{
  "id": "gamma-matrices",
  "title": "Gamma Matrices",
  "lang": "en",
  "source_lang": "en",
  "summary": "Gamma matrices represent Clifford algebra generators and encode the geometry of spacetime in the Dirac equation and the mathematics of spinors.",
  "category": "mathematics",
  "subcategory": "algebra",
  "type": "concept",
  "aliases": [
    "Dirac matrices",
    "Dirac gamma matrices"
  ],
  "generated_at": "2026-11-05",
  "model": "pro/gpt-6.1-sol",
  "url": "https://aiwiki.page/en/gamma-matrices/",
  "translations": {
    "zh-Hans": {
      "url": "https://aiwiki.page/zh/gamma-matrices/",
      "json": "https://aiwiki.page/zh/gamma-matrices.json"
    }
  },
  "sections": [
    {
      "heading": "",
      "text": "Gamma matrices are sets of matrices whose anticommutation relations encode a nondegenerate quadratic form. They provide matrix representations of a Clifford algebra and are central to the description of spinors. Their best-known application is the Dirac equation, where four complex \\(4\\times4\\) matrices describe the relativistic dynamics of spin-\\(\\tfrac12\\) particles such as the electron. The individual matrices depend on a choice of representation; their defining algebraic relations, rather than their particular entries, carry the geometric content."
    },
    {
      "heading": "Defining relations",
      "text": "In a space of dimension \\(d\\), gamma matrices \\(\\gamma^a\\), with \\(a=0,\\ldots,d-1\\), satisfy\n\n\\[\\{\\gamma^a,\\gamma^b\\} \\equiv \\gamma^a\\gamma^b+\\gamma^b\\gamma^a = 2g^{ab}I.\\]\n\nHere \\(g^{ab}\\) are the inverse components of a nondegenerate metric tensor, and \\(I\\) is the identity matrix on the spinor space. The braces denote the anticommutator.\n\nIn an orthonormal frame, the metric is diagonal. Distinct gamma matrices then anticommute, while each matrix squares to either \\(+I\\) or \\(-I\\), according to the corresponding metric sign. Thus gamma matrices cannot ordinarily be replaced by commuting numbers.\n\nFor four-dimensional Minkowski spacetime, a common convention is\n\n\\[\\eta^{\\mu\\nu}=\\operatorname{diag}(1,-1,-1,-1), \\qquad \\mu,\\nu=0,1,2,3.\\]\n\nThe defining relations become\n\n\\[(\\gamma^0)^2=I,\\qquad (\\gamma^i)^2=-I\\quad(i=1,2,3),\\]\n\nand\n\n\\[\\gamma^\\mu\\gamma^\\nu=-\\gamma^\\nu\\gamma^\\mu \\quad\\text{when }\\mu\\ne\\nu.\\]\n\nIndices are raised and lowered with the metric:\n\n\\[\\gamma_\\mu=\\eta_{\\mu\\nu}\\gamma^\\nu.\\]\n\nRepeated upper and lower indices are summed according to the Einstein summation convention. Some authors use the opposite metric signature or define Clifford multiplication with an overall minus sign. Explicit formulas must therefore be read together with their conventions.\n\nLinearizing a quadratic form\n\nFor a covector \\(p_\\mu\\), define the matrix\n\n\\[\\slashed p=\\gamma^\\mu p_\\mu.\\]\n\nThe slash notation means contraction with gamma matrices. The anticommutation relation gives\n\n\\[(\\slashed p)^2 = \\frac12\\{\\gamma^\\mu,\\gamma^\\nu\\}p_\\mu p_\\nu = p_\\mu p^\\mu I.\\]\n\nThus a matrix-valued expression linear in \\(p_\\mu\\) squares to the scalar quadratic form determined by the metric. This is the essential algebraic mechanism behind the Dirac equation."
    },
    {
      "heading": "Origin and the Dirac equation",
      "text": "Paul Dirac introduced the matrices underlying the modern gamma-matrix formalism in his 1928 relativistic theory of the electron. The aim was to construct a wave equation first order in both time and spatial derivatives while remaining compatible with the relativistic energy–momentum relation\n\n\\[E^2=c^2\\mathbf p^2+m^2c^4.\\]\n\nIn the metric convention above, the free Dirac equation is\n\n\\[\\left(i\\hbar c\\,\\gamma^\\mu\\partial_\\mu-mc^2\\right)\\psi=0,\\]\n\nwhere \\(x^0=ct\\), \\(\\hbar\\) is the reduced Planck constant, and \\(\\psi\\) is a four-component spinor field. The matrices act on the spinor components, not on the spacetime coordinates.\n\nMultiplication by the operator with the opposite mass sign yields\n\n\\[\\left(\\hbar^2\\partial_\\mu\\partial^\\mu+m^2c^2\\right)\\psi=0.\\]\n\nConsequently, each component of a free Dirac solution satisfies the Klein–Gordon equation. The converse does not generally hold: the first-order Dirac equation imposes additional relations among the components.\n\nDirac originally used a Hamiltonian form involving matrices \\(\\alpha^i\\) and \\(\\beta\\):\n\n\\[i\\hbar\\partial_t\\psi = \\left(-i\\hbar c\\,\\alpha^i\\partial_i+\\beta mc^2\\right)\\psi.\\]\n\nThese are related to the covariant gamma matrices by\n\n\\[\\beta=\\gamma^0,\\qquad \\alpha^i=\\gamma^0\\gamma^i.\\]\n\nThe term “Dirac matrices” can refer either to the gamma matrices or, in some contexts, to this Hamiltonian set."
    },
    {
      "heading": "Explicit four-dimensional representations",
      "text": "Pauli matrices\n\nCommon representations are built from the Pauli matrices:\n\n\\[\\sigma^1= \\begin{pmatrix} 0&1\\\\ 1&0 \\end{pmatrix}, \\qquad \\sigma^2= \\begin{pmatrix} 0&-i\\\\ i&0 \\end{pmatrix}, \\qquad \\sigma^3= \\begin{pmatrix} 1&0\\\\ 0&-1 \\end{pmatrix}.\\]\n\nThey satisfy\n\n\\[\\{\\sigma^i,\\sigma^j\\}=2\\delta^{ij}I_2.\\]\n\nThey therefore also provide gamma matrices for three-dimensional positive-definite Euclidean geometry.\n\nDirac representation\n\nFor signature \\((+,-,-,-)\\), the Dirac representation is\n\n\\[\\gamma^0= \\begin{pmatrix} I_2&0\\\\ 0&-I_2 \\end{pmatrix}, \\qquad \\gamma^i= \\begin{pmatrix} 0&\\sigma^i\\\\ -\\sigma^i&0 \\end{pmatrix}.\\]\n\nThis representation diagonalizes \\(\\gamma^0\\) and is useful in studying the nonrelativistic limit of the Dirac equation.\n\nThe matrices obey\n\n\\[(\\gamma^0)^\\dagger=\\gamma^0, \\qquad (\\gamma^i)^\\dagger=-\\gamma^i,\\]\n\nwhere \\(\\dagger\\) denotes the conjugate transpose. Equivalently,\n\n\\[(\\gamma^\\mu)^\\dagger = \\gamma^0\\gamma^\\mu\\gamma^0.\\]\n\nWeyl representation\n\nThe Weyl, or chiral, representation is\n\n\\[\\gamma^0= \\begin{pmatrix} 0&I_2\\\\ I_2&0 \\end{pmatrix}, \\qquad \\gamma^i= \\begin{pmatrix} 0&\\sigma^i\\\\ -\\sigma^i&0 \\end{pmatrix}.\\]\n\nIt makes the decomposition into two-component chiral spinors explicit.\n\nEquivalent representations\n\nIf \\(S\\) is invertible, then\n\n\\[\\gamma'^\\mu=S\\gamma^\\mu S^{-1}\\]\n\nsatisfies the same Clifford relations. Transforming the spinor simultaneously as \\(\\psi'=S\\psi\\) gives an equivalent description. This is a similarity transformation, not a change in the underlying spacetime geometry.\n\nOver the complex numbers, every irreducible four-dimensional spacetime representation of the full Clifford algebra is equivalent in this sense. An arbitrary similarity transformation need not preserve the displayed Hermiticity conditions relative to the ordinary component inner product; unitary changes of basis do preserve them."
    },
    {
      "heading": "Chirality and the matrix \\gamma^5",
      "text": "In four-dimensional Minkowski spacetime, define\n\n\\[\\gamma^5=i\\gamma^0\\gamma^1\\gamma^2\\gamma^3.\\]\n\nThe superscript \\(5\\) is a conventional label, not an exponent. This matrix satisfies\n\n\\[(\\gamma^5)^2=I,\\qquad \\{\\gamma^5,\\gamma^\\mu\\}=0,\\qquad (\\gamma^5)^\\dagger=\\gamma^5.\\]\n\nIts eigenvalues are \\(+1\\) and \\(-1\\). The operators\n\n\\[P_L=\\frac{I-\\gamma^5}{2}, \\qquad P_R=\\frac{I+\\gamma^5}{2}\\]\n\nare complementary projectors:\n\n\\[P_L^2=P_L,\\qquad P_R^2=P_R,\\qquad P_LP_R=0,\\qquad P_L+P_R=I.\\]\n\nThey define the left- and right-chiral components of a Dirac spinor.\n\nIn the Weyl representation,\n\n\\[\\gamma^5= \\begin{pmatrix} -I_2&0\\\\ 0&I_2 \\end{pmatrix}.\\]\n\nFor a free massless field, the two chiral components satisfy separate equations. A Dirac mass term couples them.\n\nChirality is distinct from helicity, the projection of spin along momentum. The two are closely related for massless states, but they are not interchangeable concepts for massive particles."
    },
    {
      "heading": "Lorentz transformations and spinors",
      "text": "Gamma matrices connect spacetime geometry to the representation theory of the Lorentz group. If coordinates transform as\n\n\\[x'^\\mu=\\Lambda^\\mu{}_\\nu x^\\nu,\\]\n\na corresponding spinor transformation \\(S(\\Lambda)\\) satisfies\n\n\\[S(\\Lambda)^{-1}\\gamma^\\mu S(\\Lambda) = \\Lambda^\\mu{}_\\nu\\gamma^\\nu.\\]\n\nThis relation ensures covariance of the Dirac equation.\n\nThe matrices\n\n\\[\\sigma^{\\mu\\nu} = \\frac{i}{2}[\\gamma^\\mu,\\gamma^\\nu], \\qquad [A,B]=AB-BA,\\]\n\nsupply generators of the Lorentz Lie algebra in the spinor representation, up to the conventional normalization used in exponentiation.\n\nMore precisely, spinors carry representations of the spin group, a double cover of the proper orthochronous Lorentz group. A spatial rotation through \\(2\\pi\\) acts as \\(-I\\) on a spin-\\(\\tfrac12\\) spinor; a rotation through \\(4\\pi\\) acts as \\(I\\).\n\nIn the standard Hermiticity convention, the Dirac adjoint is\n\n\\[\\bar\\psi=\\psi^\\dagger\\gamma^0.\\]\n\nIt allows the construction of quantities with definite Lorentz transformation laws. Examples include the scalar \\(\\bar\\psi\\psi\\), vector \\(\\bar\\psi\\gamma^\\mu\\psi\\), and axial vector \\(\\bar\\psi\\gamma^\\mu\\gamma^5\\psi\\)."
    },
    {
      "heading": "Algebraic identities and completeness",
      "text": "Gamma-matrix calculations can be reduced using the anticommutation relation and the cyclicity of the matrix trace. For four-component gamma matrices in four dimensions,\n\n\\[\\operatorname{tr}(I)=4, \\qquad \\operatorname{tr}(\\gamma^\\mu)=0,\\]\n\n\\[\\operatorname{tr}(\\gamma^\\mu\\gamma^\\nu) = 4\\eta^{\\mu\\nu},\\]\n\nand\n\n\\[\\operatorname{tr} (\\gamma^\\mu\\gamma^\\nu\\gamma^\\rho\\gamma^\\sigma) = 4\\left( \\eta^{\\mu\\nu}\\eta^{\\rho\\sigma} -\\eta^{\\mu\\rho}\\eta^{\\nu\\sigma} +\\eta^{\\mu\\sigma}\\eta^{\\nu\\rho} \\right).\\]\n\nThe trace of any product containing an odd number of ordinary gamma matrices vanishes in four dimensions.\n\nUseful contraction identities in dimension \\(d\\) include\n\n\\[\\gamma^a\\gamma_a=dI, \\qquad \\gamma^a\\gamma^b\\gamma_a=(2-d)\\gamma^b.\\]\n\nHere the indices range over the \\(d\\)-dimensional metric space.\n\nIn four dimensions, the 16 matrices\n\n\\[I,\\qquad \\gamma^\\mu,\\qquad \\sigma^{\\mu\\nu}\\;(\\mu<\\nu),\\qquad \\gamma^\\mu\\gamma^5,\\qquad \\gamma^5\\]\n\nform a basis for the complex vector space of all \\(4\\times4\\) matrices. Their respective numbers are\n\n\\[1+4+6+4+1=16.\\]\n\nThis completeness underlies Fierz identities, which rearrange products of spinor bilinears."
    },
    {
      "heading": "Other dimensions and signatures",
      "text": "Gamma matrices are not restricted to four-dimensional spacetime. For a nondegenerate metric in dimension \\(d\\), the smallest irreducible complex representation has dimension\n\n\\[N=2^{\\lfloor d/2\\rfloor}.\\]\n\nFor even \\(d\\), the complex Clifford algebra is isomorphic to the full matrix algebra \\(M_N(\\mathbb C)\\), and its irreducible representation is unique up to equivalence. For odd \\(d\\), it is a direct sum of two full matrix algebras and has two inequivalent irreducible representations. A single irreducible representation in odd dimension is therefore not faithful to the entire Clifford algebra.\n\nThese statements concern the full complex Clifford algebra. Restricting to its even subalgebra can change the decomposition: in even dimensions, this restriction is what permits chiral spinor representations.\n\nOver the real numbers, the classification depends on the metric signature. Real Clifford algebras exhibit an eightfold periodic pattern involving real, complex, and quaternionic matrix algebras. This structure governs the availability of reality conditions on spinors, including Majorana conditions.\n\nFor Euclidean signature, one commonly uses Hermitian matrices \\(\\Gamma^a\\) satisfying\n\n\\[\\{\\Gamma^a,\\Gamma^b\\}=2\\delta^{ab}I.\\]\n\nStarting from the Minkowski convention above, a four-dimensional Euclidean set is\n\n\\[\\Gamma^4=\\gamma^0,\\qquad \\Gamma^k=i\\gamma^k\\quad(k=1,2,3).\\]\n\nAll four then square to \\(I\\)."
    },
    {
      "heading": "Curved spacetime",
      "text": "On a curved manifold, constant flat-space gamma matrices are combined with a local orthonormal frame. If \\(e^\\mu{}_a(x)\\) is an inverse vielbein, define\n\n\\[\\gamma^\\mu(x)=e^\\mu{}_a(x)\\gamma^a.\\]\n\nThey satisfy\n\n\\[\\{\\gamma^\\mu(x),\\gamma^\\nu(x)\\} = 2g^{\\mu\\nu}(x)I.\\]\n\nThe spinor derivative must include a spin connection. One common convention is\n\n\\[\\nabla_\\mu\\psi = \\left( \\partial_\\mu +\\frac14\\omega_{\\mu ab}\\gamma^a\\gamma^b \\right)\\psi,\\]\n\nwhere \\(\\omega_{\\mu ab}\\) is antisymmetric in the frame indices. The resulting Dirac operator couples Clifford multiplication to the covariant derivative.\n\nA local matrix construction does not by itself establish the existence of globally defined spinors. The manifold must support an appropriate global structure, such as a spin structure or, in suitable settings, a spin\\(^{c}\\) structure."
    },
    {
      "heading": "Applications and convention-dependent issues",
      "text": "In quantum field theory, gamma matrices occur in fermion propagators, interaction vertices, spin sums, and scattering calculations. In differential geometry, Dirac operators connect Clifford algebras with geometric analysis and index theory. In condensed-matter physics, effective Dirac Hamiltonians use related matrix algebras to describe low-energy excitations.\n\nSeveral distinctions are important in calculations:\n\n- Metric and index conventions: changing the signature changes matrix squares, index-lowering signs, and related formulas.\n- Representation versus geometry: a change of spinor basis changes matrix entries without changing the Clifford relations.\n- Fixed versus continued dimension: four-dimensional identities cannot automatically be used unchanged in a calculation that analytically continues the spacetime dimension.\n- Treatment of \\(\\gamma^5\\): in dimensional regularization, the four-dimensional definition of \\(\\gamma^5\\) cannot be extended while retaining every familiar trace and anticommutation property without qualification. Different prescriptions specify which identities are retained and how consistency is restored.\n- Gamma matrices versus vector components: the spacetime index labels Clifford multiplication directions, while the suppressed matrix indices act on spinors. Gamma matrices are not simply a collection of ordinary vector-valued scalar functions."
    }
  ],
  "links": [
    "matrix",
    "quadratic-form",
    "clifford-algebra",
    "spinor",
    "dirac-equation",
    "electron",
    "metric-tensor",
    "identity-matrix",
    "minkowski-space",
    "einstein-summation-convention",
    "klein-gordon-equation",
    "conjugate-transpose",
    "matrix-similarity",
    "complex-number",
    "representation-theory",
    "lorentz-group",
    "lie-algebra",
    "matrix-trace",
    "basis-linear-algebra",
    "vector-space",
    "real-number",
    "quaternion",
    "manifold",
    "covariant-derivative",
    "quantum-field-theory"
  ],
  "sources": [],
  "license": "https://creativecommons.org/licenses/by/4.0/"
}