[Paper Review] A Grassmann and graded approach to coboundary Lie bialgebras, their classification, and Yang-Baxter equations

Javier de Lucas, D. Wysocki|arXiv (Cornell University)|Oct 13, 2017

Advanced Topics in Algebra33 references4 citations

TL;DR

This paper introduces a novel Grassmann and graded algebra framework to classify finite-dimensional coboundary Lie bialgebras, leveraging $ frac{rak{g}}{}$-invariant multilinear maps, $G$-gradations on Lie algebras, and the algebraic Schouten bracket. It provides an algorithmic classification of real three-dimensional coboundary Lie bialgebras and determines $r$-matrices for $ frak{so}(2,2)$ and $ frak{so}(3,2)$, simplifying automorphism-based equivalence checks without explicit automorphism computation.

ABSTRACT

This work pioneers the systematic study and classification (up to Lie algebra automorphisms) of finite-dimensional coboundary Lie bialgebras through Grassmann algebras. Several mathematical structures on Lie algebras, e.g. Killing forms or root decompositions, are extended to the Grassmann algebras of Lie algebras. This simplifies the description of the procedures and tools appearing in the theory of Lie bialgebras and originates novel techniques for its study and classification up to Lie algebra automorphisms. As a particular case, the classification of real three-dimensional coboundary Lie bialgebras is retrieved.

Motivation & Objective

• To develop geometric and algebraic tools for classifying coboundary Lie bialgebras beyond traditional algebraic methods.
• To extend structures like Killing forms, root decompositions, and gradations to Grassmann algebras of Lie algebras.
• To provide an algorithmic classification of coboundary Lie bialgebras up to Lie algebra automorphisms.
• To determine $r$-matrices solving the modified classical Yang-Baxter equation (mCYBE) on non-semisimple and higher-dimensional Lie algebras.
• To reduce the complexity of automorphism-dependent classification by using $ frak{g}$-invariant multivectors and reduced multivector spaces.

Proposed method

• Introduce $ frak{g}$-invariant multilinear maps on $ frak{g}$-modules, generalizing Killing forms and Casimir invariants to Grassmann algebras.
• Endow Lie algebras with $G$-gradations ($ frak{g} = igoplus_{ alpha in G} frak{g}^{( alpha)}$) to induce compatible decompositions on $ Lambda^k frak{g}$.
• Prove that the algebraic Schouten bracket preserves $G$-gradation: $[( Lambda^m frak{g})^{( alpha)}, ( Lambda^l frak{g})^{( beta)}]_S \subset ( Lambda^{m+l-1} frak{g})^{( alpha \star \tbeta)}$.
• Define reduced multivector spaces $\tLambda^m_R \tfrak{g} := \tLambda^m \tfrak{g} / (\tLambda^m \tfrak{g})^{\tfrak{g}}$ to classify $r$-matrices up to equivalence.
• Use $\tfrak{g}$-invariant structures to detect equivalence of coboundary Lie bialgebras under inner automorphisms, avoiding full automorphism group computation.
• Apply the framework to classify $r$-matrices for $\tfrak{r}_{3,0}$ and $\tfrak{r}'_{3,\lambda}$, identifying inequivalent solutions via orbit analysis under automorphisms.

Experimental results

Research questions

• RQ1 How can $G$-gradations on Lie algebras be used to systematically classify coboundary Lie bialgebras?
• RQ2 What is the role of the algebraic Schouten bracket in preserving graded structures and solving the mCYBE?
• RQ3 How do $\tfrak{g}$-invariant multilinear maps on Grassmann algebras simplify the classification of $r$-matrices?
• RQ4 Can the classification of coboundary Lie bialgebras be achieved without explicitly computing the automorphism group of the underlying Lie algebra?
• RQ5 What are the complete sets of inequivalent $r$-matrices for $\tfrak{so}(2,2)$ and $\tfrak{so}(3,2)$, and how do they relate to gradation-induced decompositions?

Key findings

• The classification of real three-dimensional coboundary Lie bialgebras is retrieved using the proposed Grassmann and graded algebra methods, identifying three inequivalent non-zero $r$-matrices: $r_0 = e_{12}$, $r_y = e_{23}$, and $r_z = e_{13}$.
• For $\tfrak{r}_{3,0}$, the orbits of inner automorphisms on $r$-matrices yield three distinct classes, with solutions to the mCYBE and CYBE confirmed via gradation-induced decompositions.
• The space $(\tLambda^2 \tfrak{r}_{3,0})^{\tfrak{r}_{3,0}}$ is non-trivial and spans $\tfrak{r}_{3,0}$-invariant elements, enabling classification without full automorphism group computation.
• For $\tfrak{r}'_{3,\lambda}$ with $\lambda \neq 0$, the only $r$-matrices solving the mCYBE are $\pm e_{12}$ and $0$, due to $(\tLambda^3 \tfrak{r}'_{3,\lambda})^{\tfrak{r}'_{3,\lambda}} = 0$ and $(\tLambda^2 \tfrak{r}'_{3,\lambda})^{\tfrak{r}'_{3,\lambda}} = 0$
• The automorphism group of $\tfrak{r}'_{3,\lambda}$ is generated by $T_\alpha$ with $T_\alpha(e_1) = \alpha e_1$, $T_\alpha(e_2) = \alpha e_2$, $T_\alpha(e_3) = e_3$, and $\Lambda^2 T_\alpha(e_{12}) = \alpha^2 e_{12}$, confirming the three distinct $r$-matrix classes.
• The framework successfully determines $r$-matrices for $\tfrak{so}(2,2)$ and $\tfrak{so}(3,2)$, with solutions arising from gradation-compatible structures and Schouten bracket analysis.