Abstract
Materials that are simultaneously ferromagnetic and ferroelectric – multiferroics – promise the control of disparate ferroic orders, leading to technological advances in microwave magnetoelectric applications and next generation of spintronics. Singlephase multiferroics are challenged by the opposite dorbital occupations imposed by the two ferroics, and heterogeneous nanocomposite multiferroics demand ingredients’ structural compatibility with the resultant multiferroicity exclusively at intermaterials boundaries. Here we propose the twodimensional heterostructure multiferroics by stacking up atomic layers of ferromagnetic Cr_{2}Ge_{2}Te_{6} and ferroelectric In_{2}Se_{3}, thereby leading to allatomic multiferroicity. Through firstprinciples density functional theory calculations, we find as In_{2}Se_{3} reverses its polarization, the magnetism of Cr_{2}Ge_{2}Te_{6} is switched, and correspondingly In_{2}Se_{3} becomes a switchable magnetic semiconductor due to proximity effect. This unprecedented multiferroic duality (i.e., switchable ferromagnet and switchable magnetic semiconductor) enables both layers for logic applications. Van der Waals heterostructure multiferroics open the door for exploring the lowdimensional magnetoelectric physics and spintronic applications based on artificial superlattices.
Introduction
Multiferroics, a class of functional materials that simultaneously possess more than one ferroic orders such as ferromagnetism and ferroelectricity, hold great promise in magnetoelectric applications due to the inherent coupling between ferroic orders^{1,2,3,4,5,6}, leading to technological advances in next generation of spintronics and microwave magnetoelectric devices. However, singlephase multiferroics are challenged by the different ferroics’ contradictory preference on the dorbital occupation of metal ions: ferroelectricity arising from offcenter cations requires empty dorbitals, whereas ferromagnetism usually results from partially filled dorbitals^{7}. Conventional perovskite multiferroics (chemical formula: ABO_{3}) have lonepairactive Asites which move to offcenters of centrosymmetric crystals for electric polarization, and Bsites with unpaired electrons for magnetic order. Because the ferroelectric and magnetic order in these materials are associated with different ions, the coupling between the ferroic orders are usually weak.
Heterogeneous multiferroics, synthesized composites of two mixed phases^{8}, have the coupling between ferroelectric and magnetic order exclusively at intermaterials boundaries, with magnetoelectric effects occasionally established via interfacial magnetoelastic effect. As an example, magnetic nanopillars could be embedded in ferroelectric matrix. However, these heterogeneous multiferroics stringently demand the constituent materials on their structural similarity, lattice match and chemical compatibility, and have weak magnetoelectric effects limited by the interface/bulk ratios.
Van der Waals (vdW) crystals emerged as ideal material systems with unprecedented freedom for heterostructure construction^{9}. Recent experimental advance discovered ferromagnetism^{10,11,12} and ferroelectricity^{13} in different twodimensional vdW crystals separately. It remains a paramount challenge to realize multiple ferroic orders in a singlephase 2D material simultaneously^{14,15,16,17}, as each order encounters its own challenge (e.g., ferromagnetism in 2D systems suffers from enhanced thermal fluctuations, whereas ferroelectricity the depolarization field). Constructing heterostructures of 2D magnets and 2D ferroelectrics potentially provides a generally applicable route to create 2D multiferroics. However, the fundamental question remains regarding whether the interlayer magnetoelectric coupling can be established, given the presence of the interlayer vdW spacing. If realized, layered heterostructure multiferroics would provide completely new platforms with all atoms participating in the interferroics coupling, and largely reshape the landscape of multiferroics based on vdW superlattices.
Through firstprinciples density functional theory (DFT) calculations based on a bilayer heterostructure of ferromagnetic Cr_{2}Ge_{2}Te_{6} and ferroelectric In_{2}Se_{3} monolayers^{18,19,20,21}, we discovered a strong interlayer magnetoelectric effect: the reversal electric polarizations in In_{2}Se_{3} switches the magnetocrystalline anisotropy of Cr_{2}Ge_{2}Te_{6} between outofplane and inplane orientations. For a 2D ferromagnet, such a change in magnetic anisotropy corresponds to a switching on/off of the ferromagnetic order at finite temperatures, for easyaxis anisotropy opens spin wave excitation gap and thus suppresses the thermal fluctuations, but easyplane anisotropy does not^{10,22,23}. The switching of ferromagnetic order by electric polarization promises a novel design of magnetic memory. Detailed analysis unraveled the interfacial hybridization as the cause of interlayer magnetoelectric coupling. Furthermore, In_{2}Se_{3} becomes magnetized due to the proximity to Cr_{2}Ge_{2}Te_{6}, making In_{2}Se_{3} a singlephase multiferroics (i.e., ferromagnetic and ferroelectric orders coexist in In_{2}Se_{3}), although apparently the magnetization of In_{2}Se_{3} requires the presence of the adjacent Cr_{2}Ge_{2}Te_{6}. Such multiferroicity duality—that is, the interlayer multiferroicity and the In_{2}Se_{3} intralayer multiferroicity—provides unique solidstate system in which ferroelectric and ferromagnetic orders interplay inherently. This unusual multiferroicity duality in vdW heterostructures may open avenues for developing new concepts of magnetoelectric devices: using single knob (the orientation of electric polarization in In_{2}Se_{3}) to control the magnetic order in both In_{2}Se_{3} and Cr_{2}Ge_{2}Te_{6}. We envision the multiferroicity duality potentially enriches the freedom of layerresolved data storage and that of information processing due to the diverse magnetoelectric and magnetooptic properties of constituent layers.
Results
Material model and computational details
In this work, the lattice constant of Cr_{2}Ge_{2}Te_{6} adopted the experimental value 6.83 Å and was fixed in heterostructures for the sake of minimizing artifact effects, considering the magnetic properties of 2D Cr_{2}Ge_{2}Te_{6} are sensitive to structure parameters. It has been reported that a monolayer In_{2}Se_{3} of either zincblende or wurtzite stacking is unstable with a tendency of the lateral displacement of the top Se layer, leading to the energetically degenerate ferroelectric monolayer^{18}. Although the one relaxed from the zincblende stacking is chosen in this study, it will be also applicable to the other derived from the wurtzite, because the main mechanism to be shown is determined by the interfacial monolayers and thus independent of the detailed stacking type of the multilayer In_{2}Se_{3}. The optimized lattice constant of 1 × 1In_{2}Se_{3} (4.106 Å), is strained by −4.0% to match In_{2}Se_{3}\(\sqrt 3 \, \times \,\sqrt 3\) to Cr_{2}Ge_{2}Te_{6}1 × 1 as shown by Fig. 1a. In heterostructure, the relative spacing and registry between Cr_{2}Ge_{2}Te_{6} and In_{2}Se_{3} are adjusted to find the energy minimum configuration. The reversal of the electric polarization of the isolated In_{2}Se_{3} monolayer can be achieved via lateral displacement of the middle most Se layer, with an energy barrier as small as 0.04 eV per formula unit estimated by nudged elastic band calculation^{24}. In heterostructure, due to the large vdW spacing, the presence of Cr_{2}Ge_{2}Te_{6} does not noticeably affect the energy barrier of the electricpolarization reversal process of In_{2}Se_{3}. The total energy of the heterostructure is lowest (highest) where the interfacial Te atoms sit at the hollow (top) site of In_{2}Se_{3}, with their relative energy difference amounts to 0.31 and 0.35 eV/u.c. for upward (Fig. 1b) and downward (Fig. 1c) polarizations, respectively. The equilibrium interlayer distance between Cr_{2}Ge_{2}Te_{6} and In_{2}Se_{3} at the hollow configuration is 3.20 and 3.14 Å for up and down polarizations of In_{2}Se_{3}, respectively. The total energy of the down polarization (Fig. 1c) is lower by 0.07 eV/u.c. than that of the up polarization (Fig. 1b), due to the stronger interfacial coupling between downpolarized In_{2}Se_{3} and Cr_{2}Ge_{2}Te_{6}.
In order to reproduce the experimental magnetic properties of bulk Cr_{2}Ge_{2}Te_{6}, we used small onsite Hubbard U value 0.5 eV and Hund’s coupling J value 0.0 eV for Cr d orbital in DFT calculations (see ref. ^{10} for the choice of U = 0.5 eV, J = 0.0 eV). This small onsite Columbic interaction is consistent with the fact that Cr_{2}Ge_{2}Te_{6} is a small band gap material with less localization than Croxides. The ferromagnetic ground state is confirmed with the Cr spin magnetic moment ~3.0 μ_{B}. With the spin–orbit coupling (SOC) included, the magnetocrystalline anisotropy energy (MAE) is calculated and defined as \(E_{[100]}  E_{[001]}\), where the former and latter correspond to the total energy with the Cr spins directed inplane and outofplane, respectively. Due to the threefold rational symmetry of Cr_{2}Te_{2}Te_{6}, there is not much magnetic anisotropy within the basal plane. We checked the convergence of MAE carefully, where a large value of Kmesh (12 × 12 × 1) was enough to ensure the error <10 μeV/Cr. For the isolated monolayer Cr_{2}Ge_{2}Te_{6}, our calculated MAE is −70 μeV/Cr, favoring the inplane direction. In the heterostructures with up and downpolarized In_{2}Se_{3}, the calculated Cr MAE is −95 and 75 μeV, respectively, whose energetically favorable spin orientations are indicated by S_{Cr} in Fig. 1b, c. By modulating the polarization of the adjacent ferroelectric layer, the switching of the magnetization orientation is realized. This has significant application implications: For a 2D magnetic system with easyplane anisotropy (X–Y model), the finite temperature ferromagnetic order is inhibited, whereas for easyaxis anisotropy (Ising model), the magnetic order can be sustained at finite temperatures. Therefore, in such a heterostructure, multiferroic effect offers a potential route to switch the ferromagnetism for logic devices.
Mechanism for interfacial multiferroicity
The mechanism for electricpolarization dependent MAE is discussed in details. The calculated Cr orbital moment is small (<L_{x}> = 0.04 μ_{B}, <L_{z}> = 0.01 μ_{B} for inplane, outofplane spin directions), which is less likely the origin of MAE. The plausible mechanism is related to the detailed feature of spinresolved orbitaldecomposed band structure^{25}. Starting from the collinear spin band structures, we analyzed the energy correction by the perturbation theory about λL∙S where λ is the radial part of Cr SOC. The orbital moment quenched by the crystal field results in the vanishing first order correction. Assuming the negligible change of the electron correlation energy between [100] and [001] spin directions, one can write the second order contribution to MAE as follows^{25}:
Here the first and second summations correspond to the spinconserving Δs_{z} = 0 and spinflipping Δs_{z} = 1 transitions, and \(v,\sigma\rangle\) and \(c,\sigma\rangle\) are valence and conduction band states with spin σ, respectively, whose energy eigenvalues are \({\it{\epsilon }}_{c,\sigma }\) and \({\it{\epsilon }}_{v,\sigma }\). The angular momentum matrix elements of Lz and Lx correspond to transitions with Δm_{z} = 0 and Δm_{z} = 1, respectively, for Cr dorbitals. Therefore, for spinconserving transition, SOC elements between occupied and unoccupied states with the same (different) magnetic quantum number through the \(L_z\left( {L_x} \right)\) operator contributes to positive (negative) MAE. For spinflipping transition, the contribution to MAE is reversed^{26}.
Figure 2a, b shows the spinresolved orbitaldecomposed band structure of heterostructures for up and downpolarized In_{2}Se_{3}, respectively, where the contribution of Cr dorbitals is indicated by the circles for m_{z} = 0 (z^{2}), 1 (xz and yz), or 2 (x^{2}−y^{2} and xy). As shown by the arrows in Fig. 2a for the heterostructure with uppolarized In_{2}Se_{3}, our calculated negative MAE mainly originates from the spinconserving transition from m_{z} = 1 to m_{z} = 0 or 2, i.e., Δs_{z} = 0 and Δm_{z} = 1 related to the second term of the first sum in Eq. (1). The mechanism is further confirmed by our results that the Cr MAE changes from about −100 to 200 μeV by intentionally increasing the U value from 0.5 to 2.0 eV: the increased U lowers the energy level of the majority spin in valence bands of m_{z} = 0 or 2 (Supplementary Fig. 3), and thus the transition energy gap of Δs_{z} = 0 and Δm_{z} = 1 is increased and the associated contribution is weakened, leading to the positive MAE.
However, for the heterostructure with downpolarized In_{2}Se_{3}, as shown in Fig. 2b, the conduction band minimum of Cr d m_{z} = 1 shows a large gap (~0.2 eV) near 0.5 eV above Fermi level, which is caused by hybridization with the In_{2}Se_{3} conduction band minimum. Such hybridization results in a significant depletion of m_{z} = 1 majority spin DOS (Supplementary Fig. 2a). Hence, the negative contribution to MAE found for the case of uppolarized In_{2}Se_{3} is suppressed. Meanwhile the minority spin DOS remains almost unchanged for m_{z} = 1 (Supplementary Fig. 2a), leading to positive MAE via Δs_{z} = 1 and Δm_{z} = 1, as illustrated by the arrows in Fig. 2b. Considering the interfacial hybridization depends on the band alignment of In_{2}Se_{3} and Cr_{2}Ge_{2}Te_{6}, we employed the HeydScuseriaErnzerhof exchangecorrelation functional (HSE06) to recalculate the band properties of the heterostructures. As expected, the calculated band gaps widen compared with the GGAPBE results, but the key features while In_{2}Se_{3} reverses its electric orientation from up to down keep the same: the conduction band of In_{2}Se_{3} moves down to hybridize with the conduction band of Cr_{2}Ge_{2}Te_{6}, as clearly seen in Supplementary Fig. 4.
The sign change of Cr_{2}Ge_{2}Te_{6}’s MAE upon the electricpolarization reversal of In_{2}Se_{3} from up to down arises from the increased coupling, which causes the overall shift down of In_{2}Se_{3} bands and its enhanced hybridization with Cr_{2}Ge_{2}Te_{6}. This suggests that the positive MAE for the down polarization would be enhanced by a reduced vdW spacing. To confirm this scenario, we did interlayer spacing dependent MAE calculations for hollow configuration as shown in Fig. 3. As the interlayer distance decreases, MAE increases gradually with a slight fluctuation. The fluctuation originates from the detailed variation of energy levels in the spinpolarized band structures. From the same calculation conducted for the top configuration, it exhibits a stronger fluctuation with the interlayer spacing, due to the larger degree of interfacial orbital overlap in top configuration. The same trend of the two curves in Fig. 3 confirms that the increased interlayer hybridization tends to switch the magnetocrystalline anisotropy from easyplane to easyaxis. Detailed spinresolved orbitaldecomposed analysis in Supplementary Fig. 5 shows that the decreased spinflipping energy gap near K with decreased interlayer distance contributes to the positive MAE.
Magnetized In_{2}Se_{3} in proximity to Cr_{2}Ge_{2}Te_{6}
Remarkably, the exchange splitting in Cr dband magnetizes In_{2}Se_{3} by the proximity effect. As shown in Fig. 2, the highest valence band has a significant exchange splitting, where the majorityspin band is closer to the Fermi level near Γ. Those states are mainly caused by the surface Te atoms in Cr_{2}Ge_{2}Te_{6}, which means the interfacial Te atoms have the electron spin antiparallel to that of Cr d electrons. Our calculated spin moment per Te atom is −0.11 μ_{B}for In_{2}Se_{3} of either electric polarizations. Also, the surface In and Se atoms has nonzero spin moments parallel to Te spins induced by the proximity. The spinresolved DOS of interfacial In and Se, shown in Fig. 4a, confirmed the magnetized In_{2}Se_{3}. It is practically important to note the calculated magnetized In_{2}Se_{3} here is a groundstate property. At finite temperatures, easyplane magnetization of 2D In_{2}Se_{3} is susceptible to thermal fluctuations and longrange order does not exist, but easyaxis magnetization of 2D In_{2}Se_{3} could sustain the spin polarization at certain finite temperatures. Hence, a switching of 2D magnetic ferroelectric In_{2}Se_{3} could be realized in this heterostructure multiferroics, leading to a design of spin fieldeffect transistor^{27,28}.
The induced spin moment of surface Se is attributed to the exchange coupling J ~ t^{2}/U between Te p and Se p orbitals, with t the hopping constant and U the intraorbital Coulomb repulsion. In the limit of zero t or infinite U, the system favors the triplet state similar to the atomic Hund coupling, which is the case for Te p and Se p. For a given value of U, the t varies exponentially with the distance. Consistently, our calculations show the exponentially increasing Se spin moment with decreased interfacial distance, as shown in Fig. 4b. The correlation effect should depend on the specific nonlocal correlation functional. For different vdW functionals, the induced spin moments remain at nearly the same magnitude (Supplementary Fig. 6).
Discussion on practical experimental factors
It is of experimental guidance to remark on the possible effects of real material environments. Calculation and analysis in this work are based on the heterostructure of a bilayer system floating in vacuum. In the experimental realization, the initial anisotropy of the magnetic layer Cr_{2}Ge_{2}Te_{6} could be affected by a few factors, including contacting the materials of large dielectric constants^{22,29} or large SOC strengths^{30}, unintentional doping^{31,32,33} caused by chemicals in device fabrication process, and small amount of artificial strain induced in heterostructure preparation. These factors may affect the resultant magnetoelectric effect quantitatively, as reflected from the calculated MAE of the heterostructures based on the arbitrary sets of U and J values (see Supplementary Fig. 7): the increased U value enhances the outofplane anisotropy, while the increased J value enhances the inplane anisotropy; for any tested set of U and J values, the outofplane anisotropy is always enhanced by ~ 0.15 meV/Cr when the In_{2}Se_{3} dipole is inverted from up to down. Therefore, even if our adopted values of U and J (U = 0.5 eV, J = 0.0 eV) slightly deviate from the exact description of the real heterostructure samples because of the aforementioned complex experimental conditions, the reversal of the In_{2}Se_{3} polarization from up to down always strengthens the 2D ferromagnetic order in Cr_{2}Ge_{2}Te_{6}. This leads to a general implication: in practice, one can always set a temperature so that 2D ferromagnetism could be found in P_{dn}In_{2}Se_{3}Cr_{2}Ge_{2}Te_{6} but disappear from P_{up}In_{2}Se_{3}Cr_{2}Ge_{2}Te_{6}, leading to the practical switching experiments at finite temperatures. Therefore, the magnetoelectric effect presented here, based on the modification of MAE by the intricate interface hybridization which further relates to the electric polarization of the 2D ferroelectrics, is an intrinsic interfacial phenomenon.
Summary
We employed firstprinciples DFT calculations on a vdW heterostructure consisting of ferromagnetic Cr_{2}Ge_{2}Te_{6} and ferroelectric In_{2}Se_{3} monolayers. By reversing the electric polarization of In_{2}Se_{3}, the calculated magnetocrystalline anisotropy of Cr_{2}Ge_{2}Te_{6} changes between easyaxis and easyplane (i.e., switching on/off of the ferromagnetic order), which promises a novel design of magnetic memory. Furthermore, In_{2}Se_{3} becomes magnetic ferroelectrics, with switchable spin polarizations according to its own electric polarization. The 2D multiferroic heterostructures would tremendously enlarge the landscape of multiferroics by artificially assembling 2D layers and provide new material platforms for a plethora of emergent interfacial phenomena.
Methods
The DFT method and parameters
All the calculations were performed by the DFT method implemented in Vienna ab initio Simulation Package (VASP)^{34}, with the PerdewBurkeErnzerhof (PBE) functional^{35} in the scheme of generalized gradient approximation (GGA). The main data was calculated by GGA + U based on the Liechtenstein approach with U = 0.5 eV and J = 0.0 eV. The van der Waals interatomic forces are described by the D2 Grimme method^{36}. The Kmesh of 6 × 6 × 1 and the energy cutoff of 300 eV are used for the structural optimization. The dipole correction is included to exclude spurious dipole–dipole interaction between periodic images.
Data availability
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
 1.
Wang, J. et al. Epitaxial BiFeO_{3} multiferroic thin film heterostructures. Science 299, 1719 (2003).
 2.
Kimura, T. et al. Magnetic control of ferroelectric polarization. Nature 426, 55 (2003).
 3.
Ramesh, R. & Spaldin, N. A. Multiferroics: progress and prospects in thin films. Nat. Mater. 6, 21 (2007).
 4.
Fiebig, M. et al. Revival of the magnetoelectric effect. J. Phys. D. 38, R123 (2005).
 5.
Fiebig, M. et al. The evolution of multiferroics. Nat. Rev. Mater. 1, 16046 (2016).
 6.
Cheong, S.W. & Mostovoy, M. Multiferroics: a magnetic twist for ferroelectricity. Nat. Mater. 6, 13 (2007).
 7.
Hill, N. A. Why are there so few magnetic ferroelectrics? J. Phys. Chem. B 104, 6694 (2000).
 8.
Zheng, H. et al. Multiferroic BaTiO_{3}CoFe_{2}O_{4} nanostructures. Science 303, 661 (2004).
 9.
Geim, A. K. & Grigorieva, I. V. Van der Waals heterostructures. Nature 499, 419 (2013).
 10.
Gong, C. et al. Discovery of intrinsic ferromagnetism in twodimensional van der Waals crystals. Nature 546, 265–269 (2017).
 11.
Huang, B. et al. Layerdependent ferromagnetism in a van der Waals crystal down to the monolayer limit. Nature 546, 270–273 (2017).
 12.
Gong, C. & Zhang, X. Twodimensional magnetic crystals and emergent heterostructure devices. Science 363, eaav4450 (2019).
 13.
Liu, F. et al. Roomtemperature ferroelectricity in CuInP_{2}S_{6} ultrathin flakes. Nat. Commun. 7, 12357 (2016).
 14.
Seixas, L., Rodin, A. S., Carvalho, A. & Castro Neto, A. H. Multiferroic twodimensional materials. Phys. Rev. Lett. 116, 206803 (2016).
 15.
Wang, H. & Qian, X. Twodimensional multiferroics in monolayer group IV monochalcogenides. 2D Mater. 4, 015042 (2017).
 16.
Luo, W., Xu, K. & Xiang, H. Twodimensional hyperferroelectric metals: a different route to ferromagneticferroelectric multiferroics. Phys. Rev. B 96, 235415 (2017).
 17.
Qi, J., Wang, H., Chen, X. & Qian, X. Twodimensional multiferroic semiconductors with coexisting ferroelectricity and ferromagnetism. Appl. Phys. Lett. 113, 043102 (2018).
 18.
Ding, W. et al. Prediction of intrinsic twodimensional ferroelectrics in In_{2}Se_{3} and other III_{2}VI_{3} van der Waals materials. Nat. Commun. 8, 14956 (2017).
 19.
Zhou, Y. et al. Outofplane piezoelectricity and ferroelectricity in αlayered In_{2}Se_{3} nanoflakes. Nano Lett. 17, 5508–5513 (2017).
 20.
Cui, C. et al. Intercorrelated inplane and outofplane ferroelectricity in ultrathin twodimensional layered semiconductor In_{2}Se_{3}. Nano Lett. 18, 1253–1258 (2018).
 21.
Xiao, J. et al. Intrinsic twodimensional ferroelectricity with dipole locking. Phys. Rev. Lett. 120, 227601 (2018).
 22.
Mermin, N. D. & Wagner, H. Absence of ferromagnetism or antiferromagnetism in one or twodimensional isotropic Heisenberg models. Phys. Rev. Lett. 17, 1133 (1966).
 23.
Bruno, P. Magnetization and Curie temperature of ferromagnetic ultrathin films: the influence of magnetic anisotropy and dipolar interactions. Mater. Res. Symp. Proc. 231, 299–310 (1991).
 24.
Henkelman, G. A climbing image nudged elastic band method for finding saddle points and minimum energy paths. J. Chem. Phys. 113, 9901–9904 (2000).
 25.
Wang, D. et al. Firstprinciples theory of surface magnetocrystalline anisotropy and the diatomicpair model. Phys. Rev. B 47, 14932 (1993).
 26.
Sui, X. et al. Voltagecontrollable colossal magnetocrystalline anisotropy in singlelayer transition metal dichalcogenides. Phys. Rev. B 96, 041410 (2017).
 27.
Datta, S. & Das, B. Electronic analog of the electrooptic modulator. Appl. Phys. Lett. 56, 665–667 (1990).
 28.
Gong, S.J. et al. Electrically induced 2D halfmetallic antiferromagnets and spin field effect transistors. Proc. Natl Acad. Sci. USA 115, 8511–8516 (2018).
 29.
Jiang, S., Shan, J. & Mak, K. F. Electricfield switching of twodimensional van der Waals magnets. Nat. Mater. 17, 406–410 (2018).
 30.
Avsar, A. et al. Spinorbit proximity effect in graphene. Nat. Commun. 5, 4875 (2014).
 31.
Huang, B. et al. Electrical control of 2D magnetism in bilayer CrI_{3}. Nat. Nanotechnol. 13, 544–548 (2018).
 32.
Jiang, S., Li, L., Wang, Z., Mak, K. F. & Shan, J. Controlling magnetism in 2D CrI_{3} by electrostatic doping. Nat. Nanotechnol. 13, 549–553 (2018).
 33.
Wang, Z. et al. Electricfield control of magnetism in a fewlayered van der Waals ferromagnetic semiconductor. Nat. Nanotechnol. 13, 554–559 (2018).
 34.
Kresse, G. & Furthmuller, J. Efficiency of abinitio total energy calculations for metals and semiconductors using a planewave basis set. Comput. Mater. Sci. 6, 15–50 (1996).
 35.
Perdew, J. P., Burke, K. & Ernzerhof, M. Generalized gradient approximation made simple. Phys. Rev. Lett. 77, 3865 (1996).
 36.
Grimme, S. Semiempirical GGAtype density functional constructed with a longrange dispersion correction. J. Comp. Chem. 27, 1787 (2006).
Acknowledgements
C.G., Y.W., and X.Z. acknowledge the support from the U.S. Department of Energy, Office of Science, Basic Energy Sciences, Materials Sciences and Engineering Division under contract no. DEAC0205CH11231 within the van der Waals Heterostructures program (KCWF16) for the conceptual development and preliminary calculations of 2D heterostructure multiferroics. The support from the National Science Foundation (NSF) under Grant 1753380 for the calculation and analysis of 2D magnets and the King Abdulah University of Science and Technology (KAUST) Office of Sponsored Research (OSR) under Award OSR2016CRG5 2996 for the calculation and analysis of 2D ferroelectrics was also acknowledged. G.L. acknowledges the support by the National Research Foundation of Korea (Basic Science Research Program: 2018R1D1A1B07045983) for the systematic computational studies of 2D heterostructure multiferroics. Computation was supported by KISTI (KSC2018CRE0048).
Author information
Affiliations
Contributions
C.G. and X.Z. conceived the project. G.L. and E.M.K. performed the calculations with the close discussions with C.G. C.G., G.L., and X.Z. did data analysis and wrote the paper with the assistance from Y.W.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Peer review information: Nature Communications thanks the anonymous reviewers for their contribution to the peer review of this work
Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Gong, C., Kim, E., Wang, Y. et al. Multiferroicity in atomic van der Waals heterostructures. Nat Commun 10, 2657 (2019). https://doi.org/10.1038/s41467019106930
Received:
Accepted:
Published:
Further reading

Spinconstrained optoelectronic functionality in twodimensional ferromagnetic semiconductor heterojunctions
Materials Horizons (2021)

Understanding Microscopic Operating Mechanisms of a van der Waals Planar Ferroelectric Memristor
Advanced Functional Materials (2021)

Electricpolarizationdriven magnetic phase transition in a ferroelectric–ferromagnetic heterostructure
Applied Physics Letters (2021)

Manipulating terahertz responses of GdFeO3 ceramics by thickening and thermal methods
Ceramics International (2021)

Twodimensional topological superconductivity candidate in a van der Waals layered material
Physical Review B (2021)
Comments
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.