Abstract
The Dresselhaus and Rashba effects are wellknown phenomena in solidstate physics, in which spin–orbit coupling splits spinup and spindown energy bands of nonmagnetic noncentrosymmetric crystals. Here, we discuss a phenomenon we dub band splitting with vanishing spin polarizations (BSVSP), in which, as usual, spinorbit coupling splits the energy bands in nonmagnetic noncentrosymmetric systems. Surprisingly, however, both split bands show no net spin polarization along certain highsymmetry lines in the Brillouin zone. In order to rationalize this phenomenon, we propose a classification of point groups into pseudopolar and nonpseudopolar groups. By means of firstprinciples simulations, we demonstrate that BSVSP can take place in both symmorphic (e.g., bulk GaAs) and nonsymmorphic systems (e.g., two dimensional ferroelectric SnTe). Furthermore, we identify a linear magnetoelectric coupling in reciprocal space, which could be employed to tune the spin polarization with an external electric field. The BSVSP effect and its manipulation could therefore form the basis for future spintronic devices.
Introduction
The study of a relativistic interaction, i.e., the spinorbit coupling (SOC), has been one of the central themes in the context of spintronics, branch of electronics aiming at utilizing the electron’s spin degree of freedom for device applications^{1}. Many intriguing SOCrelated phenomena were observed, such as spin relaxation^{2,3}, optical spin orientation^{4}, spin Hall effects^{5,6,7,8}, persistent spin structures^{9,10,11}, hidden spin polarization in centrosymmetric systems^{12}, and the spin galvanic effects^{13,14}. In particular, Dresselhaus^{15} and Rashba^{16,17,18} demonstrated that SOC splits spinup and spindown bands in nonmagnetic systems lacking inversion symmetry, inducing an effective magnetic field that depends on the crystal momentum k. Although the Dresselhaus and Rashba effects lead to different spin polarizations of the energy bands, they result in a similar band dispersion: the original spin degenerate bands split so that spinup and spindown bands shift towards opposite directions in kspace, as shown in Fig. 1. Even though the Rashba effect is generally associated to heterostructures and surfaces, as due to 2D structural inversion asymmetry, a giant Rashba effect has been recently observed in bulk polar/ferroelectric systems^{19,20,21,22,23}. Sometimes, Dresselhaus and Rashba terms coexist^{24}, giving rise to interesting physics and applications. For example, if the magnitudes of Dresselhaus and Rashba terms are equal, the spin–orbit field is unidirectional, resulting in a momentumindependent spin configuration (i.e., the socalled “persistent spin texture”), which could support an extraordinarily long spin lifetime of carriers^{9}. To further develop the field of spin–orbitronics (a branch of spintronics)^{25}, it is of great interest to explore SOCrelated phenomena beyond the wellknown DresselhausRashba effects. Indeed, a first contribution in this direction has been recently provided by a careful analysis of bulk and site symmetries of crystalline materials, that allowed for a unified description of spinsplitting effects and, more importantly, to unveil hidden spinpolarization phenomena in centrosymmetric systems^{12}, recently observed in several layered compounds^{26,27,28,29,30,31,32}.
In this work, we show that band splitting with vanishing spin polarizations (BSVSP) in acentric nonmagnetic materials, as shown in Fig. 1. The group theory analysis indicates that the band splitting arises from the breaking of inversion symmetry and the vanishing spin polarization (i.e., the expectation value of spin evaluated over each Bloch wavefunction) is due to the presence of additional symmetries. We demonstrate in details why BSVSP is possible and how to achieve it in two prototypical examples, namely bulk GaAs and twodimensional ferroelectric SnTe, complementing the symmetry analysis with firstprinciples calculations.
Results
Group theory analysis on BSVSP
To facilitate our discussion, we first classify the 32 crystallographic point groups into pseudopolar and nonpseudopolar point groups (Fig. 2), where a pseudovector (such as spin, magnetization, angular moment, magnetic field) is allowed or forbidden, respectively. This classification is in the same spirit as the classification of the point groups into polar and nonpolar point groups. As we will show below, it will be very useful for checking the existence of bulk and local spin polarizations. We note that spin vectors S_{x}, S_{y}, and S_{z} will not belong to the identical representation for a nonpseudopolar point group.
The essence of BSVSP manifests in a nondegenerate band at a given crystal momentum k lacking net spin polarizations in an inversionasymmetry crystal, as shown in Fig. 1. As such, the BSVSP effect cannot be realized in nonmagnetic centrosymmetric crystals, where Kramers theorem implies that all bands are at least twofold degenerate (spin degeneracy). Note that BSVSP does not mean the Bloch states are spinless since the expectation value of S^{2} is not zero. In order to realize BSVSP at a given momentum k, two conditions should be simultaneously satisfied: (1) The eigenstate—including SOC—should be nondegenerate, suggesting that the little space group associated with k should possess at least one 1D doublevalued IR; (2) The little point group associated with k should be a nonpseudopolar point group. Detailed analysis based on group theory (see Supplementary Note 2) suggests that BSVSP can not only occur in systems with symmorphic space groups, but also in systems with nonsymmorphic space groups. In Table 1, we list all the possible 4 symmorphic space groups which might display BSVSP. The corresponding highsymmetry lines where BSVSP takes place are also given. For a given nonsymmorphic space group, one first finds out the possible kpoints on the Brillouin zone boundary where the corresponding little point group has at least one 1D doublevalued IRs. Then, among these kpoints, one finds out the subset of the kpoints whose corresponding little point group is nonpseudopolar. To detect BSVSP in experiment, one can use, for example, spinpolarized angleresolved photoelectron spectroscopy^{26} to measure the spin polarizations of the bands along a special high symmetry line.
Although the net spin polarizations are zero for both split bands in the BSVSP effect, a local spin polarization might exist, in loose analogy with the hidden spinpolarization effect. By projecting the spin polarization on atomic sites in real space, it is possible to evaluate the contribution of magnetization of each atom to a chosen band. The magnetization of each atom must satisfy the symmetry of the intersection of the site symmetry group and little point group of the considered momentum k. If the resulting symmetry group is nonpseudopolar, the local spin polarizations also vanish on the chosen atom. Otherwise, local spin polarizations could survive. We find that both situations can occur [GaAs (see below) and SnI_{4} (see Supplementary Note 4) are representative of the two cases, respectively].
To prove the existence of BSVSP, we calculated from first principles the band structures and related spin polarizations of bulk GaAs and 2D SnTe, representative of symmorphic and nonsymmorphic case, respectively.
BSVSP in zincblend GaAs
Bulk GaAs shows a zincblende structure with \(\bar F43m\) space group whose corresponding point group is T_{d}. There is no inversion symmetry in GaAs, but since the crystallographic point group is nonpolar, it may exhibit Dresselhaus effect (D1, according to the classification given in ref. ^{12}). In Fig. 3a, we show the calculated SOC band structures and the x components of spin polarizations dependence on k in the vicinity of the Γ point for GaAs. The y and z components of spin polarizations have the same behavior. It is wellknown that the top valence bands are fourfold degenerate at Γ. Along symmetry line Λ [Γ → L(0.5, 0.5, 0.5)], the top valence bands split into two nondegenerate bands (labeled by \({\tilde{\mathrm{\Lambda }}}_4\) and \({\tilde{\mathrm{\Lambda }}}_5\), respectively) and a twofold degenerate band (labeled by \({\tilde{\mathrm{\Lambda }}}_6\)). Note that the twofold degenerate bands are composed by spinup and spindown components, resulting in a zero net spin polarization. Interestingly and surprisingly, the top two nondegenerate bands also have vanishing spin polarization, i.e., displaying the BSVSP behavior. Note that all bands are at least twofold degenerate for isostructural bulk Si with the inversion symmetry. This means that inversion asymmetry leads to the splitting of the top valence band in GaAs. It should be noted that BSVSP takes place only along special highsymmetry lines. In fact, conventional Dresselhaus splitting with spin polarization takes place for other symmetry lines [e.g., symmetry line Γ → A(0.5, 0.25, 0), as shown in Fig. 3a].
The vanishing spin polarizations can be rationalized by considering the band symmetry properties. Along the Λ symmetry line, the little point group is C_{3v}. According to our previous analysis, C_{3v} is classified as nonpseudopolar. Therefore, the x, y, and z components of spin polarizations must vanish. Here, we illustrate why the spin polarization along the symmetry line Λ should vanish in another clear way. Adopting a xyzcoordinate system with the local zaxis along the [111] direction of the cubic cell of bulk GaAs, the point group C_{3v} has six symmetry operations, including a mirror symmetry M_{1} which consists of reflection about the y = 0 plane
and a threefold rotation C_{3z} which consists of π/3 rotation around the z axis:
where, S_{x}, S_{y}, and S_{z} are the cartesian components of the spin polarization. The reflection M_{1} ensures the vanishing x and z components of spin polarizations, while the threefold screw rotation C_{3z} ensures the vanishing x and y components of spin polarizations. Overall, the spin polarization should be zero.
The top valence bands of zincblend semiconductors are well described by the 4 × 4 Luttinger model expressed in the basis of total angular momentum j = 3/2^{33,34}. We further adopt this method to prove the unusual band splitting with vanishing spin polarization in GaAs (see Supplementary Note 3). The band structure of the effective Hamiltonian along Λ line are shown in Fig. 3a. The top valence bands are two nondegenerate bands. The x component of spin polarizations is zero for these nondegenerate bands, as shown in Fig. 3b. The y and z components have the same behavior. This k·p analysis confirms our DFT results and group theory analysis. In Dresselhaus/Rashba effect, the spin polarization of the split bands arises as a consequence of the effective kdependent magnetic field experienced by electrons in the presence of SOC in an acentric environment. When the symmetry requirements unveiled by our grouptheory analysis are met, however, the effect of such SOCinduced magnetic fields vanishes, thus giving rise to BSVSP effect. Our further analysis shows indeed that the SOCinduced effective magnetic fields acting on two different p orbitals (e.g., p_{x} and p_{y}) of the same atom (e.g., As) are equal in strength but opposite in direction for the kpoint along the ΓL direction, resulting in a band splitting and a net vanishing spin polarization (since local magnetic moments for different orbitals of the same atom are opposite) (see Fig. 1b and Supplementary Note 3).
Our calculations show that not only the net spin polarization, but also the local spin polarizations of Ga and As atoms for these two single degenerate bands vanish. This can be understood by symmetry arguments. The site symmetry groups of Ga and As atoms are T_{d} and the little group of Λ line is C_{3v}, which is a subgroup of T_{d}. The magnetization of Ga and As atoms must satisfy the symmetry of C_{3v} which is a nonpseudopolar point group.
We propose that one can tune spin polarization with strain. For example, an inplane strain applied to GaAs would lower the crystallographic point group to D_{2d}. The little point group along symmetry line Γ → P (also along [111] direction) is C_{2} which is pseudopolar. As a consequence, a net spin polarization is symmetry allowed, as indeed confirmed by our DFT calculations when applying an inplane compressive strain to cubic GaAs (Supplementary Fig. 2).
BSVSP in 2D ferroelectric SnTe
Recently, 2D ferroelectric SnTe with high Curie temperature has been observed^{35,36}. This material comprises two sheets of alternating Sn and Te atoms, where the inplane ferroelectric polarization—arising from the relative offset of Sn and Te sublattices—breaks the inversion symmetry. Its space group is Pmn2_{1} with C_{2v} crystallographic point group, where we choose the polarization direction as the z axis, while the direction normal to the 2D thin film is taken as the y axis (see Fig. 4b). In Fig. 4a we show the band structure around the X point, where four nondegenerate bands are found along the symmetry lines Σ (X → U) and Δ (X → Γ), merging in two twofold degenerate bands at the timereversal invariant point X, labeled as \({\tilde{\mathrm{X}}}_2\), \({\tilde{\mathrm{X}}}_5\) and \({\tilde{\mathrm{X}}}_3\), \({\tilde{\mathrm{X}}}_4\), respectively. Along the Σ symmetry line, all components of spin polarization vanish, thus realizing the BSVSP effect. On the other hand, the bands along the Δ symmetry line are fully spin polarized along the y axis, as expected for a unidirectional Rashba effect, being the bands spinpolarization perpendicular to both k and ferroelectric polarization. This can be explained by looking at the little groups of the two symmetry lines, following our previous analysis. Along symmetry line Σ, the little point group is C_{2v}, which is nonpseudopolar. Therefore, the net spin polarizations vanish. Since the little point group of Δ symmetry line is the pseudopolar C_{s}, a nonzero spin polarization is allowed along this line. We further construct wave functions using representation theory to prove the vanishing spin polarizations along the symmetry line Σ (see Supplementary Note 6). A minimal k·p model can be written for the valence top bands around X, which reads as \(H =  \Delta \tau _y\sigma _y  \alpha _B\left( {k_z\tau _z\sigma _x  k_x\tau _z\sigma _z} \right) + \beta _Rk_x\tau _0\sigma _y + \alpha _Rk_x\tau _y\sigma _0\), where 2Δ is the SOC induced splitting at X, α_{B} (α_{R} and β_{R}) measure the strength of BSVSP (Rashbalike effect), σ and τ are Pauli matrices representing the spin and pseudospin degrees of freedom, respectively (the latter spanning the twodimensional singlevalued representation \({\tilde{\mathrm{X}}}_1\), see Supplementary Note 6). The k·p model also predicts the BSVSP effect.
Spin polarizations along the Σ symmetry line projected on Sn and Te atoms are shown in Fig. 4b. For each atom, only the x component of the spin polarization does not vanish. This is rationalized by considering the site symmetries of Sn and Te atoms. The intersection of site symmetry groups (C_{s}, comprising a M_{yz} mirror operation about the x = 0 plane) and little point group of Σ line (C_{2v}) is C_{s} which is pseudopolar. Therefore, there are spin polarizations on each atom. Only the x components survive as the glide reflection M_{yz} results in the zero y and z components. However, local spin polarizations projected on different Sn and Te atoms have the same values but opposite sign. They cancel each other so the total spin polarizations vanish. Furthermore, all local spin polarizations are reversed in the conjugate split band. This behavior can be regarded as “antiferromagnetism of a Bloch state”. The antiferromagnetic order satisfies the little space group symmetry of the highsymmetry line Σ. In 2D SnTe, the magnetic fields acting on different atoms of the same kind (e.g., Sn) are opposite in direction for the kpoint along the XU direction, resulting in a band splitting and a net vanishing spin polarization (since local magnetic moments for the same kind atom at different positions are opposite) (see Fig. 1b and Supplementary Note 6). The presence of a staggered kdependent magnetic field along the XU direction clearly appears in the k·p model restricted to the Σ line, which can be recast as \(H =  \Delta \tau _y\sigma _y + B_x\left( k \right)\sigma _x\), where \(B_x(k) =  \tau _z\alpha _Bk_z\).
Tuning of spin polarization with electric field
Previously, it was proposed that the electric field can switch the electric polarization, thus tune the spin texture^{21,37,38,39}. In the 2D SnTe case, the switching of the inplane electric polarization by 90° or 180° by electric field can also result in a change of the spin texture. Here, we rather propose another electric field effect. Since the BSVSP effect is protected by symmetries (e.g., mirror plane), one can use an electric field to break these symmetries and induce a net spin polarization. For example, the SnTe vanishing spin polarizations are protected by the two reflections, y = 0 and x = 0 planes. Applying an external electric field along the y axis (normal to the 2D thin film), which can break the glide reflection y = 0 plane and reduce the symmetry of 2D SnTe thin film to the pseudopolar C_{s} symmetry, results in the emergence of the x component of spin polarizations. As shown in Fig. 5a, the dependence of the x components on k by applying a 0.1 V/Å external electric field indeed confirms our analysis. By projecting spin polarizations on atoms, one can observe that the x components of spin polarizations with different orientations do not cancel each other any more, as shown in Fig. 5b. This behavior can be regarded as “ferrimagnetism of the Bloch state” and ensures the emergence of spin polarizations. The magnitude of spin polarization at a k point is linearly proportional to the external electric field, as shown in Fig. 5c. In particular, an opposite external electric field can lead to a flop of spin polarizations, as shown in Supplementary Fig. 5. This can be understood since these two states are related by the glide reflection—y = 0 plane. Our k·p model analysis shows that this electric field effect arises from the fact that the electric field causes an energy splitting of the orbitals located at different positions along the outofplane direction (see Supplementary Note 7). In some sense, this effect can be considered as one type of linear magnetoelectric coupling^{40}, namely linear magnetoelectric coupling in reciprocal space. We recall that in the conventional linear magnetoelectric coupling, the electric field induces a change of magnetization in the whole system. On the contrary, in the linear magnetoelectric coupling in reciprocal space hereby proposed, the electric field induces a change of spin polarization for some particular Bloch eigenstates.
The BSVSP effect might lead to some spintronic applications. By applying strain (e.g., via a piezoelectric material) or electric field to the system displaying the BSVSP effect, the spin polarization can be adjusted, thus the spin transport behavior can be tuned, which might lead to spintronic devices (such as spinFET^{41,42,43,44}).
Discussion
In summary, we put forward a phenomenon (dubbed as BSVSP) in nonmagnetic inversionasymmetric systems: band splitting induced by inversionsymmetry breaking, while the net spin polarization is zero. This phenomenon is rationalized in terms of protection by nonpseudopolar symmetries. The BSVSP behavior can occur not only in symmorphic but also in nonsymmorphic systems, as shown by our density functional theory calculations demonstrating that BSVSP takes place in bulk GaAs and 2D ferroelectric SnTe. BSVSP is not only interesting from the physics point of view, but also promising for applications. For instance, we propose a linear magnetoelectric coupling in reciprocal space, which can be adopted to tune the spin polarizations of Bloch states by means of an external electric field.
Methods
DFT calculations
To compute the band structure, our firstprinciples density functional theory (DFT) calculations are performed on the basis of the projector augmentedwave method^{45} encoded in the Vienna ab initio simulation package (VASP)^{46,47} using the generalizedgradient approximation (GGA) of Perdew et al.^{48}. The planewave cutoff energy was set to 400 eV. A 12 × 12 × 12 (12 × 12 × 1) MonkhorstPack kmesh is employed to sample the Brillouin zone of bulk GaAs (2D SnTe thin film).
Data availability
The data that support the findings of this study are available from the corresponding authors upon reasonable request.
References
Manchon, A., Koo, H. C., Nitta, J., Frolov, S. M. & Duine, R. A. New perspectives for Rashba spinorbit coupling. Nat. Mater. 14, 871–882 (2015).
Fabian, J. & Das Sarma, S. Spin relaxation of conduction electrons. J. Vac. Sci. Technol. B 17, 1708–1715 (1999).
Averkiev, N. S. & Golub, L. E. Giant spin relaxation anisotropy in zincblende heterostructures. Phys. Rev. B 60, 15582–15584 (1999).
Meier, F. & Zakharchenya, B. P. Optical Orientation (Elsevier, 2012).
Hirsch, J. E. Spin Hall effect. Phys. Rev. Lett. 83, 1834–1837 (1999).
Sinova, J. et al. Universal intrinsic spin Hall effect. Phys. Rev. Lett. 92, 126603 (2004).
Sinova, J., Valenzuela, S. O., Wunderlich, J., Back, C. H. & Jungwirth, T. Spin Hall effects. Rev. Mod. Phys. 87, 1213–1259 (2015).
Wunderlich, J., Kaestner, B., Sinova, J. & Jungwirth, T. Experimental observation of the spinHall effect in a twodimensional spinorbit coupled semiconductor system. Phys. Rev. Lett. 94, 047204 (2005).
Schliemann, J. Colloquium: persistent spin textures in semiconductor nanostructures. Rev. Mod. Phys. 89, 011001 (2017).
Koralek, J. D. et al. Emergence of the persistent spin helix in semiconductor quantum wells. Nature 458, 610–613 (2009).
Sasaki, A. et al. Direct determination of spinorbit interaction coefficients and realization of the persistent spin helix symmetry. Nat. Nanotech. 9, 703–709 (2014).
Zhang, X., Liu, Q., Luo, J.W., Freeman, A. J. & Zunger, A. Hidden spin polarization in inversionsymmetric bulk crystals. Nat. Phys. 10, 387–393 (2014).
Ganichev, S. D. et al. Spingalvanic effect. Nature 417, 153–156 (2002).
Ganichev, S. D. Spingalvanic effect and spin orientation by current in nonmagnetic semiconductors. Int. J. Mod. Phys. B 22, 1–26 (2008).
Dresselhaus, G. Spinorbit coupling effects in zinc blende structures. Phys. Rev. 100, 580–586 (1955).
Bychkov, Y. A. & Rashba, E. I. Properties of a 2D electron gas with lifted spectral degeneracy. J. Exp. Theor. Phys. Lett. 39, 78–81 (1984).
Bychkov, Y. A. & Rashba, E. I. Oscillatory effects and the magnetic susceptibility of carriers in inversion layers. J. Phys. 17, 6039–6045 (1984).
Rashba, E. I. Properties of semiconductors with an extremum loop. 1. Cyclotron and combinational resonance in a magnetic field perpendicular to the plane of the loop. Sov. Phys. Solid State 2, 1109–1122 (1960).
Ishizaka, K. et al. Giant Rashbatype spin splitting in bulk BiTeI. Nat. Mater. 10, 521–526 (2011).
Bahramy, M. S., Arita, R. & Nagaosa, N. Origin of giant bulk Rashba splitting: application to BiTeI. Phys. Rev. B 84, 041202 (2011).
Picozzi, S. Ferroelectric Rashba semiconductors as a novel class of multifunctional materials. Front. Phys. 2, 10 (2014).
Liebmann, M. et al. Giant Rashbatype spin splitting in ferroelectric GeTe(111). Adv. Mater. 28, 560–565 (2016).
Rinaldi, C. et al. Ferroelectric control of the spin texture in GeTe. Nano Lett. 18, 2751–2758 (2018).
Stroppa, A. et al. Tunable ferroelectric polarization and its interplay with spinorbit coupling in tin iodide perovskites. Nat. Commun. 5, 5900 (2014).
Kuschel, T. & Reiss, G. SPIN ORBITRONICS Charges ride the spin wave. Nat. Nanotech. 10, 22–24 (2015).
Riley, J. M. et al. Direct observation of spinpolarized bulk bands in an inversionsymmetric semiconductor. Nat. Phys. 10, 835–839 (2014).
Bawden, L. et al. Spinvalley locking in the normal state of a transitionmetal dichalcogenide superconductor. Nat. Commun. 7, 11711 (2016).
Yao, W. et al. Direct observation of spinlayer locking by local Rashba effect in monolayer semiconducting PtSe_{2} film. Nat. Commun. 8, 14216 (2017).
SantosCottin, D. et al. Rashba coupling amplification by a staggered crystal field. Nat. Commun. 7, 11258 (2016).
Slawinska, J., Narayan, A. & Picozzi, S. Hidden spin polarization in nonmagnetic centrosymmetric BaNiS_{2} crystal: Signatures from first principles. Phys. Rev. B 94, 241114 (2016).
Wu, S.L. et al. Direct evidence of hidden local spin polarization in a centrosymmetric superconductor LaO_{0.55}F_{0.45}BiS_{2}. Nat. Commun. 8, 1919 (2017).
Gotlieb, K. et al. Revealing hidden spinmomentum locking in a hightemperature cuprate superconductor. Science 362, 1271–1275 (2018).
Luttinger, J. M. Quantum theory of cyclotron resonance in semiconductors: general theory. Phys. Rev. 102, 1030–1041 (1956).
Pidgeon, C. R. & Groves, S. H. Inversionasymmetry and warpinginduced interband magnetooptical transitions in InSb. Phys. Rev. 186, 824–833 (1969).
Chang, K. et al. Discovery of robust inplane ferroelectricity in atomicthick SnTe. Science 353, 274–278 (2016).
Liu, K., Lu, J., Picozzi, S., Bellaiche, L. & Xiang, H. Intrinsic origin of enhancement of ferroelectricity in SnTe ultrathin films. Phys. Rev. Lett. 121, 027601 (2018).
Di Sante, D., Barone, P., Bertacco, R. & Picozzi, S. Electric control of the giant rashba effect in bulk GeTe. Adv. Mater. 25, 509–513 (2013).
da Silveira, L. G. D., Barone, P. & Picozzi, S. RashbaDresselhaus spinsplitting in the bulk ferroelectric oxide BiAlO_{3}. Phys. Rev. B 93, 245159 (2016).
Tao, L. L., Paudel, T. R., Kovalev, A. A. & Tsymbal, E. Y. Reversible spin texture in ferroelectric HfO_{2}. Phys. Rev. B 95, 245141 (2017).
Spaldin, N. A. & Fiebig, M. The renaissance of magnetoelectric multiferroics. Science 309, 391–392 (2005).
Datta, S. & Das, B. Electronic analog of the electrooptic modulator. Appl. Phys. Lett. 56, 665–667 (1990).
Koo, H. C. et al. Control of spin precession in a spininjected field effect transistor. Science 325, 1515–1518 (2009).
Chuang, P. et al. Allelectric allsemiconductor spin fieldeffect transistors. Nat. Nanotech. 10, 35–39 (2015).
Schliemann, J., Egues, J. C. & Loss, D. Nonballistic spinfieldeffect transistor. Phys. Rev. Lett. 90, 146801 (2003).
Blöchl, P. E. Projector augmentedwave method. Phys. Rev. B 50, 17953–17979 (1994).
Kresse, G. & Furthmuller, J. Efficiency of abinitio total energy calculations for metals and semiconductors using a planewave basis set. Comp. Mater. Sci. 6, 15–50 (1996).
Kresse, G. & Furthmuller, J. Efficient iterative schemes for ab initio totalenergy calculations using a planewave basis set. Phys. Rev. B 54, 11169–11186 (1996).
Perdew, J. P., Burke, K. & Ernzerhof, M. Generalized gradient approximation made simple. Phys. Rev. Lett. 77, 3865–3868 (1996).
Acknowledgements
Word at Fudan is supported by NSFC 11825403, the Special Funds for Major State Basic Research (Grant No. 2015CB921700), the Program for Professor of Special Appointment (Eastern Scholar), the Qing Nian Ba Jian Program, and the Fok Ying Tung Education Foundation. K.L. thanks Junsheng Feng for useful discussions.
Author information
Authors and Affiliations
Contributions
H.X. and S.P. proposed the concept and supervised the project. K.L. performed the firstprinciples calculations with the help from W.L. The group theory and k·p analysis were carried out by K.L. with the help of W.L., J.J. and P.B. K.L. prepared the initial draft of the paper. All authors contributed to the writing and revision of the paper.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Peer review information Nature Communications thanks Hena Das, JunWei Luo and the other, anonymous, reviewer(s) for their contribution to the peer review of this work
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Liu, K., Luo, W., Ji, J. et al. Band splitting with vanishing spin polarizations in noncentrosymmetric crystals. Nat Commun 10, 5144 (2019). https://doi.org/10.1038/s4146701913197z
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s4146701913197z
Further reading

Progress of hidden spin polarization in inversionsymmetric crystals
Science China Physics, Mechanics & Astronomy (2022)
Comments
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.