---
title: 'Mn3Si2Te6: Ferrimagnetic Semiconductor'
url: https://www.emergentmind.com/topics/mn3si2te6
type: topic
---

# Mn3Si2Te6: Ferrimagnetic Semiconductor

Mn\(_3\)Si\(_2\)Te\(_6\) is a layered manganese silicon telluride that is widely studied as a ferrimagnetic semiconductor with strong magnetic anisotropy, frustrated antiferromagnetic exchange, and unusually strong coupling among spin, orbital, lattice, and transport degrees of freedom. Across the recent literature, it is described as a trigonal compound built from MnTe\(_6\) octahedra, Si–Si dimers, and two inequivalent Mn sublattices, and it has emerged as a model system for pressure-tunable ferrimagnetism, spin-orientation-controlled electronic structure, colossal magnetoresistance, chiral orbital currents, and short-range magnetic order that persists well above the long-range ordering temperature [2211.13321][2504.07869].

## 1. Crystal structure and lattice architecture

Mn\(_3\)Si\(_2\)Te\(_6\) is consistently described as a trigonal layered material built from edge-sharing MnTe\(_6\) octahedra in the \(ab\) plane together with Si–Si dimers, with two crystallographically distinct Mn sites usually denoted Mn1 and Mn2 [2211.13321]. In the common structural description, Mn1 forms a 2D honeycomb network in the \(ab\) plane, while Mn2 forms a triangular layer between Mn1-based layers, yielding what has been called a honeycomb–triangular or “trimer–honeycomb” network [2504.07869]. This arrangement makes the compound quasi-2D in the sense of layered bonding, but still 3D-connected through the Mn2 sublattice.

Most studies in the supplied literature describe Mn\(_3\)Si\(_2\)Te\(_6\) as crystallizing in space group \(P\bar{3}1c\) (No. 163) [2110.10911]. A density-functional study reports optimized lattice parameters \(a=b=7.058~\text{\AA}\) and \(c=14.145~\text{\AA}\), close to low-temperature experimental values \(a=b=7.017\) Å and \(c=14.172\) Å, while an earlier crystallographic study reports room-temperature values \(a=7.046(2)\,\text{\AA}\) and \(c=14.278(2)\,\text{\AA}\) [2211.13321][1808.09949]. One pressure study instead describes a trigonal crystal structure, space group No. 163 \((R\bar{3})\), and one later transport paper states that powder XRD is well fitted with trigonal space group \(P\bar{3}m\); these differing notations represent a literature-level reporting discrepancy within the supplied sources rather than a resolved structural revision [2302.03170][2512.14298].

The local coordination environment is dominated by MnTe\(_6\) octahedra. Synchrotron and EXAFS analysis report Mn1–Te distances of \(3 \times 2.901(4)\) Å and \(3 \times 3.022(4)\) Å, Mn2–Te distances of \(6 \times 2.943(3)\) Å, and a Si–Si dimer bond length of \(2.21(1)\) Å [2110.10911]. A transport-focused study further emphasizes Mn1–Mn1 edge-sharing in-plane distance \(4.0520\) Å and Mn1–Mn2 face-sharing distance \(3.5200\) Å at 80 K, highlighting comparatively strong exchange along \(c\) despite the layered morphology [2103.02764]. This layered-but-interconnected architecture is central to the material’s coexistence of van der Waals-like anisotropy, ferrimagnetism, and substantial three-dimensional magnetic coupling.

## 2. Ferrimagnetic order, anisotropy, and competing magnetic states

Mn\(_3\)Si\(_2\)Te\(_6\) is an ambient-pressure ferrimagnet with a long-range ordering temperature typically reported near \(T_c \approx 73\)–78 K [2302.03170]. Magnetization, neutron diffraction, and thermodynamic studies consistently identify an in-plane easy direction and a hard \(c\)-axis, with moments lying predominantly within the basal plane [1808.09949]. A representative low-temperature in-plane saturation moment is \(M_s \approx 1.6\,\mu_B/\text{Mn}\), much smaller than a fully aligned high-spin Mn\(^{2+}\) moment because the Mn1 and Mn2 sublattices are antiparallel and unequal [2509.02901].

Single-crystal neutron diffraction refines a noncollinear ferrimagnetic structure below \(T_c\) in magnetic space group \(C2'/c'\), with Mn1 and Mn2 moments predominantly in-plane but tilted toward \(c\) by about \(10^\circ\) at ambient conditions [2209.08672]. At 5 K, refined moments are \(|m_{\rm Mn1}| = 4.55(3)\,\mu_B\) and \(|m_{\rm Mn2}| = 4.20(3)\,\mu_B\), with Mn1 carrying components \(m_a=-3.42(9)\,\mu_B\), \(m_b=1.50(8)\,\mu_B\), \(m_c=0.95(3)\,\mu_B\), and Mn2 carrying \(m_a=4.13(8)\,\mu_B\), \(m_b=0\), \(m_c=-0.73(3)\,\mu_B\) [2209.08672]. This establishes that the ground state is ferrimagnetic but not strictly collinear.

Several works characterize the anisotropy as unusually strong for a Mn\(^{2+}\) system. A transport study reports that \(M_{ab}\) fully saturates by \(\mu_0H \approx 0.05\,\text{T}\), whereas \(M_c(H)\) approaches \(M_s\) only near \(\mu_0H \approx 13\,\text{T}\), giving an anisotropy field \(\mu_0H_A \approx 13\,\text{T}\) [2103.02764]. A later microwave-resonance study likewise finds rapid saturation for \(H\parallel ab\) around \(\mu_0H \approx 60\) mT and non-saturation for \(H\parallel c\) up to \(\mu_0H \approx 3.5\) T, consistent with easy-plane anisotropy [2509.02901]. The anisotropy is not exhausted by a single second-order term: torque magnetometry and ESR reconstruct an anisotropy energy
\[
F_{\rm ani}(\theta_M)=K_1\sin^2\theta_M+K_2\sin^4\theta_M+K_3\sin^6\theta_M,
\]
with \(K_1<0\), \(K_2>0\), and a substantial \(K_3>0\), yielding what that work calls a coherent canted ferrimagnet at low temperature [2603.19673].

The ferrimagnetism itself emerges from competing antiferromagnetic exchanges on inequivalent Mn sites. A localized-spin description uses
\[
H=\sum_{\langle ij\rangle} J_{ij}\,\mathbf S_i\cdot\mathbf S_j,
\]
with dominant couplings \(J_1\), \(J_2\), and \(J_3\) all antiferromagnetic in density-functional treatments [2302.03170]. Because Mn1 has twice the multiplicity of Mn2, these AF couplings do not cancel completely, and the ground state is ferrimagnetic rather than collinear antiferromagnetic [2211.13321]. This frustrated ferrimagnetic framework is the common magnetic foundation for the material’s transport, topological, and nonequilibrium responses.

## 3. Electronic structure, semiconducting transport, and angular magnetotransport

Electronic-structure calculations place the low-energy physics of Mn\(_3\)Si\(_2\)Te\(_6\) in a charge-transfer regime where Te \(5p\) states dominate the valence edge and hybridize with Mn \(3d\) states [2211.13321]. In the ferrimagnetic in-plane spin configuration FiM[110], the calculated band gap is \(E_g^{\rm FiM[110]} \approx 161.2~\text{meV}\), whereas the out-of-plane FiM[001] configuration is metallic in the same calculation [2211.13321]. Rotating the spin quantization axis from in-plane toward \(c\) continuously suppresses the gap, providing a microscopic route to the giant angular dependence of transport.

Bulk transport studies describe Mn\(_3\)Si\(_2\)Te\(_6\) as a semiconductor or insulator with very low carrier density. One report finds \(\rho_{ab}\) rising by 10 orders of magnitude between 380 K and 3 K, reaching \(\rho_{ab}\approx 3\times10^{10}\,\Omega\cdot\text{cm}\) at 3 K, while Hall analysis gives \(n\sim10^{23}\)–\(10^{24}\,\text{m}^{-3}\) [2103.02764]. Another study gives \(\rho(300~\text{K})\approx0.46\,\Omega\cdot\text{cm}\), semiconducting \(d\rho/dT<0\), a sharp dip near \(T_c\), and a low-temperature upturn consistent with localization [2110.10911]. In both cases the transport is strongly spin-coupled.

A recurring transport signature is colossal magnetoresistance for magnetic field along the hard \(c\)-axis. At 10 K, one study reports that \(\rho_{ab}(H\parallel c)\) decreases by seven orders of magnitude above 9 T, producing an insulator–metal transition up to about 130 K, while the same field applied along the easy plane produces only about 20% reduction [2103.02764]. The effect is therefore strongest when magnetic polarization along \(c\) is incomplete, not when easy-axis saturation is achieved. That anisotropy is one of the defining anomalies of the material’s magnetotransport.

The literature distinguishes at least two related but not identical angular transport phenomena. “Colossal angular magnetoresistance” is explained in one first-principles study by the difference between insulating FiM[110] and metallic FiM[001] states, with the gap driven to zero as the spin orientation approaches the \(c\)-axis [2211.13321]. A later anisotropy study reparameterizes the same angular transport in terms of the magnetization angle \(\theta_M\) rather than the field angle \(\theta_H\), using the experimentally extracted anisotropy energy to show that sharp angular resistivity features can emerge from the nonlinear mapping between \(\theta_H\) and \(\theta_M\) near the in-plane configuration [2603.19673]. In that treatment, the near-plane response is consistent with an activated form \(\rho(\theta_M)\propto \exp[\Delta(\theta_M)/k_BT]\).

Polaronic transport is also a recurrent theme. Resistivity and thermopower in single crystals are described by adiabatic small-polaron hopping,
\[
\rho(T)=AT\exp\!\left(\frac{E_\rho}{k_B T}\right),
\qquad
S(T)=\frac{k_B}{e}\left(\alpha+\frac{E_S}{k_B T}\right),
\]
with \(E_\rho>E_S\) in both high- and intermediate-temperature regimes, which that study interprets as a hallmark of small-polaron transport [2110.10911]. Ultrafast THz spectroscopy later reports a transient finite-frequency photoconductivity peak and fits it with a small-polaron conductivity form, extracting a transient binding energy that grows from about 3 meV to about 5 meV within a few picoseconds [2311.14673]. This suggests that polaronic carrier–lattice coupling is relevant both in equilibrium transport and in nonequilibrium optical response.

## 4. Frustration, short-range order, and microscopic magnetic models

The magnetic exchange network of Mn\(_3\)Si\(_2\)Te\(_6\) is frustrated. A DFT-plus-Monte-Carlo study maps several collinear states onto an \(XY\) model
\[
H=-J_1\sum_{\langle ij\rangle}\mathbf S_i\!\cdot\!\mathbf S_j
   -J_2\sum_{[kl]}\mathbf S_k\!\cdot\!\mathbf S_l
   -J_3\sum_{\{mn\}}\mathbf S_m\!\cdot\!\mathbf S_n,
\]
with \(J_1=-33.8~\text{meV}\), \(J_2=-11.4~\text{meV}\), and \(J_3=-13.4~\text{meV}\) at \(U_{\rm eff}=0.5\) eV, all antiferromagnetic in that sign convention [2211.13321]. The same work places the material in the ferrimagnetic region of a frustrated phase diagram, near noncollinear phases. A neutron-based study gives exchange estimates \(J_1\sim-402\) K, \(J_2\sim-73\) K, and \(J_3\sim-172\) K per Mn, again all antiferromagnetic and competing on the honeycomb–triangular network [2209.08672]. The common message is that ferrimagnetism is not the consequence of a simple ferromagnetic interaction but of frustrated AF exchange on inequivalent sublattices.

This frustration extends well above \(T_c\). Total neutron scattering and magnetic pair distribution function analysis reveal short-range magnetic order over a frustrated trimer of two Mn1 and one Mn2 sites that persists to at least 126 K [2504.07869]. Above \(T_c\), the first three neighbor correlations remain AFM, FM, and AFM for \(n=1,2,3\), respectively, reproducing the ferrimagnetic pattern locally despite the absence of long-range order [2504.07869]. The local spins remain essentially confined to the \(ab\)-plane over the full 5–126 K range, and polarized neutron refinement at 90 K and 3 T gives a susceptibility tensor
\[
\hat\chi_{\rm Mn}(90\,\text{K},3\,\text{T})=
\begin{pmatrix}
0.28(3)&0&0\\
0&0.28(3)&0\\
0&0&0.09(5)
\end{pmatrix}\mu_B/\text{T},
\]
demonstrating \(\chi_{ab}/\chi_c\sim3\) and robust planar anisotropy even in the paramagnetic regime [2504.07869].

The correlation length above \(T_c\) decreases monotonically with temperature and fits a BKT-type expression
\[
\xi(T)=A\cdot\exp\!\left[b\left(\frac{T_{KT}}{T-T_{KT}}\right)^{1/2}\right],
\]
with \(A=0.119\pm0.056\), \(b=2.067\pm0.485\), and \(T_{KT}=71\pm3\,\text{K}\) [2504.07869]. Because \(T_{KT}/T_c\approx0.91\), that work argues for quasi-2D XY character in the fluctuation regime. A plausible implication is that the material’s short-range magnetism is not merely a broad precursor to ferrimagnetism but a quantitatively structured regime with trimer correlations, planar anisotropy, and quasi-2D criticality.

Critical-scaling studies likewise place Mn\(_3\)Si\(_2\)Te\(_6\) between mean-field and 3D Heisenberg limits rather than in a simple short-range universality class. One study finds \(\beta=0.41\pm0.01\), \(\gamma=1.21\pm0.02\), and an exchange decay \(J(r)\propto r^{-4.79}\) [1808.09949]. A proton-irradiation study finds exponents near mean-field values across several fluences and interprets the interactions as long-range with effective spatial dimensionality \(d=3\), while the inferred spin dimensionality evolves from \(n=1\) to \(n=2\) or \(n=3\) depending on fluence [2104.06564]. These results collectively support a picture in which layered geometry, frustrated AF exchange, and long-range interaction effects coexist.

## 5. External tuning: pressure, doping, irradiation, and ultrafast excitation

Hydrostatic pressure provides a disorder-free tuning parameter. Magnetometry up to 0.90 GPa shows that the ferrimagnetic transition temperature increases under pressure, reaching \(T_c(P=0.9~\text{GPa})=84.52~\text{K}\), an increase of about 11 K relative to ambient pressure, while the saturation magnetization decreases [2302.03170]. The same study attributes the trend primarily to strengthening of the dominant AF Mn1–Mn2 exchange \(J_1\) under pressure, with weaker pressure dependence of \(J_2\) and \(J_3\) [2302.03170]. By contrast, the easy-plane anisotropy persists and the magnetocrystalline anisotropy energy decreases by about 15% between 0 and 1 GPa [2302.03170].

Chemical substitution perturbs the electronic structure and magnetic anisotropy in a site-selective manner. Virtual-crystal calculations for Mn\(_3\)Si\(_2\)(Te\(_{1-x}\)Se\(_x\))\(_6\) and Mn\(_3\)(Si\(_{1-x}\)Ge\(_x\))\(_2\)Te\(_6\) retain the ferrimagnetic ground state over the investigated ranges [2211.13321]. Se substitution increases the gap for both spin orientations and removes the spin-orientation-induced insulator–metal transition; at \(x=0.2\), \(E_g^{\rm FiM[110]}\approx323.3~\text{meV}\) and \(E_g^{\rm FiM[001]}\approx166.9~\text{meV}\), implying insulating behavior for both orientations and strong suppression of colossal angular magnetoresistance [2211.13321]. Ge substitution has a much smaller effect because Si states lie far below \(E_F\); at \(x=0.05\), \(E_g^{\rm FiM[110]}\approx163.4~\text{meV}\) and \(E_g^{\rm FiM[001]}\approx2.8~\text{meV}\), so the out-of-plane state becomes a very small-gap semiconductor rather than strictly metallic [2211.13321].

Proton irradiation modifies the critical behavior and magnetocaloric response without destroying ferrimagnetism. For a fluence of \(5\times10^{15}~\text{H}^+/\text{cm}^2\), the maximum magnetic-entropy change at 3 T reaches \(2.45~\text{J/kg\,K}\) at \(T=73.66\) K, compared with \(1.60~\text{J/kg\,K}\) at \(T=72.99\) K for pristine crystals [2104.06564]. That study interprets the enhanced magnetization and entropy change as arising from strengthening of Te-mediated ferromagnetic superexchange rather than weakening of the underlying AF component [2104.06564]. A later study reaches a different conclusion for nominally current-induced phase transitions, arguing that abrupt voltage jumps and nonlinear \(I\)–\(V\) behavior are overwhelmingly electrothermal in origin and that the intrinsic \(I\)–\(V\) response is Ohmic when Joule heating is properly accounted for [2502.11048]. Taken together, these reports show that nonequilibrium transport anomalies in Mn\(_3\)Si\(_2\)Te\(_6\) remain an actively contested area.

Ultrafast spectroscopy reveals strong spin–lattice coupling. Femtosecond pump–probe measurements identify a coherent \(E_g\) phonon near \(3.52\)–\(3.57\,\text{THz}\) whose frequency stiffens sharply below \(T_c\) and tracks the magnetization rather than standard anharmonic behavior [2308.14931]. The incoherent dynamics are modeled with a microscopic four-temperature framework that includes electrons, optical phonons, acoustic phonons, and spins, and the extracted electron–optical-phonon coupling \(g_{e\text{-po}}\sim1.75\times10^{18}\)–\(1.95\times10^{18}\,\text{W/m}^3\text{K}\) is described as large [2308.14931]. Room-temperature optical-pump/THz-probe spectroscopy further reports a transient THz conductivity peak around \(60\)–\(70~\text{cm}^{-1}\) with peak \(\sigma_1\sim1000~\Omega^{-1}\,\text{cm}^{-1}\), interpreted as a transient polaronic-like bound state with lifetime \(\tau\approx2.2\) ps [2311.14673]. These studies place Mn\(_3\)Si\(_2\)Te\(_6\) among layered magnets where lattice, spin, and charge are dynamically inseparable on ultrafast time scales.

## 6. Orbital-current physics, topological transport, and nonequilibrium functionality

A major theme of the post-2021 literature is that Mn\(_3\)Si\(_2\)Te\(_6\) hosts chiral orbital currents (COC) flowing along Te edges of MnTe\(_6\) octahedra in the \(ab\)-plane, producing orbital moments along \(c\) [2209.08672]. In that picture, the hard-axis field \(H\parallel c\) enhances a COC state \(\Psi_C\), aligning chiral domains and producing the unconventional colossal magnetoresistance, whereas fields along the easy plane do not [2209.08672]. The same work reports strong current sensitivity and time-dependent bistable switching that it interprets as a first-order melting transition of the COC state [2209.08672]. A later Hall study attributes a sharp, current-sensitive Hall peak, Hall angle up to 0.15, and anomalous scaling \(\sigma_{xy}\propto\sigma_{xx}^{\alpha}\) with \(\alpha\approx3\)–5 in the high-field COC regime to an internal field generated by the orbital-current state rather than to conventional anomalous Hall mechanisms [2309.06610].

This orbital-current interpretation has been extended to topological Hall transport. One later study decomposes the Hall signal as
\[
\rho_{xy}=R_0B+R_sM_s+\rho_{xy}^{\rm THE},
\]
and argues that the extracted \(\rho_{xy}^{\rm THE}\) is dominated by chiral orbital currents rather than chiral spin textures [2512.14298]. In that report, the topological Hall signal is enhanced in nanoflakes, suppressed by increasing current together with the COC state, and closely tracks colossal magnetoresistance in temperature and current dependence [2512.14298]. This suggests an “orbitronic” route to Berry-curvature-driven transport in which real-space orbital textures, rather than noncoplanar spin textures, are the primary emergent field.

Microwave resonance introduces a complementary orbital perspective. Broadband spectroscopy finds \(g\approx2.0\) for \(H\parallel ab\) but \(g_c\approx1.3\)–1.4 for \(H\parallel c\), together with DFT-calculated \(\langle L_z\rangle\approx-0.6\) and \(\langle L_x\rangle\approx0\) [2509.02901]. That combination is interpreted as direct evidence for a substantial orbital magnetic moment along the hard axis arising from Te \(5p\) states when spins are oriented out of plane [2509.02901]. The same study observes microwave-induced resistance modulation under resonance, with \(\Delta R/R\sim10^{-5}\)–\(10^{-6}\), showing that the magnetotransport is sensitive to out-of-plane spin and orbital polarization on GHz time scales [2509.02901].

The nonequilibrium transport literature is now bifurcated. One line of work argues that chiral orbital currents generate intrinsic reactive and memory functionalities in bulk crystals. An example is the report of emergent inductance and nonvolatile memristance under AC drive, with clockwise inductive loops at low frequency and high \(H\parallel c\), and finite remanent voltage at zero current at higher frequency and lower field, all attributed to metastable reconfiguration of chiral orbital-current domains [2604.15635]. Another contemporaneous study instead concludes that spectacular current-induced phase transitions are electrothermal, emphasizing frequency suppression near 1000 Hz, time-resolved resistance traces that mirror equilibrium \(R(T)\), and linear intrinsic \(I\)–\(V\) response after thermal effects are removed [2502.11048]. The coexistence of these interpretations is itself a significant feature of the subject: Mn\(_3\)Si\(_2\)Te\(_6\) has become a focal point for distinguishing genuine nonequilibrium orbital states from strong Joule-heating artifacts in highly nonlinear quantum-magnet transport.

Across these orbital and topological studies, a consistent core remains. Mn\(_3\)Si\(_2\)Te\(_6\) is a ferrimagnetic nodal-line semiconductor with strong easy-plane anisotropy, frustrated AF exchange, short-range order above \(T_c\), and extreme sensitivity of transport to magnetization orientation and hard-axis fields [2504.07869]. What differentiates it from more conventional layered magnets is that this magnetic framework is repeatedly implicated in orbital-moment generation, band-topology control, and current- or field-driven transport anomalies that are far larger and more anisotropic than simple spin-polarization models would predict.

Source: https://www.emergentmind.com/topics/mn3si2te6