Nonreciprocal Angular Magnetoresistance

• Nonreciprocal angular magnetoresistance is a phenomenon where resistance depends on both current direction and magnetic-field orientation due to broken inversion symmetry and magnetic order.
• It is characterized by nonlinear, second-order responses, with experimental analysis using second-harmonic voltage measurements and detailed angular scans.
• Distinct mechanisms such as asymmetric scattering, Rashba effects, and quantum-geometric contributions provide insights for optimizing magnetotransport in various material systems.

Nonreciprocal angular magnetoresistance denotes a class of magnetotransport phenomena in which the measured resistance depends not only on magnetic-field magnitude and orientation, but also on current direction, so that reversing the current changes the resistance. In the recent literature, this behavior is typically identified as a nonlinear, second-order response extracted from second-harmonic transport, and it arises when angularly resolved magnetotransport is combined with broken inversion symmetry and either magnetic order or an external magnetic field. Experimental realizations span antiferromagnetic topological-insulator heterostructures, multiferroic Rashba semiconductors, Rashba ferromagnets, semiconductor heterojunctions with interfacial Rashba spin texture, and chiral semiconductors, while theory connects the effect to generalized galvanomagnetic tensors in systems with two vector order parameters and to intrinsic quantum-geometric contributions to nonlinear conductivity (Ye et al., 2022, Yoshimi et al., 2022, Lee et al., 2021, Li et al., 2022, Wang, 2022, Kaplan et al., 2022, Li et al., 5 Dec 2025).

1. Definition and phenomenology

A minimal phenomenological description expresses nonreciprocal transport as a resistance correction odd in current direction and odd in magnetic field or magnetic order. In the MnBi$_2$Te$_4$/Pt bilayer, the voltage response is written as

$V = R_0\left(1+\gamma BI\right),$

with $R_0$ the linear resistance, $\gamma$ the nonreciprocal coefficient, $B$ the magnetic field, and $I$ the current. In that formulation, the defining signature is current-direction-dependent resistance, and the nonreciprocal contribution is detected primarily from the second-harmonic voltage $V_{2\omega}$, which is proportional to $I^2$ and linear in $B$ at low field (Ye et al., 2022).

In multiferroic (Ge,Mn)Te, the angular dependence is formulated as

$_4$0

where $_4$1 is the unit vector along the ferroelectric polarization. This expression separates ordinary magnetoresistance from the nonreciprocal term and shows that the response is maximized when $_4$2, $_4$3, and $_4$4 are mutually orthogonal. Reversing $_4$5 or the magnetization changes the sign of the effect (Yoshimi et al., 2022).

In lattice-matched InSb/CdTe, the nonlinear transport is described as a second-order charge current,

$_4$6

and the unidirectional magnetoresistance amplitude satisfies $_4$7. The nonreciprocal coefficient is defined as

$_4$8

The paper reports room-temperature operation up to $_4$9 K and a nonreciprocal coefficient $V = R_0\left(1+\gamma BI\right),$0 (Li et al., 2022).

A broader symmetry-based parametrization appears in chiral tellurium, where the resistance contains an $V = R_0\left(1+\gamma BI\right),$1-linear contribution with direction-dependent coefficients $V = R_0\left(1+\gamma BI\right),$2. That formulation makes explicit that nonreciprocal magnetoresistance can be decomposed into distinct directional channels rather than reduced to a single scalar coefficient (Li et al., 5 Dec 2025).

2. Symmetry structure and angular laws

The central symmetry requirement is the simultaneous absence of inversion-protected reciprocity and the presence of a magnetic or field-defined handedness. A general tensor theory for materials with two vector order parameters states that magnetic materials without inversion symmetry are good candidates for observing nonreciprocal dc electron transport, and that reciprocity is restored when inversion symmetry is present. In that framework, the most general linear-response relation for two vector order parameters $V = R_0\left(1+\gamma BI\right),$3 and $V = R_0\left(1+\gamma BI\right),$4 is

$V = R_0\left(1+\gamma BI\right),$5

and, in the coplanar case,

$V = R_0\left(1+\gamma BI\right),$6

$V = R_0\left(1+\gamma BI\right),$7

Unlike ordinary one-order-parameter AMR/PHR, the longitudinal and transverse responses need not have the same magnitude, their phase difference is not generally $V = R_0\left(1+\gamma BI\right),$8, and their periodicity becomes $V = R_0\left(1+\gamma BI\right),$9 rather than $R_0$0 (Wang, 2022).

For bulk polar systems, the symmetry-allowed scalar $R_0$1 gives a compact angular law. In (Ge,Mn)Te, this immediately implies that the signal is largest when $R_0$2 is orthogonal to both $R_0$3 and $R_0$4, and negligible when $R_0$5. The experimentally observed cosine-like in-plane angular dependence directly follows that invariant (Yoshimi et al., 2022).

An intrinsic nonlinear-conductivity formulation based on quantum geometry imposes a different but related set of selection rules. In that theory, both the nonlinear anomalous Hall effect and nonreciprocal magnetoresistance are different tensor projections of the same second-order conductivity, and the normalized quantum metric dipole contributes intrinsically to both. For a nonzero intrinsic quantum-metric contribution over the full Brillouin zone, both inversion $R_0$6 and time-reversal $R_0$7 must be broken. In MnBi$R_0$8Te$R_0$9, the film symmetry and $\gamma$0 rotation lead to an in-plane angular modulation

$\gamma$1

which makes the nonlinear response explicitly anisotropic under sample rotation (Kaplan et al., 2022).

3. Experimental identification by harmonic and angular analysis

The standard experimental procedure decomposes the voltage into first- and second-harmonic components under an AC current $\gamma$2. The first harmonic isolates reciprocal magnetotransport, while the second harmonic isolates the nonlinear response that changes sign with current direction. This separation is central in MnBi$\gamma$3Te$\gamma$4/Pt, where both longitudinal and transverse channels, $\gamma$5, were measured. The angular dependence was mapped in the $\gamma$6-, $\gamma$7-, and $\gamma$8-planes, and a central result is a fixed $\gamma$9 phase offset between $B$0 and $B$1 in the $B$2-plane. The response in the $B$3 planes is suppressed or inconsistent with conventional spin-orbit-torque-driven expectations, which was taken as evidence that the effect originates at the MBT surface/interface rather than mainly in the Pt layer (Ye et al., 2022).

In ultrathin PdCoO$B$4, the first-harmonic longitudinal magnetoresistance follows the reciprocal form $B$5, consistent with anisotropic magnetoresistance and quadratic magnetoresistance, whereas the second-harmonic longitudinal signal follows $B$6. The first-harmonic amplitude is current-independent and approximately quadratic in magnetic field; the second-harmonic amplitude is linear in current and nonlinear in magnetic field below the Curie point. This contrast is the diagnostic signature of unidirectional magnetoresistance in that material, although the measured $B$7 signal also contains an anomalous Nernst contribution that must be separated geometrically (Lee et al., 2021).

In InSb/CdTe, the second-harmonic longitudinal and transverse voltages show sinusoidal angular dependence under in-plane field rotation. At $B$8 K, the longitudinal UMR is maximal at $B$9, minimal at $I$0, and vanishes at $I$1 and $I$2, following a $I$3 relation with $I$4 periodicity. The first- and second-harmonic angular scans exhibit a $I$5 phase shift between longitudinal and transverse responses, and the normalized peak amplitude follows a $I$6 law, consistent with a purely in-plane Rashba spin texture (Li et al., 2022).

Chiral Te extends this harmonic methodology to three orthogonal field directions. There, the field-odd second-harmonic component is isolated as

$I$7

which allows distinct directional coefficients $I$8, $I$9, and $V_{2\omega}$0 to be extracted and compared across opposite crystal chiralities (Li et al., 5 Dec 2025).

4. Microscopic mechanisms

A recurrent microscopic picture is asymmetric scattering on spin-momentum-locked states. In the MnBi$V_{2\omega}$1Te$V_{2\omega}$2/Pt bilayer, the proposed mechanism is asymmetric electron scattering at the MBT surface mediated by magnons. The physical picture combines spin-momentum-locked Dirac surface electrons with uncompensated surface magnetization attributed to imperfect surfaces or Mn-doped Bi$V_{2\omega}$3Te$V_{2\omega}$4-like regions. Because magnon emission and absorption are not equivalent for opposite momentum states, the electron relaxation time becomes directional. The observed $V_{2\omega}$5 angular offset is interpreted in terms of magnetic-field-controlled scattering channels on the surface-state Fermi contour (Ye et al., 2022).

A closely related magnon-mediated inelastic-scattering scenario is advanced for (Ge,Mn)Te. There, the material combines ferroelectricity, ferromagnetism, strong Rashba spin splitting, and spin-momentum locking. The model Hamiltonian,

$V_{2\omega}$6

captures Rashba-split bands deformed by exchange. At low hole density, the Fermi level approaches a regime with an isolated single spin-momentum-locked Fermi surface, where ordinary backscattering is suppressed and magnon-assisted processes dominate. The reported enhancement of the nonreciprocal coefficient at lower hole density, with $V_{2\omega}$7 tuned from about $V_{2\omega}$8 to $V_{2\omega}$9, is attributed to this single-Fermi-surface regime (Yoshimi et al., 2022).

Rashba-driven unidirectional magnetoresistance is the central interpretation in ultrathin PdCoO$I^2$0 and InSb/CdTe. In PdCoO$I^2$1, surface ferromagnetism at the Pd-terminated polar surface coexists with strong Rashba spin-orbit coupling, and the Rashba coefficient is estimated as $I^2$2. Below $I^2$3 K, the Rashba-driven UMR competes with an anomalous Nernst signal generated by Joule heating (Lee et al., 2021). In InSb/CdTe, inversion symmetry is broken at the heterojunction by band bending and a built-in field $I^2$4 at $I^2$5 K, while time-reversal symmetry is broken by the applied magnetic field. The resulting interfacial Rashba channel is characterized by $I^2$6, $I^2$7, $I^2$8, and $I^2$9, and the large UMR is attributed to stable quasi-two-dimensional spin texture under strong confinement (Li et al., 2022).

An intrinsic mechanism not reducible to scattering asymmetry is provided by quantum geometry. The unified second-order-conductivity theory contains a nonlinear Drude term, a Berry-curvature-dipole term, and a quantum-metric-dipole term. The quantum metric contribution is intrinsic and $B$0-independent because it originates from electric-field-induced renormalization of Bloch states rather than nonequilibrium occupation dynamics. In even-layer MnBi$B$1Te$B$2, where $B$3 and $B$4 are broken but $B$5 is preserved, the Berry-curvature-dipole contribution is suppressed and the quantum metric contribution becomes the central intrinsic effect (Kaplan et al., 2022).

Chiral Te illustrates that a measured nonreciprocal angular signal may be a superposition of distinct microscopic channels. The $B$6 response changes sign with Te chirality and is attributed to the Edelstein effect from a radial spin texture; the chirality-independent $B$7 response is assigned to the Nernst effect; and the $B$8 response is proposed to arise from intrinsic orbital magnetization (Li et al., 5 Dec 2025).

5. Material platforms and comparative manifestations

The experimental literature does not present a single universal waveform or a single universal microscopic origin. Instead, the observed nonreciprocal angular magnetoresistance depends strongly on whether the active degrees of freedom are topological surface states, Rashba bands, multiferroic order, surface ferromagnetism, or chiral bulk states.

Platform	Key angular/nonreciprocal signature	Distinguishing condition
MnBi$B$9Te$_4$00/Pt	$_4$01 offset between $_4$02 and $_4$03	Appears only below $_4$04
(Ge,Mn)Te	Maximal when $_4$05 and $_4$06	Enhanced at lower hole density
PdCoO$_4$07	$_4$08 term $_4$09	Present below $_4$10 K
InSb/CdTe	$_4$11 UMR with $_4$12 periodicity	Observed up to $_4$13 K
Chiral Te	Different coefficients for $_4$14	Chirality flips the $_4$15-axis response

In MnBi$_4$16Te$_4$17/Pt, the active region is the MBT surface/interface even though the current flows mainly through Pt, and the nonreciprocal signal emerges only below the Néel temperature, with $_4$18 K for 4 septuple layers and $_4$19 K for 12 layers (Ye et al., 2022). In (Ge,Mn)Te, the key distinction from nonmagnetic GeTe is the coexistence of ferroelectric and ferromagnetic orders in a bulk Rashba semiconductor, together with pronounced Fermi-level dependence (Yoshimi et al., 2022). In PdCoO$_4$20, the relevant channel is a ferromagnetic Rashba surface in a 3.8-nm-thick delafossite film, and the analysis requires disentangling Rashba UMR from anomalous Nernst voltage (Lee et al., 2021). In InSb/CdTe, the emphasis is on interfacial engineering: a lattice-matched heterostructure yields a strong built-in electric field, a quasi-two-dimensional Rashba channel, and gate control, with the UMR amplitude changing from about $_4$21 to $_4$22 as $_4$23 is swept from $_4$24 V to $_4$25 V, corresponding to about a 40% modulation (Li et al., 2022). In chiral Te, the decisive experimental handle is not a single angular harmonic but the joint dependence on field direction, chirality, and thickness (Li et al., 5 Dec 2025).

This comparison suggests that “nonreciprocal angular magnetoresistance” is best treated as an umbrella category defined by transport symmetry rather than by a unique microscopic mechanism.

6. Conceptual boundaries, related effects, and common points of confusion

A persistent source of confusion is the distinction between reciprocal angular magnetoresistance and genuine nonreciprocity. The orbital Rashba-Edelstein magnetoresistance observed in Py/Cu$_4$26 is explicitly presented as a reciprocal magnetoresistance effect. Its $_4$27-scan exhibits an SMR-like angular dependence, $_4$28, but the paper emphasizes that it is a linear-response magnetoresistance controlled by the relative angle between induced orbital angular momentum and magnetization, not by current-direction asymmetry (Ding et al., 2021).

A second boundary case is transverse magnetoresistance generated by anomalous-Hall-effect-driven charge-spin conversion in an F$_4$29/N/F$_4$30 trilayer. There the angular dependence is basically identical to the planar Hall effect through the factor $_4$31, and the genuinely new term depends on $_4$32. Because this dependence is even under angle reversal and under exchange of the two magnetizations, the effect is not nonreciprocal in the strict sense; it is a relative-angle-dependent correction to a planar-Hall-like response (Taniguchi, 2018).

A third nearby but distinct topic is the angular and bias dependence of tunnel magnetoresistance in magnetic tunnel junctions. In Fe/MgO/Fe-type spin-torque nano-oscillators, the angular dependence of resistance and its bias dependence create an asymmetric spin-transfer torque that is essential for steady-state precession. The paper describes this as clear evidence of asymmetric angular transport, but not as a direct study of nonreciprocal magnetoresistance in the usual two-terminal sense (Kowalska et al., 2018).

These distinctions matter because similar angular waveforms can arise from physically different categories: reciprocal AMR- or SMR-like responses, nonlinear nonreciprocal transport, thermoelectric contamination, or dynamical magnetoresistive feedback. The contemporary literature therefore relies on a combination of harmonic decomposition, field-reversal analysis, temperature dependence, symmetry arguments, and material-specific control parameters such as chirality, hole density, thickness, or gate voltage to establish whether a given angular magnetoresistance is genuinely nonreciprocal.