This is the accepted version of the journal article: Sierra, Juan F.; Fabian, Jaroslav; Kawakami, Roland; [et al.]. «Van der Waals heterostructures for spintronics and opto-spintronics». Nature Nanotechnology, Vol. 16, issue 8 (August 2021), p. 856-868. DOI 10.1038/s41565-021-00936-x This version is available at https://ddd.uab.cat/record/269763 under the terms of the license Van der Waals heterostructures for spintronics and opto- spintronics Juan F. Sierra1*, Jaroslav Fabian2, Roland K. Kawakami3, Stephan Roche1,4 and Sergio O. Valenzuela1,4* 1 Catalan Institute of Nanoscience and Nanotechnology (ICN2), CSIC and The Barcelona Institute of Science and Technology (BIST), Campus UAB, Bellaterra, 08193 Barcelona, Spain 2 Institute for Theoretical Physics, University of Regensburg, 93040 Regensburg, Germany 3 Department of Physics, The Ohio State University, Columbus, OH 43210, USA 4 Institució Catalana de Recerca i Estudis Avançats (ICREA), 08010 Barcelona, Spain *e:mail: juan.sierra@icn2.cat; SOV@icrea.cat The large variety of 2D materials and their co-integration in van der Waals (vdW) heterostructures enable innovative device engineering. In addition, their atomically-thin nature promotes the design of artificial materials by proximity effects that originate from short-range interactions. Such a designer approach is particularly compelling for spintronics, which typically harnesses functionalities from thin layers of magnetic and non-magnetic materials and the interfaces between them. Here, we overview recent progress on 2D spintronics and opto-spintronics using vdW heterostructures. After an introduction to the forefront on spin transport research, we highlight the unique spin-related phenomena arising from spin-orbit and magnetic proximity effects. We further describe the ability to create multi-functional hybrid heterostructures based on vdW materials, combining spin, valley and excitonic degrees of freedom. We end with an outlook on perspectives and challenges for the design and production of ultra-compact all-2D spin devices and their potential applications in conventional and quantum technologies. 1 In the past few years, van der Waals (vdW) heterostructures1,2 comprising a variety of 2D layered materials have emerged as potential building blocks for future ultrafast and low-power electronic and spintronic devices. Graphene is an ideal spin channel owing to its spin diffusion length reaching several micrometres at room temperature, gate-tunable carrier concentration and extremely high carrier mobility3–5. Semiconducting TMDCs6 such as MX2 (M= W, Mo; X = S, Se, Te) and topological insulators (TIs) such as Bi2Te3 possess strong spin-orbit coupling (SOC), which allows for the electrical generation and manipulation of spins. Semiconducting TMDCs further possess a strong spin-photon coupling that enables optical spin injection, while 2D magnets7 bring capabilities for spin filtering and non-volatile data storage. Novel functionalities arise due to the atomically-thin nature of 2D materials, which facilitates much stronger electrostatic gating effects than with conventional materials to achieve, for instance, voltage-controlled magnetism. Furthermore, the integration of graphene, TMDC, TIs and 2D magnets into vdW heterostructures not only combines the respective material functionalities but also imprints properties through proximity interactions across interfaces8, enabling the design of artificial structures with unique spin characteristics. Such properties provide opportunities for memory applications, spin interconnects, spin-transistors, microwave nano-oscillators9,10, low-power reconfigurable logics and flexible or wearable spintronic platforms11,12 (see Box 1). This Review presents the state of the art and future prospects for vdW heterostructures in spintronics and optospintronics, with a special focus on magnetic and spin-orbit proximity effects and the emerging phenomena deriving from them. Covering recent experimental and theoretical developments, the Review is divided in four main sections. The first section briefly surveys the recent progress in spin injection and detection, including the integration of opto- electronic elements, and then outlines the contemporary understanding of spin dynamics in 2D materials. This description is complemented by an overview of materials that can be utilized to 2 enhance the spin properties or further create multifunctional 2D spintronic devices (Box 1). This is followed by the core of the Review on vdW heterostructures and proximity effects. First, the second section focuses on proximity-induced SOC, which is central in modern spintronics as it can enhance the magnetic properties of 2D magnets as well as provide spin filtering, spin manipulation and efficient charge to spin interconversion (CSI) functionalities. Next, the third section addresses magnetic proximity effects, which can be harnessed in memory elements, reconfigurable spin-logic circuits and novel spin-valleytronics applications. Besides the vast catalogue of material combinations, vdW heterostructures establish new concepts based on twist-angle and stacking control between crystallographic lattices that can strongly dictate the nature and strength of proximity phenomena. Finally, the fourth section discusses potential applications and future research directions and perspectives. Spin dynamics in 2D materials Recent advances on spin injection and detection. Spin dynamics is typically investigated using lateral devices in a non-local electrical configuration3,5 or, alternatively, using spectroscopic methods in optically active materials. Lateral spin devices rely on efficient spin injection and detection with tunnel barriers playing a crucial role in alleviating the conductance mismatch problem13, which limits the effective spin polarization 𝑃s of the injector and detector contacts (Fig. 1a). Early studies with graphene as a spin channel used MgO, Al2O3, TiO2, and amorphous carbon barriers14–16 but the emergence of alternative insulators could improve the device performance in various aspects. SrO barriers, grown by evaporation of Sr in the presence of molecular oxygen, lead to robust operation with high bias (~2 V) to achieve large spin accumulation17. Barriers composed of 2D materials produce high 𝑃s . Fluorinated graphene, obtained by exposure to XeF2 gas, yields 𝑃s > 40%18 while hexagonal boron nitride (hBN), using stacking and transfer methods, efficiently injects spins into graphene19,20 and black phosphorous21. With graphene, hBN-based injectors display a differential spin polarization that 3 varies with applied voltage bias, reaching absolute values above 100 % and even changing sign20. Beyond planar contacts, one-dimensional edge contacts have been utilized for spin injection into hBN-encapsulated graphene22. Box 1| Designer van der Waals (vdW) heterostructures for spintronics. Material functionalities. Key practical elements in spintronics are the injection, Optical spin injection in graphene, using transition metal transport or communication, manipulation, and detection of dichalcogenides32 (TMDCs), has also been demonstrated33,34. spin information10,23–25. Each of these aspects, represented in MTJs, combining 2DFMs and insulating hBN barriers, or the figure below, can benefit from the unique properties of tunnel filters comprising insulating (anti)ferromagnetic 2D materials. While spin communication requires low spin- materials, could be used in ultra-compact low-power memory orbit materials, spin injection, detection and manipulation elements10,24. can be achieved by creating multifunctional vdW Spin manipulation and control as well as spin injection and heterostructures, which include materials with magnetic detection can be engineered via proximity-induced SOC and exchange interaction (MEI) and/or large spin-orbit coupling proximity-induced MEI by adjacent (insulating) large-SOC (SOC). materials (e.g, TMDCs and TIs) and (anti)FMs, respectively. Spin communication over practical distances has already Magnetization switching and precession in memory elements been demonstrated in graphene and black phosphorous at could be achieved with SOTs10, using CSI in materials with room temperature21,26,27, being enhanced by vdW large-SOC, either of intrinsic origin or acquired by proximity encapsulation with hexagonal boron nitride (hBN)28. Spin effects. Novel spin-transistor configurations and spin logics and multiplexer devices have been proposed, using polarizers, which take advantage of proximitized spin-valley spin accumulation29, gate dependence of spin lifetimes coupling in graphene, have been experimentally and/or drift currents30,31. demonstrated35–37. The atomically thin nature of 2D (anti)FMs further enable the manipulation of their magnetic state using Spin injection and detection use several approaches10,24. electric fields38, while twist-angle and stacking control They involve insulating 2D ferromagnets (2DFMs), metallic between crystallographic lattices add yet other versatile 2DFMs in combination with insulating tunnel barriers, or knobs to engineer the nature and strength of proximity topological insulators (TIs) and other large-SOC materials effects. through charge-to-spin interconversion (CSI) mechanisms, such as the spin Hall and spin galvanic effects (SHE, SGE). Optical selection rules together with Faraday and Kerr microscopies, broadly used in semiconducting materials23, are valuable tools to investigate spin-valley dynamics and spin 4 coherence in TMDCs39–41. Typically, a pump laser illuminates individual crystals with right- or left-circularly polarized light at specific wavelengths to target an exciton transition, generating spin-polarized electrons and holes in the K or K′ valley, respectively (Fig. 1b). The spin polarization can be then detected by means of the optical Kerr rotation of a linearly polarized laser probe. Fig. 1 | Spin injection and spin transport. a, Illustration of a lateral graphene-hBN spin device with current and voltage leads in the non-local configuration scheme and control by electric and thermal drifts. Blue (inner) contacts are ferromagnetic, while yellow (outer) contacts are preferably non-magnetic. b, Schematic band structure and optically-excited transitions of a monolayer TMDC at the K and K’ valleys with left (𝜎 − ) and right (𝜎 + ) circularly polarized light. c, Optical spin injection from monolayer TMDC into graphene and subsequent lateral spin transport. The spin current is remotely detected using the spin-sensitive (ferromagnetic) contact F. The injected spins are initially oriented out-of-plane and rotate under the influence of an applied magnetic field 𝐵𝑦 as they diffuse towards F, eventually becoming aligned with its magnetization. d, Non-local voltage VNL measured at probe F in c as a function of By; VNL exhibits an antisymmetric spin precession signal whose sign changes by flipping the magnetization M of the detector F. e, Spin signal RNL= VNL/I as a function of the drift current Idc applied along the spin channel as shown in a. The modulation is driven by a change in the effective 𝜆s , which increases or decreases when the current flows in favour of or 𝑆 against the spin current, respectively. f, Spin signal change Δ𝑅𝑁𝐿 as a function of the graphene carrier density n induced by the presence of a thermal gradient in the spin channel as shown in a. Panel d and e adapted from Refs. 33 and 42 respectively, American Chemical Society. Panel f reproduced from Ref 43, SNL. When embedded in vdW heterostructures, TMDCs enable a new platform for optospintronics32 (Box 1). As demonstrated in MoS2- and WSe2-graphene heterostructures, the spin-polarized carriers generated in the TMDC transfer into the neighbouring graphene33,34 (Fig. 1c). The resulting antisymmetric Hanle spin precession curve (Fig. 1d) under the influence of an applied magnetic field 𝐵𝑦 provides unambiguous proof of the optical spin injection. 5 Spin dynamics and relaxation. The spin propagation is characterized by the spin relaxation length 𝜆s , given by 𝜆s = √𝐷𝑠 𝜏s with 𝐷s the spin diffusion constant, and 𝜏s the spin lifetime. In the diffusive regime, 𝐷𝑠 is obtained from transport measurements, while various possible SOC mechanisms, either intrinsic or extrinsic, introduce sources of spin relaxation and dictate the ultimate value of 𝜏s and 𝜆s 23. At room temperature, graphene displays spin-transport figures of merit for spin communication that outperform those of all other materials (Box 1). In hBN- protected graphene 𝜏s can be larger than 10 ns (with 𝜆s ~ 30 µm)28. Room-temperature spin- diffusion lengths reaching 10 µm were further achieved in chemical vapour deposition (CVD)- grown graphene on silicon oxide (SiO2) substrates27. Black phosphorous also transports spins quite efficiently; when encapsulated with hBN, 𝜏s ~ 0.7 ns and s ~ 2.5 µm at room temperature21. In TMDCs, spin relaxation has been investigated using optical orientation and time-resolved Kerr rotation. Reported spin-valley lifetimes at low temperature (a few kelvin) exceed several nanoseconds in electron-doped CVD-grown MoS2 and WS2 monolayers39,44 and are about 80 ns for holes in MOCVD WSe2 monolayers45. Although they are significantly longer in exfoliated WSe2 (100 ns for electrons40 and 1 s for holes40,41) all-temperature dependent studies show a fast decrease with temperature. A strikingly different behaviour has been observed in MoSe2, with the largest lifetime ~100 ns found at room temperature, albeit likely corresponding to non-itinerant carriers46. Efficient generation of pure and locked spin-valley diffusion current was demonstrated in exfoliated WS2-WSe2 heterostructures at 10 K by pump- probe spectroscopy47. Excitons are created in WSe2 and subsequent fast transfer of excited electrons to WS2 suppresses the exciton-valley depolarization channel. The recombination of electrons in WS2 with holes in WSe2 leaves an excess of holes in one of the WSe2 valleys, which are found to live for more than 20 µs and propagate over 20 µm. The mechanisms leading to spin relaxation in 2D materials are very rich and frequently unique to each material. This is illustrated by the case of graphene. Theoretical calculations 6 describe a wide range of possible SOC sources, through the symmetry, spatial range and strength of spin-conserving and non-spin conserving events. Intrinsic and Rashba contributions give rise to a small spin-splitting48 of tens of eV, as corroborated experimentally49. Early theoretical work indicated that 𝜏s could be in the millisecond range3. However, follow-up studies, introducing realistic descriptions of impurities (magnetic defects such as hydrogen adsorbents serving as spin-flip resonant scatterers50,51) or subtle mechanisms such as spin- pseudospin coupling52,53, have provided alternative explanations for the observed 𝜏s , in the ns and sub-ns range. These studies account for the 𝜏s energy dependence, with the most universal feature being a minimum near the charge neutrality point. The underlying origin for the spin relaxation has been described using the Elliot-Yafet54 or Dyakonov-Perel55 mechanisms, however only strictly applicable in disordered systems with short mean free paths 56. Some progress in analysing spin dynamics in the ballistic limit, as well as possible fingerprints in spin precession measurements, has been made57. In polycrystalline graphene, theoretical analysis has revealed universal spin diffusion lengths dictated by the absolute strength of the substrate- induced Rashba SOC in the Dyakonov-Perel regime56 (𝜆s = ℏ𝑣F /2𝜆R with ℏ the Planck constant, 𝑣F the Fermi velocity, and 𝜆R the Rashba SOC strength). Despite important progress, a full correspondence between theory and experiment is still missing. Indeed, the predominance of Rashba SOC in spin transport should manifest in a spin-transport anisotropy56, where the out- of-plane spin lifetime 𝜏𝑠,⊥ is half the in-plane one 𝜏𝑠,∥ . An electric-field modulation of the spin 𝜏𝑠,⊥ relaxation anisotropy ratio 𝜁 = 𝜏𝑠,∥ , consistent with the presence of Rashba SOC, was reported in graphene encapsulated with hBN58. However, it has been argued59,60 that the application of large out-of-plane magnetic fields could affect the determination of 𝜁. Recent experimental studies59–62 have failed to establish a significant spin lifetime anisotropy, suggesting that either magnetic resonant spin-flip scattering or deformation-induced gauge pseudo-magnetic fields randomize the spatial direction of the effective SOC field50,51,59,61. Remarkably, as discussed 7 below, a known SOC can be made dominant in proximitized graphene and, in contrast to graphene, the spin dynamics be well understood. Beyond graphene, the progress has been modest. Spin relaxation in few-layer black phosphorous appears to follow the Elliot-Yafet mechanism, as suggested by the similar temperature dependence of the measured 𝜏s and the momentum lifetime21. In TMDCs, the long spin-valley lifetimes confirm the expectation of spin-valley locking, which manifests more strongly in the valence band. The relaxation is expected to be mediated by intravalley decoherence mechanisms, dominating electron spin lifetime, while spin-flip processes between valleys, requiring simultaneous scattering of both valley and spin degrees of freedom, yield slow relaxation rates for holes. As temperature increases the behaviour becomes increasingly complex as relaxation pathways involving secondary valleys and different phonon-mediated intervalley scattering rates may play a role in determining the spin lifetimes46. Current and thermal spin current drift. Large 𝜆s may facilitate the realization of all-spin reprogrammable operations by controlling spin currents in lateral devices3,63 (Box 1). In this regard, a XOR magnetologic gate has been experimentally demonstrated at room temperature by electrical bias tuning of the spin injection in graphene29. This demonstration was followed by the proposal of a gate-driven demultiplexer using local voltage gates to tune the spin currents30. Further experimental progress has been achieved in the control of spin currents via carrier and thermal drift effects. Lateral drift fields in bilayer graphene (BLG), caused by a charge current (Fig. 1a), were shown to modulate the spin signal at room temperature42 (Fig. 1e). More recently, the use of thermal gradients to enhance or suppress the spin signal has been proposed and demonstrated43 (Fig. 1f). Here the spin signal modulation is driven by thermal drifts (Fig. 1a) in combination with an energy-dependent thermoelectric power, which result in a thermoelectric spin voltage. The observation of this phenomenon requires 8 sufficiently large lateral thermal gradients, which can be achieved by hot carrier generation, either by electrical current flow in graphene or through tunnel barrier injection43,64. Spin-orbit proximity effects Proximity effects represent a versatile approach to material design that can reach its full potential with vdW heterostructures, where hybridization of electronic orbitals of adjacent atomically-thin layers occurs. Despite the weak nature of vdW interaction, interlayer coupling of pure tunnelling nature can drastically change the energy dispersion and spin texture of the electronic band structure. For instance, in BLG such tunnelling turns the linear dispersion of low energy excitation to a parabolic shape, in addition to other band modifications. In a trilayer structure, as that represented in Fig. 2a, the intercalated material acquires properties from the top and bottom layers, bringing unprecedented opportunities for spintronics, particularly to imprint a SOC or magnetic exchange interaction (MEI). SOC is ubiquitous in spintronics10,23 (Box 1), playing a central role in spin relaxation and manipulation, CSI, anisotropic magnetoresistance, perpendicular magnetic anisotropy, spin-orbit torques (SOT) and the emergence of topological states. Therefore proximity SOC concepts are particularly relevant as they can potentially help engineer and control many of these phenomena. Graphene and BLG represent model materials for proximity-effect studies. In their isolated states, the SOC strength is only tens of eV and opens a very small spin-orbit gap, as shown by ab-initio calculations48,65,66 and recent experiments49,67. The Hamiltonian of isolated graphene is 𝐻 ~ 𝐻0 , with 𝐻0 characterizing Dirac carriers (see Box 2). When graphene gets in contact with other materials---of interest are 2D semiconductors and 2D insulators that preserve the Dirac cones in their band gaps---the character of 𝐻 can radically change. Surprisingly generic Hamiltonian models 𝐻 = 𝐻0 + 𝐻𝑡,𝑏 have been derived, which capture first-principles results, with 𝐻𝑡,𝑏 comprising separate orbital, spin-orbit and exchange terms (Box 2) that can be tracked back to the top (t) and/or bottom (b) 2D materials (Fig. 2a). 9 Box 2| Emergent Hamiltonian in proximitized graphene. The emergent Hamiltonian H describing Dirac electrons in The pseudospin inversion asymmetry (PIA) SOC 𝐻𝑃𝐼𝐴 = proximitized graphene comprises orbital, spin-orbit, and 𝑎(𝜆𝑃𝐼𝐴 𝜎𝑧 + 𝛥𝑃𝐼𝐴 )(𝑘𝑥 𝑠𝑦 − 𝑘𝑦 𝑠𝑥 ) , with 𝑎 the graphene lattice exchange terms: constant. The terms proportional to 𝜆𝑃𝐼𝐴 and 𝛥𝑃𝐼𝐴 lead to a 𝐻 = 𝐻0 + 𝐻Δ + 𝐻𝐼 + 𝐻𝑉𝑍 + 𝐻𝑅 + 𝐻𝑃𝐼𝐴 + 𝐻𝐸𝑋 renormalization of the Fermi velocity and a k-linear band 𝐻0 = ℏ𝑣𝐹 (𝜏𝑘𝑥 𝜎𝑥 + 𝑘𝑦 𝜎𝑦 ) is the low-energy graphene splitting, respectively. Finally, the proximity exchange coupling 𝐴𝐹 𝐴𝐹 Hamiltonian, with 𝑣𝐹 Fermi velocity, 𝑘 the state wave vector 𝐻𝐸𝑋 = 𝜆𝐸𝑋 𝑠𝑧 + 𝜆𝐸𝑋 𝜎𝑧 𝑠𝑧 is parametrized by 𝜆𝐸𝑋 , (𝜆𝐸𝑋 ) and and 𝜎 the pseudospin Pauli matrices. The factor 𝜏 is 1 (-1) at K emerges when graphene forms heterostructures with (K’) valleys. 𝐻Δ = Δ𝜎𝑧 describes orbital gap opening when the ferromagnets (antiferromagnets). Neglecting many-body effects, pseudospin symmetry is broken. Other terms include the essentially the same Hamiltonian (𝐻) describes the electronic intrinsic SOC, existing already in pristine graphene, 𝐻𝐼 = states in (proximitized) TMDCs but instead of sublattice degrees 𝜆𝐼 𝜏𝜎𝑧 𝑠𝑧 , parametrized by 𝜆𝐼 , with 𝑠 the spin Pauli matrices. of freedom, 𝜎 matrices describe the valence and conduction The valley-Zeeman SOC 𝐻𝑉𝑍 = 𝜆𝑉𝑍 𝜏𝜎0 𝑠𝑧 with strength 𝜆𝑉𝑍 bands68,69. The table presents representative graphene-based emerges, e.g., when graphene is interfaced with TMDCs or TIs. heterostructures, their electronic band structure at the K (and K’) The Rashba SOC 𝐻𝑅 = 𝜆𝑅 (𝜏𝜎𝑥 𝑠𝑦 − 𝜎𝑦 𝑠𝑥 ) with strength 𝜆𝑅 is points and the most relevant parameters for each of them by present whenever space-inversion symmetry breaks, owing to fitting the ab-initio results of the relaxed structures (see an electric field or in heterostructures. corresponding references for further details). Material Spectrum at K Parameters 𝝀𝑰 𝝀𝑹 Graphene 12 0 Graphene 12 5 (1 V/nm) Values in 𝜇eV, Ref. 48 𝚫 𝝀𝑽𝒁 𝝀𝑹 𝜟PIA MoS2 0.5 -0.3 0.1 0.5 MoSe2 0.4 -0.2 0.3 -0.5 WS2 1.3 -1.1 0.4 1.4 WSe2 0.5 -1.2 0.6 -0.1 Values in meV, check Ref. 70 for sign convention 𝝀𝑽𝒁 𝝀𝑹 𝜟PIA Bi2Se3 1.1 0.8 -0.05 Bi2Te2Se 1.4 0.7 0.05 Bi2Te3 0.1 0.2 0.7 Values in meV, check Ref. 71 for sign convention 𝚫 𝝀𝑽𝒁 𝝀𝑹 𝝀𝑬𝑿 Cr2Si2Te6 0.1 0.2 0.2 1.2 Cr2Ge2Te6 8.2 0.1 0.3 4.4 Cr2Sn2Te6 0.2 0.1 0.5 6.8 Values in meV, Ref. 72 𝚫 𝝀𝑨𝑭 𝑬𝑿 𝝀𝑬𝑿 MnPSe3 50 80 10 Values in 𝜇eV, Ref. 73 𝚫 𝝀𝑽𝒁 𝝀𝑹 𝝀𝑬𝑿 Cr2Ge2Te6 1.3 1.1 0.5 3.6 /Gr/WSe2 Values in meV, Ref. 72 10 Owing to the short range of the magnetic-exchange and spin-orbit interactions, the proximity effects are largely driven by the layer adjacent to the proximitized graphene. Therefore, the thickness of the 2D magnet or large-SOC material does not require control. In addition, the proximity effect in BLG predominantly develops only in the layer in contact with the material. According to the previous discussion, it is not surprising that proximity SOC concepts are best established for graphene. While materials such as hBN do not increase graphene’s SOC beyond tens-of-µeV74, it has been demonstrated that strong SOC materials like as TMDCs or the Bi2Se3 TI family, significantly alter it or reinforce it (Box 2 b and c). The graphene Dirac cones are preserved within the band-gaps of many TMDCs70, which allows one to exploit the advantages of graphene’s high-mobility and novel proximity spin interactions75,76. The SOC strength can reach meVs (Box 2 b) and be dominated by a valley-Zeeman SOC, which is characterized by an out-of-plane spin-orbit field that is opposite at K and K’ valleys (as in the TMDC). In addition, carriers experience a Rashba SOC, with an in-plane spin-orbit field texture perpendicular to the momentum. On the orbital level, the breaking of the pseudospin symmetry leads to the appearance of an orbital gap, described by a staggered potential. The valley-Zeeman and Rashba fields are predicted to change by twisting the graphene relative to the TMDC, with the largest SOC strength appearing at 15-20 degrees between the lattice vectors77,78. Band structures at smaller twist angles were also theoretically investigated79,80. Graphene can also be proximitized by TIs, such as Bi2Se371,81,82. These 3D TIs exhibit protected surface states with in-plane spin-orbit fields inducing spin-momentum locking. Surprisingly, the proximitized-SOC is still dominated by the out-of-plane valley-Zeeman coupling, which is not present in the TI (Box 2 c). Experimental signatures of proximity-induced SOC in graphene-TMDC heterostructures have been found in weak (anti-)localization measurements83–87. However, the results are 11 controversial in terms of the SOC strength, which ranges from ~1 to 10 meV, as well as the SOC nature, which was reported to have Rashba or valley-Zeeman character. Variations on the SOC strength may be due in part to variations in the interface properties; twist angle or the presence of trapped bubbles – all of them difficult to control during device fabrication. Nonetheless, the nature of the proximity SOC has been established by means of spin relaxation anisotropy and CSI experiments, as discussed below. Fig. 2 | Proximity-effects. a, In a trilayer vdW heterostructure, the middle layer is proximitized by the top (t) and bottom (b) layers. Its effective Hamiltonian 𝐻 is distinct from the Hamiltonian 𝐻0 of the isolated layer, acquiring properties from t and b. b, Device geometry to investigate proximity SOC in bilayer graphene (BLG) encapsulated with WSe2. The charge density 𝑛 and displacement field 𝐷 are controlled with top and bottom gates. c, Penetration field capacitance 𝐶𝑝 as a function of 𝑛 and 𝐷 showing an incompressibility peak at 𝐷 = 0 for 𝑛 = 0. Arrows indicate maxima associated to band-splitting due to proximity-SOC. d, Illustration of the Dzyaloshinskii-Moriya interaction (DMI) at the vdW interface of a 2D magnet (Fe3GeTe2) and a TMDC with large SOC (WTe2), si and sj represent the spins of neighboring atoms and dij the DMI vector. The interfacial DMI stabilizes magnetic skyrmions, as observed by Lorentz transmission electron microscopy in WTe2- Fe3GeTe2 (magnetic field: 0.51 T; scale bar: 500 nm). e, Schematics of a lateral spin device for investigating proximity SOC in graphene by means of spin transport. The device consists of a graphene spin channel, which is partially covered with a TMDC. The two attached ferromagnets (F1 and F2) act as spin injector and detector, respectively. A charge current I through F1 injects spins with an orientation parallel to the F1 magnetization direction. While diffusing towards the spin detector (F2), the spins precess under either an oblique or in-plane magnetic field B, as represented in the top and bottom insets. When they reach the TMDC, their orientation is characterized by the angle *, whose value is with the magnitude of B. f, Spin precession experiments in a graphene-WS2 device. The nonlocal signal 𝑅𝑁𝐿 is shown for B perpendicular to the plane (blue) and in-plane (red) for which spins stay in the graphene plane and precess out of it, respectively (see insets). The disparity of the curves demonstrates the highly anisotropic nature of the spin dynamics in graphene-WS2. Panels b-c and e-f are adapted from Ref.88 and Ref.37, respectively. SNL. The Lorentz microscopy image in d is from Ref.89, SNL. 12 Graphene proximitized by a TMDC can also exhibit an inverted band structure (Box 2 b), suggesting emerging topological phenomena70,73,90, distinct from isolated graphene91 and driven by the valley-Zeeman coupling. Although the band structure remains topologically trivial, protected pseudo-helical states appear at zigzag edges of proximitized nanoribbons. An inverted band structure in BLG-WSe2 stacks has been confirmed experimentally88 (Fig. 2b and 2c). Topological quantum spin Hall phases were also predicted in proximitized BLG92, while helical edge modes in BLG-WSe2 heterostructures were recently reported93. Proximity SOC is becoming increasingly important in 2D materials beyond graphene. A SOC enhancement could help stabilize the anisotropy or the magnetic order of a 2D magnet. An increase in the Curie temperature TC of Fe3GeTe2 (FGT) up to 400 K has been observed when grown onto Bi2Te3. The larger TC in thinner FGT films on Bi2Te3, when the opposite trend is observed in FGT alone, suggests the presence of an interfacial effect94, although it is unclear why a significant TC increase is observed in relatively thick FGT films (up to 10s of nm). Similarly, an increase in TC was observed95 in the Heisenberg ferromagnet V5Se8 when in contact to NbSe2. The enhanced TC was accompanied by a strong out-of-plane magnetic anisotropy and was attributed to the Zeeman SOC in NbSe2. A large SOC together with broken inversion symmetry can also favour the anti-symmetric Dzyaloshinskii–Moriya exchange interaction (DMI) and lead to topological magnetic configurations96 (Fig. 2d). Néel-type skyrmions were observed in WTe2-FGT using Lorentz transmission electron microscopy89 (Fig. 2d); the large interfacial DMI energy of ~1.0 mJm−2 was attributed to induced Rashba SOC. Anisotropic spin relaxation and spin filtering. One of the first spin-device realizations combining graphene with a TMDC comprised a graphene lateral spin device partially capped with MoS2. Using electrostatic gating, the spin current across the graphene channel was controlled between ON and OFF states, a phenomenon attributed to spin absorption at the MoS235. It is argued that in the OFF-state spins could move freely between graphene and MoS2, 13 due to the gate-induced suppression of the Schottky barrier between graphene and MoS2, leading to fast spin relaxation35,97. Further studies36,37 reported anisotropic spin relaxation in graphene-TMDC heterostructures (with TMDC = MoSe2, MoS2 and WS2), even in the absence of spin absorption37. By implementing out-of-plane spin precession techniques59–61 (Figs. 2e and 2f) 𝜏𝑠,⊥ the spin relaxation anisotropy ratio 𝜁 = 𝜏𝑠,∥ were quantified. It was observed that the in-plane spin component is strongly reduced when propagating through the graphene-TMDC region, with 𝜏𝑠,∥ in the range of a few picoseconds (two orders of magnitude smaller than in reference graphene devices37). In contrast, the out-of-plane spin component propagates much more efficiently, with 𝜏𝑠,⊥ in the range of tens of picoseconds36,37 and thus 𝜁 ~ 10. These results evidence that graphene-TMDC heterostructures act as spin filters, whose spin transmission is tailored by the spin orientation. When no spin current is absorbed by the TMDC, the anisotropy can be fully attributed to proximity-induced SOC37,60. According to theoretical predictions, the spin dynamics is controlled by the spin-valley coupling imprinted onto graphene98,99. The spin relaxation is governed by the Dyakonov-Perel mechanism, with 𝜏𝑠,⊥ and 𝜏𝑠,∥ largely determined by the momentum (𝜏𝑝 ) and intervalley (𝜏iv ) scattering times, respectively, typically with 𝜏𝑝 ≪ 𝜏iv . Because of the relatively long 𝜏iv , the in-plane spins precess under a slowly fluctuating effective (perpendicular) magnetic field between K and K’, leading to fast spin relaxation. In contrast, because of the short 𝜏𝑝 , out-of-plane spins precess under fast fluctuating Rashba fields and their relaxation is suppressed due to motional narrowing23. The spin-anisotropy ratio, derived from the emergent Hamiltonian (Box 2), is98–100 𝜆VZ 2𝜏 1 𝜆VZ 2 𝜏iv 1 iv 𝜁=( ) + ≈( ) + (1) 𝑎𝑘𝛥PIA ±𝜆R 𝜏p 2 𝜆R 𝜏p 2 14 with the approximation being valid about the Dirac point or for small 𝛥PIA (Box 2). In the 1 absence of valley-Zeeman SOC, 𝜁 = 2 with out-of-plane spins relaxing faster than in-plane spins, as expected for a 2D Rashba system56. Using an interband tunnelling description and first-principles calculations, it has been proposed that the SOC strength can be tuned with the Fermi energy, resulting in an energy-dependent anisotropy77,78. Anisotropic spin relaxation has also been discussed theoretically in graphene-TIs81 and graphene-hBN heterostructures74. Moreover, 𝜁 ~ 10 has been measured in hBN-BLG-hBN at temperatures around 100 K near the charge neutrality point101,102. The spin relaxation becomes isotropic either at large enough carrier densities or at high temperatures (at room temperature 𝜁 ~ 1). Similar to graphene-TMDC, the large 𝜁 seems to arise from the spin-valley coupling associated to the intrinsic SOC in BLG. Charge-to-spin interconversion. CSI phenomena driven by SOC are amongst the most relevant effects in modern spintronics103. Their presence can reveal subtle spin-orbit interactions and spin dynamics in the investigated materials. They are also central for next generation SOT magnetic memories (SOT-MRAM)9,10 as well as for proposals targeting energy-efficient spin- logic architectures104 (Box 1). CSI in 2D materials has been gaining increasing attention, following the report of SOTs with non-trivial (and potentially useful) symmetries using TMDCs105, the achievement of magnetization switching with TIs and TMDCs106,107, as well as the observation of spin Hall effects (SHE) and spin galvanic effects (SGE) 108–111 (Fig. 3). A recent surge of experiments on vdW heterostructures has been triggered by the use of graphene as a channel to transport a spin current from a ferromagnetic contact to the CSI region112,113. The device geometry is analogous to that developed for fully metallic systems103, consisting of a graphene Hall cross with a large SOC 2D material along one of the arms and ferromagnetic injector and detector contacts across the other (Fig. 3a). The first experiments using platinum (Pt), a well-known material with efficient CSI by the SHE103, demonstrated large 15 CSI112,113 and established the use of spin-precession to investigate the nature of the CSI in 2D heterostructures113. The analysis and interpretation of the results differ for insulating or conducting SOC materials. While in the former case it is possible to directly ascribe the CSI to proximity-induced SOC, in the latter case the overall CSI can aggregate the CSI arising from proximity effects and the CSI at the surface and/or bulk of the conducting SOC material. The anomalously large CSI in graphene-Pt could be due to such aggregation of effects, although this remains to be clarified113. Fig. 3 | Charge-to-spin interconversion (CSI) in vdW heterostructures. a, Schematics of a lateral spin device for CSI experiments. The device consists of a graphene Hall cross with a TMDC covering one of the arms, which enhances the SOC in graphene. A current I along the graphene arm creates a transversal spin current with out-of-plane spin polarization driven by the spin Hall effect (SHE). Simultaneously, a non-equilibrium spin density, with in-plane spin polarization, is built at the proximitized graphene driven by the inverse spin galvanic effect (ISGE). Spin-sensitive contacts F1 and F2 probe diffusing spin currents under an applied magnetic field B. For B applied along z (Bz) and along x (Bx) only the spins arising from the ISGE (b) and the SHE (c) precess, respectively. d, Room temperature ISHE signal RISHE as a function of magnetic field Bx in a device with MoS2 on few-layer graphene. e, SGE signal measured at different temperatures in a graphene- WS2 device. f, Experimental demonstration of the coexistence and gate tunability of the ISHE and the SGE in graphene-WS2 at room temperature. Panels a-c and f adapted from Ref.110 SNL. Panels d and e adapted from Ref.109 and Ref.111, respectively, American Chemical Society. 16 CSI driven by SOC in a vdW heterostructure was first confirmed in multilayer-graphene- MoS2109 (Fig. 3d). This report was soon followed by the simultaneous observations of the SHE and SGE in graphene–WS2110,111 (Figs. 3e and f). The CSI in graphene by proximity SOC114–116 is best established by ruling out the spin absorption in the TMDC35,97,110. Notably, the CSI can be controlled upon electrostatic gating, which tunes the graphene carrier density n. A gate dependent CSI in proximitized graphene was first observed with WS2 up to 75 K for the inverse SGE111 and up to room temperature for the inverse SGE and SHE (and reciprocal effects)110. Gate dependence of the inverse SHE and of the SGE was later on reported in graphene-WSe2, and in graphene-TaS2 and graphene-(Bi,Sb)2Te3, respectively117–119. The effective conversion efficiencies compare favourably with those of metallic systems109,110,118. Furthermore, the experimental dependence of the CSI vs n in graphene- WS2110 agrees with theoretical modelling, for both proximity-induced SGE115 and SHE114. The SOC strength has been estimated using the Kubo-Bastin formula110,114. By matching model calculations with the experimental results 𝜆I ~ 0.2 meV and 𝜆vZ ~ 2.2 meV are obtained110 (Box 2). Prior reports of SHE in graphene by proximity of WS2 suggested a much larger SOC (17 meV)120. However, these experiments used the so-called H-geometry103, which in graphene devices is sensitive to a variety of phenomena, not necessarily spin related5,121–123. CSI was also investigated in other conducting 2D materials, following the same approach as with Pt112,113. Experiments using 1T’-MoTe2 revealed an unconventional CSI in which a charge current arises parallel to the spin orientation124. It is unclear whether the CSI originates in the bulk or the surface of the material. The observation is reminiscent to the appearance of unconventional SOTs in low-symmetry WTe2105,125, suggesting that the crystalline mirror symmetry of 1T’-MoTe2 is broken, perhaps by strain introduced during device fabrication. Unconventional CSI was also observed in WTe2, with an efficiency approaching 10 %; control experiments indicate that the CSI originate in the bulk of the material126. 17 Magnetic proximity effects When non-magnetic 2D materials, such as graphene or TMDCs, are in contact with a magnetic material, they can experience a proximity-induced MEI. The induced magnetism is characterized by a net local spin polarization in equilibrium and an energy splitting of the bands, which in graphene is equal at different valleys (in the absence of SOC). The proximity MEI is parametrized by the exchange coupling strength 𝜆𝐸𝑋 when the non-magnetic 2D 𝐴𝐹 material is in contact with a ferromagnet and 𝜆𝐸𝑋 when in contact with an antiferromagnet (Box 2). Typically, the goal is to achieve a large 𝜆𝐸𝑋 (𝜆𝐴𝐹 𝐸𝑋 ) while maintaining the (spin) transport capabilities of the isolated layer. Proximity MEI in graphene. Early first-principles calculations predicted a 𝜆𝐸𝑋 of tens of meVs in graphene when proximitized by bulk materials such as EuO127 or the ferrimagnet Y3Fe5O12 (YIG)128. Similar 𝜆𝐸𝑋 were estimated with conventional ferromagnetic metals, such as Co or Fe, across a thin hBN insulating barrier8,129,130. Control of proximity exchange by electrical polarization has been predicted in graphene on multiferroic BiFeO3131. The first experimental results were also reported in graphene proximitized by bulk materials, albeit with typical 𝜆𝐸𝑋 values that were significantly smaller than expected. Charge transport experiments in graphene-YIG showed the presence of an anomalous Hall resistance132, while Zeeman SHE indicated an exchange field of up to 14 Tesla (1.5 meV) in graphene-EuS133. Subsequently, spin transport experiments, using lateral devices based on graphene134 and BLG135 onto YIG, provided more direct indications of proximity MEI and demonstrated spin current modulation (Fig. 4a and 4b), although 𝜆𝐸𝑋 was found to be even smaller134 (~20 eV). Proximity MEI was also reported in YIG-graphene-hBN through nonlocal charge transport measurements136, Co- graphene-NiFe junctions137, and gate-dependent spin-inversion in edge-contacted graphene spin valves22. 18 Fig. 4 | Magnetic proximity effects in graphene. a, Schematics of experimental device to detect exchange interaction in graphene, induced by proximity of a magnetic substrate using spin transport. A magnetic field B sets the orientation of the substrate magnetization M, and thus of the proximity-induced exchange field BEX 𝑉𝑁𝐿 in graphene. b, Representation of 𝑅𝑁𝐿 = versus  in parallel (blue) and antiparallel (red) alignment of the 𝐼 magnetizations of the ferromagnetic contacts F1 and F2. When = 0, BEX is perpendicular to the orientation of the injected spins, and the signal presents the minimum magnitude. The coercive field of the substrate is assumed to be much lower than that of F1 and F2, as in the case of YIG. Therefore M follows B while the magnetizations of the contacts remain along their long axis (black arrows). c, 𝑅𝑁𝐿 as a function of magnetic field along x (see a) in a graphene-CrSBr heterostructure. The initial arrangement of the magnetizations of the injector and detector contacts (black arrows) and the substrate (purple arrow) is achieved with B along y. In this case, the coercive filed of the contacts is smaller than that of the substrate (an antiferromagnet). The inset shows calculated precession curves with a spin-dependent graphene conductivity model. d, Schematics of graphene coupled to CrSe. The red arrows represent the remnant magnetization induced at the interface by field cooling with opposite magnetic field orientations. e, Exchange splitting in graphene on top of antiferromagnetic CrSe. Quantum Hall plateaus are shifted by field cooling in the heterostructure. The black arrow denotes the most pronounced plateau shifted by field cooling. The shift in the plateau is attributed to the change in the remnant magnetization. Panel c adapted from Ref.138. Panels e adapted from Ref.139, SNL. The small proximity 𝜆𝐸𝑋 observed with bulk magnets could be ascribed to rough interfaces, thus recent Investigations have shifted towards proximity MEI by 2D magnets, which promise atomically smooth interfaces. Relevant 2D ferromagnets include the Cr2X2Te6 (X= Si, Ge or Si) or CrX3 (X=I, Br or Cl) families with predicted 𝜆𝐸𝑋 in the range of several meVs72 (Box 2 d). The induced MEI with antiferromagnets, such as MnPSe3, a 2D Heisenberg-like antiferromagnet, could lead to sub-meV staggered exchange coupling in graphene73 (Box 2 e). A few experiments indeed indicate a significant proximity MEI140,138,139. In graphene-CrSBr, with CrSBr an interlayer antiferromagnet, charge and spin transport driven by electrical bias and thermal gradients138 indicate 𝜆𝐸𝑋 ~ 20 meV, corresponding to an exchange field of ~ 170 T at 4.5 K (Fig. 4c). Proximity MEI was also reported in graphene-CrSe heterostructures139, where CrSe is a non- 19 collinear antiferromagnet with a complex phase diagram. A magnetized interface (Fig. 4d), which does not occur in CrSe alone, is observed in graphene-CrSe with transport and magneto- optic measurements after magnetic-field cooling. The proximity exchange field was quantified from shifts in the quantum Hall plateaus and quantum oscillations (Fig. 4e), resulting in 𝜆𝐸𝑋 larger than 130 meV at 2 K. Proximity MEI beyond graphene. Proximity MEI has also been investigated in materials such as TMDCs and TIs. Experiments in semiconducting TMDCs typically rely on optical techniques. In WSe2-EuS141, and WS2-EuS142 heterostructures, the reflection and photoluminescence spectra of circularly-polarized photons probe the electronic states of the TMDC and quantify the proximity-induced exchange splitting. For WSe2, it was estimated that 𝜆𝐸𝑋 ~ 2-4 meV, corresponding to an exchange field of ~10-20 T. For WS2, the splitting was found to be much larger 𝜆𝐸𝑋 ~ 19 meV and to have opposite sign. According to theoretical modelling142, the magnitude and sign of the splitting is determined by the surface termination of EuS and the band alignment between TMDCs and EuS. As with graphene, a growing number of studies are being carried out with 2D magnets143,144. Placing a monolayer TMDC on CrI3 (Fig. 5a) results in an estimated exchange splitting in the meV range145, which should affect the exciton spectra146. Given its short range interaction, the proximity effect allows probing the magnetization of the 2D magnet adjacent layer, even in the absence of a global magnetic moment. A layer-dependent magnetic proximity effect has been observed in monolayer WSe2 on few-layer CrI3144. While magneto- optic measurements demonstrate that bilayer CrI3 is a layered antiferromagnet, circularly- polarized photoluminescence shows that the exchange splitting in WSe2 is most sensitive to the interfacial layer. The contribution of the second layer to the splitting is of substantially smaller magnitude and has an unexpected opposite sign (Figs. 5b and 5c). However, the quantitative interpretation of the exciton spectra and dynamics is not straightforward. The hybridization of 20 the TMDC orbitals with the spin-polarized CrI3 orbitals is rather complex. In particular, it is expected that twisting the two layers would lead to significant variations of the proximity MEI, both in magnitude and character145. Furthermore, photoluminescence studies in MoSe2-CrBr3 uncovered a charge dependence of the proximity effects in which the valley polarization of the MoSe2 trion state follows the local CrBr3 magnetization, while the neutral exciton state is insensitive to it147. This is attributed to spin-dependent interlayer charge transfer in timescales between the exciton and trion radiative lifetimes. Fig. 5 | Magnetic proximity effects in TMDCs. a, Schematic of monolayer WSe2 interfacing with a magnetic layer (CrI3 in this case) to investigate proximity control of valley dynamics. b, Reflexion magnetic circular dichroism (RMCD) as a function of magnetic field in WSe2-CrI3, showing typical features of a layered antiferromagnetic bilayer CrI3. c, Valley Zeeman splitting  as a function of magnetic field as determined from photoluminescence measurements. Panels b-c, are adapted from Ref.143, SNL. Magnetic proximity effects are also intensively investigated in layered 2D and 3D TIs. In the 3D TIs, such as the Bi2Te3 family, broken time-reversal symmetry induces a gap in the Dirac band dispersion of the surface states148. Tuning of the Fermi level in the gap leads to the emergence of a quantum Hall effect at zero magnetic field: the quantum anomalous Hall effect (QAHE), a phenomenon that is very promising for quantum metrology. Signatures of proximity magnetism in 3D TIs have been reported, for example, in [EuS, YIG, Tm3Fe5O12, Cr2Ge2Te6]/TI with the observation of an anomalous Hall effect149–151 or by investigating spin-polarized neutron reflectivity152. However, the origin of the magnetic signals in these type of experiments is usually not fully understood153–155. An unambiguous demonstration of proximity MEI was reported (Zn,Cr)Te/(Bi,Sb)2Te3/(Zn,Cr)Te heterostructures with the observation of the QAHE156. In 2D TIs, or quantum spin Hall insulators, such as monolayer WTe2, the conduction is 21 dominated by helical edge states157,158 with canted spin orientation due to reduced symmetries159–161. When WTe2 is placed in a heterostructure with CrI3, magnetic proximity could lead to a change in the edge state conductance that is controlled by the magnetization of the interfacial CrI3 layer162. This could ultimately result in the observation of the QAHE, depending on other phenomena, such as charge transfer at the interface. Conclusions and future perspectives Recent progress in the design of complex vdW heterostructures brings unprecedented possibilities for developing innovative ultra-compact spin devices and computing architectures. In respect to conventional spintronic applications, it is necessary to identify the best combination of 2D materials to demonstrate practical magnetic tunnel junctions (either with conducting or insulating 2D magnets) or CSI-induced switching of 2D magnets163 (using high spin-orbit materials such as WTe2 and Bi2Te3). Voltage control of magnetic properties is another promising avenue available with 2D materials (Box 1). Advances have been made in this regard7,38, while electric-field dependent proximity SOC in 2D magnets will certainly bring further opportunities. In addition, the reduced symmetries in monolayer TMDCs such as MoTe2 and WTe2 lead to persistent spin textures, which result in multicomponent SHE124,164. In combination with the externally tunable spin-orbit fields, they may enable electric control of SOTs165. Proximity effects can be further exploited for novel spin-orbit166,167 or magnetic valves168,169 comprising BLG and a large SOC material or a 2D magnet, respectively. The interplay of two factors -- short-range of the proximity effect and the layer polarization in BLG - - results in layer-polarized electronic bands and asymmetrically conduction and valence bands. Applying a transverse electric field can reverse this situation, turning the SOC (or 𝜆𝐸𝑋 ) ON or OFF by electric field or doping and leading to novel spintronic functionalities23. An additional functionality is offered by intercalating BLG between 2D magnets, forming a spin valve that 22 could resolve parallel and antiparallel magnetizations in transport170. Another exciting perspective is engineering both the SOC and MEI, and their interplay, as in the so called ex-so- tic vdW heterostructures (Box 2 f). In graphene, such an interplay is predicted to induce the QAHE171, novel topological phases73,172, proximity-based SOT72, unique signatures of anisotropic magnetoresistance, and even new functionalities based on swapping spin-orbit coupling and exchange, all in a single device173. The experimental observation of these phenomena will be key milestones in spintronics and quantum metrology. The control of interlayer twist and stacking between layers can be further exploited to tailor the spin interactions (Box 1). For instance, the atomic stacking in the Moiré pattern in twisted CrBr3 bilayers modulates the proximity SOC and MEI, as revealed by spin-polarized scanning tunnelling microscopy174. This and other emerging phenomena could become mainstream in the forthcoming years for 2D spintronics, including the control of information transfer via magnons in 2D magnets175,176 and topological magnetic structures like skyrmions and Néel spin spirals, which have been already observed in FGT and FGT-based heterostructures89,177–179 and predicted in 2D Janus materials180. Many of these technological prospects will require overcoming important challenges. A particularly critical one is the development of large-area, stable 2D magnets with magnetic order at room temperature, using scalable stacking and growth processes. Proximity SOC has shown potential to increase 𝑇𝐶 ; similarly, a significant enhancement of 𝑇𝐶 could also be achieved by coupling a 2D ferromagnets to an antiferromagnet, as in Fe3GeTe2-FePS3 heterostructures181. Taking advantage of SOTs will demand a full understanding of the mechanisms for (vertical) spin transfer across heterostructures and ways to take advantage of the SOTs induced by 2D materials with reduced symmetries. Applications relying on topological phases, such as the QAHE, also require robust magnetic properties to both applied currents and high temperatures. In regards to skyrmions, many fundamental challenges lie ahead 23 beyond material issues, such as the development of writing, processing and reading functionalities using all-electrical schemes. Moreover, increasingly realistic theoretical modelling of proximity effects in complex vdW heterostructures (combining different 2D material families) is necessary to grasp the subtle spin and exciton dynamics and to separate the contributions of the exchange interaction from spin and orbital moments. Extracting (minimum) model Hamiltonian from ab-initio calculations is becoming very challenging owing to the intertwined combination of all interactions involved, necessary for performing spin transport simulations. Precise comparison with experiments is hampered by the difficulty of reproducing interfaces and controlling the stacking, especially in multilayer heterostructures. The advances covered in this Review therefore represent only the starting point for 2D material design for spintronics and optospintronics. However, the endless possibilities offered by proximity effects promise an enduring impact in terms of innovative devices and architectures. Engineering vdW heterostructures can reveal novel classes of artificial quantum materials182, offering opportunities for both scientific discoveries and technological breakthroughs. Information and quantum computing paradigms, electrically driven light emitters, photodetectors, and sensors might emerge by harnessing the rich internal degrees of freedom of 2D materials (spin, valley, sublattice, excitonic and layer pseudospin) and the creation and manipulation of entangled states. References 1. Geim, A. K. & Grigorieva, I. V. Van der Waals heterostructures. Nature 499, 419–425 (2013). 2. Novoselov, K. S., Mishchenko, A., Carvalho, A. & Neto, A. H. C. 2D materials and van der Waals heterostructures. Science 353, (2016). 3. Han, W., Kawakami, R. K., Gmitra, M. & Fabian, J. Graphene spintronics. Nat. Nanotechnol. 9, 794–807 (2014). 24 4. Roche, S. et al. Graphene spintronics: the European Flagship perspective. 2D Mater. 2, 030202 (2015). 5. Avsar, A. et al. Colloquium: Spintronics in graphene and other two-dimensional materials. Rev. Mod. Phys. 92, 021003 (2020). 6. Manzeli, S., Ovchinnikov, D., Pasquier, D., Yazyev, O. V. & Kis, A. 2D transition metal dichalcogenides. Nat. Rev. Mater. 2, 1–15 (2017). 7. Gibertini, M., Koperski, M., Morpurgo, A. F. & Novoselov, K. S. Magnetic 2D materials and heterostructures. Nat. Nanotechnol. 14, 408–419 (2019). 8. Zutic, I., Matos-Abiague, A., Scharf, B., Dery, H. & Belaschhenko, K. Proximitized materials. Mater. Today 22, 85–107 (2019). 9. Manchon, A. et al. Current-induced spin-orbit torques in ferromagnetic and antiferromagnetic systems. Rev. Mod. Phys. 91, 035004 (2019). 10. Dieny, B. et al. Opportunities and challenges for spintronics in the microelectronics industry. Nat. Electron. 3, 446–459 (2020). 11. Loong, L. M. et al. Flexible MgO Barrier Magnetic Tunnel Junctions. Adv. Mater. 28, 4983–4990 (2016). 12. Serrano, I. G. et al. Two-Dimensional Flexible High Diffusive Spin Circuits. Nano Lett. 19, 666– 673 (2019). 13. Schmidt, G., Ferrand, D., Molenkamp, L. W., Filip, A. T. & van Wees, B. J. Fundamental obstacle for electrical spin injection from a ferromagnetic metal into a diffusive semiconductor. Phys. Rev. B 62, R4790–R4793 (2000). 14. Tombros, N., Jozsa, C., Popinciuc, M., Jonkman, H. T. & van Wees, B. J. Electronic spin transport and spin precession in single graphene layers at room temperature. Nature 448, 571–574 (2007). 15. Han, W. et al. Tunneling Spin Injection into Single Layer Graphene. Phys. Rev. Lett. 105, 167202 (2010). 25 16. Neumann, I., Costache, M. V., Bridoux, G., Sierra, J. F. & Valenzuela, S. O. Enhanced spin accumulation at room temperature in graphene spin valves with amorphous carbon interfacial layers. Appl. Phys. Lett. 103, 112401 (2013). 17. Singh, S. et al. Strontium Oxide Tunnel Barriers for High Quality Spin Transport and Large Spin Accumulation in Graphene. Nano Lett. 17, 7578–7585 (2017). 18. Friedman, A. L., van ‘t Erve, O. M. J., Li, C. H., Robinson, J. T. & Jonker, B. T. Homoepitaxial tunnel barriers with functionalized graphene-on-graphene for charge and spin transport. Nat. Commun. 5, 3161 (2014). 19. Kamalakar, M. V., Dankert, A., Bergsten, J., Ive, T. & Dash, S. P. Enhanced Tunnel Spin Injection into Graphene using Chemical Vapor Deposited Hexagonal Boron Nitride. Sci. Rep. 4, 6146 (2014). 20. Gurram, M., Omar, S. & Wees, B. J. van. Bias induced up to 100% spin-injection and detection polarizations in ferromagnet/bilayer-hBN/graphene/hBN heterostructures. Nat. Commun. 8, 248 (2017). 21. Avsar, A. et al. Gate-tunable black phosphorus spin valve with nanosecond spin lifetimes. Nat. Phys. 13, 888–893 (2017). 22. Xu, J. et al. Spin inversion in graphene spin valves by gate-tunable magnetic proximity effect at one-dimensional contacts. Nat. Commun. 9, 2869 (2018). 23. Žutić, I., Fabian, J. & Das Sarma, S. Spintronics: Fundamentals and applications. Rev. Mod. Phys. 76, 323–410 (2004). 24. Lin, X., Yang, W., Wang, K. L. & Zhao, W. Two-dimensional spintronics for low-power electronics. Nat. Electron. 2, 274–283 (2019). 25. Torres, W. S. et al. Magnetism, spin dynamics, and quantum transport in two-dimensional systems. MRS Bull. 45, 357–365 (2020). 26. Kamalakar, M. V., Groenveld, C., Dankert, A. & Dash, S. P. Long distance spin communication in chemical vapour deposited graphene. Nat. Commun. 6, 6766 (2015). 26 27. Gebeyehu, Z. M. et al. Spin communication over 30-µm long channels of chemical vapor deposited graphene on SiO 2. 2D Mater. 6, 034003 (2019). 28. Drögeler, M. et al. Spin Lifetimes Exceeding 12 ns in Graphene Nonlocal Spin Valve Devices. Nano Lett. 16, 3533–3539 (2016). 29. Wen, H. et al. Experimental Demonstration of xor Operation in Graphene Magnetologic Gates at Room Temperature. Phys. Rev. Appl. 5, 044003 (2016). 30. Lin, X. et al. Gate-Driven Pure Spin Current in Graphene. Phys. Rev. Appl. 8, 034006 (2017). 31. Ingla-Aynés, J., Kaverzin, A. A. & van Wees, B. J. Carrier Drift Control of Spin Currents in Graphene-Based Spin-Current Demultiplexers. Phys. Rev. Appl. 10, 044073 (2018). 32. Gmitra, M. & Fabian, J. Graphene on transition-metal dichalcogenides: A platform for proximity spin-orbit physics and optospintronics. Phys. Rev. B 92, 155403 (2015). 33. Luo, Y. K. et al. Opto-Valleytronic Spin Injection in Monolayer MoS2/Few-Layer Graphene Hybrid Spin Valves. Nano Lett. 17, 3877–3883 (2017). 34. Avsar, A. et al. Optospintronics in Graphene via Proximity Coupling. ACS Nano 11, 11678–11686 (2017). 35. Yan, W. et al. A two-dimensional spin field-effect switch. Nat. Commun. 7, 13372 (2016). 36. Ghiasi, T. S., Ingla-Aynés, J., Kaverzin, A. A. & van Wees, B. J. Large Proximity-Induced Spin Lifetime Anisotropy in Transition-Metal Dichalcogenide/Graphene Heterostructures. Nano Lett. 17, 7528–7532 (2017). 37. Benítez, L. A. et al. Strongly anisotropic spin relaxation in graphene–transition metal dichalcogenide heterostructures at room temperature. Nat. Phys. 14, 303–308 (2018). 38. Mak, K. F., Shan, J. & Ralph, D. C. Probing and controlling magnetic states in 2D layered magnetic materials. Nat. Rev. Phys. 1, 646–661 (2019). 39. Yang, L. et al. Long-lived nanosecond spin relaxation and spin coherence of electrons in monolayer MoS 2 and WS 2. Nat. Phys. 11, 830–834 (2015). 40. Dey, P. et al. Gate-Controlled Spin-Valley Locking of Resident Carriers in WS2 Monolayers. Phys. Rev. Lett. 119, 137401 (2017). 27 41. Kim, J. et al. Observation of ultralong valley lifetime in WSe2/MoS2 heterostructures. Sci. Adv. 3, e1700518 (2017). 42. Ingla-Aynés, J., Meijerink, R. J. & Wees, B. J. van. Eighty-Eight Percent Directional Guiding of Spin Currents with 90 μm Relaxation Length in Bilayer Graphene Using Carrier Drift. Nano Lett. 16, 4825–4830 (2016). 43. Sierra, J. F. et al. Thermoelectric spin voltage in graphene. Nat. Nanotechnol. 13, 107–111 (2018). 44. McCormick, E. J. et al. Imaging spin dynamics in monolayer WS 2 by time-resolved Kerr rotation microscopy. 2D Mater. 5, 011010 (2017). 45. Song, X., Xie, S., Kang, K., Park, J. & Sih, V. Long-Lived Hole Spin/Valley Polarization Probed by Kerr Rotation in Monolayer WSe2. Nano Lett. 16, 5010–5014 (2016). 46. Ersfeld, M. et al. Spin States Protected from Intrinsic Electron–Phonon Coupling Reaching 100 ns Lifetime at Room Temperature in MoSe2. Nano Lett. 19, 4083–4090 (2019). 47. Jin, C. et al. Imaging of pure spin-valley diffusion current in WS2-WSe2 heterostructures. Science 360, 893–896 (2018). 48. Gmitra, M., Konschuh, S., Ertler, C., Ambrosch-Draxl, C. & Fabian, J. Band-structure topologies of graphene: Spin-orbit coupling effects from first principles. Phys. Rev. B 80, 235431 (2009). 49. Sichau, J. et al. Resonance Microwave Measurements of an Intrinsic Spin-Orbit Coupling Gap in Graphene: A Possible Indication of a Topological State. Phys. Rev. Lett. 122, 046403 (2019). 50. Kochan, D., Gmitra, M. & Fabian, J. Spin Relaxation Mechanism in Graphene: Resonant Scattering by Magnetic Impurities. Phys. Rev. Lett. 112, 116602 (2014). 51. Kochan, D., Irmer, S., Gmitra, M. & Fabian, J. Resonant Scattering by Magnetic Impurities as a Model for Spin Relaxation in Bilayer Graphene. Phys. Rev. Lett. 115, 196601 (2015). 52. Tuan, D. V., Ortmann, F., Soriano, D., Valenzuela, S. O. & Roche, S. Pseudospin-driven spin relaxation mechanism in graphene. Nat. Phys. 10, 857–863 (2014). 53. Cummings, A. W. & Roche, S. Effects of Dephasing on Spin Lifetime in Ballistic Spin-Orbit Materials. Phys. Rev. Lett. 116, 086602 (2016). 28 54. Ochoa, H., Castro Neto, A. H. & Guinea, F. Elliot-Yafet Mechanism in Graphene. Phys. Rev. Lett. 108, 206808 (2012). 55. Zhang, P. & Wu, M. W. Electron spin relaxation in graphene with random Rashba field: comparison of the D\textquotesingleyakonov–Perel\textquotesingle and Elliott–Yafet-like mechanisms. New J. Phys. 14, 033015 (2012). 56. Fabian, J., Matos-Abiague, A., Ertler, C. & Zutic, Igor. Semiconductor Spintronics. Acta Physica Slovaca vol. 57 565–907 (2007). 57. Vila, M. et al. Nonlocal Spin Dynamics in the Crossover from Diffusive to Ballistic Transport. Phys. Rev. Lett. 124, 196602 (2020). 58. Guimarães, M. H. D. et al. Controlling Spin Relaxation in Hexagonal BN-Encapsulated Graphene with a Transverse Electric Field. Phys. Rev. Lett. 113, 086602 (2014). 59. Raes, B. et al. Determination of the spin-lifetime anisotropy in graphene using oblique spin precession. Nat. Commun. 7, 11444 (2016). 60. Benítez, L. A. et al. Investigating the spin-orbit interaction in van der Waals heterostructures by means of the spin relaxation anisotropy. APL Mater. 7, 120701 (2019). 61. Raes, B. et al. Spin precession in anisotropic media. Phys. Rev. B 95, 085403 (2017). 62. Ringer, S. et al. Measuring anisotropic spin relaxation in graphene. Phys. Rev. B 97, 205439 (2018). 63. Behin-Aein, B., Datta, D., Salahuddin, S. & Datta, S. Proposal for an all-spin logic device with built-in memory. Nat. Nanotechnol. 5, 266–270 (2010). 64. Sierra, J. F., Neumann, I., Costache, M. V. & Valenzuela, S. O. Hot-Carrier Seebeck Effect: Diffusion and Remote Detection of Hot Carriers in Graphene. Nano Lett. 15, 4000–4005 (2015). 65. Abdelouahed, S., Ernst, A., Henk, J., Maznichenko, I. V. & Mertig, I. Spin-split electronic states in graphene: Effects due to lattice deformation, Rashba effect, and adatoms by first principles. Phys. Rev. B 82, 125424 (2010). 66. Konschuh, S., Gmitra, M., Kochan, D. & Fabian, J. Theory of spin-orbit coupling in bilayer graphene. Phys. Rev. B 85, 115423 (2012). 29 67. Banszerus, L. et al. Observation of the Spin-Orbit Gap in Bilayer Graphene by One-Dimensional Ballistic Transport. Phys. Rev. Lett. 124, 177701 (2020). 68. Qi, J., Li, X., Niu, Q. & Feng, J. Giant and tunable valley degeneracy splitting in ${\mathrm{MoTe}}_{2}$. Phys. Rev. B 92, 121403 (2015). 69. Scharf, B., Xu, G., Matos-Abiague, A. & Žutić, I. Magnetic Proximity Effects in Transition-Metal Dichalcogenides: Converting Excitons. Phys. Rev. Lett. 119, 127403 (2017). 70. Gmitra, M., Kochan, D., Högl, P. & Fabian, J. Trivial and inverted Dirac bands and the emergence of quantum spin Hall states in graphene on transition-metal dichalcogenides. Phys. Rev. B 93, 155104 (2016). 71. Zollner, K. & Fabian, J. Heterostructures of Graphene and Topological Insulators Bi2Se3, Bi2Te3, and Sb2Te3. Phys. Status Solidi B n/a, 2000081. 72. Zollner, K. et al. Scattering-induced and highly tunable by gate damping-like spin-orbit torque in graphene doubly proximitized by two-dimensional magnet ${\mathrm{Cr}}_{2}{\mathrm{Ge}}_{2}{\mathrm{Te}}_{6}$ and monolayer ${\mathrm{WS}}_{2}$. Phys. Rev. Res. 2, 043057 (2020). 73. Högl, P. et al. Quantum Anomalous Hall Effects in Graphene from Proximity-Induced Uniform and Staggered Spin-Orbit and Exchange Coupling. Phys. Rev. Lett. 124, 136403 (2020). 74. Zollner, K., Gmitra, M. & Fabian, J. Heterostructures of graphene and hBN: Electronic, spin- orbit, and spin relaxation properties from first principles. Phys. Rev. B 99, 125151 (2019). 75. Kochan, D., Irmer, S. & Fabian, J. Model spin-orbit coupling Hamiltonians for graphene systems. Phys. Rev. B 95, 165415 (2017). 76. Rossi, E. & Triola, C. Van Der Waals Heterostructures with Spin-Orbit Coupling. Ann. Phys. 532, 1900344 (2020). 77. Li, Y. & Koshino, M. Twist-angle dependence of the proximity spin-orbit coupling in graphene on transition-metal dichalcogenides. Phys. Rev. B 99, 075438 (2019). 78. David, A. Induced spin-orbit coupling in twisted graphene–transition metal dichalcogenide heterobilayers: Twistronics meets spintronics. Phys. Rev. B 100, (2019). 30 79. Alsharari, A. M., Asmar, M. M. & Ulloa, S. E. Topological phases and twisting of graphene on a dichalcogenide monolayer. Phys. Rev. B 98, 195129 (2018). 80. Wang, T., Bultinck, N. & Zaletel, M. P. Flat-band topology of magic angle graphene on a transition metal dichalcogenide. Phys. Rev. B 102, 235146 (2020). 81. Song, K. et al. Spin Proximity Effects in Graphene/Topological Insulator Heterostructures. Nano Lett. 18, 2033–2039 (2018). 82. Zollner, K. & Fabian, J. Single and bilayer graphene on the topological insulator ${\mathrm{Bi}}_{2}{\mathrm{Se}}_{3}$: Electronic and spin-orbit properties from first principles. Phys. Rev. B 100, 165141 (2019). 83. Wang, Z. et al. Strong interface-induced spin–orbit interaction in graphene on WS 2. Nat. Commun. 6, 8339 (2015). 84. Wang, Z. et al. Origin and Magnitude of `Designer’ Spin-Orbit Interaction in Graphene on Semiconducting Transition Metal Dichalcogenides. Phys. Rev. X 6, 041020 (2016). 85. Yang, B. et al. Tunable spin–orbit coupling and symmetry-protected edge states in graphene/WS 2. 2D Mater. 3, 031012 (2016). 86. Völkl, T. et al. Magnetotransport in heterostructures of transition metal dichalcogenides and graphene. Phys. Rev. B 96, 125405 (2017). 87. Zihlmann, S. et al. Large spin relaxation anisotropy and valley-Zeeman spin-orbit coupling in WSe2/graphene/h-BN heterostructures. Phys. Rev. B 97, 075434 (2018). 88. Island, J. O. et al. Spin–orbit-driven band inversion in bilayer graphene by the van der Waals proximity effect. Nature 571, 85–89 (2019). 89. Wu, Y. et al. Néel-type skyrmion in WTe 2 /Fe 3 GeTe 2 van der Waals heterostructure. Nat. Commun. 11, 3860 (2020). 90. Frank, T., Högl, P., Gmitra, M., Kochan, D. & Fabian, J. Protected Pseudohelical Edge States in ${\mathbb{Z}}_{2}$-Trivial Proximitized Graphene. Phys. Rev. Lett. 120, 156402 (2018). 91. Kane, C. L. & Mele, E. J. Quantum Spin Hall Effect in Graphene. Phys. Rev. Lett. 95, 226801 (2005). 31 92. Alsharari, A. M., Asmar, M. M. & Ulloa, S. E. Proximity-induced topological phases in bilayer graphene. Phys. Rev. B 97, 241104 (2018). 93. Tiwari, P., Srivastav, S. K., Ray, S., Das, T. & Bid, A. Observation of Time-Reversal Invariant Helical Edge-Modes in Bilayer Graphene/WSe2 Heterostructure. ACS Nano (2020) doi:10.1021/acsnano.0c07524. 94. Wang, H. et al. Above Room-Temperature Ferromagnetism in Wafer-Scale Two-Dimensional van der Waals Fe3GeTe2 Tailored by a Topological Insulator. ACS Nano 14, 10045–10053 (2020). 95. Matsuoka, H. et al. Spin–Orbit-Induced Ising Ferromagnetism at a van der Waals Interface. Nano Lett. 21, 1807–1814 (2021). 96. Fert, A., Reyren, N. & Cros, V. Magnetic skyrmions: advances in physics and potential applications. Nat. Rev. Mater. 2, 1–15 (2017). 97. Dankert, A. & Dash, S. P. Electrical gate control of spin current in van der Waals heterostructures at room temperature. Nat. Commun. 8, 16093 (2017). 98. Cummings, A. W., Garcia, J. H., Fabian, J. & Roche, S. Giant Spin Lifetime Anisotropy in Graphene Induced by Proximity Effects. Phys. Rev. Lett. 119, 206601 (2017). 99. Offidani, M. & Ferreira, A. Microscopic theory of spin relaxation anisotropy in graphene with proximity-induced spin-orbit coupling. Phys. Rev. B 98, 245408 (2018). 100. Garcia, J. H., Vila, M., Cummings, A. W. & Roche, S. Spin transport in graphene/transition metal dichalcogenide heterostructures. Chem. Soc. Rev. 47, 3359–3379 (2018). 101. Xu, J., Zhu, T., Luo, Y. K., Lu, Y.-M. & Kawakami, R. K. Strong and Tunable Spin-Lifetime Anisotropy in Dual-Gated Bilayer Graphene. Phys. Rev. Lett. 121, 127703 (2018). 102. Leutenantsmeyer, J. C., Ingla-Aynés, J., Fabian, J. & van Wees, B. J. Observation of Spin- Valley-Coupling-Induced Large Spin-Lifetime Anisotropy in Bilayer Graphene. Phys. Rev. Lett. 121, 127702 (2018). 103. Sinova, J., Valenzuela, S. O., Wunderlich, J., Back, C. H. & Jungwirth, T. Spin Hall effects. Rev. Mod. Phys. 87, 1213–1260 (2015). 32 104. Manipatruni, S. et al. Scalable energy-efficient magnetoelectric spin–orbit logic. Nature 565, 35–42 (2019). 105. MacNeill, D. et al. Control of spin–orbit torques through crystal symmetry in WTe 2 /ferromagnet bilayers. Nat. Phys. 13, 300–305 (2017). 106. Shi, S. et al. All-electric magnetization switching and Dzyaloshinskii–Moriya interaction in WTe 2 /ferromagnet heterostructures. Nat. Nanotechnol. 14, 945–949 (2019). 107. Liang, S. et al. Spin-Orbit Torque Magnetization Switching in MoTe2/Permalloy Heterostructures. Adv. Mater. n/a, 2002799. 108. Dushenko, S. et al. Gate-Tunable Spin-Charge Conversion and the Role of Spin-Orbit Interaction in Graphene. Phys. Rev. Lett. 116, 166102 (2016). 109. Safeer, C. K. et al. Room-Temperature Spin Hall Effect in Graphene/MoS2 van der Waals Heterostructures. Nano Lett. 19, 1074–1082 (2019). 110. Benítez, L. A. et al. Tunable room-temperature spin galvanic and spin Hall effects in van der Waals heterostructures. Nat. Mater. 19, 170–175 (2020). 111. Ghiasi, T. S., Kaverzin, A. A., Blah, P. J. & van Wees, B. J. Charge-to-Spin Conversion by the Rashba–Edelstein Effect in Two-Dimensional van der Waals Heterostructures up to Room Temperature. Nano Lett. 19, 5959–5966 (2019). 112. Yan, W. et al. Large room temperature spin-to-charge conversion signals in a few-layer graphene/Pt lateral heterostructure. Nat. Commun. 8, 661 (2017). 113. Torres, W. S. et al. Spin precession and spin Hall effect in monolayer graphene/Pt nanostructures. 2D Mater. 4, 041008 (2017). 114. Garcia, J. H., Cummings, A. W. & Roche, S. Spin Hall Effect and Weak Antilocalization in Graphene/Transition Metal Dichalcogenide Heterostructures. Nano Lett. 17, 5078–5083 (2017). 115. Offidani, M., Milletarì, M., Raimondi, R. & Ferreira, A. Optimal Charge-to-Spin Conversion in Graphene on Transition-Metal Dichalcogenides. Phys. Rev. Lett. 119, 196801 (2017). 116. Milletarì, M., Offidani, M., Ferreira, A. & Raimondi, R. Covariant Conservation Laws and the Spin Hall Effect in Dirac-Rashba Systems. Phys. Rev. Lett. 119, 246801 (2017). 33 117. Li, L. et al. Gate-Tunable Reversible Rashba–Edelstein Effect in a Few-Layer Graphene/2H- TaS2 Heterostructure at Room Temperature. ACS Nano 14, 5251–5259 (2020). 118. Herling, F. et al. Gate tunability of highly efficient spin-to-charge conversion by spin Hall effect in graphene proximitized with WSe2. APL Mater. 8, 071103 (2020). 119. Khokhriakov, D., Hoque, A. M., Karpiak, B. & Dash, S. P. Gate-tunable spin-galvanic effect in graphene-topological insulator van der Waals heterostructures at room temperature. Nat. Commun. 11, 3657 (2020). 120. Avsar, A. et al. Spin–orbit proximity effect in graphene. Nat. Commun. 5, 4875 (2014). 121. Van Tuan, D. et al. Spin Hall Effect and Origins of Nonlocal Resistance in Adatom-Decorated Graphene. Phys. Rev. Lett. 117, 176602 (2016). 122. Ribeiro, M., Power, S. R., Roche, S., Hueso, L. E. & Casanova, F. Scale-invariant large nonlocality in polycrystalline graphene. Nat. Commun. 8, 2198 (2017). 123. Völkl, T. et al. Absence of a giant spin Hall effect in plasma-hydrogenated graphene. Phys. Rev. B 99, 085401 (2019). 124. Safeer, C. K. et al. Large Multidirectional Spin-to-Charge Conversion in Low-Symmetry Semimetal MoTe2 at Room Temperature. Nano Lett. 19, 8758–8766 (2019). 125. Li, P. et al. Spin-momentum locking and spin-orbit torques in magnetic nano- heterojunctions composed of Weyl semimetal WTe 2. Nat. Commun. 9, 3990 (2018). 126. Zhao, B. et al. Unconventional Charge–Spin Conversion in Weyl-Semimetal WTe2. Adv. Mater. 32, 2000818 (2020). 127. Yang, H. X. et al. Proximity Effects Induced in Graphene by Magnetic Insulators: First- Principles Calculations on Spin Filtering and Exchange-Splitting Gaps. Phys. Rev. Lett. 110, 046603 (2013). 128. Hallal, A., Ibrahim, F., Yang, H., Roche, S. & Chshiev, M. Tailoring magnetic insulator proximity effects in graphene: first-principles calculations. 2D Mater. 4, 025074 (2017). 129. Zollner, K., Gmitra, M., Frank, T. & Fabian, J. Theory of proximity-induced exchange coupling in graphene on hBN/(Co, Ni). Phys. Rev. B 94, 155441 (2016). 34 130. Lazić, P., Belashchenko, K. D. & Žutić, I. Effective gating and tunable magnetic proximity effects in two-dimensional heterostructures. Phys. Rev. B 93, 241401 (2016). 131. Ibrahim, F. et al. Unveiling multiferroic proximity effect in graphene. 2D Mater. 7, 015020 (2019). 132. Wang, Z., Tang, C., Sachs, R., Barlas, Y. & Shi, J. Proximity-Induced Ferromagnetism in Graphene Revealed by the Anomalous Hall Effect. Phys Rev Lett 114, 016603 (2015). 133. Wei, P. et al. Strong interfacial exchange field in the graphene/EuS heterostructure. Nat. Mater. 15, 711–716 (2016). 134. Leutenantsmeyer, J. C., Kaverzin, A. A., Wojtaszek, M. & van Wees, B. J. Proximity induced room temperature ferromagnetism in graphene probed with spin currents. 2D Mater 4, 014001 (2017). 135. Singh, S. et al. Strong Modulation of Spin Currents in Bilayer Graphene by Static and Fluctuating Proximity Exchange Fields. Phys. Rev. Lett. 118, 187201 (2017). 136. Tang, C. et al. Approaching quantum anomalous Hall effect in proximity-coupled YIG/graphene/h-BN sandwich structure. APL Mater. 6, 026401 (2017). 137. Asshoff, P. U. et al. Magnetoresistance of vertical Co-graphene-NiFe junctions controlled by charge transfer and proximity-induced spin splitting in graphene. 2D Mater. 4, 031004 (2017). 138. Ghiasi, T. S. et al. Electrical and Thermal Generation of Spin Currents by Magnetic Graphene. ArXiv200715597 Cond-Mat (2020). 139. Wu, Y. et al. Large exchange splitting in monolayer graphene magnetized by an antiferromagnet. Nat. Electron. 3, 604–611 (2020). 140. Ghazaryan, D. et al. Magnon-assisted tunnelling in van der Waals heterostructures based on CrBr 3. Nat. Electron. 1, 344–349 (2018). 141. Zhao, C. et al. Enhanced valley splitting in monolayer WSe 2 due to magnetic exchange field. Nat. Nanotechnol. 12, 757–762 (2017). 142. Norden, T. et al. Giant valley splitting in monolayer WS 2 by magnetic proximity effect. Nat. Commun. 10, 4163 (2019). 35 143. Zhong, D. et al. Van der Waals engineering of ferromagnetic semiconductor heterostructures for spin and valleytronics. Sci. Adv. 3, e1603113 (2017). 144. Zhong, D. et al. Layer-resolved magnetic proximity effect in van der Waals heterostructures. Nat. Nanotechnol. 15, 187–191 (2020). 145. Zollner, K., Faria Junior, P. E. & Fabian, J. Proximity exchange effects in ${\mathrm{MoSe}}_{2}$ and ${\mathrm{WSe}}_{2}$ heterostructures with ${\mathrm{CrI}}_{3}$: Twist angle, layer, and gate dependence. Phys. Rev. B 100, 085128 (2019). 146. Scharf, B., Xu, G., Matos-Abiague, A. & Žutić, I. Magnetic Proximity Effects in Transition- Metal Dichalcogenides: Converting Excitons. Phys. Rev. Lett. 119, 127403 (2017). 147. Lyons, T. P. et al. Interplay between spin proximity effect and charge-dependent exciton dynamics in MoSe 2 /CrBr 3 van der Waals heterostructures. Nat. Commun. 11, 6021 (2020). 148. Bhattacharyya, S. et al. Recent Progress in Proximity Coupling of Magnetism to Topological Insulators. ArXiv201211248 Cond-Mat (2020). 149. Jiang, Z. et al. Independent Tuning of Electronic Properties and Induced Ferromagnetism in Topological Insulators with Heterostructure Approach. Nano Lett. 15, 5835–5840 (2015). 150. Tang, C. et al. Above 400-K robust perpendicular ferromagnetic phase in a topological insulator. Sci. Adv. 3, e1700307 (2017). 151. Mogi, M. et al. Current-induced switching of proximity-induced ferromagnetic surface states in a topological insulator. Nat. Commun. 12, 1404 (2021). 152. Katmis, F. et al. A high-temperature ferromagnetic topological insulating phase by proximity coupling. Nature 533, 513–516 (2016). 153. Krieger, J. A. et al. Do topology and ferromagnetism cooperate at the $\mathrm{EuS}/{\mathrm{Bi}}_{2}{\mathrm{Se}}_{3}$ interface? Phys. Rev. B 99, 064423 (2019). 36 154. Pereira, V. M. et al. Topological insulator interfaced with ferromagnetic insulators: $\mathrm{B}{\mathrm{i}}_{2}\mathrm{T}{\mathrm{e}}_{3}$ thin films on magnetite and iron garnets. Phys. Rev. Mater. 4, 064202 (2020). 155. Figueroa, A. I. et al. Absence of Magnetic Proximity Effect at the Interface of ${\mathrm{Bi}}_{2}{\mathrm{Se}}_{3}$ and $(\mathrm{Bi},\mathrm{Sb}{)}_{2}{\mathrm{Te}}_{3}$ with EuS. Phys. Rev. Lett. 125, 226801 (2020). 156. Watanabe, R. et al. Quantum anomalous Hall effect driven by magnetic proximity coupling in all-telluride based heterostructure. Appl. Phys. Lett. 115, 102403 (2019). 157. Fei, Z. et al. Edge conduction in monolayer WTe 2. Nat. Phys. 13, 677–682 (2017). 158. Tang, S. et al. Quantum spin Hall state in monolayer 1T’-WTe2. Nat. Phys. 13, 683–687 (2017). 159. Garcia, J. H. et al. Canted Persistent Spin Texture and Quantum Spin Hall Effect in ${\mathrm{WTe}}_{2}$. Phys. Rev. Lett. 125, 256603 (2020). 160. Zhao, W. et al. Determination of the helical edge and bulk spin axis in quantum spin Hall insulator WTe2. ArXiv201009986 Cond-Mat (2020). 161. Tan, C. et al. Determination of the spin orientation of helical electrons in monolayer WTe2. ArXiv201015717 Cond-Mat (2020). 162. Zhao, W. et al. Magnetic proximity and nonreciprocal current switching in a monolayer WTe 2 helical edge. Nat. Mater. 19, 503–507 (2020). 163. Alghamdi, M. et al. Highly Efficient Spin–Orbit Torque and Switching of Layered Ferromagnet Fe3GeTe2. Nano Lett. 19, 4400–4405 (2019). 164. Zhao, B. et al. Unconventional Charge–Spin Conversion in Weyl-Semimetal WTe2. Adv. Mater. n/a, 2000818. 165. Dolui, K. et al. Proximity Spin–Orbit Torque on a Two-Dimensional Magnet within van der Waals Heterostructure: Current-Driven Antiferromagnet-to-Ferromagnet Reversible Nonequilibrium Phase Transition in Bilayer CrI3. Nano Lett. 20, 2288–2295 (2020). 37 166. Gmitra, M. & Fabian, J. Proximity Effects in Bilayer Graphene on Monolayer ${\mathrm{WSe}}_{2}$: Field-Effect Spin Valley Locking, Spin-Orbit Valve, and Spin Transistor. Phys. Rev. Lett. 119, 146401 (2017). 167. Khoo, J. Y., Morpurgo, A. F. & Levitov, L. On-Demand Spin–Orbit Interaction from Which- Layer Tunability in Bilayer Graphene. Nano Lett. 17, 7003–7008 (2017). 168. Michetti, P., Recher, P. & Iannaccone, G. Electric Field Control of Spin Rotation in Bilayer Graphene. Nano Lett. 10, 4463–4469 (2010). 169. Zollner, K., Gmitra, M. & Fabian, J. Electrically tunable exchange splitting in bilayer graphene on monolayer Cr2X2Te6 with X = Ge, Si, and Sn. New J. Phys. 20, 073007 (2018). 170. Cardoso, C., Soriano, D., García-Martínez, N. A. & Fernández-Rossier, J. Van der Waals Spin Valves. Phys. Rev. Lett. 121, 067701 (2018). 171. Liu, C.-X., Zhang, S.-C. & Qi, X.-L. The Quantum Anomalous Hall Effect: Theory and Experiment. Annu. Rev. Condens. Matter Phys. 7, 301–321 (2016). 172. Zhang, J., Zhao, B., Yao, Y. & Yang, Z. Robust quantum anomalous Hall effect in graphene- based van der Waals heterostructures. Phys. Rev. B 92, 165418 (2015). 173. Zollner, K., Gmitra, M. & Fabian, J. Swapping Exchange and Spin-Orbit Coupling in 2D van der Waals Heterostructures. Phys. Rev. Lett. 125, 196402 (2020). 174. Chen, W. et al. Direct observation of van der Waals stacking–dependent interlayer magnetism. Science 366, 983–987 (2019). 175. Xing, W. et al. Magnon Transport in Quasi-Two-Dimensional van der Waals Antiferromagnets. Phys. Rev. X 9, 011026 (2019). 176. Liu, T. et al. Spin caloritronics in a CrBr3 magnetic van der Waals heterostructure. Phys. Rev. B 101, 205407 (2020). 177. Yang, M. et al. Creation of skyrmions in van der Waals ferromagnet Fe3GeTe2 on (Co/Pd)n superlattice. Sci. Adv. 6, eabb5157 (2020). 178. Ding, B. et al. Observation of Magnetic Skyrmion Bubbles in a van der Waals Ferromagnet Fe3GeTe2. Nano Lett. 20, 868–873 (2020). 38 179. Meijer, M. J. et al. Chiral Spin Spirals at the Surface of the van der Waals Ferromagnet Fe3GeTe2. Nano Lett. 20, 8563–8568 (2020). 180. Liang, J. et al. Very large Dzyaloshinskii-Moriya interaction in two-dimensional Janus manganese dichalcogenides and its application to realize skyrmion states. Phys. Rev. B 101, 184401 (2020). 181. Zhang, L. et al. Proximity-Coupling-Induced Significant Enhancement of Coercive Field and Curie Temperature in 2D van der Waals Heterostructures. Adv. Mater. 32, 2002032 (2020). 182. Giustino, F. et al. The 2020 Quantum Materials Roadmap. J. Phys. Mater. (2020) doi:10.1088/2515-7639/abb74e. Acknowledgements We thank D. Torres (ICN2) for help in implementing the 3D device models used in the figures. J.F.S. and S.O.V. acknowledge support of the European Union's Horizon 2020 FET-PROACTIVE project TOCHA under grant agreement 824140, the King Abdullah University of Science and Technology (KAUST) through award OSR-2018-CRG7-3717, and MINECO under contracts no. PID2019-111773RB-I00/AEI/10.13039/501100011033 and no. SEV-2017-0706 Severo Ochoa. J.F., S.R. and S.O.V. acknowledge support of the European Union Horizon 2020 research and innovation program under contract 881603 (Graphene Flagship) and J.F. of the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) SFB1277 (project-id 314695032). R.K.K. acknowledges support of US DOE-BES (DE-SC0016379), AFOSR MURI 2D MAGIC (FA9550- 19-1-0390), and NSF MRSEC (DMR-2011876). Competing Interests The authors declare no competing interests Additional Information 39 Reprints and permission information is available online at www.nature.com/reprints. Correspondence should be addressed to J.F.S. or S.O.V. 40