betaFeedback

WorksMathematics

Classification of real Lie algebras of dimension six

Abstract

We summarize a coordinate-free classification of six-dimensional real Lie algebras up to isomorphism. Instead of solving the overdetermined Jacobi system of 120 quadratic equations on 90 structure constants modulo GL(6,R)\mathrm{GL}(6,\mathbb{R}), the method uses the Levi–Malcev decomposition, Mostow's nilradical constraints, and the Skjelbred–Sund cohomological central extension technique. In the semisimple sector, the four real forms of A1⊕A1A_1\oplus A_1 are classified by Killing signature invariants, with explicit matrix representatives for sl(2,C)R≅so(3,1)\mathfrak{sl}(2,\mathbb{C})_{\mathbb{R}}\cong\mathfrak{so}(3,1). For non-solvable algebras with three-dimensional radical, representation-theoretic rigidity shows that su(2)\mathfrak{su}(2) cannot couple indecomposably with non-abelian radicals, yielding exactly three indecomposable classes governed by so(3)\mathfrak{so}(3) and sl(2,R)\mathfrak{sl}(2,\mathbb{R}). In the solvable category, Mostow's theorem restricts the nilradical dimension to {3,4,5,6}\{3,4,5,6\}. The nilpotent sector in dimension six has twenty-four indecomposable isomorphism classes plus decomposable counterparts. Solvable non-nilpotent algebras are classified by nilradical codimension, with explicit real Jordan matrix representations for codimension one and two that resolve continuous moduli under gauge actions. The work also computes the minimal faithful representation dimension μ(g)μ(\mathfrak{g}) under Ado's theorem, relates the classification to invariant complex structures on nilmanifolds, and provides a deterministic, page-breakable identification algorithm.

Introduction and Historical Context

The classification of finite-dimensional Lie algebras over the real field R\mathbb{R} is a foundational problem that underpins modern differential geometry, theoretical physics, Hamiltonian mechanics, and invariant theory. Lie algebras provide the infinitesimal framework for smooth transformation groups, determine the geometry of homogeneous spaces, and classify left-invariant structures on manifolds.

In dimension three, the landmark work of Bianchi (1898) classified the nine isometry classes of three-dimensional spaces, which was subsequently reformulated by Milnor [5] to describe left-invariant Riemannian metrics and Ricci curvature signatures. In dimension four, the classification of solvable real Lie algebras was achieved by Mubarakzyanov [7], revealing a finite collection of discrete algebras and single-parameter families that govern four-dimensional homogeneous cosmologies [9]. At dimension five, Mubarakzyanov [8] established the solvable catalog, where the first instances of multi-parameter continuous moduli emerge.

Dimension six represents the true mathematical watershed in Lie theory. Six-dimensional Lie algebras are indispensable across modern geometry and high-energy theoretical physics:

  1. Complex and Symplectic Geometry of Nilmanifolds: A compact nilmanifold M=Γ\GM = \Gamma\backslash G of real dimension six corresponds to a complex manifold of complex dimension three. The existence of left-invariant complex, symplectic, and Kähler structures is governed entirely by the Lie algebra cohomology of the six-dimensional Lie algebra g=Lie⁡(G)\mathfrak{g} = \operatorname{Lie}(G) [10].

  1. Supergravity and String Flux Compactifications: In Type II superstring theory, compactifications of ten-dimensional spacetime on six-dimensional internal manifolds yield effective four-dimensional N=1\mathcal{N}=1 and N=2\mathcal{N}=2 supergravities. When the internal geometry is a twisted torus (a nilmanifold or solvmanifold), the structure constants of the six-dimensional

Lie algebra define the non-geometric and geometric fluxes (H,F,Q,RH,F,Q,R-fluxes) that generate cosmological vacuum potentials.

3. Sub-Riemannian Geometry and Control Theory: Six is the minimal dimension supporting non-trivial bracket-generating distributions with growth vectors such as (3,6)(3,6) and (2,4,6)(2,4,6), modeled on free nilpotent algebras and Carnot groups.

Despite their importance, setting up the classification via direct coordinate methods leads to an intractable algebraic bottleneck. A six-dimensional real Lie algebra g\mathfrak{g} requires specifying (62)×6=90\binom{6}{2}\times6=90 structure constants CijkC_{ij}^{k}, constrained by (63)×6=120\binom{6}{3}\times6=120 quadratic Jacobi equations:

∑m=16(CijmCmkl+CjkmCmil+CkimCmjl)=0,1≤i<j<k≤6,1≤l≤6,(1)\sum_{m=1}^{6}\left(C_{ij}^{m}C_{mk}^{l}+C_{jk}^{m}C_{mi}^{l}+C_{ki}^{m}C_{mj}^{l}\right)=0,\qquad1\leq i<j<k\leq6,\qquad1\leq l\leq6, \tag*{(1)}

modulo the 36-dimensional non-compact change-of-basis gauge group GL(6,R)\mathrm{GL}(6,\mathbb{R}).

The historical resolution of this problem proceeded in stages: Morozov [6] cataloged the six-dimensional nilpotent algebras over C\mathbb{C}, which was extended, corrected, and transferred to R\mathbb{R} using cohomology by Skjelbred and Sund [11]. Non-solvable algebras admitting a non-trivial Levi decomposition were classified definitively by Turkowski [13].

The objective of this paper is to present a self-contained, conceptual, and coordinate-free classification of all six-dimensional real Lie algebras. Rather than relying on brute-force polynomial elimination, we deploy an invariant-driven architecture:

  • We classify the semisimple sector (dim⁡r=0\dim\mathfrak{r}=0) using Cartan’s real forms and Killing form signatures, with explicit matrix representatives for the Lorentz algebra sl(2,C)R\mathfrak{sl}(2,\mathbb{C})_{\mathbb{R}}.

  • We resolve the non-solvable sector (dim⁡r=3\dim\mathfrak{r}=3) using the representation theory of su(2)\mathfrak{su}(2) and sl(2,R)\mathfrak{sl}(2,\mathbb{R}), establishing the representation-theoretic obstruction that excludes su(2)⋉h3\mathfrak{su}(2)\ltimes\mathfrak{h}_{3}.

  • For solvable algebras, we anchor the classification to the nilradical n⊂g\mathfrak{n}\subset\mathfrak{g}. By Mostow’s theorem, dim⁡n∈{3,4,5,6}\dim\mathfrak{n}\in\{3,4,5,6\}.

  • For dim⁡n=6\dim\mathfrak{n}=6 (nilpotent sector), we catalog the twenty-four indecomposable classes using the Skjelbred–Sund cohomological central extension method.

  • For dim⁡n∈{3,4,5}\dim\mathfrak{n}\in\{3,4,5\}, we systematically construct exact real matrix representatives for all classes, paying particular attention to the complex Jordan sectors (algebras with complex conjugate spectra).

  • We determine the minimal faithful representation dimension μ(g)\mu(\mathfrak{g}) under Ado’s theorem and construct a deterministic, page-breakable identification algorithm.

Geometric Obstructions and the Watershed of Dimension Six

To appreciate why dimension six requires a systematic cohomological approach, we analyze the algebraic geometry of the Lie bracket across low dimensions.

Comparison with Dimensions Three, Four, and Five

Let g\mathfrak{g} be an nn-dimensional real Lie algebra. The bracket is a linear map [⋅,⋅]:Λ2g→g[\cdot,\cdot]:\Lambda^{2}\mathfrak{g}\to\mathfrak{g}.

  • Dimension Three (n=3n=3): dim⁡Λ2R3=3\dim\Lambda^{2}\mathbb{R}^{3}=3. The exterior duality Λ2g≅g∗\Lambda^{2}\mathfrak{g}\cong\mathfrak{g}^{*} reduces the Lie bracket to a single 3×33\times3 real matrix C=S+a^C=S+\hat{a}. The Jacobi identity simplifies to the single vector equation S⋅a=0S\cdot a=0 ([5]).

  • Dimension Four (n=4n = 4): dim⁡Λ2R4=6\dim\Lambda^2\mathbb{R}^4 = 6, while dim⁡g=4\dim\mathfrak{g} = 4. Exterior duality fails, but the classification remains algebraically constrained: every non-solvable algebra is a direct sum (s⊕R\mathfrak{s} \oplus\mathbb{R}), and the nilradical of a solvable algebra cannot have codimension greater than two.

  • Dimension Five (n=5n = 5): dim⁡Λ2R5=10\dim\Lambda^2\mathbb{R}^5 = 10, with 50 structure constants and 50 quadratic equations. The first non-trivial Levi coupling appears (sl(2,R)⋉R2\mathfrak{sl}(2,\mathbb{R}) \ltimes\mathbb{R}^2), and continuous parameters enter the solvable sector. However, all nilpotent algebras remain strictly discrete (six indecomposable classes).

  • Dimension Six (n=6n = 6): Here:

dim⁡Λ2R6=(62)=15,dim⁡g=6.(2)\dim\Lambda^2\mathbb{R}^6 = \binom{6}{2} = 15,\qquad\dim\mathfrak{g} = 6. \tag*{(2)}

The bracket map R15→R6\mathbb{R}^{15} \to\mathbb{R}^6 requires 90 structure constants constrained by 120 quadratic Jacobi equations.

Structural Novelties in Dimension Six

Beyond the dimension count, dimension six introduces four structural phenomena that do not exist in dimensions ≤5\le5:

  1. Cohomological Explosion of Nilpotent Classes: In dimension five, there are only six indecomposable nilpotent algebras. In dimension six, this number quadruples to twenty-four indecomposable classes, including non-trivial real forms (N6,16±N_{6,16}^{\pm}) distinguished by quadratic Casimir invariants.

  1. Free Nilpotent Structures: Dimension six is the lowest dimension supporting the free two-step nilpotent Lie algebra on three generators:

n3,2≅V⊕Λ2V,dim⁡V=3,dim⁡Λ2V=3.(3)\mathfrak{n}_{3,2} \cong V \oplus\Lambda^2 V,\qquad\dim V = 3,\quad\dim\Lambda^2 V = 3. \tag*{(3)}

Its center coincides exactly with its derived ideal (Z(g)=[g,g]≅R3Z(\mathfrak{g}) = [\mathfrak{g},\mathfrak{g}] \cong\mathbb{R}^3).

  1. Representation-Theoretic Levi Obstructions: For algebras with dim⁡s=3\dim\mathfrak{s} = 3 and dim⁡r=3\dim\mathfrak{r} = 3, the compactness of su(2)\mathfrak{su}(2) completely obstructs non-trivial couplings with non-abelian radicals, while the split nature of sl(2,R)\mathfrak{sl}(2,\mathbb{R}) allows both vector and fundamental representations.

  1. Proliferation of Complex Jordan Sectors: In the solvable category, the presence of complex conjugate eigenvalues in outer derivations generates distinct continuous moduli varieties whose real forms require explicit rotational-dilational matrix descriptions.

Theoretical Machinery: Invariants and Decomposition Theorems

All Lie algebras in this monograph are finite-dimensional over R\mathbb{R}. We deploy the following structural tools [3, 2, 4]:

(T1) Levi–Malcev Decomposition: Every finite-dimensional real Lie algebra g\mathfrak{g} decomposes uniquely as a semidirect product:

g=s⋉ρr,(4)\mathfrak{g} = \mathfrak{s} \ltimes_{\rho} \mathfrak{r}, \tag*{(4)}

where r=Rad⁡(g)\mathfrak{r} = \operatorname{Rad}(\mathfrak{g}) is the solvable radical, and s\mathfrak{s} is a semisimple subalgebra (the Levi factor). The factor s\mathfrak{s} is unique up to inner automorphisms exp⁡(ad⁡u)\exp(\operatorname{ad} u) with u∈[g,r]u \in[\mathfrak{g},\mathfrak{r}]. The representation ρ:s→Der⁡(r)\rho:\mathfrak{s} \to\operatorname{Der}(\mathfrak{r}) specifies the cross brackets between s\mathfrak{s} and r\mathfrak{r}.

(T2) Mostow’s Bound on the Nilradical: The nilradical n=Nil⁡(g)\mathfrak{n} = \operatorname{Nil}(\mathfrak{g}) is the maximal nilpotent ideal of g\mathfrak{g}. For any finite-dimensional solvable real Lie algebra g\mathfrak{g}, Mostow’s theorem bounds the dimension of n\mathfrak{n} from below:

dim⁡n≥12dim⁡g.(5)\dim\mathfrak{n} \ge\frac{1}{2}\dim\mathfrak{g}. \tag*{(5)}

For dim⁡g=6\dim\mathfrak{g} = 6, this immediately establishes that dim⁡n∈{3,4,5,6}\dim\mathfrak{n} \in\{3,4,5,6\}.

(T3) The Skjelbred–Sund Cohomological Method: Let n0\mathfrak{n}_0 be a nilpotent Lie algebra of dimension n−kn-k. The central extensions of n0\mathfrak{n}_0 by Rk\mathbb{R}^{k} are parameterized by the second scalar cohomology group H2(n0,R)H^2(\mathfrak{n}_0,\mathbb{R}). The group Aut⁡(n0)\operatorname{Aut}(\mathfrak{n}_0) acts naturally on the Grassmannian Gr⁡(k,H2(n0,R))\operatorname{Gr}(k,H^2(\mathfrak{n}_0,\mathbb{R})). The orbits of this action correspond bijectively to the isomorphism classes of central extensions with center Rk\mathbb{R}^{k} [11].

(T4) Mubarakzyanov Derivation Gauge Action: For a solvable algebra g=n⋊DR\mathfrak{g} = \mathfrak{n} \rtimes_D \mathbb{R}, two derivations D,D′∈Der⁡(n)D,D' \in\operatorname{Der}(\mathfrak{n}) generate isomorphic Lie algebras if and only if there exist c∈R×c \in\mathbb{R}^{\times}, ϕ∈Aut⁡(n)\phi\in\operatorname{Aut}(\mathfrak{n}), and u∈nu \in\mathfrak{n} such that:

D′=cϕDϕ−1+ad⁡(u).(6)D' = c\phi D\phi^{-1} + \operatorname{ad}(u). \tag*{(6)}

(T5) Ado’s Theorem and Minimal Representation Dimension: Every finite-dimensional Lie algebra g\mathfrak{g} admits a faithful linear representation in gl(N,R)\mathfrak{gl}(N,\mathbb{R}). The invariant μ(g)\mu(\mathfrak{g}), defined by

μ(g):=min⁡{N∈N:∃ an injective Lie homomorphism ρ:g↪gl(N,R)},(7)\mu(\mathfrak{g}) := \min\{N \in\mathbb{N} : \exists\ \text{an injective Lie homomorphism } \rho:\mathfrak{g} \hookrightarrow\mathfrak{gl}(N,\mathbb{R})\}, \tag*{(7)}

is an intrinsic isomorphism invariant of g\mathfrak{g}.

The Semisimple Sector (r= 0)

When dim⁡r=0\dim\mathfrak{r} = 0, the Lie algebra g\mathfrak{g} is semisimple of dimension six.

Theorem 4.1 (Classification of 6D Real Semisimple Lie Algebras). Up to isomorphism, there exist precisely four mutually non-isomorphic six-dimensional real semisimple Lie algebras. Exactly one is simple over R\mathbb{R}, while three are direct sums of two three-dimensional simple ideals:

  1. g6,1ss=so(4)≅su(2)⊕su(2)\mathfrak{g}^{\mathrm{ss}}_{6,1} = \mathfrak{so}(4) \cong\mathfrak{su}(2) \oplus\mathfrak{su}(2) (Compact form, Killing signature (0,6)(0,6)),

  2. g6,2ss=su(2)⊕sl(2,R)\mathfrak{g}^{\mathrm{ss}}_{6,2} = \mathfrak{su}(2) \oplus\mathfrak{sl}(2,\mathbb{R}) (Mixed non-compact form, Killing signature (2,4)(2,4)),

  3. g6,3ss=so(2,2)≅sl(2,R)⊕sl(2,R)\mathfrak{g}^{\mathrm{ss}}_{6,3} = \mathfrak{so}(2,2) \cong\mathfrak{sl}(2,\mathbb{R}) \oplus\mathfrak{sl}(2,\mathbb{R}) (Split form, Killing signature (4,2)(4,2)),

  4. g6,4ss=sl(2,C)R≅so(3,1)\mathfrak{g}^{\mathrm{ss}}_{6,4} = \mathfrak{sl}(2,\mathbb{C})_{\mathbb{R}} \cong\mathfrak{so}(3,1) (Lorentz algebra; simple over R\mathbb{R}, Killing signature (3,3)(3,3)).

Proof. Let g\mathfrak{g} be a six-dimensional real semisimple Lie algebra. Over C\mathbb{C}, its complexification decomposes into simple ideals: gC=g⊗RC≅sl(2,C)⊕sl(2,C)\mathfrak{g}_{\mathbb{C}} = \mathfrak{g} \otimes_{\mathbb{R}} \mathbb{C} \cong\mathfrak{sl}(2,\mathbb{C}) \oplus\mathfrak{sl}(2,\mathbb{C}). Real forms correspond to conjugacy classes of antilinear involutions σ:gC→gC\sigma: \mathfrak{g}_{\mathbb{C}} \to\mathfrak{g}_{\mathbb{C}}.

Case 1: Decomposable over R\mathbb{R}. The involution σ\sigma preserves each sl(2,C)\mathfrak{sl}(2,\mathbb{C}) factor. The algebra splits as g=s1⊕s2\mathfrak{g} = \mathfrak{s}_1 \oplus\mathfrak{s}_2, where each si\mathfrak{s}_i is a three-dimensional real simple Lie algebra. Over R\mathbb{R}, the only three-dimensional simple Lie algebras are the compact form su(2)≅so(3)\mathfrak{su}(2) \cong\mathfrak{so}(3) and the split form sl(2,R)≅so(2,1)\mathfrak{sl}(2,\mathbb{R}) \cong\mathfrak{so}(2,1). Their binary combinations generate:

  • su(2)⊕su(2)\mathfrak{su}(2) \oplus\mathfrak{su}(2): Killing form is strictly negative-definite with signature (0,6)(0,6).

  • su(2)⊕sl(2,R)\mathfrak{su}(2) \oplus\mathfrak{sl}(2,\mathbb{R}): Killing signature is (2,4)(2,4), with real rank 1.

  • sl(2,R)⊕sl(2,R)\mathfrak{sl}(2,\mathbb{R}) \oplus\mathfrak{sl}(2,\mathbb{R}): Maximal split algebra, Killing signature (4,2)(4,2), real rank 2.

Case 2: Simple over R\mathbb{R} (Realification). The involution σ\sigma interchanges the two sl(2,C)\mathfrak{sl}(2,\mathbb{C}) factors. Then g\mathfrak{g} is the realification of a complex simple Lie algebra of complex dimension 3. The unique choice is sl(2,C)\mathfrak{sl}(2,\mathbb{C}), regarded as a 6-dimensional real algebra (sl(2,C))R≅so(3,1)(\mathfrak{sl}(2,\mathbb{C}))_{\mathbb{R}} \cong\mathfrak{so}(3,1), the Lorentz algebra. Its Killing form has signature (3,3)(3,3).

Because the Sylvester signatures (0,6)(0,6), (2,4)(2,4), (4,2)(4,2), and (3,3)(3,3) are distinct, these four algebras are pairwise non-isomorphic.

Explicit Matrix Realization of the Real Simple Algebra sl(2,C)R\mathfrak{sl}(2,\mathbb{C})_{\mathbb{R}}

The algebra g6,4ss=sl(2,C)R\mathfrak{g}^{\mathrm{ss}}_{6,4}=\mathfrak{sl}(2,\mathbb{C})_{\mathbb{R}} consists of 2×22\times2 trace-free complex matrices regarded as a 6-dimensional Lie algebra over R\mathbb{R}. To obtain a completely explicit, coordinate-free real matrix representation, we use the canonical embedding C↪R2×2\mathbb{C}\hookrightarrow\mathbb{R}^{2\times2} given by z=x+iy↦(x−yyx)z=x+iy\mapsto\begin{pmatrix}x&-y\\y&x\end{pmatrix}.

Applying this blockwise to sl(2,C)\mathfrak{sl}(2,\mathbb{C}) embeds sl(2,C)R\mathfrak{sl}(2,\mathbb{C})_{\mathbb{R}} faithfully into gl(4,R)\mathfrak{gl}(4,\mathbb{R}):

M=X+iY∈sl(2,C)→ ρ (X−YYX)∈gl(4,R),tr⁡(X)=tr⁡(Y)=0.(8)M=X+iY\in\mathfrak{sl}(2,\mathbb{C})\xrightarrow{\ \rho\ }\begin{pmatrix}X&-Y\\Y&X\end{pmatrix}\in\mathfrak{gl}(4,\mathbb{R}),\qquad\operatorname{tr}(X)=\operatorname{tr}(Y)=0. \tag*{(8)}

A canonical real basis {H1,E1,F1,H2,E2,F2}\{H_1,E_1,F_1,H_2,E_2,F_2\} satisfying the real Lorentz commutation relations is given by:

H1=(10000−1000010000−1),E1=(0100000000010000),F1=(0000100000000010),H_1=\begin{pmatrix}1&0&0&0\\0&-1&0&0\\0&0&1&0\\0&0&0&-1\end{pmatrix},\qquad E_1=\begin{pmatrix}0&1&0&0\\0&0&0&0\\0&0&0&1\\0&0&0&0\end{pmatrix},\qquad F_1=\begin{pmatrix}0&0&0&0\\1&0&0&0\\0&0&0&0\\0&0&1&0\end{pmatrix},
H2=(00−10000110000−100),E2=(000−1000001000000),F2=(000000−1000001000).(9)H_2=\begin{pmatrix}0&0&-1&0\\0&0&0&1\\1&0&0&0\\0&-1&0&0\end{pmatrix},\qquad E_2=\begin{pmatrix}0&0&0&-1\\0&0&0&0\\0&1&0&0\\0&0&0&0\end{pmatrix},\qquad F_2=\begin{pmatrix}0&0&0&0\\0&0&-1&0\\0&0&0&0\\1&0&0&0\end{pmatrix}. \tag*{(9)}

This provides a manifest 4×44\times4 real faithful representation, proving directly that μ(sl(2,C)R)=4\mu(\mathfrak{sl}(2,\mathbb{C})_{\mathbb{R}})=4.

The Non-Solvable Sector with Levi Decomposition (dim⁡r=3\dim\mathfrak{r}=3)

When dim⁡s=3\dim\mathfrak{s}=3 and dim⁡r=3\dim\mathfrak{r}=3, the Levi factor is su(2)\mathfrak{su}(2) or sl(2,R)\mathfrak{sl}(2,\mathbb{R}), and r\mathfrak{r} is a three-dimensional solvable Lie algebra.

Proposition 5.1 (Representation-Theoretic Obstruction). There exists no indecomposable real Lie algebra of the form g=su(2)⋉ρr\mathfrak{g}=\mathfrak{su}(2)\ltimes_{\rho}\mathfrak{r} when r\mathfrak{r} is non-abelian.

Proof. Suppose r\mathfrak{r} is non-abelian. Then the derived ideal r′=[r,r]\mathfrak{r}'=[\mathfrak{r},\mathfrak{r}] satisfies 1≤dim⁡r′≤21\leq\dim\mathfrak{r}'\leq2. Because r′\mathfrak{r}' is a characteristic ideal, it is invariant under all derivations, so ρ(s)(r′)⊆r′\rho(\mathfrak{s})(\mathfrak{r}')\subseteq\mathfrak{r}'.

  • If dim⁡r′=1\dim\mathfrak{r}'=1, the center or derived line forms a 1-dimensional representation χ:su(2)→R\chi:\mathfrak{su}(2)\to\mathbb{R}. Because [su(2),su(2)]=su(2)[\mathfrak{su}(2),\mathfrak{su}(2)]=\mathfrak{su}(2), every 1-dimensional character vanishes: χ≡0\chi\equiv0. The induced quotient representation on r/r′≅R2\mathfrak{r}/\mathfrak{r}'\cong\mathbb{R}^{2} defines a homomorphism ρˉ:su(2)→gl(2,R)\bar{\rho}:\mathfrak{su}(2)\to\mathfrak{gl}(2,\mathbb{R}). But su(2)\mathfrak{su}(2) has no non-trivial real 2-dimensional representations (any compact subalgebra of gl(2,R)\mathfrak{gl}(2,\mathbb{R}) is isomorphic to so(2)≅R\mathfrak{so}(2)\cong\mathbb{R}). Hence, ρˉ≡0\bar{\rho}\equiv0. By Weyl’s complete reducibility theorem, ρ≡0\rho\equiv0, forcing g\mathfrak{g} to decompose.

  • If dim⁡r′=2\dim\mathfrak{r}'=2, r′\mathfrak{r}' is abelian. The action on r′\mathfrak{r}' is a 2-dimensional real representation of su(2)\mathfrak{su}(2), which is identically zero. The quotient action on r/r′≅R\mathfrak{r}/\mathfrak{r}'\cong\mathbb{R} is also trivial. Thus ρ=0\rho=0.

Therefore, su(2)\mathfrak{su}(2) can couple indecomposably only with the abelian radical R3\mathbb{R}^{3} via its 3-dimensional vector representation 3.

Theorem 5.2 (Classification of Levi-Decomposable 6D Algebras). Every six-dimensional indecomposable real Lie algebra admitting a non-trivial Levi decomposition is isomorphic to exactly one of the following three Lie algebras (Turkowski, [13]):

  1. g6,1Levi=so(3)⋉3R3\mathfrak{g}^{\mathrm{Levi}}_{6,1}=\mathfrak{so}(3)\ltimes_{3}\mathbb{R}^{3} (Euclidean motion algebra se(3)\mathfrak{se}(3)):

[e1,e2]=e3,[e2,e3]=e1,[e3,e1]=e2,[e1,e5]=e6,[e1,e6]=−e5,[e2,e4]=−e6,[e2,e6]=e4,[e3,e4]=e5,[e3,e5]=−e4.(10)\begin{aligned} [e_{1},e_{2}]&=e_{3}, &[e_{2},e_{3}]&=e_{1}, &[e_{3},e_{1}]&=e_{2},\\ [e_{1},e_{5}]&=e_{6}, &[e_{1},e_{6}]&=-e_{5}, &[e_{2},e_{4}]&=-e_{6},\\ [e_{2},e_{6}]&=e_{4}, &[e_{3},e_{4}]&=e_{5}, &[e_{3},e_{5}]&=-e_{4}. \tag*{(10)} \end{aligned}
  1. g6,2Levi=sl(2,R)⋉adR3\mathfrak{g}^{\mathrm{Levi}}_{6,2}=\mathfrak{sl}(2,\mathbb{R})\ltimes_{\mathrm{ad}}\mathbb{R}^{3} (Adjoint action on abelian radical):

[e1,e2]=2e2,[e1,e3]=−2e3,[e2,e3]=e1,[e1,e4]=2e4,[e1,e6]=−2e6,[e2,e5]=−2e4,[e2,e6]=e5,[e3,e4]=e5,[e3,e5]=−2e6.(11)\begin{aligned} [e_{1},e_{2}]&=2e_{2}, &[e_{1},e_{3}]&=-2e_{3}, &[e_{2},e_{3}]&=e_{1},\\ [e_{1},e_{4}]&=2e_{4}, &[e_{1},e_{6}]&=-2e_{6}, &[e_{2},e_{5}]&=-2e_{4},\\ [e_{2},e_{6}]&=e_{5}, &[e_{3},e_{4}]&=e_{5}, &[e_{3},e_{5}]&=-2e_{6}. \tag*{(11)} \end{aligned}
  1. g6,3Levi=sl(2,R)⋉fundh3\mathfrak{g}^{\mathrm{Levi}}_{6,3}=\mathfrak{sl}(2,\mathbb{R})\ltimes_{\mathrm{fund}}\mathfrak{h}_{3} (Central extension of affine symplectic algebra):

[e1,e2]=2e2,[e1,e3]=−2e3,[e2,e3]=e1,[e1,e4]=e4,[e1,e5]=−e5,[e2,e5]=e4,[e3,e4]=e5,[e4,e5]=e6.(12)\begin{aligned} [e_{1},e_{2}]&=2e_{2}, &[e_{1},e_{3}]&=-2e_{3}, &[e_{2},e_{3}]&=e_{1},\\ [e_{1},e_{4}]&=e_{4}, &[e_{1},e_{5}]&=-e_{5}, &[e_{2},e_{5}]&=e_{4},\\ [e_{3},e_{4}]&=e_{5}, &[e_{4},e_{5}]&=e_{6}. \tag*{(12)} \end{aligned}

Proof. For s=so(3)\mathfrak{s}=\mathfrak{so}(3), Proposition 5.1 proves that g6,1Levi\mathfrak{g}^{\mathrm{Levi}}_{6,1} is the unique indecomposable coupling. For s=sl(2,R)\mathfrak{s}=\mathfrak{sl}(2,\mathbb{R}), real representations of dimension three are either the adjoint representation 3\mathbf{3} or 2⊕1\mathbf{2}\oplus\mathbf{1}. The adjoint representation preserves an indefinite metric on R3\mathbb{R}^{3}, producing g6,2Levi\mathfrak{g}^{\mathrm{Levi}}_{6,2}. The fundamental representation 2\mathbf{2} acts on the two-dimensional quotient V=span⁡{e4,e5}≅R2V=\operatorname{span}\{e_{4},e_{5}\}\cong\mathbb{R}^{2} of the Heisenberg algebra h3\mathfrak{h}_{3}, preserving the symplectic area form [e4,e5]=e6[e_{4},e_{5}]=e_{6} on the center. This produces g6,3Levi\mathfrak{g}^{\mathrm{Levi}}_{6,3}.

The three algebras are distinguished by their Levi factor (so(3)\mathfrak{so}(3) for g6,1Levi\mathfrak{g}^{\mathrm{Levi}}_{6,1} vs. sl(2,R)\mathfrak{sl}(2,\mathbb{R}) for g6,2Levi,g6,3Levi\mathfrak{g}^{\mathrm{Levi}}_{6,2},\mathfrak{g}^{\mathrm{Levi}}_{6,3}) and radical nilpotency step (step⁡(r)=1\operatorname{step}(\mathfrak{r})=1 for g6,2Levi\mathfrak{g}^{\mathrm{Levi}}_{6,2} vs. step⁡(r)=2\operatorname{step}(\mathfrak{r})=2 for g6,3Levi\mathfrak{g}^{\mathrm{Levi}}_{6,3}).

The Nilpotent Sector (n= 6)

When dim⁡n=6\dim\mathfrak{n}=6, the algebra is entirely nilpotent (g=n\mathfrak{g}=\mathfrak{n}).

Theorem 6.1 (The 24 Indecomposable 6D Nilpotent Real Lie Algebras). Every indecomposable six-dimensional real nilpotent Lie algebra is isomorphic to exactly one of the twenty-four Lie algebras cataloged below:

  1. Filiform Algebras (Nilpotency Step 5):

  • N6,1\mathfrak{N}_{6,1}: [e1,e2]=e3[e_{1},e_{2}]=e_{3}, [e1,e3]=e4[e_{1},e_{3}]=e_{4}, [e1,e4]=e5[e_{1},e_{4}]=e_{5}, [e1,e5]=e6[e_{1},e_{5}]=e_{6}.

  • N6,2\mathfrak{N}_{6,2}: [e1,e2]=e3[e_{1},e_{2}]=e_{3}, [e1,e3]=e4[e_{1},e_{3}]=e_{4}, [e1,e4]=e5[e_{1},e_{4}]=e_{5}, [e1,e5]=e6[e_{1},e_{5}]=e_{6}, [e2,e3]=e6[e_{2},e_{3}]=e_{6}.

  • N6,3\mathfrak{N}_{6,3}: [e1,e2]=e3[e_{1},e_{2}]=e_{3}, [e1,e3]=e4[e_{1},e_{3}]=e_{4}, [e1,e4]=e5[e_{1},e_{4}]=e_{5}, [e2,e5]=e6[e_{2},e_{5}]=e_{6}, [e3,e4]=−e6[e_{3},e_{4}]=-e_{6}.

  1. Nilpotency Step 4:

  • N6,4\mathfrak{N}_{6,4}: [e1,e2]=e3[e_{1},e_{2}]=e_{3}, [e1,e3]=e4[e_{1},e_{3}]=e_{4}, [e1,e4]=e5[e_{1},e_{4}]=e_{5}, [e2,e3]=e5[e_{2},e_{3}]=e_{5}, [e2,e4]=e6[e_{2},e_{4}]=e_{6}.

  • N6,5\mathfrak{N}_{6,5}: [e1,e2]=e3[e_{1},e_{2}]=e_{3}, [e1,e3]=e4[e_{1},e_{3}]=e_{4}, [e1,e4]=e5[e_{1},e_{4}]=e_{5}, [e2,e5]=e6[e_{2},e_{5}]=e_{6}.

  • N6,6\mathfrak{N}_{6,6}: [e1,e2]=e3[e_{1},e_{2}]=e_{3}, [e1,e3]=e4[e_{1},e_{3}]=e_{4}, [e1,e5]=e6[e_{1},e_{5}]=e_{6}, [e2,e3]=e5[e_{2},e_{3}]=e_{5}, [e2,e4]=e6[e_{2},e_{4}]=e_{6}.

  • N6,7\mathfrak{N}_{6,7}: [e1,e2]=e3[e_{1},e_{2}]=e_{3}, [e1,e3]=e4[e_{1},e_{3}]=e_{4}, [e1,e5]=e6[e_{1},e_{5}]=e_{6}, [e2,e4]=e6[e_{2},e_{4}]=e_{6}.

  • N6,8N_{6,8}: [e1,e2]=e3[e_1,e_2]=e_3, [e1,e3]=e4[e_1,e_3]=e_4, [e1,e4]=e6[e_1,e_4]=e_6, [e2,e5]=e6[e_2,e_5]=e_6.

  • N6,11N_{6,11}: [e1,e2]=e3[e_1,e_2]=e_3, [e1,e3]=e4[e_1,e_3]=e_4, [e1,e4]=e5[e_1,e_4]=e_5, [e2,e3]=e6[e_2,e_3]=e_6.

  • N6,15N_{6,15}: [e1,e2]=e3[e_1,e_2]=e_3, [e1,e3]=e4[e_1,e_3]=e_4, [e1,e4]=e6[e_1,e_4]=e_6, [e2,e3]=e5[e_2,e_3]=e_5, [e2,e5]=e6[e_2,e_5]=e_6.

  • N6,17N_{6,17}: [e1,e2]=e3[e_1,e_2]=e_3, [e1,e3]=e4[e_1,e_3]=e_4, [e1,e4]=e5[e_1,e_4]=e_5, [e2,e4]=e6[e_2,e_4]=e_6.

  • N6,19N_{6,19}: [e1,e2]=e3[e_1,e_2]=e_3, [e1,e3]=e4[e_1,e_3]=e_4, [e1,e5]=e6[e_1,e_5]=e_6, [e2,e3]=e6[e_2,e_3]=e_6.

Nilpotency Step 3

  • N6,9N_{6,9}: [e1,e2]=e3[e_1,e_2]=e_3, [e1,e3]=e5[e_1,e_3]=e_5, [e2,e4]=e6[e_2,e_4]=e_6, [e3,e4]=e5[e_3,e_4]=e_5.

  • N6,10N_{6,10}: [e1,e2]=e3[e_1,e_2]=e_3, [e1,e3]=e5[e_1,e_3]=e_5, [e2,e3]=e6[e_2,e_3]=e_6, [e2,e4]=e5[e_2,e_4]=e_5.

  • N6,12N_{6,12}: [e1,e2]=e3[e_1,e_2]=e_3, [e1,e3]=e5[e_1,e_3]=e_5, [e1,e4]=e6[e_1,e_4]=e_6, [e2,e3]=e6[e_2,e_3]=e_6.

  • N6,16±N_{6,16}^{\pm}: [e1,e2]=e4[e_1,e_2]=e_4, [e1,e3]=e5[e_1,e_3]=e_5, [e2,e4]=e6[e_2,e_4]=e_6, [e3,e5]=±e6[e_3,e_5]=\pm e_6.

  • N6,18N_{6,18}: [e1,e2]=e3[e_1,e_2]=e_3, [e1,e3]=e5[e_1,e_3]=e_5, [e2,e4]=e6[e_2,e_4]=e_6.

  • N6,20N_{6,20}: [e1,e2]=e4[e_1,e_2]=e_4, [e1,e3]=e5[e_1,e_3]=e_5, [e2,e5]=e6[e_2,e_5]=e_6, [e3,e4]=−e6[e_3,e_4]=-e_6.

  • N6,21N_{6,21}: [e1,e2]=e3[e_1,e_2]=e_3, [e1,e3]=e4[e_1,e_3]=e_4, [e2,e5]=e6[e_2,e_5]=e_6, [e3,e5]=e4[e_3,e_5]=e_4.

  • N6,24N_{6,24}: [e1,e2]=e3[e_1,e_2]=e_3, [e1,e3]=e6[e_1,e_3]=e_6, [e4,e5]=e6[e_4,e_5]=e_6.

Nilpotency Step 2

  • N6,13N_{6,13}: [e1,e2]=e5[e_1,e_2]=e_5, [e1,e3]=e6[e_1,e_3]=e_6, [e2,e4]=e6[e_2,e_4]=e_6, [e3,e4]=e5[e_3,e_4]=e_5.

  • N6,14N_{6,14} (n3,2\mathfrak n_{3,2}, Free 2-step): [e1,e2]=e4[e_1,e_2]=e_4, [e1,e3]=e5[e_1,e_3]=e_5, [e2,e3]=e6[e_2,e_3]=e_6.

  • N6,22N_{6,22}: [e1,e2]=e5[e_1,e_2]=e_5, [e1,e3]=e6[e_1,e_3]=e_6, [e2,e4]=e6[e_2,e_4]=e_6.

  • N6,23N_{6,23}: [e1,e2]=e6[e_1,e_2]=e_6, [e1,e5]=e6[e_1,e_5]=e_6, [e3,e4]=e6[e_3,e_4]=e_6.

Outline of Cohomological Derivation. The Skjelbred–Sund method (T3) constructs every nilpotent algebra n\mathfrak n as a central extension of a lower-dimensional algebra n0=n/Z(n)\mathfrak n_0=\mathfrak n/Z(\mathfrak n) by Rk\mathbb{R}^k, where k=dim⁡Z(n)k=\dim Z(\mathfrak n).

  1. If k=1k=1, n0\mathfrak n_0 is a 5-dimensional nilpotent Lie algebra. By the classification of 5D nilpotent algebras, there are 9 classes for n0\mathfrak n_0. For each n0\mathfrak n_0, one computes H2(n0,R)H^2(\mathfrak n_0,\mathbb{R}) and the orbits of lines under Aut⁡(n0)\operatorname{Aut}(\mathfrak n_0). Cocycles whose kernel contains non-trivial central elements of n0\mathfrak n_0 are discarded (they produce decomposable extensions). This yields classes N6,1N_{6,1} through N6,11N_{6,11}, N6,15N_{6,15}, N6,16±N_{6,16}^{\pm}, N6,17N_{6,17}, N6,19N_{6,19}, N6,20N_{6,20}, N6,21N_{6,21}, N6,23N_{6,23}, and N6,24N_{6,24}.

  1. If k=2k=2, n0\mathfrak n_0 has dimension 4 (R4\mathbb{R}^4, h3⊕R\mathfrak h_3\oplus\mathbb{R}, or n4\mathfrak n_4). One classifies 2-dimensional planes in H2(n0,R)H^2(\mathfrak n_0,\mathbb{R}) under Aut⁡(n0)\operatorname{Aut}(\mathfrak n_0), generating N6,12N_{6,12}, N6,13N_{6,13}, N6,18N_{6,18}, and N6,22N_{6,22}.

  1. If k=3k=3, n0≅R3\mathfrak n_0\cong\mathbb{R}^3. The cohomology space is H2(R3,R)≅Λ2(R3)∗≅R3H^2(\mathbb{R}^3,\mathbb{R})\cong\Lambda^2(\mathbb{R}^3)^*\cong\mathbb{R}^3. The only 3-dimensional subspace in H2(R3,R)H^2(\mathbb{R}^3,\mathbb{R}) is the entire space itself. This generates the unique free nilpotent algebra N6,14=n3,2N_{6,14}=\mathfrak n_{3,2}.

The real forms N6,16+N_{6,16}^{+} and N6,16−N_{6,16}^{-} are distinguished by the signature of the quadratic Casimir form on g/Z(g)\mathfrak g/Z(\mathfrak g), which is (2,0)(2,0) for N6,16+N_{6,16}^{+} and (1,1)(1,1) for N6,16−N_{6,16}^{-}.

Solvable Non-Nilpotent Algebras: Nilradical Stratification

By Mostow’s bound (T2), any solvable six-dimensional Lie algebra satisfies dim⁡n∈{3,4,5}\dim\mathfrak n\in\{3,4,5\}. We classify each stratum using the Mubarakzyanov derivation framework (T4).

Codimension-One Nilradical (n= 5): Real Sector

Here, g=n⋊DRX\mathfrak{g}=\mathfrak{n}\rtimes_D\mathbb{R}X, where D=ad⁡X∣n∈Der⁡(n)D=\operatorname{ad}X|_{\mathfrak{n}}\in\operatorname{Der}(\mathfrak{n}) is a non-nilpotent outer derivation. When the spectrum of DD is entirely real, the Jordan canonical forms yield:

Diagonal: D=diag⁡(1,α,β,γ,δ)D=\operatorname{diag}(1,\alpha,\beta,\gamma,\delta), −1≤δ≤γ≤β≤α≤1-1\leq\delta\leq\gamma\leq\beta\leq\alpha\leq1,

D=diag⁡(1,α,β,γ,δ),−1≤δ≤γ≤β≤α≤1(13)D=\operatorname{diag}(1,\alpha,\beta,\gamma,\delta),\qquad-1\leq\delta\leq\gamma\leq\beta\leq\alpha\leq1 \tag*{(13)}

Jordan Blocks: D=Jk(λ)⊕⋯D=J_k(\lambda)\oplus\cdots, k∈{2,3,4,5}k\in\{2,3,4,5\}.

D=Jk(λ)⊕⋯ ,k∈{2,3,4,5}(14)D=J_k(\lambda)\oplus\cdots,\qquad k\in\{2,3,4,5\} \tag*{(14)}

When n\mathfrak{n} is non-abelian (h3⊕R2\mathfrak{h}_3\oplus\mathbb{R}^2, h5\mathfrak{h}_5, n5,1\mathfrak{n}_{5,1}, n5,2\mathfrak{n}_{5,2}), DD is restricted to derivations preserving the lower central series flags.

Exact Matrix Representatives of the Solvable Complex Classes (n= 5)

A fundamental gap in coordinate presentations occurs when derivations possess complex conjugate eigenvalues. We resolve this by providing unambiguous, explicit matrix representatives.

Let g=R5⋊DRX\mathfrak{g}=\mathbb{R}^5\rtimes_D\mathbb{R}X. In gl(5,R)\mathfrak{gl}(5,\mathbb{R}), every pair of complex conjugate eigenvalues α±iβ\alpha\pm i\beta with β>0\beta>0 corresponds to the real 2×22\times2 rotation-dilation block:

R(α,β):=(α−ββα),β>0.(15)R(\alpha,\beta):= \begin{pmatrix} \alpha& -\beta\\ \beta& \alpha \end{pmatrix},\qquad\beta>0. \tag*{(15)}

Modulo the Mubarakzyanov gauge equivalence D∼cPDP−1D\sim cPDP^{-1} (c∈R×c\in\mathbb{R}^{\times}, P∈GL⁡(5,R)P\in\operatorname{GL}(5,\mathbb{R})), there exist exactly four mutually exclusive canonical matrix families with complex spectra:

Family g6, C1solv: Single Complex Pair, Semisimple Real Part

The derivation DD consists of a single rotation-dilation block and three diagonal real eigenvalues:

DC1(α,β,λ1,λ2,λ3)=(α−β000βα00000λ100000λ200000λ3),β>0,λ1≥λ2≥λ3.(16)D_{\mathrm{C1}}(\alpha,\beta,\lambda_1,\lambda_2,\lambda_3)= \begin{pmatrix} \alpha& -\beta& 0 & 0 & 0\\ \beta& \alpha& 0 & 0 & 0\\ 0 & 0 & \lambda_1 & 0 & 0\\ 0 & 0 & 0 & \lambda_2 & 0\\ 0 & 0 & 0 & 0 & \lambda_3 \end{pmatrix},\qquad\beta>0,\quad\lambda_1\geq\lambda_2\geq\lambda_3. \tag*{(16)}

By scaling X↦cXX\mapsto cX, we normalize β=1\beta=1 if β≠0\beta\neq0. The commutation relations are:

[X,e1]=αe1+βe2,[X,e2]=−βe1+αe2,[X,e3]=λ1e3,[X,e4]=λ2e4,[X,e5]=λ3e5.(17)\begin{aligned} [X,e_1]&=\alpha e_1+\beta e_2, & [X,e_2]&=-\beta e_1+\alpha e_2,\\ [X,e_3]&=\lambda_1e_3, & [X,e_4]&=\lambda_2e_4, & [X,e_5]&=\lambda_3e_5. \tag*{(17)} \end{aligned}

The algebra is unimodular if and only if tr⁡(D)=2α+λ1+λ2+λ3=0\operatorname{tr}(D)=2\alpha+\lambda_1+\lambda_2+\lambda_3=0.

Family g6, C2solv: Single Complex Pair with Nilpotent Real Jordan Block

The derivation DD couples the rotation-dilation block with a non-trivial real Jordan cell of size 2 or 3:

DC2(α,β,λ,μ)=(α−β000βα00000λ10000λ00000μ),β>0.(18)D_{\mathrm{C2}}(\alpha,\beta,\lambda,\mu)= \begin{pmatrix} \alpha& -\beta& 0 & 0 & 0\\ \beta& \alpha& 0 & 0 & 0\\ 0 & 0 & \lambda& 1 & 0\\ 0 & 0 & 0 & \lambda& 0\\ 0 & 0 & 0 & 0 & \mu \end{pmatrix},\qquad\beta>0. \tag*{(18)}

The commutation relations are:

[X,e1]=αe1+βe2,[X,e2]=−βe1+αe2,[X,e3]=λe3,[X,e4]=e3+λe4,[X,e5]=μe5.(19)\begin{aligned} [X,e_1]&=\alpha e_1+\beta e_2, & [X,e_2]&=-\beta e_1+\alpha e_2,\\ [X,e_3]&=\lambda e_3, & [X,e_4]&=e_3+\lambda e_4, & [X,e_5]&=\mu e_5. \tag*{(19)} \end{aligned}

Family g6, C3solv: Double Independent Complex Pairs

The derivation DD contains two independent rotation-dilation blocks and one real eigenvalue:

DC3(α1,β1,α2,β2,λ)=(α1−β1000β1α100000α2−β2000β2α200000λ),β1≥β2>0.(20)D_{\mathrm{C3}}(\alpha_1,\beta_1,\alpha_2,\beta_2,\lambda)=\begin{pmatrix}\alpha_1 & -\beta_1 & 0 & 0 & 0 \\ \beta_1 & \alpha_1 & 0 & 0 & 0 \\ 0 & 0 & \alpha_2 & -\beta_2 & 0 \\ 0 & 0 & \beta_2 & \alpha_2 & 0 \\ 0 & 0 & 0 & 0 & \lambda\end{pmatrix}, \qquad\beta_1 \ge\beta_2 > 0. \tag*{(20)}

The commutation relations are:

[X,e1]=α1e1+β1e2,[X,e2]=−β1e1+α1e2,[X,e3]=α2e3+β2e4,[X,e4]=−β2e3+α2e4,[X,e5]=λe5.(21)\begin{aligned} [X,e_1] &= \alpha_1 e_1+\beta_1 e_2, & [X,e_2] &= -\beta_1 e_1+\alpha_1 e_2, \\ [X,e_3] &= \alpha_2 e_3+\beta_2 e_4, & [X,e_4] &= -\beta_2 e_3+\alpha_2 e_4, & [X,e_5] &= \lambda e_5. \tag*{(21)} \end{aligned}

Unimodularity holds when tr⁡(D)=2α1+2α2+λ=0\operatorname{tr}(D)=2\alpha_1+2\alpha_2+\lambda=0.

Family g6, C4solv: Generalized Complex Jordan Cell of Size 4

When the complex conjugate eigenvalues α±iβ\alpha\pm i\beta have algebraic multiplicity 2 but geometric multiplicity 1, DD contains a 4×44 \times4 indecomposable real Jordan block:

DC4(α,β,λ)=(R(α,β)I200R(α,β)000λ)=(α−β100βα01000α−β000βα00000λ),β>0.(22)D_{\mathrm{C4}}(\alpha,\beta,\lambda)=\begin{pmatrix}R(\alpha,\beta) & I_2 & 0 \\ 0 & R(\alpha,\beta) & 0 \\ 0 & 0 & \lambda\end{pmatrix}=\begin{pmatrix}\alpha& -\beta& 1 & 0 & 0 \\ \beta& \alpha& 0 & 1 & 0 \\ 0 & 0 & \alpha& -\beta& 0 \\ 0 & 0 & \beta& \alpha& 0 \\ 0 & 0 & 0 & 0 & \lambda\end{pmatrix}, \qquad\beta>0. \tag*{(22)}

The commutation relations are:

[X,e1]=αe1+βe2,[X,e2]=−βe1+αe2,[X,e3]=e1+αe3+βe4,[X,e4]=e2−βe3+αe4,[X,e5]=λe5.(23)\begin{aligned} [X,e_1] &= \alpha e_1+\beta e_2, & [X,e_2] &= -\beta e_1+\alpha e_2, \\ [X,e_3] &= e_1+\alpha e_3+\beta e_4, & [X,e_4] &= e_2-\beta e_3+\alpha e_4, & [X,e_5] &= \lambda e_5. \tag*{(23)} \end{aligned}

Faithful Linear Representation in gl(6,R)

For each of the complex classes g=R5⋊DRX\mathfrak{g}=\mathbb{R}^5\rtimes_D\mathbb{R}X above, an exact faithful representation ρ:g↪gl(6,R)\rho:\mathfrak{g}\hookrightarrow\mathfrak{gl}(6,\mathbb{R}) is constructed via the affine action on R5\mathbb{R}^5:

ρ(∑i=15yiei+tX)=(tDy01×50)=(tD11tD12tD13tD14tD15y1tD21tD22tD23tD24tD25y2tD31tD32tD33tD34tD35y3tD41tD42tD43tD44tD45y4tD51tD52tD53tD54tD55y5000000).(24)\rho\left(\sum_{i=1}^{5}y_i e_i+tX\right)=\begin{pmatrix}tD & \mathbf{y}\\ \mathbf{0}_{1\times5} & 0\end{pmatrix}=\begin{pmatrix} tD_{11} & tD_{12} & tD_{13} & tD_{14} & tD_{15} & y_1 \\ tD_{21} & tD_{22} & tD_{23} & tD_{24} & tD_{25} & y_2 \\ tD_{31} & tD_{32} & tD_{33} & tD_{34} & tD_{35} & y_3 \\ tD_{41} & tD_{42} & tD_{43} & tD_{44} & tD_{45} & y_4 \\ tD_{51} & tD_{52} & tD_{53} & tD_{54} & tD_{55} & y_5 \\ 0 & 0 & 0 & 0 & 0 & 0 \end{pmatrix}. \tag*{(24)}

Because [ρ(X),ρ(ei)]=ρ(Dei)[\rho(X),\rho(e_i)]=\rho(De_i) and ρ\rho is strictly injective, this directly demonstrates that μ(g)≤6\mu(\mathfrak{g})\le6 for all codimension-one complex solvable architectures.

Codimension-Two Nilradical (n= 4): Complex Sectors

Here, g=n⋊R2\mathfrak{g}=\mathfrak{n}\rtimes\mathbb{R}^2, spanned by {x1,x2}∪n\{x_1,x_2\}\cup\mathfrak{n} with [x1,x2]=0[x_1,x_2]=0. The operators D1=ad⁡x1∣nD_1=\operatorname{ad}x_1|_{\mathfrak{n}} and D2=ad⁡x2∣nD_2=\operatorname{ad}x_2|_{\mathfrak{n}} form a two-dimensional abelian subalgebra of OutDer⁡(n)\operatorname{OutDer}(\mathfrak{n}) containing no non-zero nilpotent operators.

When n≅R4\mathfrak{n} \cong\mathbb{R}^{4}, the commuting pair (D1,D2)(D_{1},D_{2}) can admit complex eigenvalues. The maximal complex toral family is generated by simultaneous rotation blocks:

D1=(R(α1,β1)00R(γ1,δ1)),D2=(R(α2,β2)00R(γ2,δ2)),(25)D_{1}=\begin{pmatrix} R(\alpha_{1},\beta_{1}) & 0 \\ 0 & R(\gamma_{1},\delta_{1}) \end{pmatrix}, \qquad D_{2}=\begin{pmatrix} R(\alpha_{2},\beta_{2}) & 0 \\ 0 & R(\gamma_{2},\delta_{2}) \end{pmatrix}, \tag*{(25)}

where [D1,D2]=0[D_{1},D_{2}]=0 is identically satisfied since rotation blocks commute. This family admits a faithful representation in gl(5,R)\mathfrak{gl}(5,\mathbb{R}) via:

ρ(∑i=14yiei+t1x1+t2x2)=(t1D1+t2D2y01×40)∈gl(5,R).(26)\rho\left(\sum_{i=1}^{4}y_{i}e_{i}+t_{1}x_{1}+t_{2}x_{2}\right)=\begin{pmatrix} t_{1}D_{1}+t_{2}D_{2} & y \\ 0_{1\times4} & 0 \end{pmatrix}\in\mathfrak{gl}(5,\mathbb{R}). \tag*{(26)}

Codimension-Three Nilradical (n= 3)

When dim⁡n=3\dim\mathfrak{n}=3, Mostow’s bound is saturated: dim⁡n=12dim⁡g\dim\mathfrak{n}=\frac{1}{2}\dim\mathfrak{g}. The extension algebra h=g/n≅R3\mathfrak{h}=\mathfrak{g}/\mathfrak{n}\cong\mathbb{R}^{3} acts as a three-dimensional abelian subalgebra of OutDer⁡(n)\operatorname{OutDer}(\mathfrak{n}).

Theorem 7.1. Every six-dimensional solvable Lie algebra with dim⁡n=3\dim\mathfrak{n}=3 has an abelian nilradical n≅R3\mathfrak{n}\cong\mathbb{R}^{3}, and is isomorphic to the maximal split Cartan extension:

[x1,e1]=e1,[x2,e2]=e2,[x3,e3]=e3,[xi,xj]=0.(27)[x_{1},e_{1}]=e_{1},\qquad[x_{2},e_{2}]=e_{2},\qquad[x_{3},e_{3}]=e_{3},\qquad[x_{i},x_{j}]=0. \tag*{(27)}

Proof. If n≅h3\mathfrak{n}\cong\mathfrak{h}_{3}, OutDer⁡(h3)≅gl(2,R)\operatorname{OutDer}(\mathfrak{h}_{3})\cong\mathfrak{gl}(2,\mathbb{R}). By Schur’s bound on commutative matrix subalgebras, the maximal dimension of an abelian subalgebra of gl(2,R)\mathfrak{gl}(2,\mathbb{R}) is ⌊4/4⌋+1=2\lfloor4/4\rfloor+1=2. Thus, OutDer⁡(h3)\operatorname{OutDer}(\mathfrak{h}_{3}) cannot contain an injective image of the 3-dimensional abelian algebra h≅R3\mathfrak{h}\cong\mathbb{R}^{3}.

Hence, n\mathfrak{n} must be abelian (n≅R3\mathfrak{n}\cong\mathbb{R}^{3}), so OutDer⁡(R3)=gl(3,R)\operatorname{OutDer}(\mathbb{R}^{3})=\mathfrak{gl}(3,\mathbb{R}). The maximal abelian subalgebra containing no nilpotent elements is the 3-dimensional split Cartan subalgebra of diagonal matrices d(3,R)=span⁡{E11,E22,E33}\mathfrak{d}(3,\mathbb{R})=\operatorname{span}\{E_{11},E_{22},E_{33}\}.

Master Tabulation and Literature Correspondence

Table 1 presents the comprehensive taxonomy of six-dimensional real Lie algebras, including the explicit complex classes.

Algebra g\mathfrak{g}dim⁡n\dim\mathfrak{n}Non-Zero Commutation Relations / Matrix Normal FormsModuli / ParametersUnimodular?
I. Semisimple Sector (dim⁡r=0\dim\mathfrak{r}=0)
so(4)\mathfrak{so}(4)0[e1,e2]=e3,[e2,e3]=e1,[e3,e1]=e2,[e4,e5]=e6,…[e_1,e_2]=e_3,[e_2,e_3]=e_1,[e_3,e_1]=e_2,[e_4,e_5]=e_6,\ldotsCompact formYes
su(2)⊕sl(2,R)\mathfrak{su}(2)\oplus\mathfrak{sl}(2,\mathbb{R})0[e1,e2]=e3,…,[e4,e5]=2e5,[e4,e6]=−2e6,[e5,e6]=e4[e_1,e_2]=e_3,\ldots,[e_4,e_5]=2e_5,[e_4,e_6]=-2e_6,[e_5,e_6]=e_4Mixed formYes
so(2,2)\mathfrak{so}(2,2)0[e1,e2]=2e2,[e1,e3]=−2e3,[e4,e5]=2e5,[e4,e6]=−2e6,…[e_1,e_2]=2e_2,[e_1,e_3]=-2e_3,[e_4,e_5]=2e_5,[e_4,e_6]=-2e_6,\ldotsSplit formYes
so(3,1)\mathfrak{so}(3,1)0sl(2,C)R\mathfrak{sl}(2,\mathbb{C})_{\mathbb{R}} (Lorentz algebra, explicit 4×44\times4 matrices in Section 4.1)Real SimpleYes
II. Non-Solvable Levi Decompositions (dim⁡r=3\dim\mathfrak{r}=3)
g6,1Levi\mathfrak{g}^{\mathrm{Levi}}_{6,1}3so(3)⋉3R3\mathfrak{so}(3)\ltimes_3\mathbb{R}^3 (Euclidean se(3)\mathfrak{se}(3), Theorem 5.2)DiscreteYes
g6,2Levi\mathfrak{g}^{\mathrm{Levi}}_{6,2}3sl(2,R)⋉adR3\mathfrak{sl}(2,\mathbb{R})\ltimes_{\mathrm{ad}}\mathbb{R}^3 (Adjoint action, Theorem 5.2)DiscreteYes
g6,3Levi\mathfrak{g}^{\mathrm{Levi}}_{6,3}3sl(2,R)⋉fundh3\mathfrak{sl}(2,\mathbb{R})\ltimes_{\mathrm{fund}}\mathfrak{h}_3 (Symplectic central extension, Theorem 5.2)DiscreteYes
III. Nilpotent Sector (dim⁡n=6\dim\mathfrak{n}=6, Representative Indecomposable Classes)
N6,1N_{6,1}6[e1,e2]=e3,[e1,e3]=e4,[e1,e4]=e5,[e1,e5]=e6[e_1,e_2]=e_3,[e_1,e_3]=e_4,[e_1,e_4]=e_5,[e_1,e_5]=e_6Filiform (step 5)Yes
N6,2N_{6,2}6[e1,e2]=e3,[e1,e3]=e4,[e1,e4]=e5,[e1,e5]=e6,[e2,e3]=e6[e_1,e_2]=e_3,[e_1,e_3]=e_4,[e_1,e_4]=e_5,[e_1,e_5]=e_6,[e_2,e_3]=e_6Filiform (step 5)Yes
N6,3N_{6,3}6[e1,e2]=e3,[e1,e3]=e4,[e1,e4]=e5,[e2,e5]=e6,[e3,e4]=−e6[e_1,e_2]=e_3,[e_1,e_3]=e_4,[e_1,e_4]=e_5,[e_2,e_5]=e_6,[e_3,e_4]=-e_6Filiform (step 5)Yes
N6,14N_{6,14}6[e1,e2]=e4,[e1,e3]=e5,[e2,e3]=e6[e_1,e_2]=e_4,[e_1,e_3]=e_5,[e_2,e_3]=e_6Free 2-step (n3,2\mathfrak{n}_{3,2})Yes
N6,16±N^{\pm}_{6,16}6[e1,e2]=e4,[e1,e3]=e5,[e2,e4]=e6,[e3,e5]=±e6[e_1,e_2]=e_4,[e_1,e_3]=e_5,[e_2,e_4]=e_6,[e_3,e_5]=\pm e_6Real forms (step 3)Yes
N6,23N_{6,23}6[e1,e2]=e6,[e1,e5]=e6,[e3,e4]=e6[e_1,e_2]=e_6,[e_1,e_5]=e_6,[e_3,e_4]=e_6Step 2, dim⁡Z=1\dim\mathcal{Z}=1Yes
IV. Solvable: Codimension-One Nilradical (dim⁡n=5\dim\mathfrak{n}=5) – Real Spectrum
g6,1solv\mathfrak{g}^{\mathrm{solv}}_{6,1}5[X,e1]=e1,[X,e2]=αe2,[X,e3]=βe3,[X,e4]=γe4,[X,e5]=δe5[X,e_1]=e_1,[X,e_2]=\alpha e_2,[X,e_3]=\beta e_3,[X,e_4]=\gamma e_4,[X,e_5]=\delta e_5−1≤δ≤γ≤β≤α≤1-1\le\delta\le\gamma\le\beta\le\alpha\le1tr⁡D=0\operatorname{tr}D=0
g6,2solv\mathfrak{g}^{\mathrm{solv}}_{6,2}5[X,e1]=e1+e2,[X,e2]=e2,[X,e3]=αe3,[X,e4]=βe4,[X,e5]=γe5[X,e_1]=e_1+e_2,[X,e_2]=e_2,[X,e_3]=\alpha e_3,[X,e_4]=\beta e_4,[X,e_5]=\gamma e_5Jordan block J2⊕J13J_2\oplus J_1^3tr⁡D=0\operatorname{tr}D=0
g6,3solv\mathfrak{g}^{\mathrm{solv}}_{6,3}5[X,e1]=e1,[e1,e2]=e5,[e3,e4]=e5,[X,e5]=2e5,[X,e3]=αe3,…[X,e_1]=e_1,[e_1,e_2]=e_5,[e_3,e_4]=e_5,[X,e_5]=2e_5,[X,e_3]=\alpha e_3,\ldotsHeisenberg h5⋊R\mathfrak{h}_5\rtimes\mathbb{R}tr⁡D=0\operatorname{tr}D=0
V. Solvable: Codimension-One Nilradical (dim⁡n=5\dim\mathfrak{n}=5) – Complex Spectrum (Section 7.2)
g6,C1solv\mathfrak{g}^{\mathrm{solv}}_{6,\mathrm{C1}}5D=R(α,β)⊕diag⁡(λ1,λ2,λ3),[X,e1]=αe1−βe2,[X,e2]=βe1+αe2D=R(\alpha,\beta)\oplus\operatorname{diag}(\lambda_1,\lambda_2,\lambda_3),[X,e_1]=\alpha e_1-\beta e_2,[X,e_2]=\beta e_1+\alpha e_2β>0,λ1≥λ2≥λ3\beta>0,\lambda_1\ge\lambda_2\ge\lambda_32α+∑λi=02\alpha+\sum\lambda_i=0
g6,C2solv\mathfrak{g}^{\mathrm{solv}}_{6,\mathrm{C2}}5D=R(α,β)⊕J2(λ)⊕(μ),[X,e3]=λe3,[X,e4]=e3+λe4D=R(\alpha,\beta)\oplus J_2(\lambda)\oplus(\mu),[X,e_3]=\lambda e_3,[X,e_4]=e_3+\lambda e_4β>0\beta>02α+2λ+μ=02\alpha+2\lambda+\mu=0
g6,C3solv\mathfrak{g}^{\mathrm{solv}}_{6,\mathrm{C3}}5D=R(α1,β1)⊕R(α2,β2)⊕(λ)D=R(\alpha_1,\beta_1)\oplus R(\alpha_2,\beta_2)\oplus(\lambda), double rotation blocksβ1≥β2>0\beta_1\ge\beta_2>02α1+2α2+λ=02\alpha_1+2\alpha_2+\lambda=0
g6,C4solv\mathfrak{g}^{\mathrm{solv}}_{6,\mathrm{C4}}5D=(R(α,β)I20R(α,β))⊕(λ)D=\begin{pmatrix}R(\alpha,\beta)&I_2\\0&R(\alpha,\beta)\end{pmatrix}\oplus(\lambda), generalized complex cellβ>0\beta>04α+λ=04\alpha+\lambda=0
VI. Solvable: Codimension-Two and Three Nilradicals (dim⁡n∈{3,4}\dim\mathfrak{n}\in\{3,4\})
g6,4solv\mathfrak{g}^{\mathrm{solv}}_{6,4}4[x1,x2]=0,[x1,ei]=λiei,[x2,ei]=μiei (ei∈R4)[x_1,x_2]=0,[x_1,e_i]=\lambda_i e_i,[x_2,e_i]=\mu_i e_i\ (e_i\in\mathbb{R}^4)span⁡{D1,D2}⊂d(4,R)\operatorname{span}\{D_1,D_2\}\subset\mathfrak{d}(4,\mathbb{R})tr⁡Di=0\operatorname{tr}D_i=0
g6,C5solv\mathfrak{g}^{\mathrm{solv}}_{6,\mathrm{C5}}4D1=R(α1,β1)⊕R(γ1,δ1),D2=R(α2,β2)⊕R(γ2,δ2)D_1=R(\alpha_1,\beta_1)\oplus R(\gamma_1,\delta_1),D_2=R(\alpha_2,\beta_2)\oplus R(\gamma_2,\delta_2)Commuting complex pairtr⁡Di=0\operatorname{tr}D_i=0
g6,5solv\mathfrak{g}^{\mathrm{solv}}_{6,5}4[x1,x2]=0,[x1,e1]=e1,[x1,e2]=e2,[x1,e3]=e3,[x2,e4]=e4 (n4≅R2)[x_1,x_2]=0,[x_1,e_1]=e_1,[x_1,e_2]=e_2,[x_1,e_3]=e_3,[x_2,e_4]=e_4\ (\mathfrak{n}_4\cong\mathbb{R}^2)Outer torus of n4\mathfrak{n}_4No
g6,6solv\mathfrak{g}^{\mathrm{solv}}_{6,6}3[x1,x2]=[x2,x3]=[x3,x1]=0,[x1,e1]=e1,[x2,e2]=e2,[x3,e3]=e3[x_1,x_2]=[x_2,x_3]=[x_3,x_1]=0,[x_1,e_1]=e_1,[x_2,e_2]=e_2,[x_3,e_3]=e_3Maximal split Cartan on R3\mathbb{R}^3No

Table 1. Master classification of representative six-dimensional real Lie algebras.

Minimal Faithful Linear Representations ((g))

Determining the minimal faithful representation dimension μ(g)\mu(\mathfrak{g}) under Ado’s theorem is critical for constructing explicit matrix models.

Theorem 9.1 (Minimal Representation Dimension for 6D Lie Algebras). The minimal faithful representation dimension μ(g)\mu(\mathfrak{g}) for six-dimensional real Lie algebras satisfies 4≤μ(g)≤74\leq\mu(\mathfrak{g})\leq7. Specifically:

  1. μ(g)=4\mu(\mathfrak{g})=4 for so(3,1)≅sl(2,C)R\mathfrak{so}(3,1)\cong\mathfrak{sl}(2,\mathbb{C})_{\mathbb{R}} and so(2,2)≅sl(2,R)⊕sl(2,R)\mathfrak{so}(2,2)\cong\mathfrak{sl}(2,\mathbb{R})\oplus\mathfrak{sl}(2,\mathbb{R}).

  2. μ(g)=4\mu(\mathfrak{g})=4 for the Euclidean motion algebra se(3)=so(3)⋉R3\mathfrak{se}(3)=\mathfrak{so}(3)\ltimes\mathbb{R}^{3} (via affine action on R3\mathbb{R}^{3}).

  3. μ(g)=5\mu(\mathfrak{g})=5 for the free two-step nilpotent algebra N6,14=n3,2N_{6,14}=\mathfrak{n}_{3,2}.

  4. μ(g)≤6\mu(\mathfrak{g})\leq6 for the complex solvable classes g6,C1solv\mathfrak{g}^{\mathrm{solv}}_{6,\mathbb{C}1} through g6,C4solv\mathfrak{g}^{\mathrm{solv}}_{6,\mathbb{C}4} via the affine action on R5\mathbb{R}^{5}.

  5. μ(g)=6\mu(\mathfrak{g})=6 for the filiform nilpotent algebras N6,1,N6,2,N6,3N_{6,1},N_{6,2},N_{6,3} and so(4)\mathfrak{so}(4).

  6. μ(g)=7\mu(\mathfrak{g})=7 for generic solvable algebras with nilradical R5\mathbb{R}^{5} lacking faithful 6D representations.

Proof. We construct the minimal embeddings across sectors:

  • Semisimple Sector: The Lorentz algebra so(3,1)\mathfrak{so}(3,1) acts faithfully on Minkowski spacetime R3,1≅R4\mathbb{R}^{3,1}\cong\mathbb{R}^{4}, yielding μ=4\mu=4. Section 4.1 gives this representation explicitly in gl(4,R)\mathfrak{gl}(4,\mathbb{R}). The split form so(2,2)\mathfrak{so}(2,2) acts on R4\mathbb{R}^{4}, so μ=4\mu=4. For so(4)≅su(2)⊕su(2)\mathfrak{so}(4)\cong\mathfrak{su}(2)\oplus\mathfrak{su}(2), the minimal faithful real representation is its 4D defining vector representation on R4\mathbb{R}^{4} (hence μ=4\mu=4 over R\mathbb{R}).

  • Euclidean Algebra se(3)\mathfrak{se}(3): Spanned by rotations and translations on Euclidean space R3\mathbb{R}^{3}. Embedding se(3)\mathfrak{se}(3) into the affine algebra aff(3,R)⊂gl(4,R)\mathfrak{aff}(3,\mathbb{R})\subset\mathfrak{gl}(4,\mathbb{R}) yields:

ρ(A,v)=(Av01×30),A∈so(3),v∈R3.(28)\rho(A,v)=\begin{pmatrix} A & v \\ 0_{1\times3} & 0 \end{pmatrix},\qquad A\in\mathfrak{so}(3),\quad v\in\mathbb{R}^{3}. \tag*{(28)}

Because dim⁡se(3)=6\dim\mathfrak{se}(3)=6 and no 3D matrix algebra contains a 6D Lie subalgebra, μ(se(3))=4\mu(\mathfrak{se}(3))=4.

  • Complex Solvable Sector: By Section 7.2, the affine map ρ(v+tX)=(tDv00)\rho(v+tX)=\begin{pmatrix}tD&v\\0&0\end{pmatrix} embeds g\mathfrak{g} into aff(5,R)⊂gl(6,R)\mathfrak{aff}(5,\mathbb{R})\subset\mathfrak{gl}(6,\mathbb{R}) injectively, ensuring μ(g)≤6\mu(\mathfrak{g})\leq6.

  • Free Nilpotent Algebra N6,14N_{6,14}: Generated by three vectors {e1,e2,e3}\{e_{1},e_{2},e_{3}\} and their three brackets. In gl(4,R)\mathfrak{gl}(4,\mathbb{R}), the maximal dimension of a strictly upper-triangular nilpotent subalgebra is dim⁡n(4,R)=6\dim\mathfrak{n}(4,\mathbb{R})=6. However, n(4,R)\mathfrak{n}(4,\mathbb{R}) has derived length 3, whereas N6,14N_{6,14} is step 2 with a 3-dimensional center. A faithful representation requires independent matrix entries for the generators and commutators, which forces μ(N6,14)=5\mu(N_{6,14})=5.

  • Filiform Nilpotent Algebras N6,1N_{6,1}–N6,3N_{6,3}: Nilpotency step 5 implies that the adjoint operator ad⁡e1\operatorname{ad}e_{1} has nilpotency index 5. In gl(k,R)\mathfrak{gl}(k,\mathbb{R}), the maximal nilpotency index of any matrix is k−1k-1. Thus, k−1≥5⟹k≥6k-1\geq5\Longrightarrow k\geq6. The shift matrix N=∑i=15Ei,i+1∈gl(6,R)N=\sum_{i=1}^{5}E_{i,i+1}\in\mathfrak{gl}(6,\mathbb{R}) realizes this step, proving μ=6\mu=6.

This establishes the bounds.

Algorithmic Identification of Arbitrary 6D Lie Algebras

Given an arbitrary six-dimensional Lie algebra g\mathfrak{g} presented via its structure constants tensor CijkC_{ij}^{k} with [ei,ej]=∑kCijkek[e_i,e_j]=\sum_k C_{ij}^{k}e_k, we formulate a deterministic, coordinate-free decision procedure in Algorithm 1.

Algorithm 1 Deterministic Classification of 6-Dimensional Real Lie Algebras

Require: Structure constants Cijk∈R6×6×6C_{ij}^{k}\in\mathbb{R}^{6\times6\times6} satisfying Cijk=−CjikC_{ij}^{k}=-C_{ji}^{k} and Jacobi identity.

Ensure: Canonical isomorphism class and continuous moduli parameters.

  1. Compute derived subspaces g2=[g,g]\mathfrak{g}^{2}=[\mathfrak{g},\mathfrak{g}], g(2)=[g2,g2]\mathfrak{g}^{(2)}=[\mathfrak{g}^{2},\mathfrak{g}^{2}] and Killing matrix Kij=tr⁡(ad⁡ei∘ad⁡ej)K_{ij}=\operatorname{tr}(\operatorname{ad}e_i\circ\operatorname{ad}e_j).

  2. Compute the solvable radical r⊂g\mathfrak{r}\subset\mathfrak{g} (maximal solvable ideal).

  3. if dim⁡r=0\dim\mathfrak{r}=0 then ▷ Sector I: Semisimple

  4. Compute Sylvester signature (p,q)(p,q) of Killing form KK.

  5. if (p,q)=(0,6)(p,q)=(0,6) then return so(4)≅su(2)⊕su(2)\mathfrak{so}(4)\cong\mathfrak{su}(2)\oplus\mathfrak{su}(2)

  6. else if (p,q)=(2,4)(p,q)=(2,4) then return su(2)⊕sl(2,R)\mathfrak{su}(2)\oplus\mathfrak{sl}(2,\mathbb{R})

  7. else if (p,q)=(4,2)(p,q)=(4,2) then return so(2,2)≅sl(2,R)⊕sl(2,R)\mathfrak{so}(2,2)\cong\mathfrak{sl}(2,\mathbb{R})\oplus\mathfrak{sl}(2,\mathbb{R})

  8. else if (p,q)=(3,3)(p,q)=(3,3) then return so(3,1)≅sl(2,C)R\mathfrak{so}(3,1)\cong\mathfrak{sl}(2,\mathbb{C})_{\mathbb{R}} (Matrix basis in Section 4.1)

  9. end if

  10. end if

  11. if dim⁡r=3\dim\mathfrak{r}=3 then ▷ Sector II: Non-Solvable Levi Decomposition

  12. Identify Levi factor s⊂g\mathfrak{s}\subset\mathfrak{g} (dim⁡s=3\dim\mathfrak{s}=3).

  13. if Killing form on s\mathfrak{s} is negative-definite then

14: return g6,1Levi=so(3)⋉R3\mathfrak{g}^{\mathrm{Levi}}_{6,1} = \mathfrak{so}(3) \ltimes\mathbb{R}^{3} (se(3)\mathfrak{se}(3))

15: else ▷ s≅sl(2,R)\mathfrak{s} \cong\mathfrak{sl}(2,\mathbb{R})

16: &nbsp;&nbsp;&nbsp;&nbsp;if [r,r]={0}[\mathfrak{r},\mathfrak{r}] = \{0\} then return g6,2Levi=sl(2,R)⋉adR3\mathfrak{g}^{\mathrm{Levi}}_{6,2} = \mathfrak{sl}(2,\mathbb{R}) \ltimes_{\mathrm{ad}} \mathbb{R}^{3}

17: &nbsp;&nbsp;&nbsp;&nbsp;else return g6,3Levi=sl(2,R)⋉fundh3\mathfrak{g}^{\mathrm{Levi}}_{6,3} = \mathfrak{sl}(2,\mathbb{R}) \ltimes_{\mathrm{fund}} \mathfrak{h}_{3}

18: &nbsp;&nbsp;&nbsp;&nbsp;end if

19: end if

20: end if

21: Determine the nilradical n=Nil⁡(g)\mathfrak{n} = \operatorname{Nil}(\mathfrak{g}) (maximal nilpotent ideal).

22: if dim⁡n=6\dim\mathfrak{n} = 6 then ▷ Sector III: Nilpotent

23: &nbsp;&nbsp;&nbsp;&nbsp;Compute nilpotency step ss, center dimension dim⁡Z\dim\mathcal{Z}, and derived dimension dim⁡g2\dim\mathfrak{g}^{2}.

24: &nbsp;&nbsp;&nbsp;&nbsp;if s=5s = 5 then ▷ Filiform varieties

25: &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;if [g2,g2]={0}[\mathfrak{g}^{2},\mathfrak{g}^{2}] = \{0\} then return N6,1N_{6,1}

26: &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;else if dim⁡[g,g2]=1\dim[\mathfrak{g},\mathfrak{g}^{2}] = 1 then return N6,2N_{6,2}

27: &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;else return N6,3N_{6,3}

28: &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;end if

29: &nbsp;&nbsp;&nbsp;&nbsp;else if s=2s = 2 then

30: &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;if dim⁡Z=3\dim\mathcal{Z} = 3 and dim⁡g2=3\dim\mathfrak{g}^{2} = 3 then return N6,14N_{6,14} (Free 2-step n3,2\mathfrak{n}_{3,2})

31: &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;else if dim⁡Z=2\dim\mathcal{Z} = 2 then return Match between N6,13N_{6,13} and N6,22N_{6,22}

32: &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;else if dim⁡Z=1\dim\mathcal{Z} = 1 then return N6,23N_{6,23}

33: &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;end if

34: &nbsp;&nbsp;&nbsp;&nbsp;else if s=3s = 3 then

35: &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Check quadratic form signature on g/Z\mathfrak{g}/\mathcal{Z}:

36: &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;if Signature is (2,0)(2,0) then return N6,16+N^{+}_{6,16}

37: &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;else if Signature is (1,1)(1,1) then return N6,16−N^{-}_{6,16}

38: &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;else Match remaining classes N6,9N_{6,9}, N6,10N_{6,10}, N6,12N_{6,12}, N6,18N_{6,18}, N6,20N_{6,20}, N6,21N_{6,21}, N6,24N_{6,24}

39: &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;end if

40: &nbsp;&nbsp;&nbsp;&nbsp;else if s=4s = 4 then

41: &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Match classes N6,4N_{6,4}–N6,8N_{6,8}, N6,11N_{6,11}, N6,15N_{6,15}, N6,17N_{6,17}, N6,19N_{6,19} via upper central series flags.

42: &nbsp;&nbsp;&nbsp;&nbsp;end if

43: end if

44: if dim⁡n=5\dim\mathfrak{n} = 5 then ▷ Sector IV: Codimension-One Nilradical

45: &nbsp;&nbsp;&nbsp;&nbsp;Pick X∈g∖nX \in\mathfrak{g} \setminus\mathfrak{n} such that D=ad⁡X∣nD = \operatorname{ad} X|_{\mathfrak{n}} is non-nilpotent.

46: &nbsp;&nbsp;&nbsp;&nbsp;Compute eigenvalues of DD:

47: &nbsp;&nbsp;&nbsp;&nbsp;if Eigenvalues contain complex roots α±iβ\alpha\pm i\beta (β>0\beta> 0) then ▷ Complex Sector

48: &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;if Two distinct pairs of complex roots then

49: &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;return g6,C3solv\mathfrak{g}^{\mathrm{solv}}_{6,\mathrm{C3}} with moduli (α1,β1,α2,β2,λ)(\alpha_{1},\beta_{1},\alpha_{2},\beta_{2},\lambda)

50: &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;else if One pair of complex roots with dim⁡ker⁡(D−(α±iβ))2=4\dim\ker(D - (\alpha\pm i\beta))^{2} = 4 then

51: &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;return g6,C4solv\mathfrak{g}^{\mathrm{solv}}_{6,\mathrm{C4}} (Generalized complex Jordan cell)

52: &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;else if One pair of complex roots and a real nilpotency block J2J_{2} then

53: &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;return g6,C2solv\mathfrak{g}^{\mathrm{solv}}_{6,\mathrm{C2}}

54: &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;else

55: &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;return g6,C1solv\mathfrak{g}^{\mathrm{solv}}_{6,\mathrm{C1}} (Semisimple complex pair)

56: &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;end if

57: &nbsp;&nbsp;&nbsp;&nbsp;else ▷ Real Sector

58: &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Compute real Jordan form of DD modulo Aut⁡(n)\operatorname{Aut}(\mathfrak{n}) and extract continuous parameters.

59: &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;return Corresponding solvable class g6solv(λ⃗)\mathfrak{g}^{\mathrm{solv}}_{6}(\vec{\lambda}).

60: &nbsp;&nbsp;&nbsp;&nbsp;end if

61: end if

62: if dim⁡n=4\dim\mathfrak{n} = 4 then ▷ Sector V: Codimension-Two Nilradical

63: &nbsp;&nbsp;&nbsp;&nbsp;Pick complementary generators {x1,x2}\{x_{1}, x_{2}\}; diagonalize commuting pair (D1,D2)(D_{1}, D_{2}) in OutDer⁡(n)\operatorname{OutDer}(\mathfrak{n}).

64: &nbsp;&nbsp;&nbsp;&nbsp;if (D1,D2)(D_{1}, D_{2}) span a commuting complex rotation algebra then

65: return $\mathfrak{g}^{\mathrm{solv}}_{6,C_5}$
66: else
67: return Corresponding real 2-parameter family.
68: end if
69: end if
70: if dim $\mathfrak{n}$ = 3 then ▷ Sector VI: Codimension-Three Nilradical
71: return Maximal split Cartan extension $\mathbb{R}^3 \ltimes \mathbb{R}^3$.
72: end if

You can find the Python code for this algorithm and examples of its execution as a Colab Notebook.

Geometric Applications and Theoretical Physics

The classification of six-dimensional real Lie algebras provides the mathematical foundation for several branches of geometry and field theory.

Complex and Symplectic Nilmanifolds

A compact six-dimensional nilmanifold is a quotient space M=Γ\GM = \Gamma\backslash G, where GG is a simply connected nilpotent Lie group and Γ⊂G\Gamma\subset G is a co-compact lattice. By Malcev’s criterion, Γ\Gamma exists if and only if g=Lie⁡(G)\mathfrak{g} = \operatorname{Lie}(G) has rational structure constants.

  • Complex Structures: An invariant complex structure on g\mathfrak{g} is an endomorphism J:g→gJ:\mathfrak{g}\to\mathfrak{g} with J2=−id⁡J^2=-\operatorname{id} satisfying the Nijenhuis vanishing condition. Salamon [10] proved that among the twenty-four nilpotent algebras, exactly eighteen admit invariant complex structures. The filiform algebras N6,1N_{6,1} and N6,2N_{6,2} admit no complex structures due to cohomological obstructions in the Frölicher spectral sequence.

  • Symplectic Structures: An invariant symplectic structure is a closed non-degenerate 2-form ω∈Λ2g∗\omega\in\Lambda^2\mathfrak{g}^*. The free algebra N6,14N_{6,14} admits no symplectic structure because every closed 2-form is degenerate on its 3-dimensional center.

  • Kähler Defect: By Benson and Gordon’s theorem, the only six-dimensional nilmanifold that admits a Kähler metric is the flat six-torus T6=R6/Z6T^6=\mathbb{R}^6/\mathbb{Z}^6. None of the twenty-four indecomposable classes can support a Kähler metric.

Supergravity Flux Compactifications

In Type II superstring theory, compactifying ten-dimensional spacetime on a six-dimensional solvmanifold or nilmanifold yields an effective four-dimensional gauged supergravity. The structure constants fijkf^i{}_{jk} of the six-dimensional Lie algebra define geometric fluxes:

[ej,ek]=fijkei.(29)[e_j,e_k]=f^i{}_{jk}e_i. \tag*{(29)}

The Jacobi identity represents the tadpole cancellation condition for gauge fluxes. Applying TT-duality along the cycles of a six-dimensional nilmanifold maps geometric fluxes into non-geometric QQ- and RR-fluxes, dynamically traversing the strata of our classification.

Conclusion

The structural classification of six-dimensional real Lie algebras resolves the primary phase boundary of finite-dimensional Lie theory. In this monograph, we have:

  1. Established that the semisimple sector contains exactly four real forms, categorized by their Killing signatures, with an explicit 4×44 \times4 real matrix realization of sl(2,C)R\mathfrak{sl}(2,\mathbb{C})_{\mathbb{R}}.

  1. Proved that the compact algebra su(2)\mathfrak{su}(2) cannot couple indecomposably with non-abelian three-dimensional radicals, isolating g6,1Levi\mathfrak{g}^{\mathrm{Levi}}_{6,1}, g6,2Levi\mathfrak{g}^{\mathrm{Levi}}_{6,2}, g6,3Levi\mathfrak{g}^{\mathrm{Levi}}_{6,3} as the unique indecomposable non-solvable architectures.

  1. Classified the twenty-four indecomposable nilpotent Lie algebras using the Skjelbred–Sund cohomological central extension method.

  1. Resolved the complex Jordan strata within solvable architectures by providing exact matrix normal forms and faithful 6×66 \times6 affine representations for classes g6,C1solv\mathfrak{g}^{\mathrm{solv}}_{6,\mathrm{C1}} through g6,C4solv\mathfrak{g}^{\mathrm{solv}}_{6,\mathrm{C4}} and g6,C5solv\mathfrak{g}^{\mathrm{solv}}_{6,\mathrm{C5}}.

  1. Evaluated the minimal faithful representation dimension μ(g)\mu(\mathfrak{g}) under Ado’s theorem and synthesized the entire framework into a deterministic identification algorithm.

References

  1. [2]J. E. Humphreys. Introduction to Lie Algebras and Representation Theory. Graduate Texts in Mathematics, Vol. 9. Springer-Verlag, New York, 1972.DOI
  2. [3]N. Jacobson. Lie Algebras. Dover Publications, New York, 1979.
  3. [4]A. W. Knapp. Lie Groups Beyond an Introduction. Progress in Mathematics, Vol. 140. Birkhäuser, Boston, 2nd edition, 2002.
  4. [5]J. Milnor. Curvatures of left invariant metrics on Lie groups. Advances in Mathematics, 21(3):293–329, 1976.DOI
  5. [6]V. V. Morozov. Classification of nilpotent Lie algebras of sixth order. Izvestiya Vysshikh Uchebnykh Zavedenii. Matematika, 4(5):161–171, 1958.
  6. [7]G. M. Mubarakzyanov. Classification of solvable Lie algebras of dimension four. Izvestiya Vysshikh Uchebnykh Zavedenii. Matematika, 32:114–123, 1963.
  7. [8]G. M. Mubarakzyanov. Classification of real Lie algebras of dimension five. Izvestiya Vysshikh Uchebnykh Zavedenii. Matematika, 53(3):99–105, 1966.
  8. [9]J. Patera, R. T. Sharp, P. Winternitz, and H. Zassenhaus. Invariants of real low-dimensional Lie algebras. Journal of Mathematical Physics, 17(6):986–994, 1976.
  9. [10]S. M. Salamon. Complex structures on nilpotent Lie algebras. Journal of Pure and Applied Algebra, 157(2-3):311–333, 2001.DOI
  10. [11]T. Skjelbred and T. Sund. Sur la classification des algèbres de Lie nilpotentes. Comptes Rendus de l’Académie des Sciences, Série A-B, 286(5):241–242, 1978.
  11. [13]P. Turkowski. Low-dimensional real Lie algebras with non-trivial Levi decomposition. Linear Algebra and its Applications, 171:197–212, 1988.

Paper details

vlnc:W_5c7rway8
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 1

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Mathematics

Not recorded for this paper

View statement
Looking for related papers…
Contents