## Abstract

Spin-momentum locked surface states in topological insulators (TIs) provide a promising route for achieving high spin-orbit torque (SOT) efficiency beyond the bulk spin-orbit coupling in heavy metals (HMs). However, in previous works, there is a huge discrepancy among the quantitative SOTs from TIs in various systems determined by different methods. Here, we systematically study the SOT in the TI(HM)/Ti/CoFeB/MgO systems by the same method, and make a conclusive assessment of SOT efficiency for TIs and HMs. Our results demonstrate that TIs show more than one order of magnitude higher SOT efficiency than HMs even at room temperature, at the same time the switching current density as low as 5.2×105 A cm-2 is achieved with (Bi1-xSbx)2Te3. Furthermore, we investigate the relationship between SOT efficiency and the position of Fermi level in (Bi1-xSbx)2Te3, where the SOT efficiency is significantly enhanced near the Dirac point, with the most insulating bulk and conducting surface states, indicating the dominating SOT contribution from topological surface states. This work unambiguously demonstrates the ultrahigh SOT efficiency from topological surface states.

Original language | English |
---|---|

Article number | 207205 |

Journal | Physical review letters |

Volume | 123 |

Issue number | 20 |

DOIs | |

Publication status | Published - 2019 Nov 15 |

## All Science Journal Classification (ASJC) codes

- Physics and Astronomy(all)

## Access to Document

## Fingerprint Dive into the research topics of 'Room-Temperature Spin-Orbit Torque from Topological Surface States'. Together they form a unique fingerprint.

## Cite this

- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver

*Physical review letters*,

*123*(20), [207205]. https://doi.org/10.1103/PhysRevLett.123.207205

**Room-Temperature Spin-Orbit Torque from Topological Surface States**. In: Physical review letters. 2019 ; Vol. 123, No. 20.

}

*Physical review letters*, vol. 123, no. 20, 207205. https://doi.org/10.1103/PhysRevLett.123.207205

**Room-Temperature Spin-Orbit Torque from Topological Surface States.** / Wu, Hao; Zhang, Peng; Deng, Peng; Lan, Qianqian; Pan, Quanjun; Razavi, Seyed Armin; Che, Xiaoyu; Huang, Li; Dai, Bingqian; Wong, Kin; Han, Xiufeng; Wang, Kang L.

Research output: Contribution to journal › Article › peer-review

TY - JOUR

T1 - Room-Temperature Spin-Orbit Torque from Topological Surface States

AU - Wu, Hao

AU - Zhang, Peng

AU - Deng, Peng

AU - Lan, Qianqian

AU - Pan, Quanjun

AU - Razavi, Seyed Armin

AU - Che, Xiaoyu

AU - Huang, Li

AU - Dai, Bingqian

AU - Wong, Kin

AU - Han, Xiufeng

AU - Wang, Kang L.

N1 - Funding Information: https://orcid.org/0000-0001-8608-3035 Wu Hao 1 ,* Zhang Peng 1 Deng Peng 1 Lan Qianqian 2 Pan Quanjun 1 Razavi Seyed Armin 1 Che Xiaoyu 1 Huang Li 3 Dai Bingqian 1 Wong Kin 1 Han Xiufeng 3 Wang Kang L. 1 ,† Department of Electrical and Computer Engineering, and Department of Physics and Astronomy, 1 University of California , Los Angeles, California 90095, USA Ernst Ruska-Centre for Microscopy and Spectroscopy with Electrons and Peter Grünberg Institute, 2 Forschungszentrum Jülich , Jülich 52425, Germany Beijing National Laboratory for Condensed Matter Physics, Institute of Physics, 3 Chinese Academy of Sciences , Beijing 100190, China * Corresponding author. wuhaophysics@ucla.edu † Corresponding author. wang@ee.ucla.edu 15 November 2019 15 November 2019 123 20 207205 17 April 2019 © 2019 American Physical Society 2019 American Physical Society Spin-momentum locked surface states in topological insulators (TIs) provide a promising route for achieving high spin-orbit torque (SOT) efficiency beyond the bulk spin-orbit coupling in heavy metals (HMs). However, in previous works, there is a huge discrepancy among the quantitative SOTs from TIs in various systems determined by different methods. Here, we systematically study the SOT in the TI ( HM ) / Ti / CoFeB / MgO systems by the same method, and make a conclusive assessment of SOT efficiency for TIs and HMs. Our results demonstrate that TIs show more than one order of magnitude higher SOT efficiency than HMs even at room temperature, at the same time the switching current density as low as 5.2 × 10 5 A cm − 2 is achieved with ( Bi 1 − x Sb x ) 2 Te 3 . Furthermore, we investigate the relationship between SOT efficiency and the position of Fermi level in ( Bi 1 − x Sb x ) 2 Te 3 , where the SOT efficiency is significantly enhanced near the Dirac point, with the most insulating bulk and conducting surface states, indicating the dominating SOT contribution from topological surface states. This work unambiguously demonstrates the ultrahigh SOT efficiency from topological surface states. National Science Foundation 10.13039/100000001 1611570 Army Research Office 10.13039/100000183 W911NF-16-1-0472 W911NF-15-1-10561 U.S. Department of Energy 10.13039/100000015 DE-SC0012670 National Key Research and Development Program 2017YFA0206200 National Natural Science Foundation of China 10.13039/501100001809 11434014 Spin-orbit torque (SOT) [1–3] provides an efficient way to electrically manipulate the magnetic order. In general, SOT originates from the charge-spin conversion in materials with large spin-orbit coupling (SOC), which can be quantified as θ SH = J s 3 D / J e 3 D or q ICS = J s 3 D / J e 2 D = θ SH / t s , where J s 3 D represents the three-dimensional (3D) spin current density; J e 3 D and J e 2 D represent the 3D and two-dimensional (2D) electric (charge) current density, respectively; and t s represents the effective thickness of the SOC layer. SOT based on heavy metals (HMs) with bulk SOC has been widely studied; however, due to the limited θ SH (typically around 0.1) [1,4,5] , the switching current density J c remains ultrahigh [6–8] ; therefore, improvement of SOT efficiency is still a major challenge for further reducing the power dissipation of SOT-based devices. Spin-momentum locked surface states in topological insulators (TIs) are expected to be a promising candidate to break through the limited θ SH , and previous works have reported the very large θ SH (425) [9] and the SOT-induced magnetization switching [9–11] with TIs at low temperature. Recently, several works reported the room-temperature SOT switching by TIs [12–15] , which is a crucial step towards practical applications. However, there exists a huge discrepancy among the reported θ SH of TIs in various systems characterized by different methods, such as 0.047 in Bi 2 Se 3 / CoFeB [16] and 3.5 in Bi 2 Se 3 / NiFe [17] by the spin torque ferromagnetic resonance measurement, 0.16 in Bi 2 Se 3 / CoTb [12] and 0.40 in ( BiSb ) 2 Te 3 / CoTb [12] by the hysteresis loop shift measurement, 18.6 in Bi 2 Se 3 / CoFeB [13] by the planar Hall measurement, and 52 in Bi 0.9 Sb 0.1 / MnGa [14] by the coercivity field shift measurement. Moreover, some groups argue that the bulk SOC [18,19] and interfacial Rashba effect [20,21] could also be involved in the SOT from TIs. Therefore, a conclusive study of SOT in TIs is in great need to clarify the actual spin current source and obtain the reliable SOT efficiency. Here, we systematically investigate the SOT in a series of TIs and HMs, based on the similar TI ( HM ) / Ti / CoFeB / MgO heterostructures. We demonstrate the room-temperature SOT-induced magnetization switching, and the critical switching current density in ( Bi 1 − x Sb x ) 2 Te 3 ( 5.2 × 10 5 A cm − 2 ) could be 1–2 orders of magnitude smaller than those in HMs. The charge-spin conversion efficiency θ SH (or q ICS ) is obtained by the harmonic Hall method, which shows that TIs [ θ SH = 2.5 for ( Bi 1 − x Sb x ) 2 Te 3 ] can break through the θ SH < 1 limitation in HMs. By engineering the band structure of ( Bi 1 − x Sb x ) 2 Te 3 , we find that the SOT strongly depends on the Fermi level position and reaches the maximum near the Dirac point with the most insulating bulk and conducting surface states, indicating the dominating SOT from topological surface states. Figure 1(a) shows the schematic of SOT in the TI / Ti / CoFeB / MgO system. In topological surface states, the spin and momentum directions are locked [22–24] ; therefore, the electrical current flowing possess a net spin polarization. The spin current injection from the TI exerts a torque on the adjacent magnetic moment of CoFeB, and the antidamping torque τ SOT = m × ( m × σ ) can switch the magnetization at a sufficient current density, where m and σ represent the magnetic and spin vectors, respectively. A constant external in-plane magnetic field H ext is applied to break the mirror symmetry between + M z and − M z states and break the chiral domain walls to induce the domain wall expansion [25–27] for deterministic SOT switching. Six quintuple layers (QLs) TIs [ ( Bi 1 − x Sb x ) 2 Te 3 , Bi 2 Te 3 , and SnTe] are grown on Al 2 O 3 ( 0001 ) substrates by using the molecular beam epitaxy method, and Ti ( 2 nm ) / CoFeB ( 1.4 nm ) / MgO ( 2 nm ) multilayers are deposited by the magnetron sputtering method. The layer-by-layer growth of TIs is monitored by the reflection high-energy electron diffraction (RHEED). The sharp RHEED patterns show the high crystal quality and the flat surface, and the thickness of 6 QLs is determined by the periods of RHEED oscillations, as shown in Fig. 1(b) . The nonmagnetic interlayer Ti is used to provide the perpendicular magnetic anisotropy (PMA) of CoFeB. At the same time, the SOT contribution from Ti is negligible due to the extremely small θ SH (0.0004) [28] . The films are patterned to 20 μ m × 130 μ m Hall bar devices. Figure 1(c) shows the M - H z and R x y - H z loops in the ( Bi 1 − x Sb x ) 2 Te 3 / Ti / CoFeB / MgO sample, which show the strong perpendicular anisotropy (PMA) and the saturation magnetization M s of 868 emu / cm 3 . Figure 1(d) shows the high-angle annular dark field (HAADF) image and energy dispersive x-ray (EDX) mapping in the ( Bi 1 − x Sb x ) 2 Te 3 / Ti / CoFeB / MgO structure, indicating both the ( Bi 1 − x Sb x ) 2 Te 3 / Ti and Ti / CoFeB interfaces are clear and sharp, which promises the high interfacial spin transparency [29] . 1 10.1103/PhysRevLett.123.207205.f1 FIG. 1. (a) Schematic of SOT-induced magnetization switching in TI / Ti / CoFeB / MgO heterostructures. The electrical current flowing in topological surface states is spin-polarized by the spin-momentum locking, and this spin accumulation exerts a spin torque τ SOT on the adjacent magnetic moment M of CoFeB. (b) The RHEED oscillations show the layer-by-layer growth mode of ( Bi 1 − x Sb x ) 2 Te 3 . (c) Magnetization M and Hall resistance R x y as a function of H z . (d) High-angle annular dark field image and energy dispersive x-ray mapping in the ( Bi 1 − x Sb x ) 2 Te 3 / Ti / CoFeB / MgO heterostructure. The SOT-induced magnetization switching is measured at room temperature in the TI ( HM ) / Ti / CoFeB / MgO structures with TIs of ( Bi 1 − x Sb x ) 2 Te 3 [Figs. 2(a) and 2(b) ], Bi 2 Te 3 [Figs. 2(c) and 2(d) ], and SnTe [Figs. 2(e) and 2(f) ], and HMs of Ta [Figs. 2(g) and 2(h) ], W [Figs. S4(a) and S4(b)] [30] , and Pt [Figs. S4(c) and S4(d)] [30] . The nonvolatile SOT switching is measured by the pulsed current, where a 1-ms writing current pulse J W is applied to provide the SOT, followed by another 1-ms reading current pulse J R to read the R x y , where the reversed switching chirality at ± 100 Oe H x shows the typical SOT characteristic. It is known that the switching current density J c depends on the magnitude of the in-plane magnetic field H x [33] , and 100 Oe is enough to overcome the Dzyaloshinskii-Moriya interaction and obtain the saturation (minimum) value of J c in our systems. We calculate the current density J e in TIs and HMs by the parallel circuit model. Compared to HMs (Ta, W, and Pt), the switching current density J c of TIs [ ( Bi 1 − x Sb x ) 2 Te 3 , Bi 2 Te 3 , and SnTe] is much smaller. In ( Bi 1 − x Sb x ) 2 Te 3 , J c of 5.2 × 10 5 A cm − 2 is 1–2 orders of magnitude smaller than the typical value of 10 6 – 10 7 A cm − 2 in HMs, which indicates the high charge-spin conversion efficiency in topological surface states. It is worth noting that the chirality of SOT switching in Bi 2 Te 3 is opposite to that in ( Bi 1 − x Sb x ) 2 Te 3 and SnTe, which comes from the bulk states contribution, as to be discussed later. 2 10.1103/PhysRevLett.123.207205.f2 FIG. 2. Room-temperature SOT-induced magnetization switching in TI ( HM ) / Ti / CoFeB / MgO heterostructures. The R x y - J e curves are shown in the figures for the TI ( HM ) / Ti / CoFeB / MgO heterostructures with ( Bi 1 − x Sb x ) 2 Te 3 (a),(b); Bi 2 Te 3 (c),(d); SnTe (e),(f); and Ta (g),(h), respectively, where the in-plane magnetic field H x of ± 100 Oe is applied for the deterministic SOT switching, respectively. The harmonic Hall method [34,35] is employed to quantify the SOT: when we apply an ac current density J e = J 0 sin ω t , the SOT-induced effective field H SOT = H 0 sin ω t exerts the oscillation of M , which contributes to the 2 ω Hall signal R x y 2 ω . When H x is larger than the magnetic anisotropy field H k , R x y 2 ω can be written as R x y 2 ω = R A 2 H DL | H x | − H k + R P H FL | H x | + R SSE + ANE H x | H x | + R offset , (1) where H DL and H FL represent the effective field from the dampinglike [ m × ( m × σ ) ] and fieldlike ( m × σ ) torques, respectively; R A and R P represent the anomalous Hall and planar Hall resistances [36] , respectively; R SSE + ANE is the thermal contribution [37,38] , and R offset is the offset signal. The harmonic Hall signals R x y 1 ω and R x y 2 ω as a function of H x are measured in the TI ( HM ) / Ti / CoFeB / MgO structures with TIs of ( Bi 1 − x Sb x ) 2 Te 3 [Figs. 3(a) and 3(b) ], Bi 2 Te 3 [Figs. 3(c) and 3(d) ] and SnTe [Figs. 3(e) and 3(f) ], and HMs of Ta [Figs. 3(g) and 3(h) ], W [Figs. S5(a) and S5(b)] [30] , and Pt [Figs. S5(c) and S5(d)] [30] . By fitting the R x y 2 ω - H x curve with Eq. (1) , we can obtain H DL and χ SOT = H DL / J e . θ SH is calculated by θ SH = ( 2 | e | M s t F / ℏ ) χ SOT , where e is the electron charge, t F is the magnetic film thickness, and ℏ is the reduced Planck constant. The polar magneto-optic Kerr effect (MOKE) is also employed to measure χ SOT [30] , where χ SOT obtained by the optical MOKE method is consistent with that from the harmonic Hall method, indicating the asymmetric magnon scattering [10] contribution is negligible. Moreover, due to the shunting effect and the small magnetic field we use, the second harmonic contribution of the bilinear magnetoelectric resistance from TIs themselves is also negligible [39] . 3 10.1103/PhysRevLett.123.207205.f3 FIG. 3. Harmonic Hall measurement in TI ( HM ) / Ti / CoFeB / MgO heterostructures. The 1 ω and 2 ω harmonic Hall resistances ( R x y 1 ω and R x y 2 ω ) as a function of in-plane magnetic field H x for the TI ( HM ) / Ti / CoFeB / MgO heterostructures with ( Bi 1 − x Sb x ) 2 Te 3 (a),(b); Bi 2 Te 3 (c),(d); SnTe (e),(f); and Ta (g),(h), respectively. Table I summarizes | J c | , | χ SOT | , | θ SH | , magnetic anisotropy field H k , electrical conductivity σ c , t s , | q ICS | , and power dissipation density P D in TIs [ ( Bi 1 − x Sb x ) 2 Te 3 , Bi 2 Te 3 , and SnTe] and HMs (Ta, W, and Pt). The t s in TIs is estimated from the half hybridization thickness of the top and bottom surface states [40] , and in HMs is obtained by the spin diffusion length [41] . | θ SH | and | q ICS | in ( Bi 1 − x Sb x ) 2 Te 3 are more than one order of magnitude larger than those in HMs, which is consistent with the ultralow J c of 5.2 × 10 5 A cm − 2 ( 10 6 – 10 7 A cm − 2 in HMs). P D is proportional to J c 2 / σ c , by considering the smaller σ c in ( Bi 1 − x Sb x ) 2 Te 3 that increases the Ohmic loss, P D in ( Bi 1 − x Sb x ) 2 Te 3 ( 0.15 × 10 16 W m − 3 ) is still much reduced compared to HMs ( 0.35 – 12.8 × 10 16 W m − 3 ). I 10.1103/PhysRevLett.123.207205.t1 TABLE I. Room-temperature | J c | , | χ SOT | , | θ SH | , H k , σ c , t s , | q ICS | , and P D of TIs and HMs in this work. | J c | ( 10 6 A cm − 2 ) | χ SOT | ( 10 − 6 Oe A − 1 cm 2 ) | θ SH | H k (kOe) σ c ( 10 4 Ω − 1 m − 1 ) t s (nm) | q ICS | ( nm − 1 ) P D ( 10 16 W m − 3 ) ( BiSb ) 2 Te 3 0.52 67.6 2.50 2.24 1.83 1.5 1.67 0.15 Bi 2 Te 3 2.43 2.20 0.08 2.06 15.0 1.5 0.05 0.39 SnTe 1.46 38.1 1.41 2.18 5.45 1.5 0.94 0.39 Ta 3.44 4.94 0.19 2.10 34.3 1.9 0.10 0.35 W 4.57 3.65 0.13 1.88 38.8 2.1 0.06 0.54 Pt 33.5 0.30 0.01 1.84 87.2 7.3 0.001 12.8 The sign of θ SH in ( Bi 1 − x Sb x ) 2 Te 3 and SnTe is the same as that in Ta (negative), indicating the bottom surface states of TIs dominate the SOT, because the work function difference in top TI/metal interface shifts the Fermi level away from the surface states and thus smears out the helical spin structure in the top surface [42] . This can also be proven by the TI thickness dependence [30] and voltage gating measurements [11] . The signs of θ SH and q ICS in Bi 2 Te 3 are opposite to those in ( Bi 1 − x Sb x ) 2 Te 3 and SnTe, indicating their different SOT origins: bulk states in Bi 2 Te 3 and surface states in ( Bi 1 − x Sb x ) 2 Te 3 and SnTe. It is worth noting that | θ SH | and | q ICS | are enhanced when the topological surface states dominate as in ( Bi 1 − x Sb x ) 2 Te 3 and SnTe, which contributes to a smaller σ c , while the dominating bulk states in Bi 2 Te 3 contribute to a much reduced | θ SH | and | q ICS | . Tuning the band structure of TIs to eliminate the bulk states contribution is very crucial for the intrinsic quantum transport of topological surface states [43,44] . Moreover, SOC from bulk states in TIs could also contribute to the SOT. Therefore, tuning the Fermi level of TIs by band engineering and investigating the Fermi level dependence are very essential to figure out the dominating SOT contribution [45] . We tune the Fermi level E F by changing the Sb ratios ( x ) in ( Bi 1 − x Sb x ) 2 Te 3 [46] , as shown in Fig. 4(a) . From the two-dimensional carrier density | n 2 D | and resistivity ρ x x as a function of Sb ratios x in Fig. 4(b) , we can obtain that E F starts from the bulk conduction band ( x = 0 , n type), to the topological surface band ( x = 0.7 – 0.93 , n type), and then to the bulk valence band ( x = 1.0 , p type). E F is close to the Dirac point at x = 0.85 and 0.93, which shows the ideal topological properties with much insulating bulk. 4 10.1103/PhysRevLett.123.207205.f4 FIG. 4. (a) Schematic of the Fermi level positions for different Sb ratios ( x = 0 , 0.7, 0.78, 0.85, 0.93, 1.0) of ( Bi 1 − x Sb x ) 2 Te 3 , which are estimated from the two-dimensional (2D) carrier density | n 2 D | and resistivity ρ x x . (b) | n 2 D | and ρ x x as a function of Sb ratios in ( Bi 1 − x Sb x ) 2 Te 3 . (c) Switching current density | J c | and SOT-induced effective field | χ SOT | as a function of Sb ratios. Then we investigate the SOT in these samples [ ( Bi 1 − x Sb x ) 2 Te 3 / Ti / CoFeB / MgO ] with varied E F , and the details are shown in the Supplemental Material [30] . J c and χ SOT as a function of Sb ratios are shown in Fig. 4(c) , which show that ( Bi 1 − x Sb x ) 2 Te 3 with much insulating bulk and conducting surface states contribute to larger χ SOT and smaller J c . By comparing the results of x = 0 ( E F in the bulk conduction band), x = 0.7 – 0.93 ( E F in the topological surface band), and x = 1.0 ( E F in the bulk valence band), we obtain that χ SOT from topological surface states can be more than one order of magnitude larger than that from the bulk states. χ SOT reaches the maximum value ( 67.6 × 10 − 6 Oe A − 1 cm 2 ) near the Dirac point ( x = 0.93 ), while J c is significantly reduced ( 5.2 × 10 5 A cm − 2 ) at the same time. In our work, the main purpose of tuning the Fermi level near the Dirac point is to minimize the bulk states contribution, and the most dominating topological surface states contribute to the maximal SOT. In this Letter, we systematically investigate the SOT from TIs and HMs in TI ( HM ) / Ti / CoFeB / MgO systems. One of the main purposes of our work is to resolve the huge discrepancy ( θ SH from 0.047 to 425) of SOT from TIs in different systems characterized by different methods, and to make a conclusive comparison between TIs and HMs. By using the same method and the same structures, our results clearly show that TIs have much higher SOT and energy efficiency than HMs even at room temperature. By tuning the Fermi level of TIs, we show that the SOT is significantly enhanced when the topological surface states dominate, while the bulk states contribute to a very small SOT. These findings indicate that the discrepancy of SOT efficiency from TIs in previous works comes from the different contributions from the bulk and topological surface states. Our work unambiguously demonstrates the giant SOT from topological surface states at room temperature. This work is supported by the NSF Grant No. 1611570, the Nanosystems Engineering Research Center for Translational Applications of Nanoscale Multiferroic Systems (TANMS), the U.S. Army Research Office MURI program under Grants No. W911NF-16-1-0472 and No. W911NF-15-1-10561, and the Spins and Heat in Nanoscale Electronic Systems (SHINES) Center funded by the U.S. Department of Energy (DOE), under Award No. DE-SC0012670. We are also grateful to the support from the Function Accelerated nanoMaterial Engineering (FAME) Center, and a Semiconductor Research Corporation (SRC) program sponsored by Microelectronics Advanced Research Corporation (MARCO) and Defense Advanced Research Projects Agency (DARPA). X. F. H acknowledges the National Key Research and Development Program of China [Grant No. 2017YFA0206200], and the NSFC of China [Grant No. 11434014]. [1] 1 L. Liu , C.-F. Pai , Y. Li , H. W. Tseng , D. C. Ralph , and R. A. Buhrman , Science 336 , 555 ( 2012 ). SCIEAS 0036-8075 10.1126/science.1218197 [2] 2 I. M. Miron , K. Garello , G. Gaudin , P.-J. Zermatten , M. V. Costache , S. Auffret , S. Bandiera , B. Rodmacq , A. Schuhl , and P. Gambardella , Nature (London) 476 , 189 ( 2011 ). NATUAS 0028-0836 10.1038/nature10309 [3] 3 L. Liu , O. J. Lee , T. J. Gudmundsen , D. C. Ralph , and R. A. Buhrman , Phys. Rev. Lett. 109 , 096602 ( 2012 ). PRLTAO 0031-9007 10.1103/PhysRevLett.109.096602 [4] 4 C.-F. Pai , L. Liu , Y. Li , H. W. Tseng , D. C. Ralph , and R. A. Buhrman , Appl. Phys. Lett. 101 , 122404 ( 2012 ). APPLAB 0003-6951 10.1063/1.4753947 [5] 5 C.-F. Pai , Y. Ou , L. H. Vilela-Leão , D. C. Ralph , and R. A. Buhrman , Phys. Rev. B 92 , 064426 ( 2015 ). PRBMDO 1098-0121 10.1103/PhysRevB.92.064426 [6] 6 X. Zhang , C. H. Wan , Z. H. Yuan , Q. T. Zhang , H. Wu , L. Huang , W. J. Kong , C. Fang , U. Khan , and X. F. Han , Phys. Rev. B 94 , 174434 ( 2016 ). PRBMDO 2469-9950 10.1103/PhysRevB.94.174434 [7] 7 X. Wang , C. Wan , W. Kong , X. Zhang , Y. Xing , C. Fang , B. Tao , W. Yang , L. Huang , H. Wu , M. Irfan , and X. Han , Adv. Mater. 30 , 1801318 ( 2018 ). ADVMEW 0935-9648 10.1002/adma.201801318 [8] 8 D. Wu , G. Yu , Q. Shao , X. Li , H. Wu , K. L. Wong , Z. Zhang , X. Han , P. K. Amiri , and K. L. Wang , Appl. Phys. Lett. 108 , 212406 ( 2016 ). APPLAB 0003-6951 10.1063/1.4952771 [9] 9 Y. Fan , P. Upadhyaya , X. Kou , M. Lang , S. Takei , Z. Wang , J. Tang , L. He , L.-T. Chang , M. Montazeri , G. Yu , W. Jiang , T. Nie , R. N. Schwartz , Y. Tserkovnyak , and K. L. Wang , Nat. Mater. 13 , 699 ( 2014 ). NMAACR 1476-1122 10.1038/nmat3973 [10] 10 K. Yasuda , A. Tsukazaki , R. Yoshimi , K. Kondou , K. S. Takahashi , Y. Otani , M. Kawasaki , and Y. Tokura , Phys. Rev. Lett. 119 , 137204 ( 2017 ). PRLTAO 0031-9007 10.1103/PhysRevLett.119.137204 [11] 11 Y. Fan , X. Kou , P. Upadhyaya , Q. Shao , L. Pan , M. Lang , X. Che , J. Tang , M. Montazeri , K. Murata , L.-T. Chang , M. Akyol , G. Yu , T. Nie , K. L. Wong , J. Liu , Y. Wang , Y. Tserkovnyak , and K. L. Wang , Nat. Nanotechnol. 11 , 352 ( 2016 ). NNAABX 1748-3387 10.1038/nnano.2015.294 [12] 12 J. Han , A. Richardella , S. A. Siddiqui , J. Finley , N. Samarth , and L. Liu , Phys. Rev. Lett. 119 , 077702 ( 2017 ). PRLTAO 0031-9007 10.1103/PhysRevLett.119.077702 [13] 13 M. Dc , R. Grassi , J.-Y. Chen , M. Jamali , D. Reifsnyder Hickey , D. Zhang , Z. Zhao , H. Li , P. Quarterman , Y. Lv , M. Li , A. Manchon , K. A. Mkhoyan , T. Low , and J.-P. Wang , Nat. Mater. 17 , 800 ( 2018 ). NMAACR 1476-1122 10.1038/s41563-018-0136-z [14] 14 N. H. D. Khang , Y. Ueda , and P. N. Hai , Nat. Mater. 17 , 808 ( 2018 ). NMAACR 1476-1122 10.1038/s41563-018-0137-y [15] 15 Y. Wang , D. Zhu , Y. Wu , Y. Yang , J. Yu , R. Ramaswamy , R. Mishra , S. Shi , M. Elyasi , K.-L. Teo , Y. Wu , and H. Yang , Nat. Commun. 8 , 1364 ( 2017 ). NCAOBW 2041-1723 10.1038/s41467-017-01583-4 [16] 16 Y. Wang , P. Deorani , K. Banerjee , N. Koirala , M. Brahlek , S. Oh , and H. Yang , Phys. Rev. Lett. 114 , 257202 ( 2015 ). PRLTAO 0031-9007 10.1103/PhysRevLett.114.257202 [17] 17 A. R. Mellnik , J. S. Lee , A. Richardella , J. L. Grab , P. J. Mintun , M. H. Fischer , A. Vaezi , A. Manchon , E. A. Kim , N. Samarth , and D. C. Ralph , Nature (London) 511 , 449 ( 2014 ). NATUAS 0028-0836 10.1038/nature13534 [18] 18 Y. Liu , J. Besbas , Y. Wang , P. He , M. Chen , D. Zhu , Y. Wu , J. M. Lee , L. Wang , J. Moon , N. Koirala , S. Oh , and H. Yang , Nat. Commun. 9 , 2492 ( 2018 ). NCAOBW 2041-1723 10.1038/s41467-018-04939-6 [19] 19 Y. Zhang , K. He , C.-Z. Chang , C.-L. Song , L.-L. Wang , X. Chen , J.-F. Jia , Z. Fang , X. Dai , W.-Y. Shan , S.-Q. Shen , Q. Niu , X.-L. Qi , S.-C. Zhang , X.-C. Ma , and Q.-K. Xue , Nat. Phys. 6 , 584 ( 2010 ). NPAHAX 1745-2473 10.1038/nphys1689 [20] 20 S. Shi , A. Wang , Y. Wang , R. Ramaswamy , L. Shen , J. Moon , D. Zhu , J. Yu , S. Oh , Y. Feng , and H. Yang , Phys. Rev. B 97 , 041115(R) ( 2018 ). PRBMDO 2469-9950 10.1103/PhysRevB.97.041115 [21] 21 J. C. R. Sánchez , L. Vila , G. Desfonds , S. Gambarelli , J. P. Attané , J. M. De Teresa , C. Magén , and A. Fert , Nat. Commun. 4 , 2944 ( 2013 ). NCAOBW 2041-1723 10.1038/ncomms3944 [22] 22 D. Hsieh , Y. Xia , D. Qian , L. Wray , J. H. Dil , F. Meier , J. Osterwalder , L. Patthey , J. G. Checkelsky , N. P. Ong , A. V. Fedorov , H. Lin , A. Bansil , D. Grauer , Y. S. Hor , R. J. Cava , and M. Z. Hasan , Nature (London) 460 , 1101 ( 2009 ). NATUAS 0028-0836 10.1038/nature08234 [23] 23 Z. H. Pan , E. Vescovo , A. V. Fedorov , D. Gardner , Y. S. Lee , S. Chu , G. D. Gu , and T. Valla , Phys. Rev. Lett. 106 , 257004 ( 2011 ). PRLTAO 0031-9007 10.1103/PhysRevLett.106.257004 [24] 24 C. H. Li , O. M. J. van ‘t Erve , J. T. Robinson , Y. Liu , L. Li , and B. T. Jonker , Nat. Nanotechnol. 9 , 218 ( 2014 ). NNAABX 1748-3387 10.1038/nnano.2014.16 [25] 25 G. Yu , P. Upadhyaya , Y. Fan , J. G. Alzate , W. Jiang , K. L. Wong , S. Takei , S. A. Bender , L.-T. Chang , Y. Jiang , M. Lang , J. Tang , Y. Wang , Y. Tserkovnyak , P. K. Amiri , and K. L. Wang , Nat. Nanotechnol. 9 , 548 ( 2014 ). NNAABX 1748-3387 10.1038/nnano.2014.94 [26] 26 O. J. Lee , L. Q. Liu , C. F. Pai , Y. Li , H. W. Tseng , P. G. Gowtham , J. P. Park , D. C. Ralph , and R. A. Buhrman , Phys. Rev. B 89 , 024418 ( 2014 ). PRBMDO 1098-0121 10.1103/PhysRevB.89.024418 [27] 27 S. Emori , U. Bauer , S.-M. Ahn , E. Martinez , and G. S. D. Beach , Nat. Mater. 12 , 611 ( 2013 ). NMAACR 1476-1122 10.1038/nmat3675 [28] 28 C. Du , H. Wang , F. Yang , and P. C. Hammel , Phys. Rev. B 90 , 140407(R) ( 2014 ). PRBMDO 1098-0121 10.1103/PhysRevB.90.140407 [29] 29 W. Zhang , W. Han , X. Jiang , S.-H. Yang , and S. S. P. Parkin , Nat. Phys. 11 , 496 ( 2015 ). NPAHAX 1745-2473 10.1038/nphys3304 [30] 30 See Supplemental Material at http://link.aps.org/supplemental/10.1103/PhysRevLett.123.207205 for the experimental methods, magnetic properties, SOT switching and harmonic Hall measurements in ( W , Pt ) / Ti / CoFeB / MgO structures, optical MOKE method to obtain χ SOT , temperature dependence, and TI thickness dependence of SOT, which includes Refs. [31–32]. [31] 31 M. Montazeri , P. Upadhyaya , M. C. Onbasli , G. Yu , K. L. Wong , M. Lang , Y. Fan , X. Li , P. Khalili Amiri , R. N. Schwartz , C. A. Ross , and K. L. Wang , Nat. Commun. 6 , 8958 ( 2015 ). NCAOBW 2041-1723 10.1038/ncomms9958 [32] 32 X. Fan , H. Celik , J. Wu , C. Ni , K.-J. Lee , V. O. Lorenz , and J. Q. Xiao , Nat. Commun. 5 , 3042 ( 2014 ). NCAOBW 2041-1723 10.1038/ncomms4042 [33] 33 Q. Hao and G. Xiao , Phys. Rev. Applied 3 , 034009 ( 2015 ). PRAHB2 2331-7019 10.1103/PhysRevApplied.3.034009 [34] 34 M. Hayashi , J. Kim , M. Yamanouchi , and H. Ohno , Phys. Rev. B 89 , 144425 ( 2014 ). PRBMDO 1098-0121 10.1103/PhysRevB.89.144425 [35] 35 Q. Shao , G. Yu , Y.-W. Lan , Y. Shi , M.-Y. Li , C. Zheng , X. Zhu , L.-J. Li , P. K. Amiri , and K. L. Wang , Nano Lett. 16 , 7514 ( 2016 ). NALEFD 1530-6984 10.1021/acs.nanolett.6b03300 [36] 36 S. Cho and B.-G. Park , Curr. Appl. Phys. 15 , 902 ( 2015 ). 1567-1739 10.1016/j.cap.2015.04.002 [37] 37 G. E. W. Bauer , E. Saitoh , and B. J. van Wees , Nat. Mater. 11 , 391 ( 2012 ). NMAACR 1476-1122 10.1038/nmat3301 [38] 38 C. O. Avci , K. Garello , M. Gabureac , A. Ghosh , A. Fuhrer , S. F. Alvarado , and P. Gambardella , Phys. Rev. B 90 , 224427 ( 2014 ). PRBMDO 1098-0121 10.1103/PhysRevB.90.224427 [39] 39 P. He , S. S. L. Zhang , D. Zhu , Y. Liu , Y. Wang , J. Yu , G. Vignale , and H. Yang , Nat. Phys. 14 , 495 ( 2018 ). NPAHAX 1745-2473 10.1038/s41567-017-0039-y [40] 40 K. Park , J. J. Heremans , V. W. Scarola , and D. Minic , Phys. Rev. Lett. 105 , 186801 ( 2010 ). PRLTAO 0031-9007 10.1103/PhysRevLett.105.186801 [41] 41 H. L. Wang , C. H. Du , Y. Pu , R. Adur , P. C. Hammel , and F. Y. Yang , Phys. Rev. Lett. 112 , 197201 ( 2014 ). PRLTAO 0031-9007 10.1103/PhysRevLett.112.197201 [42] 42 J. Zhang , J. P. Velev , X. Dang , and E. Y. Tsymbal , Phys. Rev. B 94 , 014435 ( 2016 ). PRBMDO 2469-9950 10.1103/PhysRevB.94.014435 [43] 43 C.-Z. Chang , J. Zhang , X. Feng , J. Shen , Z. Zhang , M. Guo , K. Li , Y. Ou , P. Wei , L.-L. Wang , Z.-Q. Ji , Y. Feng , S. Ji , X. Chen , J. Jia , X. Dai , Z. Fang , S.-C. Zhang , K. He , Y. Wang , L. Lu , X.-C. Ma , and Q.-K. Xue , Science 340 , 167 ( 2013 ). SCIEAS 0036-8075 10.1126/science.1234414 [44] 44 A. J. Bestwick , E. J. Fox , X. Kou , L. Pan , K. L. Wang , and D. Goldhaber-Gordon , Phys. Rev. Lett. 114 , 187201 ( 2015 ). PRLTAO 0031-9007 10.1103/PhysRevLett.114.187201 [45] 45 K. Kondou , R. Yoshimi , A. Tsukazaki , Y. Fukuma , J. Matsuno , K. S. Takahashi , M. Kawasaki , Y. Tokura , and Y. Otani , Nat. Phys. 12 , 1027 ( 2016 ). NPAHAX 1745-2473 10.1038/nphys3833 [46] 46 J. Zhang , C.-Z. Chang , Z. Zhang , J. Wen , X. Feng , K. Li , M. Liu , K. He , L. Wang , X. Chen , Q.-K. Xue , X. Ma , and Y. Wang , Nat. Commun. 2 , 574 ( 2011 ). NCAOBW 2041-1723 10.1038/ncomms1588

PY - 2019/11/15

Y1 - 2019/11/15

N2 - Spin-momentum locked surface states in topological insulators (TIs) provide a promising route for achieving high spin-orbit torque (SOT) efficiency beyond the bulk spin-orbit coupling in heavy metals (HMs). However, in previous works, there is a huge discrepancy among the quantitative SOTs from TIs in various systems determined by different methods. Here, we systematically study the SOT in the TI(HM)/Ti/CoFeB/MgO systems by the same method, and make a conclusive assessment of SOT efficiency for TIs and HMs. Our results demonstrate that TIs show more than one order of magnitude higher SOT efficiency than HMs even at room temperature, at the same time the switching current density as low as 5.2×105 A cm-2 is achieved with (Bi1-xSbx)2Te3. Furthermore, we investigate the relationship between SOT efficiency and the position of Fermi level in (Bi1-xSbx)2Te3, where the SOT efficiency is significantly enhanced near the Dirac point, with the most insulating bulk and conducting surface states, indicating the dominating SOT contribution from topological surface states. This work unambiguously demonstrates the ultrahigh SOT efficiency from topological surface states.

AB - Spin-momentum locked surface states in topological insulators (TIs) provide a promising route for achieving high spin-orbit torque (SOT) efficiency beyond the bulk spin-orbit coupling in heavy metals (HMs). However, in previous works, there is a huge discrepancy among the quantitative SOTs from TIs in various systems determined by different methods. Here, we systematically study the SOT in the TI(HM)/Ti/CoFeB/MgO systems by the same method, and make a conclusive assessment of SOT efficiency for TIs and HMs. Our results demonstrate that TIs show more than one order of magnitude higher SOT efficiency than HMs even at room temperature, at the same time the switching current density as low as 5.2×105 A cm-2 is achieved with (Bi1-xSbx)2Te3. Furthermore, we investigate the relationship between SOT efficiency and the position of Fermi level in (Bi1-xSbx)2Te3, where the SOT efficiency is significantly enhanced near the Dirac point, with the most insulating bulk and conducting surface states, indicating the dominating SOT contribution from topological surface states. This work unambiguously demonstrates the ultrahigh SOT efficiency from topological surface states.

UR - http://www.scopus.com/inward/record.url?scp=85075113013&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85075113013&partnerID=8YFLogxK

U2 - 10.1103/PhysRevLett.123.207205

DO - 10.1103/PhysRevLett.123.207205

M3 - Article

C2 - 31809108

AN - SCOPUS:85075113013

VL - 123

JO - Physical Review Letters

JF - Physical Review Letters

SN - 0031-9007

IS - 20

M1 - 207205

ER -