Manipulating spin-polarized photocurrents in 2Dtransition metal dichalcogenidesLu Xiea and Xiaodong Cuia,1

aPhysics Department, University of Hong Kong, Hong Kong

Edited by Philip Kim, Harvard University, Cambridge, MA, and accepted by the Editorial Board February 26, 2016 (received for review November 21, 2015)

Manipulating spin polarization of electrons in nonmagnetic semi-conductors by means of electric fields or optical fields is an essentialtheme of the conceptual nonmagnetic semiconductor-based spin-tronics. Here we experimentally demonstrate an electric method ofdetecting spin polarization in monolayer transition metal dichalco-genides (TMDs) generated by circularly polarized optical pumping.The spin-polarized photocurrent is achieved through the valley-dependent optical selection rules and the spin–valley locking inmonolayerWS2, and electrically detected by a lateral spin–valve struc-ture with ferromagnetic contacts. The demonstrated long spin–valley lifetime, the unique valley-contrasted physics, and the spin–valley locking make monolayer WS2 an unprecedented candidatefor semiconductor-based spintronics.

monolayer transition metal dichalcogenides | spin–valley coupling |spintronics | spin lifetime | valley lifetime

Alongtime focus in nonmagnetic semiconductor spintronicsresearch is to explore methods to generate and manipulate

spin of electrons by means of electric fields or optical fields in-stead of magnetic fields, enabling scalable and integrated devices(1). The present efforts follow two distinct paths. One uses spinHall effect or optical pumping in III-V semiconductors whichfeature a significant spin–orbit coupling in a form of Dresselhausand/or Rashba terms (2–4); the other focuses on spin transport[usually generated by spin injection from ferromagnetic (FM)electrodes] in semiconductor structures made of silicon (5), carbonnanotube (6), graphene (7), etc. which have long spin-coherencelength due to weak spin–orbit coupling. The emergence of atomictwo-dimensional group VI transition metal dichalcogenides(TMDs) MX2 (M = Mo, W; X = S, Se), featuring nonzero butcontrasting Berry curvatures at inequivalent K and K′ (equivalentto −K) valleys and unique spin–valley locking, provides an al-ternative pathway toward spintronics (8).Valleys refer to the energy extremes around the high symmetry

points of the Brillouin zone, either a “valley” in the conductionband or a “hill” in the valence band. Owing to their hexagonallattices, the family of TMDs has degenerate but inequivalentK(K ′) valleys well separated in the first Brillouin zone. This giveselectrons an extra valley degree of freedom, in addition to chargeand spin. In monolayer TMDs the inversion symmetry breakingof crystal structures gives rise to nonzero but contrasting Berrycurvatures at K and K′ valley which are a characteristic of theBloch bands and could be recognized as a form of orbital mag-netic moment of Bloch electrons (9–11). These contrasting Berrycurvatures of electrons (holes) at K(K′) valleys lead to con-trasting response to certain stimulus (9–19). One example is thevalley Hall effect: An electric field would drive the electrons atdifferent valleys (K and K′) toward opposite transverse direc-tions, in a similar way as in spin Hall effect (10, 20). A morepronounced manifestation is valley-dependent circular opticalselection rules in K and K ′ valleys. Namely, the interband opticaltransitions at K(K ′) only couple with circularly polarized light ofσ+(σ−) helicity. Consequently the valley polarization could berealized by the polarization field of optical excitations (10–13).On the other side, the band edge at K(K′) valleys mainly con-structed from d orbits of the heavy metal atoms inherits the strong

spin–orbit coupling (SOC) of atomic orbits. And, the Zeeman-likeSOC originating from D1

3h symmetry of monolayer TMDs lifts theout-of-plane spin degeneracy of the band edges at K and K′ valleysby a significant amount, around 0.16 and 0.45 eV in the valencebands of molybdenum dichalcogenides and tungsten dichal-cogenides, respectively, and about 1 order of magnitude smaller inconduction bands (10, 21–27). Owing to the presence of time-reversal symmetry ðK ↔ −KÞ, the spin splitting has opposite signbetween K and K′ valleys at monolayer TMDs as illustrated inFig. 1A. The Kramer doublet, spin-up state Sz = Z=2 at K valleyjK↑i, and spin-down state Sz =−ðZ=2Þ at K′ valleyjK′↓i, are sep-arated from the other doublet jK↓i and jK′↑i by the SOC energy.This strong SOC and the explicit inversion symmetry breaking lockthe spin and valley degrees of freedom in monolayer TMDs andthis interplay leads to sophisticated consequences. First the spinand valley relaxation are dramatically squelched due to the si-multaneously requirements of spin flip and momentum conser-vation. The intrinsic mirror symmetry with respect to out-of-planedirection further suppresses spin relaxation via the D’yakonov–Perel’ mechanism, which usually plays an important role for spinrelaxation in III-V semiconductors (28). Subsequently the valleyand spin polarization are expected to be robust against low-energyperturbation (10, 29). Second, the spin–valley locking offers aversatile measure to manipulate spin degree of freedom via con-trol of valley degree of freedom or vice versa (8, 30–32). Thiscould lead to an integrated and complementary approach of val-leytronics and spintronics in monolayer TMDs.Here we report an experimental demonstration of spin po-

larization via valley-dependent optical selection rules in mono-layer WS2. The valley polarization is realized by controlling thepolarization field of interband optical excitations and the spinpolarization is simultaneously generated via spin–valley locking inmonolayer WS2. The spin polarization is electrically detected bythe lateral spin–valve structure consisting of a tunneling barrier

Significance

Monolayer group VI transition metal dichalcogenides (TMDs)feature a massive Dirac fermion system with strong spin–valleylocking. It provides a route to manipulate quantum states viathe interplay of spin and valley degrees of freedom. Here wereport that the spin polarization and spin–valley lifetime offree carriers are electrically detected via a spin–valve-like struc-ture in monolayer TMDs. The long spin–valley lifetime (∼102 ns)of free carriers is electrically probed, contrasting to that of exci-tons (∼101–102 ps) probed by optical spectroscopy. It demon-strates the potential application of 2D TMDs in nonmagneticsemiconductor-based spintronics.

Author contributions: X.C. designed research; L.X. performed research; L.X. and X.C. an-alyzed data; and L.X. and X.C. wrote the paper.

The authors declare no conflict of interest.

This article is a PNAS Direct Submission. P.K. is a guest editor invited by the EditorialBoard.1To whom correspondence should be addressed. Email: xdcui@hku.hk.

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1523012113/-/DCSupplemental.

3746–3750 | PNAS | April 5, 2016 | vol. 113 | no. 14 www.pnas.org/cgi/doi/10.1073/pnas.1523012113

Dow

nloa

ded

by g

uest

on

Mar

ch 1

2, 2

020

of Al2O3 and superlattice-structured cobalt–palladium (Co/Pd)ferromagnetic electrodes with perpendicular magnetization an-isotropy (PMA). A high spin polarization of diffusive photo-currents is observed and a micrometer-size spin-free pathand spin lifetime of free carriers in the range of 101∼102 nsare estimated.

MethodsThe photocurrent measurements were conducted on a 5-μm-channel field-effect transistor (FET) structure of mechanically exfoliated monolayer WS2on a silicon substrate capped with 300-nm oxide. To overcome the conduc-tance mismatch for efficient spin filtering, an ultrathin Al2O3 (1.2 nm) wasdeposited between the monolayer and ferromagnetic electrodes, which aremade of 20 periods of alternating Co (4.5 Å)/Pd(15 Å) layers deposited with adeposition rate of 1 Å/min (Co) and 0.25 Å/min (Pd) in a metal molecularbeam epitaxy system. Owing to the intrinsic mirror symmetry (D1

3hÞ with re-spect to the plane of metal atoms, the spin projection is along the out-of-planedirection and SZ is a good quantum number in monolayer WS2. To electricallydetect the spin polarization along the z direction, a spin analyzer with PMA isthe key, which is realized with a superlattice of ultrathin Co (4.5 Å)/Pd (15 Å)multilayers (33). The in situ polar magnetooptic Kerr effect spectroscopy(MOKE) demonstrates a clear ferro-magnetization along the~z direction with acoercive force around 30 Oe, as shown in Fig. 2B (SI Appendix). Standardelectric characterization as shown in Fig. 2A shows a slightly n-type FET be-havior in all of the devices, which might be induced from the defects, vacancy,and/or substrate effects. The source–drain conductance is at a tens of nano-siemens level at maximum within the back-gate bias range of VG =±80 V,showing the Fermi level falls deep in the band gap.

Results and DiscussionThe photocurrent was generated at a source–drain bias underthe near-resonance excitation of 2.09 eV. The source–draincurrent was fed to a preamplifier with input impedance of 100 KΩclose to the sample side. The laser was focused through a 50×

objective lens onto a spot of 1 μm and the excitation power waskept below 150 μW. The photocurrents and the circular dichroismwere monitored simultaneously with a photoelastic modulator(50 KHz) and two sets of lock-in amplifiers which extract both thephotocurrents and the difference between two helicities. So, thepotential effects due to sample inhomogeneity were minimized.The drain current rises by 1–3 orders of magnitude when the

near-resonance excitation scans across the monolayer, similar tothe reported photocurrent experiments on multilayer WS2 (34).As demonstrated in Fig. 3B, the scanning photocurrent distrib-utes inhomogeneously across the channel, concentrating aroundcharge traps/defects and electrode contacts where local electricfields are strong enough to disassociate excitons, quasiparticlesof Coulomb-bounded electron–hole pairs, into free carriers. Togenerate significant photocurrents with a minimum backgroundelectric current (dark current), a gate VG =−80 V pulls the FETto the “off” state and a source–drain bias VDS =−5 V is applied toaccelerate the photocarriers. Once the FM electrodes are ferro-magnetized by the external magnetic field, the photocurrent showsa distinct pattern of optical-polarization responses at zero magneticfield as demonstrated in Fig. 3 C and D. For the excitation close tothe electrode–TMD contacts, the strength of the photocurrentexhibits a strong dependence on the combination of the FM elec-trode magnetization and the polarization of optical excitations.Depending on the magnetization of FM electrodes, the photocur-rent at the same location shows a clear circular dichroism for thecircularly polarized optical excitations with opposite helicities.Namely under one magnetization direction, for example, alongpositive ~z ðM↑Þ, the excitation with polarization of σ+ induceshigher photocurrent than that of σ−. If the FM magnetizationis reversed, the photocurrent difference (σ+ − σ−Þ between oppo-site helicities also switches the sign. The nonzero photocurrent

Fig. 1. (A) Schematics of valley-dependent optical selection rules at K and K′ valleys in the momentum space of monolayer TMDs and spin–valley locking, andthe proposed mechanism of the spin-resolved photocurrent measurement with the ferromagnetic electrodes. The spin-splitting at conduction band (∼0.03 eV)and valence band (∼0.45 eV) are disproportionally sketched for clarity. (B) Schematic of monolayer WS2 devices for spin-polarized photocurrent measure-ments. (Inset) Optical image of the representative devices.

Fig. 2. Transport characteristics (A) Ids −VG curve and standard I–V characteristic of the device at 10 K. (B) Polar MOKEmeasurement of Co/Pd layered FM electrodesat 10 K with external magnetic field perpendicular to the sample surface. The magnetic hysteresis loop clearly shows a ferromagnetic behavior with PMA.

Xie and Cui PNAS | April 5, 2016 | vol. 113 | no. 14 | 3747

PHYS

ICS

Dow

nloa

ded

by g

uest

on

Mar

ch 1

2, 2

020

difference shows a clear dependence on the magnetization of FMelectrodes as shown in Fig. 4B, which is consistent with the mag-netization of the FM electrodes demonstrated in the MOKEmeasurements. The photocurrent difference has a clear spatialdistribution pattern: It generally rises upon the excitation spotbeing close to FM electrodes and it vanishes when the excitation isfar away from the FM electrodes. This scenario is well understoodwith the valley-dependent circular optical selection rules and thespin–valley locking in monolayer WS2. The excitation of σ+ðσ−   Þ

selectively pumps the excitons at K(K′) valley and the electrons arefully spin-polarized to Sz = Z=2 (Sz =−ðZ=2Þ) due to spin–valleylocking. If local electric fields break the excitons into free carriers,these free carriers are accelerated by the source–drain bias togenerate photocurrents while the spin polarization remains. If thespin polarization survives when the photocarriers reach the FMelectrodes, the spin alignment with the FM electrodes yields thedifferent effective resistance. Fig. 3 A–D also shows that the photo-currents and the photocurrent difference Iσ+ − Iσ− are uncorrelated.

Fig. 3. (A) Laser scanning reflection image of the photocurrent device. The areas outlined with red dashed line are FM electrodes. (Inset) Correspondingoptical image. (B) Photocurrent map with a scanning excitation under bias VG =−80 V and VDS =−5 V. (C and D) Differential photocurrents ðIσ+ − Iσ−Þbetween the σ+ and σ−. circularly polarized excitations through the FM electrodes with opposite magnetization M↑ and M↓ under zero magnetic field. Thedifference changes sign at opposite magnetizations. The photocurrent difference Iσ+ − Iσ− keeps the same polarity at both source and drain electrodes underthe same FM magnetization. (E and F) Degree of photocurrent polarization P = ðIσ+ − Iσ−Þ=ðIσ+ + Iσ−Þ through the FM electrodes with opposite magnetizationM↑ and M↓.

Fig. 4. (A) Photocurrent and degree of photocurrent polarization P as a function of the excitation intensity. (B) Photocurrent difference between circularlypolarized excitations with opposite helicities as a function of external magnetic field along the out-of-plane direction. The photocurrent difference shows anFM-like loop which is consistent with the magnetization of the FM electrodes. (C) Representative photocurrent polarization P as a function of the distancefrom the FM electrodes with opposite magnetization M↑ and M↓. The hatched area labels the FM electrodes. The fit curve (blue) assuming P ∼ P0e−ðx−x0Þ=ls

yields peak polarization P0 =0.15± 0.02 and spin-free path ls =1.7± 0.2 μm for holes, and P0 = 0.07± 0.01 and ls = 1.3± 0.1 μm for electrons, respectively.

3748 | www.pnas.org/cgi/doi/10.1073/pnas.1523012113 Xie and Cui

Dow

nloa

ded

by g

uest

on

Mar

ch 1

2, 2

020

It is because the scanning photocurrent directly reflects the strengthof local electric fields, whereas the photocurrent difference Iσ+ − Iσ−also depends on the photocarriers’ spin polarization arriving at theFM electrodes and the efficiency of the electrode–TMD junctionfor spin filtering.To quantitatively evaluate the spin polarization of the photo-

current, we define the degree of the photocurrent polarizationP= ðIσ+ − Iσ−Þ=ðIσ+ + Iσ−Þ, where Iσ+(Iσ−) is the photocurrent un-der the excitation of σ+(σ−). Given that the photocurrent Iσ+(Iσ−)at minimum is around several nanoamperes, which is far beyondthe dark current and the noise level of tens of picoamperes in thesystem, artifacts in calculating polarization P are safely excluded.The photocurrent polarization P peaks around 0.15 at electrode–TMD junctions and decays to a negligible level when the opticalexcitation scans away from the FM electrodes as shown in Figs. 3 Eand F and 4C. The polarization P reverses the sign if the magne-tization of FM electrodes switches, showing a signature of efficientspin–valve structure. As a result of valley-dependent opticalselection rules and spin–valley locking in monolayer TMD, thephotocurrent polarization P reflects the spin polarization ofthe electrons (holes) arriving at the electrode–TMD junction.At the experimental conditions the photocurrent is dominatedby the diffusive drift current (SI Appendix), and the electrons/holes’ trajectory could be simplified as a collective movementwith a drift velocity. Without considering many-body interac-tions, the spin polarization exponentially decays with a char-acteristic time, equivalently distributing with a characteristicspin-polarization-free path in space. Consequently the profileof the spin polarization in the scanning photocurrent measurementsfollows P∼P0e−ðx=lsÞ, where P0, x, and ls denote the peak polari-zation, the distance between the optical excitation and theelectrode–TMD junction, and the spin-free path, respectively(SI Appendix). The representative contour demonstrated inFig. 4C yields P0 = 0.15± 0.02 and ls = 1.7± 0.2 μm for holes, andP0 = 0.07± 0.01 and ls = 1.3± 0.1 μm for electrons, respectively.The peak polarization P0 is attributed to the spin polarization ofthe photocarriers and the anisotropic magnetization resistanceof the FM electrodes superimposed by the efficiency of the spin-injection junction. If we assume that the detected polarization isa simple product of the spin polarization of electrons at Fermilevel of cobalt electrodes at 0.4, the (up-bound) efficiency of thespin injection η at 0.7 (35), and the photocurrent spin polari-zation, the spin polarization of the photocurrent is estimated tobe 54% (low-bound), surpassing all demonstrated in conventionalsemiconductors.The micrometer-size spin-free path of electrons also implies a

sizable spin-splitting in the conductance band edge which wastheoretically predicted to be around 30 meV (26, 36). The sim-ilar spin-free paths of electrons (1.3 μm) and holes (1.7 μm) couldbe interpreted as the result of the close effective masses ofelectrons and holes and the large spin-splitting gaps at the con-duction and valence band edges with respect to the thermalenergy (10 K ∼ 0.86 meV) and the Fermi energy (around zeroat intrinsic state) at the experimental conditions. Meanwhile Fig.3 E and F shows that the degree of spin polarization of holes issignificantly higher than that of electrons, 15% vs. 7%. This is

consistent with the calculations that the spin-splitting carries thesame sign between conduction band and valence band mono-layer WS2 as shown in Fig. 1A (26, 27). As spin is conserved inthe optical interband transition, electrons are pumped to thespin-split upper subband under near-resonant excitations. Unlikephotogenerated holes which are around the band edge with aspin-splitting in valence band of around 0.45 eV, the electronrelaxation could take place through two channels, intravalleyscattering (to the spin-split lower subband in the same valley)where spin-flip is required, or intervalley scattering (to the spin-split lower subband in the opposite valley) where spin is con-served. This explains why the spin polarization of electrons (atthe source side) is weaker than that of holes (at the drain side).Fig. 3 C–F shows that the photocurrent difference Iσ+ − Iσ− andthe degree of spin polarization carry the same polarity withcomparable strength (15% vs. 7%) at both drain and sourceelectrodes under the same FM magnetization. It implies that thespin-conserved intervalley scattering predominates the electronrelaxation process.We also could estimate the magnitude of the spin lifetime from

the spin-free path. Given that the effective bias added on thechannel is on the order of Vds − ðEg +Eexciton  bindingÞ∼ 2.5 eVwherewe assume the band bending at both contacts is roughly of theelectronic band gap at most and the mobility of 0.1–1 cm2 · v=s ofthe devices (SI Appendix), the spin-free path indicates the esti-mated spin–valley lifetime around 101∼102 ns. This estimate isorders of magnitude larger than the valley lifetime estimated frompolarization-resolved photoluminescence and pump–probe spec-troscopy in which the exciton effect predominates the opticalproperties and consequently the valley lifetime of excitons insteadof free carriers is probed (37, 38). Note that the electron–holeexchange interaction provides the major channel for excitons’ spin–valley depolarization (39), whereas the exchange interactions aregreatly suppressed in oppositely drifting free carriers in a nearlyintrinsic state, and consequently the free carriers presumably showsignificantly longer spin–valley lifetime. The capability of moni-toring spin–valley lifetime of free carriers makes the presenttechnique complementary to optical approaches.

SummaryIn summary, we have demonstrated the highly spin-polarizedphotocurrents in monolayer WS2 by controlling the polarizationfield of optical excitations. The spin polarization is well achievedas the result of valley-dependent optical selection rules and spin–valley locking in monolayer TMDs. The spin polarization of freecarriers could be electrically detected with a lateral spin–valvestructure. The demonstrated micrometer-size spin-free path andspin lifetime in the range of 101∼102 ns, and the unique spin–valley locking make monolayer TMDs a promising candidate forspintronics applications.

ACKNOWLEDGMENTS. The work is supported by General Research Fund(17300415), Area of Excellency (AoE/P-04/08), Collective Research Fund(HKU9/CRF/13G) of Hong Kong Research Grant Council, and StrategicResearch Theme on New Materials of The University of Hong Kong.

1. Wolf SA, et al. (2001) Spintronics: A spin-based electronics vision for the future.Science 294(5546):1488–1495.

2. Kato YK, Myers RC, Gossard AC, Awschalom DD (2004) Observation of the spin Halleffect in semiconductors. Science 306(5703):1910–1913.

3. Wunderlich J, Kaestner B, Sinova J, Jungwirth T (2005) Experimental observation ofthe spin-Hall effect in a two-dimensional spin-orbit coupled semiconductor system.Phys Rev Lett 94(4):047204.

4. Ganichev SD, Prettl W (2003) Spin photocurrents in quantum wells. J Phys CondensMatter 15(20):R935–R983.

5. Appelbaum I, Huang B, Monsma DJ (2007) Electronic measurement and control ofspin transport in silicon. Nature 447(7142):295–298.

6. Tsukagoshi K, Alphenaar BW, Ago H (1999) Coherent transport of electron spin in aferromagnetically contacted carbon nanotube. Nature 401(6753):572–574.

7. Tombros N, Jozsa C, Popinciuc M, Jonkman HT, van Wees BJ (2007) Electronic spintransport and spin precession in single graphene layers at room temperature. Nature448(7153):571–574.

8. Xu X, YaoW, Xiao D, Heinz TF (2014) Spin and pseudospins in layered transition metaldichalcogenides. Nat Phys 10(5):343–350.

9. Xiao D, Yao W, Niu Q (2007) Valley-contrasting physics in graphene: Magnetic mo-ment and topological transport. Phys Rev Lett 99(23):236809.

10. Xiao D, Liu G-B, Feng W, Xu X, Yao W (2012) Coupled spin and valley physics in mono-layers of MoS2 and other group-VI dichalcogenides. Phys Rev Lett 108(19):196802.

11. Cao T, et al. (2012) Valley-selective circular dichroism of monolayer molybdenumdisulphide. Nat Commun 3:887.

12. Zeng H, Dai J, Yao W, Xiao D, Cui X (2012) Valley polarization in MoS2 monolayers byoptical pumping. Nat Nanotechnol 7(8):490–493.

Xie and Cui PNAS | April 5, 2016 | vol. 113 | no. 14 | 3749

PHYS

ICS

Dow

nloa

ded

by g

uest

on

Mar

ch 1

2, 2

020

13. Mak KF, He K, Shan J, Heinz TF (2012) Control of valley polarization in monolayerMoS2 by optical helicity. Nat Nanotechnol 7(8):494–498.

14. Zhang YJ, Oka T, Suzuki R, Ye JT, Iwasa Y (2014) Electrically switchable chiral light-emitting transistor. Science 344(6185):725–728.

15. Kim J, et al. (2014) Ultrafast generation of pseudo-magnetic field for valley excitons inWSe₂ monolayers. Science 346(6214):1205–1208.

16. Aivazian G, et al. (2015) Magnetic control of valley pseudospin in monolayer WSe2.Nat Phys 11(2):148–152.

17. MacNeill D, et al. (2015) Breaking of valley degeneracy by magnetic field in mono-layer MoSe2. Phys Rev Lett 114(3):037401.

18. Srivastava A, et al. (2015) Valley Zeeman effect in elementary optical excitations ofmonolayer WSe2. Nat Phys 11(2):141–147.

19. Li Y, et al. (2014) Valley splitting and polarization by the Zeeman effect in monolayerMoSe2. Phys Rev Lett 113(26):266804.

20. Mak KF, McGill KL, Park J, McEuen PL (2014) Valleytronics. The valley Hall effect inMoS₂ transistors. Science 344(6191):1489–1492.

21. Mak KF, Lee C, Hone J, Shan J, Heinz TF (2010) Atomically thin MoS₂: A new direct-gapsemiconductor. Phys Rev Lett 105(13):136805.

22. Splendiani A, et al. (2010) Emerging photoluminescence in monolayer MoS2. NanoLett 10(4):1271–1275.

23. Zeng H, et al. (2013) Optical signature of symmetry variations and spin-valley couplingin atomically thin tungsten dichalcogenides. Sci Rep 3:1608.

24. Zhao W, et al. (2013) Evolution of electronic structure in atomically thin sheets of WS2and WSe2. ACS Nano 7(1):791–797.

25. Cheiwchanchamnangij T, Lambrecht WRL (2012) Quasiparticle band structure calcu-lation of monolayer, bilayer, and bulk MoS2. Phys Rev B 85(20):205302.

26. Liu G-B, Shan W-Y, Yao Y, Yao W, Xiao D (2013) Three-band tight-binding model formonolayers of group-VIB transition metal dichalcogenides. Phys Rev B 88(8):085433.

27. Ko�smider K, González JW, Fernández-Rossier J (2013) Large spin splitting in theconduction band of transition metal dichalcogenide monolayers. Phys Rev B 88(24):245436.

28. D’yakonov M, Perel’ V (1971) Spin orientation of electrons associated with interbandabsorption of light in semiconductors. J Exp Theor Phys 33(5):1053.

29. Ochoa H, Roldán R (2013) Spin-orbit-mediated spin relaxation in monolayer MoS2.Phys Rev B 87(24):245421.

30. Gong Z, et al. (2013) Magnetoelectric effects and valley-controlled spin quantumgates in transition metal dichalcogenide bilayers. Nat Commun 4:2053.

31. Yuan H, et al. (2014) Generation and electric control of spin-valley-coupled circularphotogalvanic current in WSe2. Nat Nanotechnol 9(10):851–857.

32. Eginligil M, et al. (2015) Dichroic spin-valley photocurrent in monolayer molybdenumdisulphide. Nat Commun 6:7636.

33. Carcia PF, Meinhaldt AD, Suna A (1985) Perpendicular magnetic anisotropy in Pd/Cothin film layered structures. Appl Phys Lett 47(2):178–180.

34. Ubrig N, Jo S, Berger H, Morpurgo AF, Kuzmenko AB (2014) Scanning photocurrentmicroscopy reveals electron-hole asymmetry in ionic liquid-gated WS2 transistors.Appl Phys Lett 104(17):171112.

35. Salis G, et al. (2005) Temperature independence of the spin-injection efficiency of aMgO-based tunnel spin injector. Appl Phys Lett 87(26):262503.

36. Song Y, Dery H (2013) Transport theory of monolayer transition-metal dichalcoge-nides through symmetry. Phys Rev Lett 111(2):026601.

37. Mai C, et al. (2014) Exciton valley relaxation in a single layer of WS2 measured byultrafast spectroscopy. Phys Rev B 90(4):041414.

38. Zhu B, Zeng H, Dai J, Gong Z, Cui X (2014) Anomalously robust valley polarization andvalley coherence in bilayer WS2. Proc Natl Acad Sci USA 111(32):11606–11611.

39. Yu T, WuMW (2014) Valley depolarization due to intervalley and intravalley electron-hole exchange interactions in monolayer MoS2. Phys Rev B 89(20):205303.

3750 | www.pnas.org/cgi/doi/10.1073/pnas.1523012113 Xie and Cui

Dow

nloa

ded

by g

uest

on

Mar

ch 1

2, 2

020