This paper was converted on www.awesomepapers.org from LaTeX by an anonymous user.
Want to know more? Visit the Converter page.

Magnetic anisotropy of the van der Waals ferromagnet Cr2Ge2Te6 studied by angular-dependent XMCD

M. Suzuki Department of Physics, The University of Tokyo, Bunkyo-ku, Tokyo 113-0033, Japan    B. Gao Rutgers Center for Emergent Materials and Department of Physics and Astronomy, Rutgers University, Piscataway, New Jersey 08854, USA    G. Shibata Department of Physics, The University of Tokyo, Bunkyo-ku, Tokyo 113-0033, Japan Department of Applied Physics, Tokyo University of Science, Katsushika-ku, Tokyo 125-8585, Japan    S. Sakamoto Department of Physics, The University of Tokyo, Bunkyo-ku, Tokyo 113-0033, Japan    Y. Nonaka Department of Physics, The University of Tokyo, Bunkyo-ku, Tokyo 113-0033, Japan    K. Ikeda Department of Physics, The University of Tokyo, Bunkyo-ku, Tokyo 113-0033, Japan    Z. Chi Department of Physics, The University of Tokyo, Bunkyo-ku, Tokyo 113-0033, Japan    Y.-X. Wan Department of Physics, The University of Tokyo, Bunkyo-ku, Tokyo 113-0033, Japan    T. Takeda Department of Electrical Engineering and Information Systems, The University of Tokyo, Bunkyo-ku, Tokyo 113-8656, Japan    Y. Takeda Materials Sciences Research Center, Japan Atomic Energy Agency (JAEA), Sayo, Hyogo 679-5148, Japan    T. Koide Photon Factory, Institute of Materials Structure Science, High Energy Accelerator Research Organization, Tsukuba, Ibaraki 305-0801, Japan    A. Tanaka Department of Quantum Matter, ADSM, Hiroshima University, Higashi-hiroshima 739-8527, Japan    M. Kobayashi Department of Electrical Engineering and Information Systems, The University of Tokyo, Bunkyo-ku, Tokyo 113-8656, Japan Center for Spintronics Research Network, The University of Tokyo, Bunkyo-ku, Tokyo 113-0033, Japan    S.-W. Cheong Rutgers Center for Emergent Materials and Department of Physics and Astronomy, Rutgers University, Piscataway, New Jersey 08854, USA    A. Fujimori Department of Physics, The University of Tokyo, Bunkyo-ku, Tokyo 113-0033, Japan Materials Sciences Research Center, Japan Atomic Energy Agency (JAEA), Sayo, Hyogo 679-5148, Japan Department of Applied Physics, Waseda University, Shinjuku-ku, Tokyo 169-8555, Japan
Abstract

The van der Waals ferromagnet Cr2Ge2Te6 (CGT) has a two-dimensional crystal structure where each layer is stacked through van der Waals force. We have investigated the nature of the ferromagnetism and the weak perpendicular magnetic anisotropy (PMA) of CGT by means of X-ray absorption spectroscopy and X-ray magnetic circular dichroism (XMCD) studies of CGT single crystals. The XMCD spectra at the Cr L2,3L_{2,3} edge for different magnetic field directions were analyzed on the basis of the cluster-model multiplet calculation. The Cr valence is confirmed to be 3+ and the orbital magnetic moment is found to be nearly quenched, as expected for the high-spin t2gt_{2g}3 configuration of the Cr3+ ion. A large (0.2\sim 0.2 eV) trigonal crystal-field splitting of the t2gt_{2g} level caused by the distortion of the CrTe6 octahedron has been revealed, while the single-ion anisotropy (SIA) of the Cr atom is found to have a sign opposite to the observed PMA and too weak compared to the reported anisotropy energy. The present result suggests that anisotropic exchange coupling between the Cr atoms through the ligand Te 5p5p orbitals having strong spin-orbit coupling has to be invoked to explain the weak PMA of CGT, as in the case of the strong PMA of CrI3.

I Introduction

The discovery of graphene has invoked renewed interest in the studies on two-dimensional (2D) materials including layered transition-metal chalcogenides graphene_first ; review_2 ; review_3 ; TMD_review . The 2D materials show various fascinating properties that may lead to applications for new spintronics devices. In addition, the relation between dimensionality and various physical properties has long been a major subject from a fundamental point of view. For instance, the Mermin-Wagner theorem states that a long-range magnetic order in the 2D isotropic Heisenberg model cannot be realized at finite temperature due to long-range magnetic fluctuations Mermin . Recently, van der Waals (vdW) ferromagnets, which have 2D crystal structures but show ferromagnetic order vdWFerro_review , such as Cr2Ge2Te6 (CGT) CGT_oldcalc_prop ; CGT_oldcalc_2 ; CGT_tricritical_exp ; CGT_aniso_exp ; CGT_MH_MT ; CGT_bilayer ; CGT_hydropressure ; CGT_thinfilm_mag ; ARPES1 ; ARPES2 ; EM_19_Kang and CrI3 CrI3_mono ; CrI3_giant have attracted great interest. Although the magnetism of a few to several-monolayer CGT shows soft behaviors well described by the 2D Heisenberg model CGT_bilayer , the finite Curie temperature of 61 K CGT_aniso_exp ; CGT_oldcalc_prop ; CGT_tricritical_exp and the easy magnetization axis along the cc-axis (out-of-plane) direction CGT_aniso_exp ; CGT_oldcalc_prop ; CGT_MH_MT indicate a finite perpendicular magnetic anisotropy (PMA) and hence a departure from the 2D Heisenberg model. On the other hand, CrI3 exhibits ferromagnetism down to monolayer thickness, which suggests a stronger PMA than CGT and is well described by the 2D Ising model CrI3_mono .

The magnetic anisotropy of a 2D ferromagnet arises primarily from (i) the single-ion anisotropy (SIA) of the magnetic ion and (ii) anisotropic exchange interaction between the magnetic ions in the 2D layer. The SIA of the Cr-based vdW ferromagnets is considered to be small because of the t2gt_{2g\uparrow}3 configuration of the Cr3+ ion, which has no orbital degree of freedom, as discussed, e.g., by Lado and Fernández-Rossi CrI3_aniso_calc . On the contrary, first-principles estimates of exchange coupling by Xu et al. CrI3 Kitaev vs SIA have shown that the SIA and Kitaev-type anisotropic magnetic coupling in the Cr-based vdW ferromagnets are of comparable magnitudes and compete with each other. Furthermore, Kim et al. PRL_19_Kim have performed multi-site cluster-model multiplet calculation and attributed the strong PMA of CrI3 to anisotropic exchange coupling between Cr atoms through the iodine 5p5p ligand orbitals, which has strong spin-orbit coupling (SOC). In fact, X-ray magnetic circular dichroism (XMCD) measurements at the iodine L1L_{1} edge of CrI3 have revealed a finite orbital magnetic moment of the I 5p5p electrons, implying the importance of the ligand pp orbitals in the PMA of the vdW ferromagnets Choi I XMCD . As for CGT, which shows a weaker PMA than CrI3, the situation is complicated and it is far from clear which of the SIA or the anisotropic exchange is responsible for the PMA PRL_19_Kim ; PCCP_19_Wang .

In order to evaluate the contribution of the SIA of Cr to the magnetic anisotropy, XMCD studies at the Cr L2,3L_{2,3} edge are expected to give essential information. So far Cr L2,3L_{2,3}-edge XMCD has been reported for CGT Fumega XMCD and CrI3 Frisk_XMCD with line-shape analysis using cluster-model calculation, but no information about the contributions of the Cr 3d3d electrons to the PMA has been explored. In the present work, we have carried out Cr L2,3L_{2,3}-edge XMCD measurements of CGT single crystal under magnetic fields applied to different crystallographic directions and subsequent cluster-model multiplet calculation to estimate the SIA energy of the Cr atom centered in the CrTe6 octahedron.

II methods

Single crystals of CGT were synthesized using a flux method. High purity Cr, Ge and Te powders were mixed in a molar ratio of 2:6:36; the extra Ge and Te were used as a flux. The mixture was loaded in an alumina crucible and sealed in an evacuated quartz tube. A ball of alumina fiber was placed on top of the crucible. The quartz tube was placed in a furnace, heated to 700 C, held for 5 hours, and then slowly cooled down to 480 C over a period of 3 days. This was followed by centrifugation to remove the flux.

X-ray absorption spectroscopy (XAS) and XMCD measurements at the Cr L2,3L_{2,3} edge were performed at beam line BL23SU of SPring-8 and beam line BL-16A of Photon Factory at High Energy Accelerator Research Organization (KEK-PF). The measurement at SPring-8 was performed under a magnetic field of μ0H=3\mu_{0}H=3 T parallel to the cc-axis of the sample. The XAS and XMCD spectra were recorded in the total electron yield mode. Right- (σ+\sigma^{+}) and left (σ\sigma^{-}) circularly polarized x-rays were switched at a frequency of 1 Hz in order to eliminate time-dependent background from the obtained XMCD spectra. The sample temperature (TT) during the measurement was set to 20 K, well below the Curie temperature. The sample was cleaved in the vacuum chamber prior to the measurements to obtain fresh surfaces. Magnetic-field-direction dependence of XMCD spectra was measured at the BL-16A of KEK-PF. Polarization switching was made at a frequency of 10 Hz. Magnetic field of μ0H=1\mu_{0}H=1 T was applied parallel and perpendicular to the cc-axis. Signals were recorded in the total electron-yield mode.

To investigate the electronic structure which contributes to the ferromagnetism, configuration-interaction cluster-model multiplet calculations have been conducted. Figure 1(a) shows the crystal structure of CGT. Each Cr atom is surrounded by six Te atoms, forming a CrTe6 cluster. It constitutes the CrGeTe3 layer, which are stacked through vdW force. The CrTe6 cluster is slightly elongated along the cc-axis CGT_oldcalc_prop , as shown in Fig. 1(b), which means that the Cr electrons are under a nearly octahedral, trigonally distorted crystal field. The following parameters of the cluster model are used to reproduce the experimental XMCD spectra: the Cr 3dd-3dd Coulomb energy UddU_{dd}, the Cr 2pp-3dd Coulomb energy Upd1.2UddU_{pd}\approx 1.2U_{dd}, the charge-transfer energy Δ\Delta from the Te 5pp to Cr 3dd levels, the Slater-Koster parameter (pdσpd\sigma) between the Cr 3dd and Te 5pp orbitals, the crystal-field splitting 10DqDq (0.7\approx 0.7 eV) between the Cr ege_{g} and t2gt_{2g} levels (apart from the pp-dd covalency contribution), and the trigonal crystal-field splitting DtDt between the a1ga_{1g} and egπe^{\pi}_{g} orbitals. [For the definition of 10DqDq and DtDt, see Fig. 1(c).] As for the parameter values, we follow the empirical relationship fujimori_para and employ Δ=4.5\Delta=4.5 eV, Udd=6.0U_{dd}=6.0 eV, and (pdσ)=2.6(pd\sigma)=2.6 eV as being relevant to Cr3+ tellurides. The SOC parameter of the Cr 3d3d orbital is taken to be 35 meV.

III results and discussion

Figure 2 shows Cr L2,3L_{2,3}-edge XAS and XMCD spectra of CGT under the out-of-plane magnetic field of μ0H=\mu_{0}H= 3 T at TT = 20 K. The applied magnetic field was high enough to saturate the magnetization of CGT because the magnetization saturates below 0.5\sim 0.5 T CGT_tricritical_exp . The tail of the Te M4,5M_{4,5} XAS peak overlapping the Cr L2,3L_{2,3} XAS spectrum has been subtracted assuming a hyperbola. Arctangent steps at the L2L_{2} and L3L_{3} edges have also been subtracted from the Cr L2,3L_{2,3} XAS as usual. On the other hand, the XMCD signals predominantly come from the magnetic Cr atoms, and there is no need to subtract signals from the Te atoms.

Figure 3 shows comparison of the measured Cr L2,3L_{2,3} XMCD spectrum with those calculated using the cluster model assuming the Cr valences of 4+, 3+, 2+ high-spin states, and 2+ low-spin states. The experimental spectrum shows three negative peaks and one positive peak at the L3L_{3} edge, and one negative peak and two positive peaks at the L2L_{2} edge. The comparison shows that the calculation for the Cr3+ state reproduces the experimental spectrum better than those for the Cr2+ and Cr4+ states. Thus, one can conclude that the Cr atom in CGT is in the 3+3+ state (with the high-spin t2g3t_{2g}^{3} configuration), and not in the 2+2+ nor 4+4+ state.

III.1 Spin and orbital magnetic moments

The orbital and effective spin magnetic moments are quantitatively estimated from the XAS and XMCD spectra using the XMCD sum rules Carra ; Thole . For the L2,3L_{2,3} edge (2p3d2p\rm\rightarrow 3d transition), the sum rules are given by:

morb\displaystyle m_{\rm orb} =\displaystyle= 4nh3L3+L2(μ+μ)𝑑ωL3+L2(μ++μ)𝑑ω,\displaystyle-\frac{4n_{\rm h}}{3}\frac{\int_{L_{3}+L_{2}}(\mu_{+}-\mu_{-})d\omega}{\int_{L_{3}+L_{2}}(\mu_{+}+\mu_{-})d\omega},
mspineff\displaystyle m_{\rm spin}^{\rm eff} =\displaystyle= mspin+7mT\displaystyle m_{\rm spin}+7m_{\rm T}
=\displaystyle= 2nhL3(μ+μ)𝑑ω2L2(μ+μ)𝑑ωL3+L2(μ++μ)𝑑ω,\displaystyle-{2n_{\rm h}}\frac{\int_{L_{3}}(\mu_{+}-\mu_{-})d\omega-2\int_{L_{2}}(\mu_{+}-\mu_{-})d\omega}{\int_{L_{3}+L_{2}}(\mu_{+}+\mu_{-})d\omega},

where μ+\mu_{+} (μ\mu_{-}) is an XAS spectrum taken with σ+\sigma^{+} (σ\sigma^{-}), morbm_{\rm orb}, mspinm_{\rm spin} and mTm_{\rm T} are the orbital magnetic moment, spin magnetic moment, and magnetic dipole moment, respectively, in units of Bohr magneton μB\mu_{\rm B}/Cr atom. μ+\mu_{+} (μ\mu_{-}) stands for the absorption coefficient for photon helicity parallel (antiparallel) to the Cr majority spin direction. nhn_{h} is the number of holes in the valence shell. By applying the sum rules to the spectra in Fig. 2, the orbital magnetic moment of morb=0.00±0.05m_{\rm orb}=0.00\pm 0.05 μB\mu_{\rm B}/Cr atoms has been obtained, consistent with the quenched orbital magnetic moment in the d3d^{3} configuration of Cr3+. The effective spin magnetic moment of mspineffmspin+7mT=2.2±0.1m_{\rm spin}^{\rm eff}\equiv m_{\rm spin}+7m_{\rm T}=2.2\pm 0.1 μB\mu_{\rm B}/Cr atom has been obtained, where the mspinm_{\rm spin} and mTm_{\rm T} are the spin magnetic moment and the magnetic dipole moment, respectively. In a high-symmetry environment of the Cr atom as in CGT, mTm_{\rm T} is negligibly small compared to mspinm_{\rm spin} dipole_1 ; dipole_2 . Therefore, since morbm_{\rm orb} is nearly zero, we regard mspineffm_{\rm spin}^{\rm eff} as the total magnetic moment mspinm_{\rm spin}. The value mspinmspineff=2.2±0.1μBm_{\rm spin}\sim m_{\rm spin}^{\rm eff}=2.2\pm 0.1\mu_{\rm B} is consistent with that of the reported saturation magnetization of CGT, 2.2 - 2.9 μB\mu_{\rm B}/Cr atom CGT_MH_MT ; CGT_aniso_exp ; CGT_oldcalc_prop .

III.2 Crystal-field splitting

In order to study the anisotropies of the electronic and magnetic properties of CGT, XMCD measurements under different magnetic field directions have been made. Figure 4 shows the Cr L2,3L_{2,3} XAS and XMCD spectra taken with magnetic fields applied to the out-of-plane (HcH\parallel c-axis) and in-plane (HabH\parallel ab-plane) directions. The XMCD spectra show small but clear differences between the different magnetic-field directions, while the spectral line shapes of the XAS spectra are relatively insensitive to the field direction. To identify the origin of the spectral differences between the different field directions, cluster-model multiplet calculations have been conducted assuming the Cr valence to be 3+. The effects of the elongation of the CrTe6 octahedron along the cc-axis CGT_oldcalc_prop are taken into account as a trigonal crystal-field splitting DtDt of the t2gt_{2g} level as shown in Fig. 1(c), where the triply-degenerate t2gt_{2g} level is split into the non-degenerate a1ga_{1g} and doubly-degenerate egπe^{\pi}_{g} levels. Here, for a positive (negative) trigonal crystal field DtDt, the a1ga_{1g} level is located above (below) the egπe^{\pi}_{g} level. Figures 5(b) and (c) show the calculated spectra without and with the trigonal crystal field, respectively, compared with experiment [Fig. 5(a)]. Without the trigonal crystal field, i.e., Dt=0Dt=0 [Fig. 5(b)], there is no difference in the XMCD spectra between the out-of-plain and in-plane magnetic fields. On the other hand, the calculated XMCD spectra under the trigonal crystal field DtDt of 0.2 eV show finite differences between the out-of-plain and in-plane magnetic fields, as shown in Fig. 5(c), and well explain the experiment [Fig. 5(a)]. Thus we conclude that the differences in the XMCD spectra taken with the different magnetic-field directions are attributed to the elongation of the CrTe6 octahedron along the cc-axis.

III.3 Magnetic anisotropy energies

To evaluate the SIA of the Cr ion, the ground-state energy of the cluster is calculated using the cluster model for the different magnetic field directions. Thus we find that the energy for the in-plane magnetic field is lower than the energy for the out-of-plane magnetic field by 5×106\sim 5\times 10^{-6} eV/Cr atom, that is, the Cr atom has the easy magnetization axis in the abab-plane. This is opposite to the observed PMA of CGT and furthermore its magnitude is 1-2 orders of magnitude smaller than the experimental value of 14×105\sim 1-4\times 10^{-5} eV/Cr atom CGT_aniso_exp ; CGT_bilayer as well as the first-principles-theoretical value of SIA, 3×104\sim 3\times 10^{-4} eV/Cr atom EM_19_Kang . The present result, on the other hand, agrees with the first-principles calculation by Wang et al. PCCP_19_Wang in that the SIA of Cr is weak and acts against PMA.

In spite of the large trigonal crystal field Dt0.2Dt\sim 0.2 eV and the resulting anisotropy in the XMCD spectra, the calculated magnetic anisotropy energy is small because of the high-spin t2g3t_{2g}^{3} configuration, which has no orbital degree of freedom. Therefore, we consider that effects beyond the SIA evaluated using the cluster model, where only the spin-orbit coupling of the Co 3d3d electrons are taken into account, may play important roles in realizing the PMA of CGT. Such effects would include anisotropic exchange coupling between Cr ions through the ligand Te 5p5p electrons with strong SOC, in analogy to CrI3. In the case of CrI3, which has a stronger PMA than CGT, density functional theory (DFT) calculation CrI3_structure_info ; CrI3_aniso_calc and multi-Cr site cluster-model multiplet calculation PRL_19_Kim have clearly shown that exchange coupling between Cr ions through I 5p5p orbitals explains the stronger PMA of CrI3.

In addition to the SIA of Cr and anisotropic exchange between Cr discussed above, the magnetic anisotropy of the Te 5p5p electrons themselves may contribute to the magnetic anisotropy of CGT PCCP_19_Wang . Such an anisotropy could arise because the Te 5p5p electrons are partially spin-polarized due to hybridization with the Cr 3d3d electrons, in analogy to the magnetic anisotropy of the Pt 5d5d electrons in L10L1_{0}-type FePt FePt_Solovyev ; FePt_Ikeda . In order to address this issue, further theoretical and experimental (XMCD at the Te absorption edges) studies would be necessary.

IV conclusion

We have performed XAS and XMCD measurements at the Cr L2,3L_{2,3} edge and subsequent cluster-model calculation on CGT to investigate the electronic structure which contributes to the ferromagnetism and magnetic anisotropy. The effective spin magnetic moment obtained using the sum rules is 2.2 μB\rm\mu_{B}, consistent with previous reports, while the orbital magnetic moment is nearly zero. Subtle changes of the XMCD spectra are observed by changing the magnetic field direction, and are attributed to the distortion of the CrTe6 octahedron from comparison with cluster-model calculation including the trigonal crystal field of the cluster elongated along the cc axis. The magnetic anisotropy energy calculated using the cluster model is too small and has an opposite sign compared to experiment. This implies that the magnetic anisotropy of CGT cannot be explained by the SIA of Cr described by the cluster model, and that one has to take into account interaction between Cr atoms through Te 5p5p ligands which have strong SOC.

Acknowledgements.
We would like to thank I. Solovyev, D. Khomskii, S. Streltsov, and K. Kugel for enlightening discussion, and M. Suzuki-Sakamaki and K. Amemiya for valuable technical support at KEK-PF. This work was supported by a Grant-in-Aid for Scientific Research from JSPS (grant Nos. 15H02109 and and 19K03741), by Center for Spintronics Research Network (CSRN), the University of Tokyo, under Spintronics Research Network of Japan (Spin-RNJ), and by the Shared Use Program of Japan Atomic Energy Agency (JAEA) Facilities (Proposal No. 2018A-E24) supported by JAEA Advanced Characterization Nanotechnology Platform as a program of ”Nanotechnology Platform” of MEXT (Proposal No. A-18-AE-0019). The synchrotron experiment at KEK-PF was done under the approval of the Program Advisory Committee (Proposal Nos. 2016S2-005), and the experiment at SPring-8 was done at the JAEA beam line BL23SU (Proposal Nos. 2018A3841). The work at Rutgers University was funded by the Gordon and Betty Moore Foundation’s EPiQS Initiative through Grant GBMF4413 to the Rutgers Center for Emergent Materials.

References

  • (1) K. Novoselov, A. Geim, S. Morozov, D. Jiang, Y. Zhang, S. Dubonos, I. Grigorieva, and A. Firsov, Science 306, 666 (2004).
  • (2) S. Z. Butler, S. M. Hollen, L. Cao, Y. Cui, J. A. Gupta, H. R. Gutierrez, T. F. Heinz, S. S. Hong, J. Huang, A. F. Ismach, E. Johnston-Halperin, M. Kuno, V. V. Plashnitsa, R. D. Robinson, R. S. Ruo, S. Salahuddin, J. Shan, L. Shi, M. G. Spencer, M. Terrones, W. Windl, and J. E. Goldberger, Progress, challenges, and opportunities in two-dimensional materials beyond graphene, ACS Nano 7, 2898 (2013).
  • (3) M. Chhowalla, H. S. Shin, G. Eda, L.-J. Li, K. P. Loh, and H. Zhang, The chemistry of two-dimensional layered transition metal dichalcogenide nanosheets, Nat. Chem. 5, 263 (2013)
  • (4) Q. H. Wang, K. Kalantar-Zadeh, A. Kis, J. N. Coleman, and M. S. Strano, Electronics and optoelectronics of two-dimensional transition metal dichalcogenides, Nat. Nanotechnol. 7, 699 (2012).
  • (5) N. D. Mermin and H. Wagner, Absence of ferromagnetism or antiferromagnetism in one- or two-dimensional isotropic Heisenberg models, Phys. Rev. Lett. 17, 1133 (1966).
  • (6) C. Gong and X. Zhang, Two-dimensional magnetic crystals and emergent heterostructure devices, Science 363, 706 (2019).
  • (7) V. Carteaux, D. Brunet, G. Ouvrard, and G. Andre, Crystallographic, magnetic and electronic structures of a new layered ferromagnetic compound Cr2Ge2Te6, J. Phys. Condens. Matter 7, 69 (1995).
  • (8) B. Siberchicot, S. Jobic, V. Carteaux, P. Gressier, and G. Ouvrard, Band structure calculations of ferromagnetic chromium tellurides CrSiTe3 and CrGeTe3, J. Phys. Chem. 100, 5863 (1996).
  • (9) H. Ji, R. A. Stokes, L. D. Alegria, E. C. Blomberg, M. A. Tanatar, A. Reijnders, L. M. Schoop, T. Liang, R. Prozorov, K. S. Burch, N. P. Ong, J. R. Petta, and R. J. Cava, A ferromagnetic insulating substrate for the epitaxial growth of topological insulators, J. Appl. Phys. 114, 114907 (2013).
  • (10) G. T. Lin, H. L. Zhuang, X. Luo, B. J. Liu, F. C. Chen, J. Yan, Y. Sun, J. Zhou, W. J. Lu, P. Tong, Z. G. Sheng, Z. Qu, W. H. Song, X. B. Zhu, and Y. P. Sun, Tricritical behavior of the two-dimensional intrinsically ferromagnetic semiconductor CrGeTe3, Phys. Rev. B 95, 245212 (2017).
  • (11) X. Zhang, Y. Zhao, Q. Song, S. Jia, J. Shi, and W. Han, Magnetic anisotropy of the single-crystalline ferromagnetic insulator Cr2Ge2Te6, Jpn. J. Appl. Phys. 55, 033001 (2016).
  • (12) C. Gong, L. Li, Z. Li, H. Ji, A. Stern, Y. Xia, T. Cao, W. Bao, C. Wang, Y. Wang, Z. Q. Qiu, R. J. Cava, S. G. Louie, J. Xia, and X. Zhang, Discovery of intrinsic ferromagnetism in two-dimensional van der Waals crystals, Nature 546, 265 (2017).
  • (13) Y. Sun, R. C. Xiao, G. T. Lin, R. R. Zhang, L. S. Ling, Z. W. Ma, X. Luo, W. J. Lu, Y. P. Sun, and Z. G. Sheng, Effects of hydrostatic pressure on spin-lattice coupling in two-dimensional ferromagnetic Cr2Ge2Te6, Appl. Phys. Lett. 112, 072409 (2018).
  • (14) M. Mogi, A. Tsukazaki, Y. Kaneko, R. Yoshimi, K. S. Takahashi, M. Kawasaki, and Y. Tokura, Ferromagnetic insulator Cr2Ge2Te6 thin films with perpendicular remanence, APL Mater. 6, 091104 (2018).
  • (15) M. Suzuki, B. Gao, K. Koshiishi, S. Nakata, K. Hagiwara, C. Lin, Y. X. Wan, H. Kumigashira, K. Ono, Sungmo Kang, Seungjin Kang, J. Yu, M. Kobayashi, S.-W. Cheong, and A. Fujimori, Coulomb-interaction effect on the two-dimensional electronic structure of the van derWaals ferromagnet Cr2Ge2Te6, Phys. Rev. B 99, 161401(R) (2019).
  • (16) J. Zhang, X. Cai, W. Xia, A. Liang, J. Huang, C. Wang, L. Yang, H. Yuan, Y. Chen, S. Zhang, Y. Guo, Z. Liu, and G. Li, Unveiling electronic correlation and the ferromagnetic superexchange mechanism in the van der Waals crystal CrSiTe3, Phys. Rev. Lett. 123, 047203 (2019).
  • (17) S. Kang, S. Kang, and J. Yu, Effect of coulomb interactions on the electronic and magnetic properties of two-dimensional CrSiTe3 and CrGeTe3 materials, J. Electronic Mater. 48, 1441 (2019).
  • (18) B. Huang, G. Clark, E. Navarro-Moratalla, D. R. Klein, R. Cheng, K. L. Seyler, D. Zhong, E. Schmidgall, M. A. McGuire, D. H. Cobden, W. Yao, D. Xiao, P. Jarillo-Herrero, and X. Xu, Layer-dependent ferromagnetism in a van der Waals crystal down to the monolayer limit, Nature 546, 270 (2017).
  • (19) T. Song, X. Cai, M. W.-Y. Tu, X. Zhang, B. Huang, N. P. Wilson, K. L. Seyler, L. Zhu, T. Taniguchi, K. Watanabe, M. A. McGuire, D. H. Cobden, D. Xiao, W. Yao, and X. Xu, Giant tunneling magnetoresistance in spin-filter van der Waals heterostructures, Science 360, 1214 (2018).
  • (20) J. L. Lado and J. Fernández-Rossier, On the origin of magnetic anisotropy in two dimensional CrI3, 2D Mater. 4, 035002 (2017).
  • (21) C. Xu, J. Feng, H. Xiang, and L. Bellaiche, Interplay between Kitaev interaction and single ion anisotropy in ferromagnetic CrI3 and CrGeTe3 monolayers, npj Comput. Mater. 4, 57 (2018). Toward intrinsic room-temperature ferromagnetism in two-dimensional semiconductors,
  • (22) D.-H. Kim, K. Kim, K.-T. Ko, J. H. Seo, J. S. Kim, T.-H. Jang, Y. Kim, J.-Y. Kim, S.-W. Cheong, and J. H. Park, Giant Magnetic anisotropy induced by ligand LSLS coupling in layered Cr compounds, Phys. Rev. Lett. 122, 207201 (2019).
  • (23) K. Wang, S. Nikolaev, W. Ren, and I. Solovyev, Giant contribution of the ligand states to the magnetic properties of the Cr2Ge2Te6 monolayer, Phys. Chem. Chem. Phys. 21, 9597 (2019).
  • (24) Y. Choi, P. J. Ryan, D. Haskel, J. L. McChesney, G. Fabbris, M. A. McGuire, and J.-W. Kim, Iodine orbital moment and chromium anisotropy contributions to CrI3 magnetism, Appl. Phys. Lett. 117, 022411 (2020).
  • (25) A. O. Fumega, S. Blanco-Canosa, H. Babu-Vasili, P. Gargiani, H. Li, J.-S. Zhou, F. Rivadulla, and V. Pardo, Electronic structure and magnetic exchange interactions of Cr-based van der Waals ferromagnets. A comparative study between CrBr3 and Cr2Ge2Te6, J. Mater. Chem. C 8, 13582, (2020).
  • (26) A. Frisk, L. B. Duffy, S. Zhang, G. van der Laan, and T. Hesjedal, Magnetic x-ray spectroscopy of two-dimensional CrI3 layers, Mater. Lett. 232 5, (2018).
  • (27) A. E. Bocquet, T. Mizokawa, T. Saitoh, H. Namatame, and A. Fujimori, Electronic structure of 3d3d-transition-metal compounds by analysis of the 2p2p core-level pbotoemission spectra, Phys. Rev. B 46, 3771 (1992).
  • (28) S. Seong, E. Lee, H. W. Kim, B. Min, S. Lee, J. Dho, Y. Kim, J.-Y. Kim, and J.-S. Kang, Experimental evidence for mixed-valent Cr ions in half-metallic CrO2: Temperature-dependent XMCD study, J. Magn. Magn. Mater. 452, 447 (2018).
  • (29) N. D. Telling, V. S. Coker, R. S. Cutting, G. van der Laan, C. I. Pearce, R. A. D. Pattrick, E. Arenholz, and J. R. Lloyd, Remediation of Cr(VI) by biogenic magnetic nanoparticles: An x-ray magnetic circular dichroism study, Appl. Phys. Lett. 95, 163701 (2009).
  • (30) P. Carra, B. T. Thole, M. Altarelli, and X. Wang, X-ray circular dichroism and local magnetic fields, Phys. Rev. Lett. 70, 694 (1993).
  • (31) B. T. Thole, P. Carra, and F. Sette, X-ray circular dichroism as a probe of orbital magnetization, Phys. Rev. Lett. 68, 1943 (1992).
  • (32) H. Ebert and G. Schutz, Spin-Orbit Influenced Spectroscopies of Magnetic Solids (Springer-Verlag, 1996).
  • (33) A. S. Schlachter and F. J. Wuilleumier, New Directions in Research with Third-Generation Soft X-ray Synchrotron Radiation Sources (NATO ASI Series 254, Kluwer, 1994).
  • (34) M. A. McGuire, H. Dixit, V. R. Cooper, and B. C. Sales, Coupling of crystal structure and magnetism in the layered, ferromagnetic insulator CrI3, Chem. Mater. 27, 612 (2015).
  • (35) I. V. Solovyev, P. H. Dederichs, and I. Mertig, Origin of orbital magnetization and magnetocrystalline anisotropy in TXTX ordered alloys (where TT = Fe, Co and XX = Pd, Pt) Phys. Rev. B 52, 13419 (1995).
  • (36) K. Ikeda, T. Seki, G. Shibata, T. Kadono, K. Ishigami, Y. Takahashi, M. Horio, S. Sakamoto, Y. Nonaka, M. Sakamaki, K. Amemiya, N. Kawamura, M. Suzuki, K. Takanashi, and A. Fujimori, Magnetic anisotropy of L10L1_{0}-ordered FePt thin films studied by Fe and Pt L2,3L_{2,3}-edges x-ray magnetic circular dichroism Appl. Phys. Lett. 111, 142402 (2017).
Refer to caption
Figure 1: Crystal structure of Cr2Ge2Te6 (CGT) (a) Schematic drawing of the crystal structure. (b) CrTe6 octahedron which is elongated along the cc-axis. (c) Crystal-field splitting of the Cr 3dd orbital in the distorted octahedron to trigonal symmetry.
Refer to caption
Figure 2: Cr L2,3L_{2,3} XAS and XMCD spectra of CGT at μ0H\mu_{0}H = 3 T and TT = 20 K. The red curve shows the average of XAS spectra obtained with left- and right-circularly polarized x-rays. The green curve shows the XMCD spectrum, which is the intensity difference between the above two XAS spectra. The blue dotted curve shows the integral of the XMCD spectrum. Magnetic field is applied perpendicular to the sample surface (HcH\parallel c-axis), that is, along the easy magnetization axis.
Refer to caption
Figure 3: Experimental and calculated Cr L2,3L_{2,3} XMCD spectra of CGT. The experimental spectrum is the XMCD spectrum taken with magnetic field perpendicular to the abab plane. The orange, green, light-blue, and blue curves show XMCD spectra calculated using the cluster model, assuming that the valence of Cr is 3+, 4+, 2+ high-spin (HS), and 2+ low-spin (LS) states, respectively.
Refer to caption
Figure 4: Experimental Cr L2,3L_{2,3} XAS and XMCD spectra of CGT obtained with different magnetic field directions. (a) Red and black curves show XAS spectra taken with the out-of-plain and in-plane magnetic fields, respectively. (b) XMCD spectra under out-of-plane and in-plane magnetic fields. Red and black spectra show the out-of-plane and in-plane XMCD spectra, respectively. The strength of the magnetic field was 1 T and the measurement temperature was 50 K.
Refer to caption
Figure 5: Experimental and calculated Cr L2,3L_{2,3} XMCD spectra of CGT for different magnetic field directions. The red and black curves show XMCD spectra taken under magnetic fields perpendicular (out-of-plane) and parallel (in-plane) to the surface, respectively. Blue curves show the difference between the XMCD spectra measured with different magnetic field directions. (a) Experimental XMCD spectra. (b), (c) XMCD spectra obtained by cluster-model calculation with and without the trigonal crystal field. DtDt denotes the trigonal-field splitting. Clear differences in the line shapes the different magnetic field directions are indicated by black arrows.