Free commuting involutions on closed two-dimensional surfaces
Tatiana Neretina

TL;DR
This paper investigates the maximum number of free commuting involutions on orientable surfaces, identifying minimal genus surfaces with a given number of involutions as specific real moment-angle complexes.
Contribution
It establishes a precise relationship between the number of involutions and the minimal genus of surfaces, characterizing these surfaces as real moment-angle complexes derived from polygons.
Findings
Minimal genus surfaces with n involutions are real moment-angle complexes.
The genus of these surfaces is given by g = 1 + 2^{n-1}(n-2).
Identifies the boundary of an (n+2)-gon as key in the surface structure.
Abstract
We consider the function that assigns to an orientable surface of genus the maximal number of free commuting independent involutions on . We show that the surface of minimal genus with is a real moment-angle complex , where is the boundary of an -gon. The genus is given by the formula .
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsGeometric and Algebraic Topology · Geometry and complex manifolds · Advanced Combinatorial Mathematics
Free commuting involutions on closed two-dimensional surfaces
Tatiana Neretina
Lomonosov Moscow State University
Abstract.
We consider the function that assigns to an orientable surface of genus the maximal number of free commuting independent involutions on . We show that the surface of minimal genus with is a real moment-angle complex , where is the boundary of an -gon. The genus is given by the formula .
1. Introduction
One of the main objects of study in toric topology is the moment-angle-complex , which is a cell complex with a torus action constructed from a simplicial complex , see [1]. Along with the moment-angle complex , its real analog is considered and has many interesting properties. If is a simplicial subdivision of an -dimensional sphere with vertices, then is an -dimensional (closed) manifold, and is an -dimensional manifold. In particular, if is the boundary of an -gon, then is an orientable closed two-dimensional manifold (a surface). Given a positive integer , let be the maximal number of free commuting independent involutions on a closed orientable surface of genus (that is, is the largest possible such that acts on freely). In this paper, it is proved that the function attains a local maximum on the surfaces see Fig. 1. In other words, for any integer , the surface of minimum genus that supports a free action of has the form , where is an -gon. The genus is given by the formula .
The author is grateful to her supervisor Taras Panov for suggesting the problem and attention to this work.
2. Basic notions and preliminary statements
A moment-angle complex is a special case of the following construction:
Construction 2.1** (polyhedral product).**
Let be a simplicial complex on the set and let
[TABLE]
be a collection of pairs of topological spaces, . For each subset denote
[TABLE]
and define the polyhedral product of corresponding to a simplicial complex by
[TABLE]
Here the union is a subset of .
In the case when all pairs are the same, i.e. and for , we use the notation for .
If then is called the moment-angle complex. We will consider its real analogue:
[TABLE]
where is the a line segment, and is the pair of points.
If is a simplicial decomposition of -dimensional sphere, then is an -dimensional manifold with the action of group (see [1, Theorem 4.1.7]). In other words, we have commuting involutions on . Of these involutions no more than act freely. For completeness, below we present a proof of these facts for being the boundary of an -gon. The surfaces corresponding to -gons appear as ‘‘regular topological skew polyhedra’’ in the 1937 work of Coxeter [2].
Proposition 2.2**.**
Let be the boundary of an -gon. Then is an orientable closed surface of genus
[TABLE]
Proof.
Consider the cube with the standard structure of cubical complex. Its two-dimensional skeleton consists of faces of the form
[TABLE]
By definition, is a cubical subcomplex of . Indeed, is a union of two-dimensional faces of cube of the following form:
[TABLE]
(here and below we consider subscripts modulo , i.e., ). Each one-dimensional face is the boundary of precisely two squares, namely and . Moreover, the intersection of two squares is either empty, or a vertex, or a one-dimensional face (since this is valid for the whole cube and one-dimensional faces of form the whole one-dimensional skeleton of the cube). Each vertex is contained in precisely squares. This implies that is a closed two-dimensional orientable manifold.
The surface is glued of squares. Each square has edges, and each edge is contained in two squares. Therefore the number of edges is . All vertices of the cube are vertices of our surface, therefore the number of vertices is . Thus, the Euler characteristic is
[TABLE]
which immedially implies the formula for genus of the surface. ∎
Example 2.3**.**
Let be the boundary of a triangle. Then is the boundary of a -dimensional cube and therefore is homeomorphic to a sphere, i.e., the genus is [math].
Lemma 2.4**.**
If is the boundary of an -gon, then there ia a free action of on .
Proof.
We describe a freely acting subgroup by explicitly defining its generators. Let be the involution sending to and fixing the other coordinates. The involutions , , commute and therefore generate a -action. However, this action is not free as fixes the points whose -th coordinate is zero. Consider the composition , where (it corresponds to a diagonal in the polygon ). The set of fixed points of the involution is the set with coordinates , but does not contain such points, since are not consecutive. If , then consider the involutions , , , and , , , . These involutions commute pairwise and generate a free action of . Similarly, for an odd consider the involutions , , , , , , , and . ∎
Remark*.*
For even , each element of the freely acting group defined above preserves the orientation of , since it is a composition of an even number of elementary involutions . For odd , the involution reverses the orientation of .
Next, consider an arbitrary closed two-dimensional surface . We ask the following question: find the maximal such that there is a free action of on . Obviously, depends only on the genus of the surface . (The Euler characteristic of an orientable surface of genus is , and the Euler characteristic of a nonorientable surface of genus is .) Define the function
[TABLE]
Let . Then is a closed two-dimensional manifold with Euler characteristic .
Proposition 2.5**.**
Let be the genus of . If the surface is orientable, then . If is nonorientable, then .
Proof.
Consider the case of orientable . The fundamental group of is
[TABLE]
The fundamental group of the manifold is a normal subgroup of the group (since the action of is free) and
[TABLE]
Therefore the square of each coset is 1,
[TABLE]
for all . Moreover, all cosets (i.e., elements of ) commute. The quotient group is generated by the cosets . Since they commute and their squares equal the unit, we can compose at most words of them. Therefore the order of the group is at most . Hence, . For a nonorientable surface the argument is similar: the fundamental group
[TABLE]
has generators and the order of the quotient is at most . ∎
Proposition 2.6**.**
- a)
For any orientable surface of genus and an integer , there exists an orientable surface with a free action of such that .
- b)
For any nonorientable surface of genus and an integer , there exists a surface with a free action of such that .
Proof.
Let be orientable. Consider the subgroup in the fundamental group generated by the squares of all elements
[TABLE]
and consider its normalizer subgroup . Then is a normal subgroup and , since contains only elements with even number of letters. The relations in the quotient group imply that . Hence there exists a regular covering of , which gives a free action of (see [3]). Now add one generator to : and consider . Then is a proper normal subgroup of (since in each of its elements the generator occurs even number of times), and the quotient group is . The corresponding covering of is regular and gives a free action of . Continuing this process, we add to the other generators , and obtain regular coverings of corresponding to free actions of for all .
The nonorientable case is considered similarly. ∎
3. Main results
Let be a surface of genus .
Proposition 3.1**.**
Let be the maximal integer such that
[TABLE]
Then if is even, and if is odd.
Proof.
Let act freely on a surface . Then where , and by Proposition 2.5. By definition, is the maximum of such , hence . Now we estimate from below. Consider the two cases:
Case 1: is even. Consider an orientable surface with Euler characteristic and genus . Since , by Proposition 2.6 there exists an orientable surface such that acts freely on and . The Euler characteristic is , therefore has genus and .
Case 2: is odd. Write . Consider an orientable surface of Euler characteristic , i.e., of genus . We have , since and . By Proposition 2.6 there exists an orientable surface such that acts freely on and . The Euler characteristic is , therefore has genus and . ∎
Define as the inverse function to , that is,
[TABLE]
where denotes the Lambert function (i.e., the function inverse to ).
Theorem 3.2**.**
Theorem , and an equality is attained on real moment-angle manifolds and only on them.
Proof.
Substituting and in (3.1) we obtain
[TABLE]
By Proposition 3.1, for the maximal satisfying the conditions above we have . On the other hand we have
[TABLE]
or equivalently . This implies the required inequality . An equality is attained if and only if the genus can be written as Compairing this expession with the formula from Proposition 2.2, we obtain that , where is the boundary of -gon. ∎
The values of the function are shown in Fig. 1, where the dashed line is the graph of .
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1[1] V. M. Buchstaber, T. E.Panov. Toric Topology . Mathematical Surveys and Monographs, vol. 204, American Mathematical Society, Providence, RI, 2015.
- 2[2] H.S.M. Coxeter. Regular skew polyhedra in three and four dimensions and their topological analogues Proc. London Math. Soc., 43 (2) (1937), pp. 33-62
- 3[3] A. Hatcher. Algebraic topology . Cambridge University Press, 2002.
- 4[4] T. E. Panov. Geometric structures on moment-angle manifolds . Russian Math. Surveys, 68:3 (2013), 503–568.
