Exploiting the spin degree of freedom to store and manipulate information provides a paradigm for future microelectronics. The development of van der Waals (vdW) heterostructures has created a fascinating platform for exploring spintronic properties in the two-dimensional (2D) limit. Transition-metal dichalcogenides such as tungsten diselenide (WSe2) have electronic band structures that are ideal for hosting many exotic spin–orbit phenomena. Here, we report the spin-filtering effect in all-vdW heterostructures with WSe2 barrier. Combining 2D-perpendicular magnetic anisotropy Fe3GeTe2 (FGT) with different thicknesses of WSe2, the FGT/WSe2/FGT spin valve shows distinct charge and spin transport behavior. Moreover, the negative magnetoresistance (−4.3%) could be inverted into positive magnetoresistance (up to +25.8%) with decreasing the WSe2 thickness. Furthermore, we proposed a spin-filtering model based on Δ-symmetry electrons tunneling to explain the crossover from negative to positive MR signal through ab initio calculation. These experimental and theoretical results illustrate the rich potential of the families of TMDC materials to control spin currents in 2D spintronic devices.
Transition-metal dichalcogenide (TMDC) semiconductors are layered materials with a two-dimensional honeycomb crystal structure. These materials offer an opportunity to integrate heterogeneous systems due to the weak van der Waals (vdW) interactions between individual layers, without dangling bonds on the surfaces1,2. Moreover, TMDCs are semiconductors with a strong spin–orbit coupling (SOC), several 10 meV in the conduction band and 100 meV in the valence band3,4,5. Especially, the maximal spin splitting of the uppermost valence band (UVB) at the K-point amounts to 456 meV for the WSe2 monolayer4. Recently, such high-SOC TMDCs also showed fascinating charge transport and spin dynamics properties6,7,8,9,10,11. Furthermore, the SOC in such vdW heterostructures can be utilized to electrically control the spin polarization. However, the use of ferromagnetic film in the vertical spintronic device is now explored for few decades, which are mainly fabricated by physical growth methods such as molecular beam epitaxy, sputtering (d.c. or a.c.), pulsed laser deposition, or electron beam evaporation. Interface roughness results in local variations of the barrier thickness, local magnetic fringe field, and inhomogeneous electronic configuration and coupling at the interfaces that may strongly affect the spin-dependent conductance12,13,14.
Meanwhile, with the recent discovery of 2D magnets, the control of the spin degree of freedom can be integrated to realize 2D spintronics with spontaneous time-reversal symmetry breaking15,16. However, in situ preparation of all vdW heterostructures in a controlled environment could circumvent most of the problems mentioned above. Owing to the ideal interface between 2D materials, the junctions incorporating all vdW materials avoid the associated intermixing effect and defect-induced gap states, showing performance exceeding that of covalently bonded magnetic multilayers17,18,19,20,21. The expanding 2D materials family is now the focus of several on-going vertical spin transport studies, taking advantage of, for instance, tunneling (h-BN, etc.)19, semiconducting (MoS222, WS223, InSe10, etc.), and magnetic (CrI324, CrBr325, etc.) behaviors. Theoretical studies further suggest that the nature of different nonmagnetic 2D spacers can effectively possess tunneling or metallic spin transport behaviors26. Despite that a few of spin valve devices with TMDCs including WSe2 as the nonmagnetic spacer layer have been reported22,23,27, WSe2-based all-vdWs spin valve remains a relatively unexplored area. Herein, for the WSe2-based all-vdWs Fe3GeTe2(FGT)/WSe2/FGT structure, the distinct charge and spin transport characteristics were observed as a function of WSe2 thicknesses, and more importantly, a spin-filtering physical model based on the Δ-symmetry electrons tunneling was proposed to explain the negative MR based on ab initio calculations.
Here, we have demonstrated the charge and spin-dependent transport behaviors in all-vdW FGT/WSe2/FGT spin valve devices, and the spin-filtering effect is observed and explained by ab initio calculation. We find that the charge and spin transport show distinct behaviors in the spin valve devices with varying thicknesses of WSe2. In particular, the crossover of magnetoresistance (MR) has been found. For a typical spin valve, the MR can be defined as MR = (Rap − Rp)/Rp, where Rap and Rp denote the resistance in the antiparallel (AP) and parallel (P) configurations of ferromagnetic electrodes magnetization direction, respectively. The sign of the MR changes from negative to positive, and the value changes from −4.3% to +25.8% with the decreasing thicknesses of WSe2. Supported by ab initio calculations, we proposed the spin-filtering model based on the Δ-symmetry electrons tunneling to illustrate the negative MR physical mechanism. The estimated spin polarization up to 33.8% at the FGT/WSe2 interface is obtained with 10 nA at 10 K, corresponding to the highest values reported to date based on WSe2-based spin valves. The evolution of the WSe2 with spin-filter function is thus shown to lead to promising perspectives for tailoring spintronics devices using TMDC materials.
Device structure and characterization
Figure 1a shows the schematic crystal structure geometry of FGT. It is worth noting that the 2D ferromagnetic FGT crystals exhibit relatively large Hc, high Tc (~220 K), and strong PMA17,28,29. Figure 1b shows the side view of the trigonal prismatic coordination of the atoms in 2D semiconducting WSe2. The layer-to-layer vdWs gap is about 12.98 Å, and the electronic structure of the top-lying valence bands of WSe2 is derived from the W 5d and Se 4p orbitals30. The schematic of the spin valve structure based on all-vdWs materials and the experimental set-up are displayed in Fig. 1c. The device consists of two ferromagnetic FGT electrodes and a nonmagnetic WSe2 spacer. The magneto-transport measurements were performed using two-terminal measurements with an out-of-plane magnetic field applied parallel to the easy axis of FGT. Figure 1d shows the optical image of FGT/WSe2/FGT spin valve, and every layer of the device has been marked by dotted lines. In addition, the high-resolution interface between FGT and WSe2 characterized by scanning transmission electron microscope (STEM) is shown in Fig. 1e. The enlarged STEM image of the FGT/WSe2 clean interface reveals that the vdWs heterostructure-based spin valve does not exist penetration (Fig. 1f).
Electric and magneto-transport of FGT/WSe2/FGT
We first investigated 12 nm WSe2-based spin valve with FGT electrodes. For the FGT/WSe2 (12 nm)/FGT junction (see Supplementary Fig. 1), as shown in Fig. 2a, the spin valve shows a symmetric and nonlinear I–V curve at 10 K, which points to typical tunneling behavior. As shown in Fig. 2b, a representative spin valve signal with 10 nA bias current at 10 K was observed, and the inset shows the direct V–H curve of the device. A sharp and large decrease in resistance was observed close to μ0H = +0.19 T, followed by the same magnitude increase at μ0H = +0.33 T with sweeping the out-of-plane magnetic field μ0H from negative to positive orientation. In addition, an analogous decrease and a subsequent increase were also observed at μ0H = −0.19 T and −0.33 T as the magnetic field was swept back. In this situation, a negative MR value of −4.3% was obtained, that is to say that the resistance at antiparallel alignment was smaller than the resistance at parallel state. The magnitude of the observed MR is about four times higher than the MR value of the previously reported WSe2-based spin valve27, and this value is also much higher than WS2- and MoS2-based spin valves22,23.
The bias current intensity dependence of the MR at 10 K was investigated. As shown in Fig. 2c, the extracted MR decreases from −4.3 to −0.3% approximately in a linear way in the bias current ranging from 10 nA to 1 µA. The decrease of the MR with increasing the bias current is related to the higher energy electrons, and this has a larger probability to be scattered by localized trap states in the WSe2 layer or at the interface31. In order to understand the temperature dependence of the MR in the WSe2 spin valve devices, we performed the device measurements from 10 to 200 K. The extracted MR values are presented in Fig. 2d, where the MR decreases from −4.3% to −0.8% monotonically with increasing the temperature from 10 to 200 K. This can be attributed to the decrease of the spin polarization of FGT with increasing the temperature, which becomes fully un-polarized above the Curie temperature of FGT (see Supplementary Fig. 2).
Notably, the MR behaviors at different currents and temperatures show standard rectangle configuration compared to the reported WSe2-based devices using traditional ferromagnetic electrodes27, indicating that the resistance is very stable when the spin valve system lies in the parallel and antiparallel state, and the all-vdW spin valve devices have an advantage of better spinterface than the common interface between depositing ferromagnetic metal and 2D materials.
After considering the spin scattering in the WSe2 layer, the spin polarization of the device can be approximately calculated using Julliere model, MR = 2P2/(1-P2)22. The spin polarization P of FGT can be estimated to be 14.5% at 10 K and 6.3% at 200 K. The maximum P (14.5%) is larger than that of the Co (7.1%)32, NiFe (4.9%)33, and Fe3O4 (3.2%)34 which sandwiched 2D material, and is also larger than that of the FGT/graphite/FGT20, FGT/MoS2/FGT22, and FGT/WS2/FGT23 junctions. However, this value is still much lower than that in FGT/InSe/FGT10 and FGT/h-BN/FGT19 structure. A logical extension of Bloch’s spin wave theory is to adopt a spin polarization that varies with temperature as does the magnetization35, and the spin polarization can be described as P(T) = P0(1 − αT3/2). The parameter α is related to concrete materials and fitted to be α = 1.99 × 10−4 K−3/2, and which is close to the values as reported (see Supplementary Fig. 3)22,27,29,36.
After the measurements of spin valve with 12 nm WSe2 spacer, we characterized the other spin valves with different thicknesses of WSe2. However, the charge and spin transport show distinct behaviors while the thickness of WSe2 is below than 9.6 nm. In short, a strong thickness dependence of the charge and spin signal is observed, with a striking MR sign reversal for WSe2 layers less than 9.6 nm. Herein, we chose the spin valve with a thickness of 6.9 nm WSe2 as the example. For the FGT/WSe2 (6.9 nm)/FGT spin valve (see Supplementary Fig. 4), the symmetric I–V curve shows a linear feature as shown in Fig. 3a. Figure 3b shows a typical MR trace for 6.9 nm WSe2 layer spin valve at a constant current of 10 nA at 10 K. Different to the thicker spacer layer device, the spin valve shows positive MR, namely the resistance at antiparallel state was larger than that at parallel state. Figure 3c shows that the spin valve MR values decrease from 25 to 13% at various current intensities ranging from 5 nA to 50 µA at 10 K. As observed similarly with 12 nm WSe2 spin valve, the extracted MR decreases as the bias current intensity increases. As shown in Fig. 3d, the magnitude of the MR signal decays steeply with increasing temperature, decreasing from 25% to 3.3% (see Supplementary Fig. 5). In addition, the symmetrical interface for spin transport based on FGT/WSe2/FGT junction has been found through reversing the applied bias current direction, and which has very little influence on the magnitude of MR (see the Fig. 4a, b). For the spinterface-based FGT/WSe2, we calculated the spin polarization P, and the value varies 33.8 to 13% from 10 to 200 K. In addition, the parameter α with Bloch’s law based on 6.9 nm WSe2 spin valve is about α = 2.3 × 10−4 K−3/2, which is much larger than that obtained from 12 nm WSe2 device (see Supplementary Fig. 6)22,29.
To clarify the WSe2 spin valve thickness-dependent charge and spin transport, we fabricated more than ten devices and studied their charge and spin transport behaviors. Figure 5a shows the summary of spin valve MR and resistance-area products (RA) with changing the thicknesses of the WSe2 spacer. The charge and spin measurement results show that the MR signal changes from positive (left blue section) to negative (right green section) with the increase of the WSe2 thickness, and the spin valves I–V curves show linear and nonlinear I–V characteristics, respectively. Obviously, for the device with WSe2 thickness of less than 9.6 nm, the spin valve shows a positive MR effect, and the RA is less than 5 kΩ μm2. However, when the WSe2 thickness reaches 9.6 nm and above, the MR signal turns into negative and the junction shows tunneling transport behavior. According to the measurement results of all devices, the RA value of right section devices (>9.6 nm) is several dozens of times larger than that of left section devices (<9.6 nm).
The metallic behaviors of vdW heterostructure spin valve are often attributed to the strong hybridization between ferromagnetic electrodes and semiconducting spacer22,23. To understand the hybridization effect and barrier height between FGT and WSe2, the electronic structure and projected density of state (PDOS) of the FGT/WSe2 interface can be obtained based on the DFT theory. For the FGT/WSe2 interface as shown in Fig. 5b, the Fermi energy is close to the conduction band of WSe2, indicating that FGT/WSe2 interface performs n-type contact. In addition, the location of Fermi level is an important factor for the transport of TMDs devices, and which may be tuned by doping37,38,39,40. Meanwhile, from the perspective of the smoothness of PDOS, the WSe2 layer closest to FGT has a significant hybridization effect. However, the CBM of 2nd to 4th layer of WSe2 exhibits weaker hybridization states, and the 5th layer of WSe2 is no longer affected by FGT. For the positive MR effect and linear I–V curve of thin WSe2, the hybridization effect plays a vital, but not the only role. The actual situation may be affected by many factors, such as the synergistic effect combining hybridization and the pinhole10 or impurities. With increasing the thickness of the WSe2 layer, the nonlinear I–V curve shows the formation of the Schottky barrier with larger RA values (above 1 MΩ μm2) at WSe2/FGT interface, and the hybridization and pinhole effect do not play the vital role. In contrast, the tunneling effect dominates the transport mechanism of the devices, and there is the emergence of a negative MR signal.
Model of symmetry-state selection
Indeed, the MR is tightly related to the different symmetry-related carriers filtering for spin-polarized electrons tunneling in spin valve. Based on the experimental characterization of spin value thickness-dependent MR, we proposed a model of symmetry-state selection for the tunneling current in the FGT/WSe2/FGT through ab initio calculation. The bandgap is very distinct for WSe2 with different thicknesses, as shown in Fig. 6a–c, which shows the transition from direct-to-indirect bandgap of WSe2. Herein, the spin–orbital coupling (SOC) effect has been taken into account in the band structure calculation, which leads to the obvious splitting of valence band (see Supplementary Figs. 7 and 8). For monolayer WSe2, the bandgap is about 1.45 eV without considering the hybridization effect. The monolayer WSe2 shows direct bandgap properties (K-K), and the conduction band minimum (CBM) and valence band maximum (VBM) of WSe2 are located at K-point of the Brillouin zone, where the carriers transport along the CBM of K-point with ∆1 state. With the increment of WSe2 layers, the bandgap of the bilayer and trilayer of WSe2 is about 1.37 and 1.29 eV (see Supplementary Fig. 8), switching from direct bandgap (monolayer) to indirect bandgap. As the layers of WSe2 continue increasing, more conductance bands at Q-point and valence band at Γ-points are occurring, and the bandgap decreases as a function of the WSe2 layer. In the case of four layers of WSe2, the CBM at the Q-point and K-point have almost the same energy level, and the carriers transport along with CBM at Q/K-point (Fig. 6b). However, with more additional layers of WSe2 (more than five layers), as shown in Fig. 6c, the minimum of its conduction band shifts toward the Q-point of the Brillouin zone, and the electrons only transport at the CBM of Q-point with three symmetries: ∆1, ∆5, and ∆2’ states. For the FGT/WSe2/FGT spin valve, the transport occurs in multichannels. The ferromagnetic electrode FGT emits electrons with different symmetries, and these electrons are subsequently injected into the transport layer WSe2. Figure 6d shows the band structure of bulk FGT for the spin-up and spin-down electron states. Considering that the spin-polarized electrons propagate along the Q-point based on thick WSe2 layers, we assume that the spin-polarized electrons crossing the Fermi level near the Q-point can be used as an effective transport channel. Therefore, we extracted the spin-dependent band structure of the bulk FGT near the Q-point in the Brillouin region (the shadow part of the FGT band structure). As shown in Fig. 6e and f, it is found that there are two kinds of spin-up channels, one of which is related to ∆5 state, labeled as C1, and the other one is related to ∆1 state, labeled as C2. Moreover, there are also two spin-down channels, labeled C3 and C4, respectively. Both channels contain ∆1, ∆5, ∆2, and ∆2’ states. Based on the ab initio calculations, we proposed a simplified spin-related transport model to explain the negative MR phenomenon based on FGT/WSe2/FGT. Following Fig. 6g and h, for the parallel (P) state, ∆5 state and ∆1 state electrons are injected through the spin-up channels C1 and C2 from the left electrode, respectively. After the electrons tunnel across the WSe2 spacer, the right electrode C1 cannot receive the ∆1 state electrons, and C2 cannot receive the ∆5 state electrons, indicating that multiple channels are not all valid. The electrons tunneling from C1 (C2) of the left electrode to the right electrode channel C2 (C1) is spin forbidden, and this will drastically reduce the tunneling current. Furthermore, for the antiparallel (AP) state, both the ∆1 and ∆5 states injected by the left electrode can find the equivalent state on the right electrode with the opposite spin state, C3 or C4. This shows that all possible transport channels in AP state are not spin forbidden for the thick WSe2 layer-based spin valve, while the two transport channels in P state are spin forbidden, and which causes the tunneling current in AP state to be higher than that in P state. The model proposed here outlines the tunable spin transport properties and spin-filtering effect dependent on the thicknesses of WSe2 spacer. For thicker WSe2 above 15 nm, the MR effect disappeared even at 10 K. It is probable that thick flakes enhance the probability of spin scattering, in turn limiting the extracted spin signal.
Our study explores the spin-filtering effect and tunable spin transport in vertical all-van der Waals spin valves of FGT/WSe2/FGT. We report the MR crossover from negative (−4.3%) to positive (up to +25.8%) with decreasing the thickness of WSe2, and the calculated maximum spin polarization for the spinterface is about 14.5% and 33.8%, which corresponds to negative and positive MR spin valve, respectively, and that is larger than that obtained in conventional ferromagnetic electrodes sandwiching 2D material structures. Furthermore, we proposed a spin-filtering model based on WSe2 by ab initio calculations. The FGT/WSe2 interface spin polarization sign can be expected to be tuned with WSe2 thicknesses, which should be common and adapt to many TMDCs. Our work highlights the potential of the rich TMDC family of 2D materials to control spin transport in 2D spintronic devices.
Materials and device fabrication
All the high-quality vdW bulk single-crystals (FGT, WSe2, h-BN obtained from HQ Graphene) were exfoliated into few layer flakes through the mechanical exfoliation method inside a nitrogen-filled glovebox. Then the flakes of FGT, WSe2, FGT, and h-BN were stacked sequentially on the prepared Cr (5 nm)/Au (50 nm) electrodes by the dry transfer method. The obtained FGT/WSe2/FGT spin valve devices were baked at 120 °C for 10 min to eliminate the air gaps.
Electric and magneto-transport measurements
The electrical and magneto-transport were performed in the cryogenic probe station (Lake Shore Cryotronics, Inc). The instrument operation temperature varies from 10 to 300 K. The superconducting magnet could provide an out-of-plane magnetic field up to 2.25 T. The transport was measured by the combination of Keithley model 2602B sourcemeter and Keithley model 2182 A nanovoltmeter. The thickness of WSe2 flake was determined by an atomic force microscope. The FGT/WSe2/FGT device optical image was obtained by Olympus optical microscope. The FGT/WSe2/FGT device interface image was characterized by aberration-corrected scanning transmission electron microscopy.
The Schottky barrier, band structure, and electron symmetry are carried out by Quantum ATK41 based on the density functional theory. Hexagonal structures are selected for the primitive cells. The lattice constant of WSe2 is 3.29 Å, and that of FGT is 3.991 Å. The band structure of bulk WSe2 and bulk FGT obtained by self-consistently calculated are in good agreement with other calculated results42,43. The FGT/WSe2 interface structure is constructed with 1 × 1 FGT and 1 × 1 WSe2 by using the interface builder tool of Quantum ATK41, which can automatically and optimally match the two different materials and calculate the strain percentage. The in-plane lattice constant of the FGT/WSe2 interface is 6.58 Å, and the lattice mismatch between the FGT and WSe2 of ~3.2% is tolerable. The spin-polarized generalized gradient approximation (SGGA) exchange-correlation potential using Perfew–Burke–Ernzerh (PBE) function, and the spin–orbit coupling is considered for theory calculations, with 7 Å k-point density and 100 Hartree density mesh cut-off. All structures have been fully optimized for the FGT/WSe2 interface and bulk materials until the total energy is converged to less than 10−5 eV. The maximum force on each atom is less than 0.05 eV/Å.
The data that support the findings of this study are available from the corresponding authors upon reasonable request.
Liu, Y., Huang, Y. & Duan, X. Van der Waals integration before and beyond two-dimensional materials. Nature 567, 323–333 (2019).
Sierra, J. F., Fabian, J., Kawakami, R. K., Roche, S. & Valenzuela, S. O. Van der Waals heterostructures for spintronics and opto-spintronics. Nat. Nanotechnol. 16, 856–868 (2021).
Mak, K. F., Lee, C., Hone, J., Shan, J. & Heinz, T. F. Atomically thin MoS2: a new direct-gap semiconductor. Phys. Rev. Lett. 105, 136805 (2010).
Zhu, Z. Y., Cheng, Y. C. & Schwingenschlögl, U. Giant spin-orbit-induced spin splitting in two-dimensional transition-metal dichalcogenide semiconductors. Phys. Rev. B 84, 153402 (2011).
Dankert, A. & Dash, S. P. Electrical gate control of spin current in van der Waals heterostructures at room temperature. Nat. Commun. 8, 16093 (2017).
Yang, L. et al. Long-lived nanosecond spin relaxation and spin coherence of electrons in monolayer MoS2 and WS2. Nat. Phys. 11, 830–834 (2015).
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).
Benítez, L. A. et al. Strongly anisotropic spin relaxation in graphene–transition metal dichalcogenide heterostructures at room temperature. Nat. Phys. 14, 303–308 (2018).
Han, W. Perspectives for spintronics in 2D materials. APL Mater. 4, 032401 (2016).
Zhu, W. et al. Large tunneling magnetoresistance in van der Waals ferromagnet/semiconductor heterojunctions. Adv. Mater. 33, 2104658 (2021).
Chen, P. et al. Approaching the intrinsic exciton physics limit in two-dimensional semiconductor diodes. Nature 599, 404–410 (2021).
Wang, S. et al. Room-temperature spin valve effects in La0.67Sr0.33MnO3/Alq3/Co devices. Synth. Met. 161, 1738–1741 (2011).
Fourmental, C. et al. Deep metallic interdiffusion in a model ferromagnetic/molecular system. Phys. Rev. Mater. 3, 083603 (2019).
Zheng, Y. et al. Magnetoresistance and spinterface of organic spin valves based on diketopyrrolopyrrole polymers. Adv. Electron. Mater. 5, 1900318 (2019).
Gong, C. et al. Discovery of intrinsic ferromagnetism in two-dimensional van der Waals crystals. Nature 546, 265–269 (2017).
Huang, B. et al. Layer-dependent ferromagnetism in a van der Waals crystal down to the monolayer limit. Nature 546, 270–273 (2017).
Gong, C. & Zhang, X. Two-dimensional magnetic crystals and emergent heterostructure devices. Science 363, eaav4450 (2019).
Song, T. et al. Giant tunneling magnetoresistance in spin-filter van der Waals heterostructures. Science 360, 1214–1218 (2018).
Wang, Z. et al. Tunneling spin valves based on Fe3GeTe2/hBN/Fe3GeTe2 van der Waals heterostructures. Nano Lett. 18, 4303–4308 (2018).
Albarakati, S. et al. Antisymmetric magnetoresistance in van der Waals Fe3GeTe2/graphite/Fe3GeTe2 trilayer heterostructures. Sci. Adv. 5, eaaw0409 (2019).
Novoselov, K. S., Mishchenko, A., Carvalho, A. & Castro Neto, A. H. 2D materials and van der Waals heterostructures. Science 353, aac9439 (2016).
Lin, H. et al. Spin-valve effect in Fe3GeTe2/MoS2/Fe3GeTe2 van der Waals heterostructures. ACS Appl. Mater. Interfaces 12, 43921–43926 (2020).
Hu, C., Yan, F., Li, Y. & Wang, K. Vertical WS2 spin valve with ohmic property based on Fe3GeTe2 electrodes. Chin. Phys. B 30, 097505 (2021).
Klein, D. R. et al. Probing magnetism in 2D van der Waals crystalline insulators via electron tunneling. Science 360, 1218–1222 (2018).
Wang, Z. et al. Magnetization dependent tunneling conductance of ferromagnetic barriers. Nat. Commun. 12, 6659 (2021).
Zhang, H. et al. Magnetoresistance in Co/2D MoS2/Co and Ni/2D MoS2/Ni junctions. Phys. Chem. Chem. Phys. 18, 16367–16376 (2016).
Zhao, K. et al. Magnetic transport property of NiFe/WSe2/NiFe spin valve structure. J. Magn. Magn. Mater. 432, 10–13 (2017).
Deng, Y. et al. Gate-tunable room-temperature ferromagnetism in two-dimensional Fe3GeTe2. Nature 563, 94–99 (2018).
Hu, C. et al. From two- to multi-state vertical spin valves without spacer layer based on Fe3GeTe2 van der Waals homo-junctions. Sci. Bull. 65, 1072–1077 (2020).
Yeh, P.-C. et al. Layer-dependent electronic structure of an atomically heavy two-dimensional dichalcogenide. Phys. Rev. B 91, 041407 (2015).
Evgeny, Y. T., Oleg, N. M. & Patrick, R. L. Spin-dependent tunnelling in magnetic tunnel junctions. J. Phys.: Condens. Matter 15, R109–R142 (2003).
Meng, J., Chen, J.-J., Yan, Y., Yu, D.-P. & Liao, Z.-M. Vertical graphene spin valve with Ohmic contacts. Nanoscale 5, 8894–8898 (2013).
Iqbal, M. Z. et al. Spin valve effect of NiFe/graphene/NiFe junctions. Nano Res. 6, 373–380 (2013).
Wu, H.-C. et al. Spin-dependent transport properties of Fe3O4/MoS2/Fe3O4 junctions. Sci. Rep. 5, 15984 (2015).
Shang, C. H., Nowak, J., Jansen, R. & Moodera, J. S. Temperature dependence of magnetoresistance and surface magnetization in ferromagnetic tunnel junctions. Phys. Rev. B 58, R2917–R2920 (1998).
Wang, W. et al. Spin-valve effect in NiFe/MoS2/NiFe junctions. Nano Lett. 15, 5261–5267 (2015).
Wang, X. H. et al. Spin manipulation by giant valley-Zeeman spin-orbit field in atom-thick WSe2. Appl. Phys. Rev. 9, 031402 (2022).
Zatko, V. et al. Band-structure spin-filtering in vertical spin valves based on chemical vapor deposited WS2. ACS Nano. 13, 14468–14476 (2019).
Khan, M. F., Kim, H., Nazir, G., Jung, S. & Eom, J. Layer dependent magnetoresistance of vertical MoS2 magnetic tunnel junctions. Nanoscale 10, 16703–16710 (2018).
Dolui, K., Narayan, A., Rungger, I. & Sanvito, S. Efficient spin injection and giant magnetoresistance in Fe/MoS2/Fe junctions. Phys. Rev. B 90, 041401 (2014).
Smidstrup, S. et al. QuantumATK: an integrated platform of electronic and atomic-scale modelling tools. J. Phys.: Condens. Matter 32, 015901 (2019).
Riley, J. M. et al. Direct observation of spin-polarized bulk bands in an inversion-symmetric semiconductor. Nat. Phys. 10, 835–839 (2014).
Li, X. L. et al. Spin-dependent transport in van der Waals magnetic tunnel junctions with Fe3GeTe2 electrodes. Nano Lett. 19, 5133–5139 (2019).
We are grateful for financial support from the National Key Research and Development Program of China (Grant No. 2017YFA0303400), the National Natural Science Foundation of China (Grant Nos. 61774144, 21905285 and 62005265), the Key Research Program of Frontier Sciences (Grant No. QYZDYSSW-JSC020), the Strategic Priority Research Program of the Chinese Academy of Science (Grant Nos. XDB44000000 and XDB28000000), and the Beijing Natural Science Foundation Key Program (Grant No. Z190007).
The authors declare no competing interests.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
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
Zheng, Y., Ma, X., Yan, F. et al. Spin filtering effect in all-van der Waals heterostructures with WSe2 barriers. npj 2D Mater Appl 6, 62 (2022). https://doi.org/10.1038/s41699-022-00339-z