Free \(\mathbb{C}_+\)-actions on \(\mathbb{C}^3\) are translations (Q1879010)
From MaRDI portal
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Free \(\mathbb{C}_+\)-actions on \(\mathbb{C}^3\) are translations |
scientific article |
Statements
Free \(\mathbb{C}_+\)-actions on \(\mathbb{C}^3\) are translations (English)
0 references
22 September 2004
0 references
In understanding the polynomial automorphism group Aut \(\mathbb C^n\) it is very helpful (and interesting in itself) to describe its one-parameter subgroups. Therefore, the classification of algebraic \(\mathbb C_+\)-actions on \(\mathbb C^n\) (or, equivalently, locally nilpotent derivations) has received great attention during the last decades. It is classically known that each polynomial automorphism of the complex plane is a product of linear and triangular ones [\textit{H. Jung}, J. Reine Angew. Math. 184, 161--174 (1942; Zbl 0027.08503)] and that every \(\mathbb C_+\)-action on \(\mathbb C^2\) is triangular in a suitable polynomial coordinate system (this follows from \textit{A. Gutwirth}'s reproof of Jung's result [Proc. Am. Math. Soc. 12, 631--638 (1961, Zbl 0105.34401)] and was explicitly proven by \textit{R. Rentschler} [C. R. Acad. Sci. Paris 267, 384--387 (1968; Zbl 0165.05402)]). In particular, this implies that a free \(\mathbb C_+\)-action on \(\mathbb C^2\) can be viewed as a translation. The situation is different in higher dimensions: There is a \(\mathbb C_+\)-action on \(\mathbb C^3\) which is not triangularizable [\textit{H. Bass}, J. Pure Appl. Algebra 33, 1--5 (1984; Zbl 0555.14019)] and a free \(\mathbb C_+\)-action on \(\mathbb C^4\) which is not a translation [\textit{J. Winkelmann}, Math. Z. 204, 117--127 (1990; Zbl 0701.32014)]. It should be noted that also the structure of Aut \(\mathbb C^3\) is still unknown; \textit{U. U. Umirbaev} and \textit{I. P. Shestakov} [J. Am. Math. Soc. 17, No. 1, 197-227 (2004; Zbl 1056.14085)] proved recently that the analogue of Jung's result for \(\mathbb{C}^2\) doesn't hold. In the paper under review, the author classifies free \(\mathbb C_+\)-actions on \(\mathbb C^3\) in a more general setup: For a smooth contractible three-dimensional affine algebraic variety \(X\) with a free algebraic \(\mathbb C_+\)-action such that the quotient \(S=X/\!/\mathbb C_+\) is smooth, it is proven that \(X\) is isomorphic to \(S\times\mathbb C\) and the action is induced by a translation on the second factor. Combined with the result of \textit{M. Miyanishi} that \(\mathbb C^3/\!/\mathbb C_+\simeq \mathbb C^2\) for every nontrivial \(\mathbb C_+\)-action [Osaka J. Math. 17, 329--338 (1980; Zbl 0461.13009)], the author obtains as a corollary that every free algebraic \(\mathbb C_+\)-action on \(\mathbb C^3\) is a translation in a suitable coordinate system.
0 references
contractible affine variety
0 references
free additive action
0 references
smooth quotient
0 references