{
  "id": "spin-orbit-interaction",
  "title": "Spin–Orbit Interaction",
  "lang": "en",
  "source_lang": "en",
  "summary": "Spin–orbit interaction couples a particle’s intrinsic spin to its motion, shaping atomic spectra, chemical behavior, magnetism, and electronic transport.",
  "category": "science",
  "subcategory": "physics",
  "type": "concept",
  "aliases": [
    "Spin–orbit coupling",
    "Spin–orbital coupling",
    "LS interaction"
  ],
  "generated_at": "2026-11-05",
  "model": "pro/gpt-6.1-sol",
  "url": "https://aiwiki.page/en/spin-orbit-interaction/",
  "translations": {
    "zh-Hans": {
      "url": "https://aiwiki.page/zh/spin-orbit-interaction/",
      "json": "https://aiwiki.page/zh/spin-orbit-interaction.json"
    }
  },
  "sections": [
    {
      "heading": "",
      "text": "Spin–orbit interaction is a coupling between a particle’s intrinsic spin and degrees of freedom associated with its motion. For an electron in an atom, it links spin to orbital angular momentum in the electric field of the nucleus. In a crystal, it connects spin to electronic motion through the microscopic electric fields and symmetries of the solid. It is a relativistic contribution to quantum mechanics, rather than a separate fundamental force.\n\nThe interaction explains important features of atomic spectra and influences chemical bonding, magnetic anisotropy, spin relaxation, and the electronic structure of solids. It also underlies methods for manipulating spins electrically and helps create topological electronic phases. Although often treated as a small correction, spin–orbit coupling becomes a leading energy scale in many heavy-element compounds and low-dimensional materials. Related spin–orbital couplings appear in nuclear physics and optics, but their microscopic origins and mathematical descriptions must be distinguished from those of electrons in electric fields."
    },
    {
      "heading": "Historical development and physical meaning",
      "text": "The interpretation of spin–orbit interaction emerged from efforts to explain the fine structure of atomic spectra: closely spaced spectral lines that could not be accounted for by the simplest Bohr model. Arnold Sommerfeld introduced relativistic corrections to that model in 1916, obtaining important aspects of the hydrogen spectrum without a modern account of electron spin. Alkali-metal doublets and the anomalous Zeeman effect nevertheless exposed limitations of orbital descriptions. Spectral lines appeared to require an additional, two-valued electronic degree of freedom.\n\nIn 1925, George Uhlenbeck and Samuel Goudsmit proposed electron spin as the source of this degree of freedom. A spinning electron carries a magnetic moment, suggesting that its orientation could interact with the magnetic field associated with orbital motion. A straightforward semiclassical calculation, however, predicted an interaction twice as large as required. Llewellyn Thomas resolved this discrepancy in 1926 by accounting for the relativistic rotation of the electron’s instantaneous rest frame, now called Thomas precession. The resulting factor of one-half remains an essential feature of the leading electronic spin–orbit Hamiltonian.\n\nPaul Dirac’s relativistic electron theory of 1928 placed these results on a more systematic foundation. The Dirac equation incorporates spin and, when expanded at low velocities, yields the familiar spin–orbit term along with other relativistic corrections. This development changed the interpretation of spin from an added mechanical model into an intrinsic property of the relativistic quantum electron. Spin does not mean that an electron is literally a small rotating charged sphere.\n\nThe language of “spin interacting with its own orbit” is useful but incomplete. In a central atomic potential, orbital angular momentum is well defined and the interaction can be written in terms of \\(\\mathbf L\\cdot\\mathbf S\\). In a crystal, an electron generally lacks a conserved orbital angular momentum about a single center. Spin–orbit coupling then appears as a matrix-valued contribution to its band structure, constrained by crystal symmetry. Effective couplings can also involve pseudospins, which label pairs of states rather than pure physical-spin orientations.\n\nThe interaction does not require an externally applied magnetic field. An electrostatic environment alone is sufficient because relativistic transformations connect electric and magnetic fields, and because the quantum dynamics of spin reflects that connection. This distinction is central to electrically controlled spin devices: an applied voltage can alter orbital motion or confinement and thereby affect spin without applying a large magnetic field."
    },
    {
      "heading": "Relativistic origin and the electronic Hamiltonian",
      "text": "Consider an electron moving through a static electric field. In its instantaneous rest frame, the Lorentz transformation produces a magnetic-field contribution associated with that motion. The electron’s magnetic moment interacts with this field. This picture captures the basic mechanism, but the rest frame is accelerating and cannot be treated as a single global inertial reference frame. Including Thomas precession gives the correct leading coefficient. A derivation from relativistic quantum mechanics incorporates these effects without separating them into successive classical pictures.\n\nFor an electron of mass \\(m\\), let \\(V(\\mathbf r)\\) denote its electrostatic potential energy, rather than the electric potential itself. In the absence of an applied magnetic vector potential, the leading spin–orbit contribution to the Hamiltonian is\n\n\\[H_{\\mathrm{SO}} = \\frac{\\hbar}{4m^2c^2} \\boldsymbol{\\sigma}\\cdot \\left(\\nabla V\\times\\mathbf p\\right) = \\frac{1}{2m^2c^2} \\mathbf S\\cdot \\left(\\nabla V\\times\\mathbf p\\right),\\]\n\nwhere \\(\\mathbf p\\) is the momentum operator, \\(\\mathbf S=\\hbar\\boldsymbol{\\sigma}/2\\), and \\(\\boldsymbol{\\sigma}\\) denotes the three Pauli matrices. Here \\(c\\) is the speed of light and \\(\\hbar\\) is the reduced Planck constant. Defining \\(V\\) explicitly matters because using electric potential instead introduces the electron charge into the coefficient.\n\nFor a spherically symmetric potential, the relation \\(\\nabla V=(dV/dr)\\mathbf r/r\\) gives\n\n\\[H_{\\mathrm{SO}} = \\xi(r)\\,\\mathbf L\\cdot\\mathbf S, \\qquad \\xi(r)= \\frac{1}{2m^2c^2r}\\frac{dV}{dr}.\\]\n\nFor an attractive Coulomb potential, \\(V(r)=-Ze^2/(4\\pi\\varepsilon_0r)\\), the radial coefficient is positive. At fixed radial state, the level with \\(j=l+1/2\\) therefore receives a higher spin–orbit energy shift than the level with \\(j=l-1/2\\). More complicated configurations can have different overall level orderings because several electrons, exchange, and shell occupancy enter.\n\nSpin–orbit interaction is only one term in the low-energy expansion of the Dirac theory. At the same order in \\(1/c^2\\), the Hamiltonian contains a relativistic correction to kinetic energy and the Darwin term, which is sensitive to rapid spatial variation of the potential. The hydrogen fine structure cannot consistently be attributed to spin–orbit coupling alone. External magnetic fields additionally require the appropriate kinetic momentum and magnetic interaction terms.\n\nThe expansion is most transparent when electronic speeds are small compared with \\(c\\). Near a heavy nucleus, relativistic effects become substantial, and treating spin–orbit coupling as a weak perturbation may be inadequate. Fully relativistic or carefully constructed effective Hamiltonians are then needed. In solids, replacing the free-electron mass by a band effective mass in the vacuum expression is generally insufficient: coupling to other bands and the orbital character of the wave functions determine the effective interaction."
    },
    {
      "heading": "Atomic angular momentum and spectral structure",
      "text": "In a central potential, orbital angular momentum and spin combine to form total angular momentum,\n\n\\[\\mathbf J=\\mathbf L+\\mathbf S.\\]\n\nThe spin–orbit operator can be rewritten as\n\n\\[\\mathbf L\\cdot\\mathbf S = \\frac{1}{2} \\left(\\mathbf J^2-\\mathbf L^2-\\mathbf S^2\\right).\\]\n\nConsequently, a state labeled by quantum numbers \\(l\\), \\(s\\), and \\(j\\) has the expectation value\n\n\\[\\langle\\mathbf L\\cdot\\mathbf S\\rangle = \\frac{\\hbar^2}{2} \\left[j(j+1)-l(l+1)-s(s+1)\\right].\\]\n\nFor an electron, \\(s=1/2\\), so a level with \\(l>0\\) splits into \\(j=l+1/2\\) and \\(j=l-1/2\\) components. An \\(s\\) orbital has \\(l=0\\), and its direct central-field \\(\\mathbf L\\cdot\\mathbf S\\) contribution vanishes.\n\nFor a fixed radial state, the corresponding eigenvalues of \\(\\mathbf L\\cdot\\mathbf S\\) are \\(\\hbar^2l/2\\) and \\(-\\hbar^2(l+1)/2\\). Their difference explains the characteristic doublet structure of many one-valence-electron atoms. The sodium D lines near 589.0 and 589.6 nanometres arise from transitions between the split \\(3p\\,{}^2P_{3/2,1/2}\\) levels and the \\(3s\\,{}^2S_{1/2}\\) level. These familiar yellow lines are a direct spectroscopic manifestation of spin–orbit coupling.\n\nIn hydrogen-like ions, the characteristic fine-structure energy scales approximately as \\((Z\\alpha)^4mc^2\\), with additional dependence on the principal and angular quantum numbers. Here \\(Z\\) is the nuclear charge and \\(\\alpha\\) is the fine-structure constant. This scaling explains why relativistic splittings grow rapidly along a hydrogenic sequence. In neutral many-electron atoms, screening, radial penetration, and configuration mixing prevent a universal \\(Z^4\\) rule from accurately describing all measured splittings.\n\nFor many light atoms, electrostatic interactions between electrons are stronger than spin–orbit coupling. Individual orbital angular momenta first combine into total \\(\\mathbf L\\), and individual spins into total \\(\\mathbf S\\); these then combine into \\(\\mathbf J\\). This is LS coupling, or Russell–Saunders coupling. Within an approximately isolated term, an effective Hamiltonian proportional to \\(\\mathbf L\\cdot\\mathbf S\\) yields the Landé interval rule: adjacent \\(J\\) levels are separated by an amount proportional to the larger \\(J\\). Configuration interaction and stronger relativistic mixing produce departures from that rule.\n\nIn heavier atoms, it can be more useful to couple each electron’s \\(\\mathbf l_i\\) and \\(\\mathbf s_i\\) into \\(\\mathbf j_i\\) before combining the individual \\(\\mathbf j_i\\) values. This is the \\(jj\\)-coupling limit; many actual atoms lie between the two limits. Neither coupling scheme should be confused with hyperfine interaction, which involves nuclear moments. An external magnetic field introduces another competition: in the strong-field Paschen–Back regime, orbital and spin angular momenta become approximately decoupled from each other."
    },
    {
      "heading": "Molecular states, chemical bonding, and photophysics",
      "text": "In a molecule, the electric field is not generally spherically symmetric, so atomic \\(L\\) and \\(S\\) labels cease to provide a complete description. Nevertheless, the short-range electric fields near nuclei often make atom-centered spin–orbit contributions especially important. The interaction mixes electronic states with compatible total symmetries, including states that would have different total spin in a nonrelativistic treatment. The amount of mixing depends on both the coupling matrix element and the energy separation of the states.\n\nThis mixing modifies the selection rules of spectroscopy. In a spin-independent approximation, electric-dipole transitions conserve total spin, making transitions between pure singlet and triplet states forbidden. Spin–orbit coupling gives those states small components of other spin character, allowing otherwise weak transitions to acquire intensity. “Spin-forbidden” therefore usually means suppressed in an approximation, not rigorously impossible in a real molecule. The effect depends on orbital symmetry and cannot be inferred from spin multiplicity alone.\n\nA major consequence is intersystem crossing, the nonradiative transfer of population between electronic states of different predominant spin multiplicity. Spin–orbit coupling provides a mechanism for changing spin character, while nuclear motion supplies access to favorable geometries and an appropriate density of final vibrational states. A large coupling matrix element does not by itself guarantee rapid transfer: energy gaps, vibrational overlap, and the topology of potential-energy surfaces also matter. The same spin mixing can enable phosphorescence from a predominantly triplet excited state to a predominantly singlet ground state.\n\nHeavy atoms often enhance these effects because their electrons experience stronger relativistic interactions near the nucleus. Platinum- and iridium-containing complexes exploit this behavior in phosphorescent emitters, including materials used in organic light-emitting devices. The relevant excited states may involve metal-centered orbitals, ligand-centered orbitals, and metal-to-ligand charge transfer. Their different orbital compositions determine how effectively atomic spin–orbit coupling enters the molecular excited-state dynamics. The presence of a heavy atom alone does not specify a unique lifetime or emission efficiency.\n\nSpin–orbit coupling also affects the electronic states and magnetic properties of coordination compounds, especially those containing heavy transition metals or lanthanides. It competes with ligand fields, electron–electron interactions, and distortions of the molecular geometry. These competitions influence magnetic anisotropy and the suitability of spin-only descriptions. In chemical calculations, scalar relativistic effects—relativistic changes that do not explicitly couple spin and orbital motion—must be distinguished from spin–orbit effects. For example, attributing every unusual property of gold or mercury solely to spin–orbit coupling would ignore important scalar relativistic changes in orbital energies and radial distributions."
    },
    {
      "heading": "Crystal symmetry and spin-dependent band structure",
      "text": "In a periodic solid, electrons occupy Bloch states labeled by crystal wave vector \\(\\mathbf k\\) and a band index. Spin–orbit coupling mixes spin and orbital character within these states. A band eigenstate need not be an eigenstate of any fixed spin component, and its spin expectation value may have a magnitude smaller than \\(\\hbar/2\\). The effective spin structure depends on the atomic orbitals participating in the bands, the crystal environment, and couplings between bands, rather than simply on the heaviest element present.\n\nTime-reversal symmetry places a crucial restriction on the spectrum. For spinful electrons in a time-reversal-invariant system, a state at \\(\\mathbf k\\) has a partner of equal energy at \\(-\\mathbf k\\). At momenta equivalent to their negatives modulo a reciprocal lattice vector, Kramers’ theorem guarantees a degeneracy. Time-reversal symmetry alone does not generally require twofold degeneracy at every arbitrary \\(\\mathbf k\\), because the partner may occur at a different momentum.\n\nSpatial inversion changes \\(\\mathbf k\\) to \\(-\\mathbf k\\) while leaving spin, an axial vector, unchanged. If a nonmagnetic crystal has both inversion and time-reversal symmetry, their combination enforces twofold degeneracy at every \\(\\mathbf k\\). Spin–orbit coupling can still be strong in such a crystal: it can rearrange orbital energies, change wave functions, and open gaps at crossings without splitting each doubly degenerate band. The absence of visible spin splitting is therefore not evidence for negligible spin–orbit interaction.\n\nWhen inversion symmetry is absent, spin splitting is generally allowed away from time-reversal-invariant momenta. Crystal symmetries determine the permitted direction and momentum dependence of the spin texture and can also protect special degeneracies. Surfaces naturally break inversion symmetry perpendicular to the boundary, even when the bulk crystal is centrosymmetric. Interfaces, polar crystals, and asymmetric quantum wells thus provide important environments for spin-dependent band structures. Magnetically ordered systems require a separate analysis because ordinary time-reversal symmetry may be broken.\n\nThese symmetry statements also clarify the meaning of “spin polarization.” A split Bloch band can exhibit a momentum-dependent spin expectation while the equilibrium sample has zero net magnetization, because time-reversed states contribute oppositely. Conversely, the orbital composition of a wave function can produce unusual spin patterns that are not captured by a simple up-or-down label. Experiments and calculations must distinguish physical spin, orbital angular momentum, and pseudospin when interpreting such textures."
    },
    {
      "heading": "Rashba and Dresselhaus interactions",
      "text": "The Rashba effect describes a common form of spin–orbit coupling in systems with structural inversion asymmetry, particularly surfaces and semiconductor quantum wells. For an ideal two-dimensional electron system in the \\(xy\\) plane, a frequently used Hamiltonian is\n\n\\[H_R=\\alpha_R(\\sigma_x k_y-\\sigma_y k_x),\\]\n\nwhere \\(\\alpha_R\\) has units of energy times length. Combined with a parabolic spin-independent band, this gives\n\n\\[E_\\pm(k)=\\frac{\\hbar^2k^2}{2m^\\ast}\\pm\\alpha_R k.\\]\n\nThe two branches have opposite in-plane spin textures, perpendicular to momentum in this idealized model. They remain degenerate at \\(k=0\\) if time-reversal symmetry is preserved.\n\nEmmanuel Rashba developed the relevant spin-splitting theory in 1960; Yuri Bychkov and Rashba formulated the widely used two-dimensional version in 1984. Structural asymmetry can arise from unequal confining barriers, electrostatic band bending, or an interface between dissimilar materials. A gate voltage can modify confinement, carrier density, and orbital mixing, thereby changing the effective coupling. The coefficient is not determined by the average macroscopic electric field alone: atomic-scale potential gradients and interband matrix elements are essential.\n\nThe Dresselhaus effect, identified by Gene Dresselhaus in 1955, instead originates from inversion asymmetry of the bulk crystal structure. In a zinc-blende semiconductor, the leading bulk conduction-band contribution is cubic in wave vector and contains terms such as\n\n\\[H_D^{\\mathrm{bulk}} = \\gamma\\left[ \\sigma_x k_x(k_y^2-k_z^2) +\\sigma_y k_y(k_z^2-k_x^2) +\\sigma_z k_z(k_x^2-k_y^2) \\right].\\]\n\nConfinement in a quantum well converts part of this interaction into terms linear in the in-plane wave vector. For a commonly used orientation and coordinate convention, these take the form \\(H_D=\\beta(\\sigma_xk_x-\\sigma_yk_y)\\), with additional cubic corrections.\n\nRashba and Dresselhaus contributions can coexist. Their interference produces anisotropic spin splitting and direction-dependent spin dynamics. In the ideal linear model, matching their strengths creates an enhanced symmetry and supports a persistent spin helix: a spatially rotating spin pattern with an unusually long lifetime. The effect was predicted in 2006 and observed experimentally in semiconductor quantum wells in 2009. Cubic Dresselhaus terms, imperfect tuning, and other relaxation processes limit the lifetime in real samples.\n\nThese effective Hamiltonians are local descriptions near selected band extrema, not universal equations for every surface or semiconductor. Strongly warped surface bands, multiple occupied subbands, and substantial orbital mixing require more elaborate models. Their value lies in making symmetry-allowed couplings and electrically tunable spin dynamics explicit, while retaining a transparent connection between momentum and spin."
    },
    {
      "heading": "Magnetism and correlated electronic states",
      "text": "Spin–orbit coupling connects spin orientation to the spatial structure of a crystal. Without it, an ideal isotropic exchange model can often rotate all spins together without changing the energy. Once spin becomes linked to orbitals that experience the crystal environment, different magnetization directions generally have different energies. This is magnetic anisotropy, which determines preferred magnetic axes and helps control the stability of magnetic domains, permanent magnets, and nanoscale magnetic elements.\n\nIn many transition-metal compounds, the crystal field partly suppresses the orbital angular momentum, a phenomenon called orbital quenching. Spin–orbit coupling can nevertheless admix excited orbital states and generate appreciable anisotropy. Where ordinary nondegenerate perturbation theory applies, characteristic anisotropy contributions often scale as \\(\\lambda^2/\\Delta\\), where \\(\\lambda\\) is a spin–orbit energy and \\(\\Delta\\) an orbital excitation energy. Near degeneracies, this estimate can fail, and first-order effects may become important. The sign and magnitude depend on orbital occupancy and crystal symmetry.\n\nSpin–orbit coupling also permits anisotropic interactions between neighboring magnetic moments. An important example is the Dzyaloshinskii–Moriya interaction,\n\n\\[H_{\\mathrm{DM}} = \\mathbf D_{ij}\\cdot (\\mathbf S_i\\times\\mathbf S_j).\\]\n\nProposed phenomenologically by Igor Dzyaloshinskii in 1958 and derived microscopically by Toru Moriya in 1960, it can favor canting or twisting rather than purely parallel or antiparallel alignment. An inversion center at the midpoint of a magnetic bond forbids this interaction for that bond. Other spatial symmetries constrain the allowed direction of \\(\\mathbf D_{ij}\\).\n\nCompetition among exchange, anisotropy, magnetic fields, and Dzyaloshinskii–Moriya interactions can stabilize spiral orders and magnetic skyrmions. These textures are not produced by spin–orbit coupling alone, but their handedness and stability often depend strongly on it. At interfaces between ferromagnets and heavy metals, broken inversion symmetry and strong relativistic coupling can generate sizable interfacial chiral interactions. Their strength is sensitive to atomic arrangement, hybridization, and interface chemistry.\n\nIn correlated \\(4d\\)- and \\(5d\\)-electron compounds, spin–orbit coupling can be comparable to crystal-field splittings and electron–electron interaction energies. In an ideal octahedral description of a \\(5d^5\\) ion, the \\(t_{2g}\\) manifold can reorganize into effective \\(j_{\\mathrm{eff}}=1/2\\) and \\(j_{\\mathrm{eff}}=3/2\\) states. This framework has been influential in understanding layered iridates such as strontium iridate. Real materials depart from the ideal limit through covalency, distortions, and intersite hopping. In suitable lattices, the resulting entangled spin–orbital states can support strongly bond-dependent exchange interactions rather than a simple isotropic spin model."
    },
    {
      "heading": "Spin relaxation, spin currents, and electrical control",
      "text": "A central goal of spintronics is to generate, transport, and detect spin information electrically. Spin–orbit coupling assists these operations by connecting spin to charge motion, but the same connection can also destroy a prepared spin polarization. It therefore plays a dual role: a resource for manipulating spins and a mechanism limiting their persistence. Which role dominates depends on band structure, disorder, phonons, confinement, and the timescale of the experiment.\n\nThe Elliott–Yafet mechanism arises because spin–orbit coupling makes Bloch states mixtures of nominal up- and down-spin components. Scattering by impurities or phonons can then change the spin content of the occupied state. In a simplified regime, the spin-relaxation rate scales approximately as \\(b^2/\\tau_p\\), where \\(b\\) measures spin mixing and \\(\\tau_p\\) is the momentum-relaxation time. The proportionality is material-dependent, and strongly mixed regions of a Fermi surface can contribute disproportionately.\n\nThe D’yakonov–Perel’ mechanism is important in systems with momentum-dependent spin splitting. Between scattering events, an electron’s spin precesses about an effective field \\(\\boldsymbol{\\Omega}(\\mathbf k)\\). Scattering changes \\(\\mathbf k\\) and hence the precession axis. In the rapid-scattering limit, \\(\\Omega\\tau_p\\ll1\\), a characteristic relaxation rate behaves as\n\n\\[\\frac{1}{\\tau_s}\\sim \\langle\\Omega^2\\rangle\\tau_p.\\]\n\nUnlike the simplest Elliott–Yafet scaling, more frequent momentum scattering can slow this form of spin dephasing through motional narrowing. Outside that limit, or in spatially structured spin states, the dynamics require a more detailed treatment.\n\nSpin–orbit coupling also produces transverse transport responses. In the spin Hall effect, an applied electric field can drive oppositely oriented spin components toward opposite transverse directions without requiring a ferromagnet. Intrinsic contributions arise from the band structure, while extrinsic contributions include asymmetric impurity scattering and scattering-induced coordinate shifts. The inverse spin Hall effect converts spin flow into a transverse electrical signal. Because spin is generally not conserved under spin–orbit coupling, defining a bulk spin-current operator requires care; measured spin accumulation and torque provide complementary physical observables.\n\nAt an interface with a magnet, a nonequilibrium spin accumulation or spin current can transfer angular momentum to the magnetic order. The resulting spin–orbit torques can drive precession, move domain walls, or reverse magnetization. Current-induced spin polarization in inversion-asymmetric systems supplies another route to such torques. Proposed and demonstrated device concepts exploit these processes for switching and microwave generation, but their energy cost and reliability depend on current distribution, heat dissipation, magnetic structure, and the details of spin transmission across interfaces."
    },
    {
      "heading": "Topological phases and superconducting systems",
      "text": "Spin–orbit coupling can change not just band energies but the global organization of electronic wave functions. In some materials it reverses the ordering of bands with different orbital character and opens a gap in a way associated with a nontrivial topological invariant. Such systems can be topological insulators: electrically insulating in the bulk but supporting boundary states constrained by topology and symmetry. Strong spin–orbit coupling is an important route to these phases, although it does not guarantee them.\n\nThe two-dimensional quantum spin Hall effect provides a key example. The Kane–Mele model, introduced in 2005, showed how a spin-dependent coupling on a honeycomb lattice could produce counterpropagating edge channels related by time reversal. Graphene’s intrinsic coupling is too small for this proposal to provide a readily accessible large-gap platform. A different route, predicted for mercury telluride quantum wells in 2006, led to experimental evidence for the quantum spin Hall regime in 2007 when the well thickness crossed the band-inversion threshold.\n\nThree-dimensional materials such as bismuth selenide subsequently became prominent examples of strong topological insulators. Their surfaces can host a Dirac-like band with spin and momentum linked in a characteristic texture. The surface states connect bulk valence and conduction bands rather than constituting an ordinary isolated two-dimensional band. Defects can place the chemical potential inside a bulk band, however, so observing a topological surface state does not establish that electrical transport is dominated by the surface.\n\nTime-reversal symmetry restricts scattering between exact Kramers partners, but topological protection is not equivalent to immunity from every source of resistance. Scattering through other available states, inelastic processes, interactions, magnetic perturbations, and coupling between nearby boundaries can affect transport. Likewise, spin–momentum locking does not require every state to have a perfectly polarized physical spin. Orbital entanglement and crystal anisotropy modify the measured texture while preserving the underlying topological classification.\n\nSpin–orbit interaction also figures prominently in proposals involving superconductivity. A semiconductor wire with Rashba coupling, a Zeeman field, and proximity-induced conventional pairing can, in an appropriate parameter regime, enter a one-dimensional topological superconducting phase. Ideal models predict Majorana zero modes at its ends. Spin–orbit coupling supplies a crucial link between spin splitting and pairing, but disorder, multiband occupancy, and spatial inhomogeneity complicate experimental identification. In noncentrosymmetric superconductors, the absence of inversion symmetry also allows singlet-like and triplet-like pairing components to coexist within the spin–orbit-split band structure."
    },
    {
      "heading": "Experimental determination, calculation, and related interactions",
      "text": "Atomic spin–orbit splittings are measured through high-resolution optical, ultraviolet, and X-ray spectroscopy. The observed intervals must be interpreted alongside configuration interaction, hyperfine structure, external-field effects, and radiative corrections. In solids, angle-resolved photoemission spectroscopy maps occupied band dispersions, and spin-resolved variants measure components of their spin polarization. Photoemission matrix elements and the experimental geometry influence the signal, so a measured photoelectron polarization cannot always be identified directly with the complete spin expectation value of the initial Bloch state.\n\nTransport and time-resolved measurements provide complementary information. Oscillations in magnetoresistance can reveal multiple Fermi-surface areas associated with split bands, although density differences alone do not identify the microscopic origin of the splitting. Weak antilocalization reflects spin-dependent interference between scattered electron trajectories and is often used to estimate spin–orbit-related times or lengths. Such estimates depend on the applicable dimensionality, disorder regime, and theoretical model. Time-resolved optical measurements can instead follow spin precession and decay, separating some effects that are combined in a steady-state transport measurement.\n\nElectronic-structure calculations range from adding spin–orbit coupling perturbatively to using fully relativistic spinor methods. In density functional theory, a common approach first obtains scalar relativistic states and then diagonalizes a spin–orbit Hamiltonian in that basis. Fully relativistic pseudopotentials and all-electron methods offer other routes. Small anisotropy energies require particularly careful numerical convergence because they are differences between much larger total energies. Correlated materials may require additional treatments of local interactions and many-body dynamics beyond a standard band calculation.\n\nThe phrase “spin–orbit interaction” also has an important nuclear meaning. The nuclear shell model developed by Maria Goeppert Mayer and independently by J. Hans D. Jensen and collaborators in 1949 incorporated a strong spin–orbit term to explain nuclear magic numbers, notably 28, 50, 82, and 126. The effective ordering favors the higher-\\(j\\) partner in the usual nuclear convention, unlike the simple electronic Coulomb case. This nuclear interaction arises from effective strong-interaction dynamics and nuclear mean fields; it is not the weak electromagnetic correction derived for an electron orbiting a nucleus.\n\nIn optics, spin–orbit interactions of light describe coupling between polarization-related spin and spatial or orbital properties of an electromagnetic field. Tight focusing, inhomogeneous media, interfaces, and geometric phases can convert or correlate these forms of angular momentum. These effects have a different microscopic setting from electronic spin–orbit coupling: photons are massless spin-one particles, and polarization is constrained by the transversality of electromagnetic waves. Synthetic spin–orbit couplings in ultracold atoms provide another distinct realization, using laser-induced couplings between internal states and atomic momentum to engineer Hamiltonians resembling those of spin–orbit-coupled electrons."
    }
  ],
  "links": [
    "spin-angular-momentum",
    "electron",
    "atom",
    "angular-momentum",
    "quantum-mechanics",
    "fine-structure",
    "dirac-equation",
    "electronic-band-structure",
    "magnetic-field",
    "electric-field",
    "inertial-reference-frame",
    "potential-energy",
    "hamiltonian-operator",
    "momentum",
    "speed-of-light",
    "planck-constant",
    "kinetic-energy",
    "fine-structure-constant",
    "molecule",
    "spectroscopy",
    "reciprocal-lattice",
    "magnetization",
    "spintronics",
    "phonon",
    "topological-insulator",
    "superconductivity",
    "angle-resolved-photoemission-spectroscopy",
    "density-functional-theory"
  ],
  "sources": [],
  "license": "https://creativecommons.org/licenses/by/4.0/"
}