Fixed-point-free automorphisms of solvable Lie algebras

Dietrich Burde, Karel Dekimpe · 2026

Read Fixed-point-free automorphisms of solvable Lie algebras by Dietrich Burde, Karel Dekimpe (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2604-27916.txt.

Introduction

Burde and Dekimpe study Lie algebras that admit an automorphism moving every nonzero element, a symmetry with no fixed points. They show any such algebra must be strongly unimodular, then give a full criterion for complex almost abelian cases. For filiform algebras they prove the symmetry exists exactly when the algebra is not characteristically nilpotent.

Catalog record

CREATORDietrich Burde, Karel Dekimpe
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE54982
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2604.27916

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Fixed-point-free automorphisms of solvable Lie algebras Dietrich Burde, Karel Dekimpe Abstract . In this paper, we investigate the existence of fixed-point-free automorphisms for finite-dimensional Lie algebras. By a result of Jacobson, a Lie algebra admitting a fixed-point-free automorphism is solvable. We prove that such a Lie algebra must be even strongly unimodular. We find a necessary and sufficient criterion such that a complex almost abelian Lie algebra admits a fixed-point-free automorphism. For complex filiform Lie algebras we show that the existence of a fixed-point-free automorphism is equivalent to not being characteristically nilpotent. 2000 Mathematics Subject Classification Primary 17B30, 17B08 1. Introduction Fixed-point-free automorphisms of groups and Lie rings have been studied by many authors. An automorphism \varphi of a group G is called f.p.f. (fixed-point-free) if C_{G}(\varphi)=1 , that is, the only fixed point of \varphi is 1 . For Lie rings, an automorphism is called f.p.f., if \varphi(x)=x implies x=0 for all x . Higman showed in 1957 that a finite-dimensional Lie algebra over a field of characteristic zero admitting an f.p.f. automorphism of prime order p is nilpotent. It was then asked for a function h(p) , which is a good bound on the nilpotency class, depending on p . In 1963 Kreknin showed that if \mathfrak{g} admits an an f.p.f. automorphism of any finite order, then \mathfrak{g} is solvable. Jacobson proved the solvability also without the assumption that \varphi has finite order. It is also known that a finite group admitting an f.p.f. automorphism is solvable. There are many other results, and a detailed survey can be found in the paper [ 7 ] by Khukhro. In this paper we present some new results on Lie algebras admitting an f.p.f. automorphism. In section 3 , we show that any such Lie algebra is strongly unimodular . Here strongly unimodular Lie algebras form a subclass of unimodular Lie algebras, arising in the context of lattices in Lie groups. Garland showed in 1966 that the Lie algebra of a (connected and simply connected) solvable Lie group admitting a lattice is strongly unimodular. In section 4 we classify all complex Lie algebras of dimension n\leq 4 admitting an f.p.f. automorphism. For such Lie algebras we also determine all possible finite orders of f.p.f. automorphisms. In dimension 4 we find a one-parameter family of Lie algebras, where all Lie algebras are strongly unimodular, but only a few of them admit an f.p.f. automorphism. This family consists of almost abelian Lie algebras. In section 5 we determine, motivated by the family of Lie algebras in dimension 4 , a necessary and sufficient criterion for the existence of an f.p.f. automorphism for complex almost abelian Lie algebras. It is related to n -cyclotomic operators, see Theorem 5.4 . Finally, in section 6 we prove that a complex filiform Lie algebra admits an f.p.f. automorphism if and only if it is not characteristically nilpotent, if and only if it admits a nonsingular derivation, and if and only if it is a derived Lie algebra, i.e., it equals [\mathfrak{h},\mathfrak{h}] for some Lie algebra \mathfrak{h} . Here we are using results from a recent paper [ 9 ] by Siciliano and Towers. 2. Preliminaries Let \mathfrak{g} be a finite-dimensional Lie algebra over a field K of characteristic zero. Denote by \Aut(\mathfrak{g}) the group of Lie algebra automorphisms of \mathfrak{g} . Definition 2.1 . A Lie algebra automorphism \varphi\in\Aut(\mathfrak{g}) is called fixed-point-free , or f.p.f., if \varphi(x)=x implies x=0 for all x\in\mathfrak{g} . Note that an automorphism \varphi\in\Aut(\mathfrak{g}) is f.p.f., if and only if \varphi-\id is nonsingular. It is well known that the existence of an f.p.f. automorphism (of finite or infinite order) has consequences for the algebraic structure of the Lie algebra \mathfrak{g} . For a survey on some results see [ 10 ] . As we have mentioned in the introduction, Higman [ 5 ] showed in 1957 , that the existence of an f.p.f. automorphism of prime order of \mathfrak{g} implies that \mathfrak{g} is nilpotent. Proposition 2.2 (Higman) . Let \mathfrak{g} be a finite-dimensional Lie algebra over a field of characteristic zero admitting an f.p.f. automorphism of order p with p prime. Then \mathfrak{g} is nilpotent. For p=2 we have the following result, which is much easier to prove, see [ 10 ] . Proposition 2.3 (Zha) . Suppose that \mathfrak{g} admits an f.p.f. automorphism of order 2 . Then \mathfrak{g} is abelian. Proof. Let \varphi be an f.p.f. automorphism of order 2 . Then the linear transformation \id-\varphi is surjective, so that every element in \mathfrak{g} can be written as x-\varphi(x) for some x . Because of \varphi(x-\varphi(x))=\varphi(x)-\varphi^{2}(x)=-(x-\varphi(x)) it follows that \varphi(x)=-x for all x\in\mathfrak{g} . This implies \varphi([x,y])=[\varphi(x),\varphi(y)]=[-x,-y]=[x,y] for all x,y\in\mathfrak{g} . Thus [x,y]=0 for all x,y\in\mathfrak{g} . ∎ A much more general result is the following one by Kreknin [ 8 ] , published in 1963 . Proposition 2.4 (Kreknin) . Let \mathfrak{g} be a finite-dimensional Lie algebra over a field of characteristic zero admitting an f.p.f. automorphism of order n . Then \mathfrak{g} is solvable. This result is still true if \mathfrak{g} admits any f.p.f. automorphism, not necessarily of finite order. This has been shown by Jacobson [ 6 ] , see Theorem 9 . Proposition 2.5 (Jacobson) . Let \mathfrak{g} be a finite-dimensional Lie algebra over a field of characteristic zero admitting an f.p.f. automorphism. Then \mathfrak{g} is solvable. The converse of this proposition is not true. Example 2.6 . The 2 -dimensional solvable non-abelian Lie algebra \mathfrak{r}_{2}(K) does not admit an f.p.f. automorphism. Indeed, let \{e_{1},e_{2}\} be a basis, with [e_{1},e_{2}]=e_{1} . Then any automorphism has the form \varphi=\begin{pmatrix}\alpha&\beta\cr 0&1\end{pmatrix} for some scalars \alpha,\beta\in K . But we have \det(\varphi-\id)=0 , and \varphi has a nonzero fixed-point. Denote by \mathfrak{n} the nilradical of \mathfrak{g} . Let \mathfrak{n}^{k}=[\mathfrak{n},\mathfrak{n}^{k-1}] be the k -th term in the descending central series of \mathfrak{n} for all k\geq 1 , with \mathfrak{n}^{1}=\mathfrak{n} . Definition 2.7 . A Lie algebra \mathfrak{g} is called strongly unimodular , if it is solvable and if for all x\in\mathfrak{g} the induced operator of {\rm ad}(x) on each subspace \mathfrak{n}^{k}/\mathfrak{n}^{k+1} has trace zero, for all k\geq 1 . A Lie algebra \mathfrak{g} is called unimodular , if \mathop{\rm tr}({\rm ad}(x))=0 for all x\in\mathfrak{g} . Strongly unimodular Lie algebras arise in the study of lattices in solvable Lie groups. The following result was proved by Garland [ 3 ] . Theorem 2.8 (Garland) . Let G be a connected and simply connected solvable Lie group admitting a cocompact lattice. Then its Lie algebra is strongly unimodular. It is well known that a strongly unimodular Lie algebra is unimodular. For completeness we will give a proof here. Lemma 2.9 . A strongly unimodular Lie algebra \mathfrak{g} is unimodular. Proof. As vector space, \mathfrak{g} corresponds to the direct sum \displaystyle\mathfrak{g}/\mathfrak{n}\oplus\mathfrak{n}/\mathfrak{n}^{2}\oplus\mathfrak{n}^{2}/\mathfrak{n}^{3}\oplus\cdots\oplus\mathfrak{n}^{c}/\mathfrak{n}^{c+1}, where \mathfrak{n}^{c+1}=0 . Since \mathfrak{g} is solvable, [\mathfrak{g},\mathfrak{g}] is nilpotent, so that [\mathfrak{g},\mathfrak{g}]\subseteq\mathfrak{n} . Hence the quotient Lie algebra \mathfrak{g}/\mathfrak{n} is abelian, and we have \displaystyle\mathop{\rm tr}({\rm ad}(x)) \displaystyle=\mathop{\rm tr}({\rm ad}(x)_{0})+\mathop{\rm tr}({\rm ad}(x)_{1})+\cdots+\mathop{\rm tr}({\rm ad}(x)_{c}) \displaystyle=0 for all x\in\mathfrak{g} , where {\rm ad}(x)_{i} ( 1\leq i\leq c ) is the induced operator on the quotient \mathfrak{n}^{i}/\mathfrak{n}^{i+1} and {\rm ad}(x_{0})=0 (which is the induced operator on \mathfrak{g}/\mathfrak{n} ). ∎ The converse statement need not

Read the complete text