betaFeedback

WorksPhysics and Astronomy

Time-resolved magneto-optical effects in the altermagnet candidate MnTe

Abstract

αα-MnTe is an antiferromagnetic semiconductor with above room temperature TNT_N = 310 K, which is promising for spintronic applications. Recently, it was reported to be an altermagnet, containing bands with momentum-dependent spin splitting; time-resolved experimental probes of MnTe are therefore important both for understanding novel magnetic properties and potential device applications. We investigate ultrafast spin dynamics in epitaxial MnTe(001)/InP(111) thin films using pump-probe magneto-optical measurements in the Kerr configuration. At room temperature, we observe an oscillation mode at 55 GHz that does not appear at zero magnetic field. Combining field and polarization dependence, we identify this mode as a magnon, likely originating from inverse stimulated Raman scattering. Magnetic field-dependent oscillations persist up to at least 335 K, which could reflect coupling to known short-range magnetic order in MnTe above TNT_N. Additionally, we observe two optical phonons at 3.6 THz and 4.2 THz, which broaden and redshift with increasing temperature.

In the past several years, antiferromagnets have been developed into active elements of spintronic devices, forming memory elements[1, 2], terahertz nano-oscillators[3, 4], and low-dissipation interconnects in hybrid quantum systems[5]. A natural route towards integrating these devices with traditional semiconductor functionality is to employ magnetic semiconductors[6]; while room-temperature ferromagnetic semiconductors are rare, antiferromagnetic semiconductors are more common and offer new methods of controlling the spin degree of freedom, such as coupling between magnons and excitons[7]. Hexagonal α\alpha-MnTe is promising due to its high Néel temperature TN≈310T_{N} \approx310 K[8], stable magnetoresistance reflecting Néel orientation[9], and spin-phonon coupling[10, 11]. Very recently, MnTe has received increased attention after the prediction[12, 13] and subsequent confirmation[14] that it is not simply an antiferromagnet, but an altermagnet. In this new magnetic phase, the spin-up and spin-down bands are neither degenerate, like conventional antiferromagnets, nor uniformly split, like ferromagnets; rather, the bands contain anisotropic spin polarization that alternates sign in momentum space while preserving zero net spin polarization.

This anisotropic spin polarization has very recently been observed via ARPES in MnTe[14], MnTe2MnTe_{2}[15], and RuO2RuO_{2}[16], and is predicted to lead to new effects of both scientific and practical importance, such as an anomalous Hall effect[12, 17], an associated magneto-optical Kerr effect (MOKE), novel forms of spin-valley control18, and spin current generation19,20. Currently there are few experimental demonstrations of these effects, mostly in RuO2RuO_{2}[16] [21, 22, 23, 24]. In MnTe, there is one report of an anomalous Hall effect attributed to altermagnetism25; however, variations in stoichiometry and strain can lead to weak ferromagnetism, which can also produce an anomalous Hall effect26. Meanwhile, MOKE in MnTe has been predicted27 but not yet observed26. Experimental studies of magnetism in MnTe are therefore important for better understanding of a potential altermagnetic phase.

In this letter, we perform time-resolved magneto-optical (MO) measurements in the Kerr geometry on hexagonal MnTe thin films as a function of temperature, polarization, and magnetic field. We identify optical phonon modes at 3.6 THz and 4.2 THz and show that they broaden and redshift with increasing temperature from 10 K to 350 K. In the presence of magnetic field, we observe an oscillation mode at 55 GHz below TNT_{N}, which we hypothesize is due to an in-plane k=0k = 0 magnon, as the oscillation is absent at zero magnetic field28. The oscillation amplitude depends on the pump polarization angle, which indicates a non-thermal origin. Surprisingly, magnetic field-dependent 55 GHz oscillations persist even above TN≈310T_{N} \approx310 K, up to the maximum measured T=335T = 335 K, which could be due to the presence of known short-range magnetic ordering above TNT_{N} in MnTe29.

The structure of hexagonal α\alpha-MnTe, with space group P63/mmcP6_{3}/mmc, is shown in Figure 1(a). It is an A-type antiferromagnet, containing ferromagnetic planes of in-plane Mn spins that alternate spin direction along the cc-axis. The easy axis is at 30∘30^{\circ} to the crystalline aa-axis. Our samples are 40 nm-thick MnTe(001) thin films, grown on InP(111) via molecular beam epitaxy.

Image not extracted

Figure 1. Time-resolved magneto-optical setup and observation of optical phonons in MnTe. (a) Crystal and magnetic structure of hexagonal MnTe (001). Gold and silver balls represent Mn and Te, respectively. (b) Schematic of two-color pump-probe MOKE setup with 1560 nm pump and 780 nm probe, both ∼100\sim100 fs in duration. (c) Probe polarization rotation Δθ(t)\Delta\theta(t) in MnTe(001) at 295 K as a function of pump polarization angle φ\varphi, and (d) associated fast Fourier transforms (FFTs).

We perform two-color pump-probe measurements, schematically illustrated in Figure 1(b). We use a 1560 nm pump at 10 mW power and a 780 nm probe at 0.7 mW power on the sample; both pump and probe pulses are ≈100\approx100 fs in duration and have 80 MHz repetition rate. Both beams are focused through a 10x microscope objective at normal incidence, forming a 10 μ\mum spot size. The reflected light is short-pass filtered to reject the pump, then the polarization rotation Δθ(t)\Delta\theta(t) is measured with a Wollaston prism and a balanced photodiode. Signal-to-noise is enhanced by modulating the pump with a photoelastic modulator at 42 kHz and measuring the second harmonic of the photodiode voltage output using a lock-in amplifier. Both pump and probe are linearly polarized.

We first show short time scans (up to 13 ps) of Δθ(t)\Delta\theta(t) at room temperature (295 K) as a function of pump polarization angle φ\varphi in Figure 1(c). φ\varphi is referenced to a InP(111) substrate crystalline axis; the MnTe crystalline axes should follow that of the substrate, likely with 60∘60^{\circ} crystalline twinning. After an initial sharp peak, we observe beating between two modes. Upon taking the FFTs in Figure 1(d), we find peaks at 3.6 and 4.2 THz, whose amplitudes depend on φ\varphi. These frequencies match modes observed in previous Raman spectroscopy of MnTe[10] that were attributed to transverse and longitudinal optical phonons, respectively. Polarization-dependent amplitudes demonstrate a non-thermal excitation mechanism, such as impulsive stimulated Raman scattering (ISRS)30 or displacive excitation of coherent phonons (DECP)31. Since DECP requires absorption of a photon (meaning above-bandgap excitation), ISRS does not, and our pump wavelength is below the bandgap of MnTe(1.15 eV = 1078 nm at 295 K)32, ISRS is the most likely excitation mechanism.

Our results differ from previous time-resolved reflectivity and MOKE on MnTe[11] [33], where only the doubly degenerate E2gE_{2g} phonon mode at 5.3 THz (also seen in Raman scattering[10]) was observed. In these works, however, the pump wavelength was 800 nm (above the bandgap), the probe wavelength was 730 nm, and the pump-probe signal was explicitly attributed to DECP. These different excitation mechanisms and probe wavelengths might cause the differences in observed phonons.

To observe spin oscillations, we fix φ=225∘\varphi= 225^{\circ} from horizontal, where both phonon modes are excited in roughly equal magnitude, and measure Δθ(t)\Delta\theta(t) while varying the applied magnetic field HappH_{\mathrm{app}}. To observe MO signal from out-of-plane magnetization or potential in-plane altermagnetism25, we apply out-of-plane field with a permanent magnet placed behind the sample. Δθ(t)\Delta\theta(t) scans are shown in Figure 2(a) and associated FFTs are shown in Figure 2(b). After the initial fast phonon oscillations, we observe slower oscillations whose amplitude depends on HappH_{\mathrm{app}}, and do not appear when Happ=0H_{\mathrm{app}} = 0. Oscillations of similar frequencies were observed in time-resolved transmissivity at 800 nm pump wavelength in MnTe/SrF2_2(111) in Ref. 34, where they were tentatively attributed to acoustic phonons35.

Image not extracted

Figure 2. Magnon and acoustic phonon oscillations at 295 K. (a) Longer scans (133 ps duration) at fixed φ=225∘\varphi= 225^\circ as a function of field HappH_{\mathrm{app}}, and (b) associated FFTs. We observe a mode at 55 GHz with field-dependent amplitude, which indicates spin oscillations, as well as modes at ∼17\sim17 and 34 GHz that may be due to acoustic phonons. (c) Pump polarization dependence of Δθ(t)\Delta\theta(t) at fixed Happ=2.2H_{\mathrm{app}} = 2.2 kG, with FFTs in (d). Oscillation amplitudes are φ\varphi-dependent, demonstrating a non-thermal origin such as phonon-induced magnons. (e) Difference and sum of Δθ(t)\Delta\theta(t) at +2.2 kG and -2.2 kG, and (f) associated FFTs. Δθ(t)\Delta\theta(t) is largely even in HappH_{\mathrm{app}}.

We fit the data from 10 ps onwards, discarding the initial phonon modes, to the sum of an exponential decay representing electron-phonon relaxation[11], two exponentially damped sinusoids representing acoustic phonons and/or magnons36, and a linear background:

Δθ(t)=Aexpe−t/t0+A1e−α1tsin⁡(2πf1t+ϕ1)+A2e−α2tsin⁡(2πf2t+ϕ2)+Bt+C(1)\begin{aligned} \Delta\theta(t) &= A_{\mathrm{exp}}e^{-t/t_0} + A_1e^{-\alpha_1 t}\sin(2\pi f_1t+\phi_1) \\ &\quad+ A_2e^{-\alpha_2 t}\sin(2\pi f_2t+\phi_2) + Bt + C \tag*{(1)} \end{aligned}

Fits are shown with the data in Figure 2(a), except at Happ=0H_{\mathrm{app}} = 0 kG, where a consistent fit could not be obtained. At this pump angle (225∘225^\circ), the features of the data are well-captured by two damped oscillations and fitting to three did not significantly improve the fit (see the supporting information (SI)). Best fit parameters are shown in Table I. We find that although both oscillations go to zero when Happ=0H_{\mathrm{app}} = 0, A1A_1 monotonically depends on the magnitude of HappH_{\mathrm{app}}, which suggests a magnetic origin, whereas A2A_2 does not.

f1f_1 (GHz)f2f_2 (GHz)f3f_3 (GHz)A1A_1 (mdeg)A2A_2 (mdeg)A3A_3 (mdeg)
295 K, +2.2 kG13.7(5)33.8(4)59.6(1.5)0.19(1)0.36(5)0.25(6)
295 K, -2.2 kGN/A33.8(2)64(1)N/A0.69(5)1.0(4)
325 K, +2.2 kGN/A31.4(4)58(2)N/A0.20(3)1.4(5)
325 K, 0 kG15.3(3)34.9(3)N/A0.07(1)0.33(4)N/A
325 K, -2.2 kG18(2)38.3(4)59.1(4)0.6(1)0.56(5)0.68(7)

Table I. Fit parameters for the field-dependent data in Figure 3(b) and Figure 3(c) of the main text.

Previous studies of magnons in MnTe are scarce[8] [37]. We can estimate the k=0k = 0 magnon frequencies from the anisotropy, which in MnTe is largely easy-plane because optics probe the $k = 0 magnons. Since the in-plane anisotropy field HAzH_{Az} is small compared to the hard-axis (easy-plane) field HAxH_{Ax}, we can use the formula for the two k=0k=0 easy-plane magnon frequencies ωα\omega_{\alpha} and ωβ\omega_{\beta}28:

ωα,β2=γ2{HE(HAx+2HAz)+H02±[4H02HE(HAx+2HAz)+HE2HAx2]1/2}(2)\omega_{\alpha,\beta}^{2}=\gamma^{2}\left\{H_{E}(H_{Ax}+2H_{Az})+H_{0}^{2}\pm\left[4H_{0}^{2}H_{E}(H_{Ax}+2H_{Az})+H_{E}^{2}H_{Ax}^{2}\right]^{1/2}\right\} \tag*{(2)}

where γ=gμB/ℏ\gamma=g\mu_{B}/\hbar is the gyromagnetic ratio and HEH_{E} is the exchange field. For HAx≫HAzH_{Ax}\gg H_{Az}, Eqn. (3) reduces to:

ωα02≈γ2(2HEHAx+3Happ2), ωβ02≈γ2(2HEHAz−Happ2),(3)\omega_{\alpha0}^{2}\approx\gamma^{2}(2H_{E}H_{Ax}+3H_{app}^{2}),\ \omega_{\beta0}^{2}\approx\gamma^{2}(2H_{E}H_{Az}-H_{app}^{2}), \tag*{(3)}

while for Happ=0H_{app}=0, regardless of the anisotropy values, the frequencies are

ωα02=γ22HE(HAz+HAx), ωβ02=γ22HEHAz.(4)\omega_{\alpha0}^{2}=\gamma^{2}2H_{E}(H_{Az}+H_{Ax}),\ \omega_{\beta0}^{2}=\gamma^{2}2H_{E}H_{Az}. \tag*{(4)}

Although the values of HAzH_{Az} and HAxH_{Ax} are not well-known, we can get an estimate by using the spin-flop field HSF≈2HEHAzH_{SF}\approx\sqrt{2H_{E}H_{Az}}, about 2 T for MnTe/InP(111)38. Assuming g=2g=2 for MnTe, the in-plane frequency ωβ0\omega_{\beta0} evaluates to 56 GHz, consistent with our measured value. From the lack of any measurable magnetization response in MnTe with OOP field up to 6 T in Ref. 9, we can estimate that HAx>6H_{Ax}>6 T, which means that the OOP magnon frequency ωα0>176\omega_{\alpha0}>176 GHz - therefore, features between 17 and 35 GHz are unlikely to be magnons. They may represent acoustic phonons, or a non-monotonic magneto-optical response to sample heating39. In previous studies of the easy-plane antiferromagnet NiO, the in-plane mode was observed with a linearly polarized probe through Faraday rotation, while the out-of-plane mode was observed with a circularly polarized probe through the Cotton-Mouton effect (linear birefringence)40. If similar mechanisms are active in MnTe, the higher-frequency magnon may only appear when using a circularly polarized probe, which will be the subject of future work.

Following the field dependence, we measure pump polarization dependence of the spin oscillations at fixed Happ=2.2H_{app}=2.2 kG in Figure 2(c), with associated FFTs in 2(d). We find that the oscillation amplitudes are pump polarization-dependent, although at the current signal-to-noise level, we cannot definitively determine the symmetry. Data at φ=75∘\varphi=75^\circ fits better to three oscillations, while data at φ=195∘\varphi=195^\circ and 225∘225^\circ seems to contain only two (see SI). This suggests that multiple acoustic phonons with different pump angle dependences may be excited by the laser.

In general, magnons measured with ultrafast optical pump-probe can be thermally or non-thermally generated. In thermal generation, laser heating temporarily changes the anisotropy field HAH_A, which reorients HeffH_{eff} and causes the spins to precess around the new HeffH_{eff}41. Non-thermal magnon generation is often caused by spin-phonon coupling in magnetic semiconductors and insulators when the laser coherently excites phonons42, and was observed in MnTe in Refs. 11 and 33. In those experiments, the off-normally incident laser excited the 5.3 THz E2gE_{2g} optical phonon, causing the Te atoms to oscillate out-of-phase. This modulated the Mn-Te-Mn bond angle, in turn causing an oscillation in the weak Te-mediated superexchange J3J_3 which generated magnons at the same frequency. In our data, however, the 3.6 THz and 4.2 THz oscillations in Δθ(t)\Delta\theta(t) show no field dependence within experimental resolution (see the supplemental information). Therefore, we attribute them entirely to optical phonons. The polarization dependence in Figure 2(c) is consistent with a hypothesis of acoustic phonon-induced spin oscillations at 55 GHz, driven by ISRS30.

In many pump-probe experiments, magnetic and non-magnetic signals - for example, from acoustic phonons or phonon-induced spin tilting43,44 - are separated by taking the sum and difference of data acquired at opposite magnetic fields45. Therefore, in Figure 2(e) we plot the sum and difference of Δθ(t)\Delta\theta(t) at Happ=+2.2H_{app}=+2.2 kG and -2.2 kG, with FFTs in 2(f). We find greater amplitude in the sum than the difference, and in fact the difference nearly cancels out after 40 ps. This means that although Δθ(t)\Delta\theta(t) is field-dependent, it does not change sign with HappH_{app} (which can be seen as well in Figure 2(a)). Δθ(t)\Delta\theta(t) that is even in HappH_{app} can occur if Δθ(t)\Delta\theta(t) is due to a Voigt effect40 or optical birefringence46, which is quadratic in magnetization mm, rather than the Kerr effect that is linear in mm. Since the net mm is zero in antiferromagnets, the dominant magneto-optical effect is often these quadratic terms. Static Kerr and circular dichroism measurements of these samples showed no hysteresis26, which is consistent with a lack of observed TR-MOKE signal.

After field and polarization dependence, we measure Δθ(t)\Delta\theta(t) as a function of temperature through TNT_N (≈300\approx300 K in our thin films26, slightly lowered from bulk TN=310T_N = 310 K due to strain), where long-range ordering disappears. Figure 3(a) shows Δθ(t)\Delta\theta(t) at fixed Happ=+2.2H_{\mathrm{app}} = +2.2 kG from 295 K to 335 K. Surprisingly, oscillations at 55 GHz seem to persist at least up to 335 K, the highest temperature measured. To determine if they are still magnetic above the nominal TNT_N, we compare Δθ(t)\Delta\theta(t) taken at Happ=+2.2H_{\mathrm{app}} = +2.2 kG, 0 kG, and -2.2 kG at 295 K and 325 K in Figure 3(b) and (c), respectively. Fit parameters are shown in the SI. At both temperatures, the 55 GHz mode appears only when Happ≠0H_{\mathrm{app}} \ne0, which indicates that it is still magnetic. Previous studies in both ferromagnets47 and antiferromagnets48 have shown that in materials with short-range ordering above the Curie or Néel temperatures, respectively, magnetic anisotropy can persist above TCT_C or TNT_N, and therefore collective spin oscillations can be observed at the same frequency through the transition. Since short-range magnetic order is known to exist in MnTe well above TNT_N29, it is therefore plausible that MnTe continues to exhibit spin oscillations for some temperature range above TNT_N.

Image not extracted

Figure 3. Measuring Δθ(t)\Delta\theta(t) through TNT_N. (a) Δθ(t)\Delta\theta(t) at fixed Happ=+2.2H_{\mathrm{app}} = +2.2 kG from 295 K to 335 K. 55 GHz oscillations appear at all temperatures. (b), (c) Δθ(t)\Delta\theta(t) at Happ=+2.2H_{\mathrm{app}} = +2.2 kG, 0 kG, and -2.2 kG OOP at 295 K and 325 K, respectively. At both temperatures, the 55 GHz mode appears only when Happ≠0H_{\mathrm{app}} \ne0, which indicates that it is magnetic.

Lastly, we measure Δθ(t)\Delta\theta(t) as a function of temperature from 10 K up to 350 K in a cryostat. In this configuration, the maximum attainable HappH_{\mathrm{app}} was ≈500\approx500 G, which we found was not sufficient to resolve spin oscillations. Instead, we focus on the optical phonon oscillations, showing Δθ(t)\Delta\theta(t) traces in Figure 4a and FFTs in 4b. (Full time traces showing no spin oscillations are shown in the SI.)

Image not extracted

Figure 4. Temperature dependence of the 3.6 THz and 4.2 THz optical phonons. (a) Δθ(t)\Delta\theta(t) traces from 10 K to 350 K and (b) corresponding FFTs. At higher temperatures, the phonons broaden in linewidth, redshift, and have shortened lifetime, in agreement with previous Raman studies. (c) Fit time scales τ1\tau_1 and τ2\tau_2 after fitting the background to a two-temperature model described in Eqn. (4). (d) Amplitudes and (e) frequencies of the two phonons, obtained by fitting to two damped sinusoids after background subtraction.

To analyze the temperature dependence of the phonons, we fit the background, then subtract it off and fit the residual oscillations to two damped sinusoids (see SI). We find that this two-step process is more robust than fitting the background and oscillations simultaneously. At low temperatures, modeling the background by a sum of decaying exponentials does not yield a satisfactory fit. Therefore, we adapt a phenomenological model developed to fit pump-probe reflectivity in Ref. 49:

ΔTel(t)=ΔTe,max⁡erf⁡(tτee)e−t/τel,ΔTl(t)=ΔTl,max⁡erf⁡(tτ0)(1−e−(t+τ0)/τel),ΔR/R(t)=AΔTe(t)+BΔTl(t),(5)\begin{aligned} \Delta T_{\mathrm{el}}(t) &= \Delta T_{e,\max}\operatorname{erf}\left(\frac{t}{\tau_{ee}}\right)e^{-t/\tau_{\mathrm{el}}},\\ \Delta T_l(t) &= \Delta T_{l,\max}\operatorname{erf}\left(\frac{t}{\tau_0}\right)\left(1-e^{-(t+\tau_0)/\tau_{\mathrm{el}}}\right),\\ \Delta R/R(t) &= A\Delta T_e(t)+B\Delta T_l(t), \tag*{(5)} \end{aligned}

Eqn. (6) approximates a two-temperature model in which ΔTel(t)\Delta T_{\mathrm{el}}(t) is the electron temperature, ΔTl(t)\Delta T_l(t) is the lattice temperature, τee\tau_{ee} is the electron-electron equilibration time, τel\tau_{\mathrm{el}} is the electron-phonon relaxation time, τ0∼100\tau_0 \sim100 fs is the pump pulse width, and AA and BB are fitting amplitudes relating temperature changes to optical response. Since Δθ(t)\Delta\theta(t) is not identical to reflectivity, we fit to:

Δθ(t)=ΔθK,1(t)+ΔθK,2(t)=C1erf⁡(tτ1)e−t/τ2+C2erf⁡(τ/τ0)(1−e−(τ+τ0)/τ2)(6)\Delta\theta(t) = \Delta\theta_{\mathrm{K},1}(t) + \Delta\theta_{\mathrm{K},2}(t) = C_1\operatorname{erf}\left(\frac{t}{\tau_1}\right)e^{-t/\tau_2} + C_2\operatorname{erf}\left(\tau/\tau_0\right)\left(1-e^{-(\tau+\tau_0)/\tau_2}\right) \tag*{(6)}

without explicitly identifying τ1\tau_1 and τ2\tau_2 with τee\tau_{ee} and τel\tau_{el}. Fitting results from the electron and lattice time scales are shown in Figure 4(c). τ1\tau_1 increases with increasing TT, while τ2\tau_2 decreases. Although the background is well-fit by this model (see SI for details), τ1\tau_1 is too long to be straightforwardly associated with τee\tau_{ee}, which is typically <1< 1 ps. However, τ2\tau_2 is on the same order as typical τel\tau_{el}, and the increase of τ2\tau_2 with TT may correspond to the known increase of τel\tau_{el} with increasing electron temperature50. Unlike Ref. 11, we do not observe a sublattice demagnetization time scale that changes dramatically above TNT_N, which suggests that the 3.6 and 4.2 THz phonons are unrelated to the spin oscillations.

After fitting the background, we fit the residual optical phonon oscillations to two damped sinusoids as in Eqn. (2): Δθ(t)=A1e−α1tsin⁡(2πf1t+ϕ1)+A2e−α2tsin⁡(2πf2t+ϕ2)\Delta\theta(t) = A_1e^{-\alpha_1t}\sin(2\pi f_1t+\phi_1)+A_2e^{-\alpha_2t}\sin(2\pi f_2t+\phi_2). The phonon fit amplitudes A1,A2A_1,A_2 and frequencies f1,f2f_1,f_2 are shown in Figure 4(d) and (e). The frequencies monotonically decrease with increasing TT, in agreement with previous Raman studies[10]. We find that the amplitude of the 4.2 THz phonon A1A_1 decreases slightly with increasing TT, while A2A_2 (the 3.6 THz phonon) increases until it remains constant above 250 K.

In conclusion, we used TR-MOKE to identify a magnon mode and two optical phonon modes in the antiferromagnetic semiconductor MnTe. Both the magnon and the optical phonon can be suppressed or enhanced by changing pump polarization, demonstrating a non-thermal origin. We continued to observe field-dependent oscillations above TNT_N, at least up to 335 K, which may reflect the persistence of magnetic anisotropy and coupling to short-range order above TNT_N and will be the subject of future work. Our results provide new insight into this unique magnetic system, and will be important as MnTe continues to be studied for spintronic and altermagnetic applications.

Acknowledgments

The authors thank Prof. Jay Kikkawa for helpful discussions. I.G. was mainly sponsored by the Army Research Office under the grant No. W911NF-20-2-0166. The construction of the pump-probe setup was supported by the Air Force Office of Scientific Research under award no. FA9550-22-1-0410. Q.D. was supported by the NSF EPM program under grant no. DMR-2213891 and the Vagelos Institute of Energy Science and Technology graduate fellowship. Q.T. was support by the US Office of Naval Research (ONR) through the grant N00014-24-1-2064. The work at ORNL was supported by the U. S. Department of Energy (DOE), Office of Science, Basic Energy Sciences (BES), Materials Sciences and Engineering Division (growth and materials characterization).

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Author Declarations

Conflict of Interest

The authors have no conflicts to disclose.

Author Contributions

Isaiah Gray: Conceptualization (equal); Data curation (lead); Formal analysis (lead); Methodology (equal); Validation (lead); Software (supporting); Visualization (lead); Writing - original draft (lead); Writing - review and editing (equal). Qinwen Deng: Data curation (supporting); Formal analysis (supporting); Methodology (equal); Software (equal). Qi Tian: Methodology (equal); Software (equal). Michael Chilcote: Resources (equal). J. Steven Dodge: Writing: review and editing (supporting). Matthew Brahlek: Resources (equal); Writing - review and editing (equal). Liang Wu: Conceptualization (equal); Funding Acquisition (lead); Methodology (equal); Project Administration (lead), Supervision (lead), Writing - review and editing (equal).

References

Supplemental Materials for "Time-resolved magneto-optical Kerr effect in the altermagnet candidate MnTe"

Magnetic field dependence of optical phonons

In Figure S1, we show short scans of Δθ(t)\Delta\theta(t) scans at Happ=+2.2H_{\mathrm{app}} = +2.2 kG, 0 kG, and −2.2-2.2 kG out-of-plane, with corresponding FFTs. The oscillations repeat nearly point-for-point, showing no change with field. Therefore, we ascribe this signal purely to optical phonons and not phonon-induced magnons. Vertical dashed lines are drawn at f=3.65f = 3.65 THz and f=4.26f = 4.26 THz in S1(b) as a visual aid to show that the peaks do not shift.

Image not extracted

Figure S1. (a) Δθ(t)\Delta\theta(t) time traces and (b) FFTs showing optical phonons at 295 K at Happ=+2.2H_{\mathrm{app}} = +2.2 kG, 0 kG, and −2.2-2.2 kG out-of-plane.

Long scans at low temperature

Here, we show long scans at low temperature at Happ=500H_{\mathrm{app}} = 500 G, demonstrating no magnon oscillations. In Figure S2, we show the full data set from Figure 4 of the main text, which restricted the time extent to 20 ps so that the optical phonon oscillations could be more easily seen. In Figure S2, we present a separate set of Δθ(t)\Delta\theta(t) scans with 133 ps time extent on the same sample, also with Happ=500H_{\mathrm{app}} = 500 G. Within experimental sensitivity (∼1 μV\sim1\,\mu\mathrm{V}), Δθ(t)\Delta\theta(t) is constant after electron-lattice relaxation.

Image not extracted

Figure S2. (a) Full data set from Figure 4(a) of the main text, with 35 ps time extent. (b) A separate temperature-dependent run with 133 ps time extent, showing no magnon oscillations.

Fitting TR-MOKE background at low temperatures

In this section, we detail the fitting process for the low-temperature data in Figure 4 of the main text. We show the fitting at 10 K; other temperatures are processed in exactly the same way. Figure S3 shows the raw data at 10 K, the individual contributions from the best fit Δθ1(t)\Delta\theta_1(t) and Δθ2(t)\Delta\theta_2(t) from Eqn. (6), and their sum Δθnet(t)\Delta\theta_{\mathrm{net}}(t). In Figure S3, we subtract the best-fit background from the data and fit the residuals to Δθ(t)=A1e−α1tsin⁡(2πf1t+ϕ1)+A2e−α2tsin⁡(2πf2t+ϕ2)\Delta\theta(t) = A_1 e^{-\alpha_1 t}\sin(2\pi f_1 t + \phi_1) + A_2 e^{-\alpha_2 t}\sin(2\pi f_2 t + \phi_2).

Image not extracted

Figure S3. Fitting 10 K data from Figure 4 of the main text. (a) Fitting the background to Eqn. (6). (b) Fitting the residual oscillations (with background subtracted) to two damped sinusoids.

Details of fitting magnetic field and pump polarization dependence

Here we provide details of the fitting of the spin oscillations as well as the optical phonon oscillations in Figure 1 and Figure 2 of the main text. In Figure S4, we show an example of fitting the magnetic field dependence at fixed 225∘225^\circ pump angle in Figure 2(a) of the main text. We fit to two damped sinusoids in Figure S4 and three damped sinusoids in S4(b). In this set, a third oscillation is not well-resolved (f1=3.9±3.6f_1=3.9\pm3.6 GHz) and the quality of the fit is not improved by including it. Data at other values of HappH_{\mathrm{app}} in Figure 2(a) shows similar behavior.

Image not extracted

Figure S4. Fitting the data at Happ=+2.2H_{\mathrm{app}}=+2.2 kG from Figure 2(a) of the main text. At this pump angle (225∘225^\circ), three damped sines do not provide an improved fit compared to two damped sines.

We then fit the pump polarization dependence in Figure S5. We fit the spin oscillation at 75∘75^\circ pump angle to two and three sinusoids in S5(a) and (b), respectively, and compare to similar fits of the oscillations at 195∘195^\circ pump angle in (c) and (d). In the 75∘75^\circ data, three frequencies, fitting to 25, 43, and 55 GHz, are better than two, while in the 195∘195^\circ data, the fits are nearly identical and f3=58±5f_3=58\pm5 GHz is not well-distinguished from f1=51.8±0.2f_1=51.8\pm0.2 GHz. This suggests that there are at least three frequencies - likely one magnon and two acoustic phonons - with different pump angle dependence.

Image not extracted

Figure S5. Fitting polarization dependence from Figure 1 and Figure 2 of the main text. (a,b) Fitting the spin oscillation data at 75° pump angle to two and three damped sinusoids, respectively. (c,d) Repeating the process at 195° pump angle. While three sinusoids fit better than two at 75°, no noticeable improvement is seen at 195°.

Fitting Δθ(t)\Delta\theta(t) above TNT_N

In this section, we show the fit parameters for the field-dependent data in Figure 3(b) and (c) of the main text. We follow the same fitting procedure as in the rest of the text: we fit the raw data to a background of a decaying exponential plus a linear function, subtract off the background, then fit the remaining oscillations to three exponentially damped sinusoids. Resulting frequencies and amplitudes for each sinusoid are shown in Table 1, with uncertainties on the last significant digit given in parentheses. As in Figure 2 of the main text, the data at 295 K and 0 kG did not exhibit sufficient oscillation amplitude to allow a consistent fit.

HappH_{app}2.2 kG1.7 kG1.3 kG-2.2 kG
A1A_1 (mdeg)0.108(5)0.063(5)0.039(4)0.049(3)
α1\alpha_1 (ps−1^{-1})0.023(1)0.023(2)0.016(2)0.026(2)
f1f_1 (GHz)53.9(2)53.3(3)53.4(3)56.0(5)
ϕ1\phi_1 (deg)13(3)12(5)5.0(1)-16(4)
A2A_2 (mdeg)0.236(8)0.22(1)0.32(1)0.345(5)
α2\alpha_2 (ps−1^{-1})0.084(3)0.078(4)0.107(4)0.073(2)
f2f_2 (GHz)36.7(5)36.2(7)32.8(6)18.2(3)
ϕ2\phi_2 (deg)14(2)16(3)20(2)52(2)

Table 1. Fit parameters for the data in Figure 2(a), fit to Eqn. (1). Uncertainties on the last significant digit are given in parentheses.

We find that at both 295 K and 325 K, the f3f_3 mode (between 58 and 64 GHz) appears only when Happ≠0H_{\mathrm{app}} \ne0, whereas the ∼15\sim15 GHz and 35 GHz modes are also present at 325 K and 0 kG. Note that no constraints on the frequency values were imposed in the fitting, so the presence or absence of the 55 GHz mode is not an artifact of bounding the frequencies.

References

  1. [1]P. Wadley, B. Howells, J. Železný, C. Andrews, V. Hills, R. P. Campion, V. Novák, K. Olejník, F. Maccherˈozzi, S. S. Dhesi, S. Y. Martin, T. Wagner, J. Wunderlich, F. Freimuth, Y. Mokrousov, J. Kuneš, J. S. Chauhan, M. J. Grzybowski, A. W. Rushforth, K. W. Edmonds, B. L. Gallagher, and T. Jungwirth, “Electrical switching of an antiferromagnet,” Science 351, 587–590 (2016).
  2. [2]T. Higo, K. Kondou, T. Nomoto, M. Shiga, S. Sakamoto, X. Chen, D. Nishio-Hamane, R. Arita, Y. Otani, S. Miwa, and S. Nakatsuji, “Perpendicular full switching of chiral antiferromagnetic order by current,” Nature 607, 474–479 (2022).DOI
  3. [3]R. Cheng, D. Xiao, and A. Brataas, “Terahertz antiferromagnetic spin hall nano-oscillator,” Phys. Rev. Lett. 116, 207603 (2016).
  4. [4]K. Olejník, T. Seifert, Z. Kašpar, V. Novák, P. Wadley, R. P. Campion, M. Baumgartner, P. Gambardella, P. Němec, J. Wunderlich, J. Sinova, P. Kužel, M. Müller, T. Kampfrath, and T. Jungwirth, “Terahertz electrical writing speed in an antiferromagnetic memory,” Science Advances 4, eaar3566 (2018).DOI
  5. [5]R. Lebrun, A. Ross, S. A. Bender, A. Qaiumzadeh, L. Baldrati, J. Cramer, A. Brataas, R. A. Duine, and M. Kläui, “Tunable long-distance spin transport in a crystalline antiferromagnetic iron oxide,” Nature 561, 222–225 (2018).DOI
  6. [6]T. Jungwirth, X. Marti, P. Wadley, and J. Wunderlich, “Antiferromagnetic spintronics,” Nature Nanotechnology 11, 231–241 (2016).
  7. [7]Y. J. Bae, J. Wang, A. Scheie, J. Xu, D. G. Chica, G. M. Diederich, J. Cenker, M. E. Ziebel, Y. Bai, H. Ren, C. R. Dean, M. Delor, X. Xu, X. Roy, A. D. Kent, and X. Zhu, “Exciton-coupled coherent magnons in a 2d semiconductor,” Nature 609, 282–286 (2022).DOI
  8. [8]W. Szuszkiewicz, B. Hennion, B. Witkowska, E. Łusakowska, and A. Mycielski, “Neutron scattering study of structural and magnetic properties of hexagonal MnTe,” physica status solidi (c) 2, 1141–1146 (2005).
  9. [9]D. Kriegner, K. Výborný, K. Olejník, H. Reichlová, V. Novák, X. Marti, J. Gazquez, V. Saidl, P. Němec, V. V. Olobuev, G. Springholz, V. Holý, and T. Jungwirth, “Multiple-stable anisotropic magnetoresistance memory in antiferromagnetic MnTe,” Nature Communications 7, 11623 (2016).DOI
  10. [10]J. Zhang, Q. Lian, Z. Pan, W. Bai, J. Yang, Y. Zhang, X. Tang, and J. Chu, “Spin-phonon coupling and two-magnons scattering behaviors in hexagonal NiAs-type antiferromagnetic MnTe epitaxial films,” Journal of Raman Spectroscopy 51, 1383–1389 (2020).DOI
  11. [11]D. Bossini, S. Dal Conte, M. Terschanski, G. Springholz, A. Bonanni, K. Deltenre, F. Anders, G. S. Uhrig, G. Cerullo, and M. Cinchetti, “Femtosecond phononic coupling to both spins and charges in a room-temperature antiferromagnetic semiconductor,” Phys. Rev. B 104, 224424 (2021).DOI
  12. [12]L. Šmejkal, J. Sinova, and T. Jungwirth, “Emerging research landscape of altermagnetism,” Phys. Rev. X 12, 040501 (2022).
  13. [13]L. Šmejkal, J. Sinova, and T. Jungwirth, “Beyond conventional ferromagnetism and antiferromagnetism: A phase with nonrelativistic spin and crystal rotation symmetry,” Phys. Rev. X 12, 031042 (2022).DOI
  14. [14]J. Krampaský, L. Šmejkal, S. W. D’Souza, M. Hajlaoui, G. Springholz, K. Uhlířová, F. Alarab, P. C. Constantinou, V. Strocov, D. Usanov, W. R. Pudelko, R. González-Hernández, A. B. Hellenes, Z. Jansa, H. Reichlová, Z. Šobáň, R. D. G. Betancourt, P. Wadley, J. Sinova, D. Kriegner, J. Minár, J. H. Dil, and T. Jungwirth, “Altermagnetic lifting of Kramers spin degeneracy,” Nature 626, 517–522 (2024).
  15. [15]Y. Zhu, X. Chen, X.-R. Liu, Y. Liu, P. Liu, H. Zha, G. Qu, C. Hong, J. Li, Z. Jiang, X.-M. Ma, Y.-J. Hao, M.-Y. Zhu, W. Liu, M. Zeng, S. Jayaram, M. Lenger, J. Ding, S. Mo, K. Tanaka, M. Arita, Z. Liu, M. Ye, D. Shen, J. Wrachtrup, Y. Huang, R.-H. He, S. Qiao, Q. Liu, and C. Liu, Nature 626, 523–528 (2024).
  16. [16]O. Fedchenko, J. Minár, A. Akashdeep, S. W. D’Souza, D. Vasilyev, O. Tkach, L. Odenbreit, Q. Nguyen, D. Kutnyakhov, N. Wind, L. Wenthaus, M. Scholz, K. Rossnagel, M. Hoesch, M. Aeschlimann, B. Stadtmüller, M. Kläui, G. Schönhense, T. Jungwirth, A. B. Hellenes, G. Jakob, L. Šmejkal, J. Sinova, and H.-J. Elmers, “Observation of time-reversal symmetry breaking in the band structure of altermagnetic RuO₂,” Science Advances 10, eadj4883 (2024), https://www.science.org/doi/pdf/10.1126/sciadv.adj4883.DOI
  17. [17]S. Hayami and H. Kusunose, “Essential role of the anisotropic magnetic dipole in the anomalous hall effect,” Phys. Rev. B 103, L180407 (2021).

Paper details

vlnc:W_3ccf5ya3
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