantization of the Hall effect in an anisotropic three-dimensional electronic system, Phys. Rev. Lett. 56, 85 (1986).pubmed.ncbi.nlm.nih.gov/10032535
  33. [37]D. P. Druist, P. J. Turley, K. D. Maranowski, E. G. Gwinn, and A. C. Gossard, Observation of chiral surface states in the integer quantum Hall effect, Phys. Rev. Lett. 80, 365 (1998).DOI
  34. [38]S. Datta, Electronic Transport in Mesoscopic Systems (Cambridge University Press, Cambridge, 1995).
  35. [39]See Supplemental material for more details about: the model hamiltonians, calculation methods and transport analyses.
  36. [40]P. W. Anderson, Absence of diffusion in certain random lattices, Phys. Rev. 109, 1492 (1958).
  37. [41]E. Abrahams, P. W. Anderson, D. C. Licciardello, and T. V. Ramakrishnan, Scaling theory of localization: Absence of quantum diffusion in two dimensions, Phys. Rev. Lett. 42, 673 (1979).
  38. [42]F. Evers and A. D. Mirlin, Anderson transitions, Rev. Mod. Phys. 80, 1355 (2008).
  39. [43]Actually, the white energy interval possesses the trivial surface states along the y direction. Nevertheless, the white energy interval is rather tiny, which can be roughly neglected.
  40. [44]K. Wang, Y. Li, H. Mei, P. Li, and Z. X. Guo, Quantum anomalous Hall and valley quantum anomalous Hall effects in two-dimensional 0 orbital XY monolayers, Phys. Rev. Materials 6, 044202 (2022).arxiv.org/abs/2204.07488
  41. [45]S. Jin, Y. Xia, W. Shi, J. Hu, R. Claessen, W. Hanke, R. Thomale, and G. Li, Large-gap quantum anomalous Hall states induced by functionalizing buckled Bi-III monolayer/Al2O3, Phys. Rev. B 106, 125151 (2022).DOI

Paper details

vlnc:W_bew2h726
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents