On prime endomorphisms of the free group of rank two
Andreas Thom · 2026
Read On prime endomorphisms of the free group of rank two by Andreas Thom (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-2606-02225.txt.
Introduction
Thom studies prime endomorphisms of the free group on two generators. Endomorphisms map the group into itself in structure preserving ways. Prime ones are singled out for study here. The work asks how common non-prime endomorphisms are. Density is measured on a logarithmic scale.
Thom proves lower and upper bounds for that density. The non-prime density lies between three quarters and 27/28. Equivalent bounds govern how fast a key probability decays. The chance that a random endomorphism is non-prime decays exponentially. Rates are pinned between the stated bounds. The results align with a heuristic argument by Ian Agol. Group theorists gain sharp limits on prime maps. The free group of rank two is the test case.
Catalog record
| CREATOR | Andreas Thom |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 62 KB |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2606.02225
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
On prime endomorphisms of the free group of rank two Andreas Thom Abstract . We study prime endomorphisms of the free group F 2 F_{2} . The main results provide lower and upper bounds for the logarithmic density δ np \delta_{\mathrm{np}} of non-prime endomorphisms: 3 4 ≤ δ np ≤ 27 28 . \frac{3}{4}\leq\delta_{\mathrm{np}}\leq\frac{27}{28}. Equivalently, we prove corresponding lower and upper bounds for the exponential decay of the probability that a random endomorphism is non-prime, aligning with a heuristic argument of Ian Agol. 2020 Mathematics Subject Classification Primary 20E05; Secondary 20P05, 60B15 Introduction Every endomorphism of the additive group \mathbb{Z} has the form x\mapsto mx for some m\in\mathbb{Z} . The prime number theorem says that the number of primes less than or equal to n is approximately equal to \frac{n}{\log n} . Thus the prime number theorem may be viewed as a counting theorem for endomorphisms of \mathbb{Z} that cannot be written as a composition of non-bijective endomorphisms. This suggests the analogous counting problem for the free group F_{2} . First of all, we say that an injective endomorphism of F_{2} is prime if any way of writing it as a composition of endomorphisms must involve an automorphism, see Definition 1.2 . Hence, the question is: How often is an endomorphism of F_{2}=F(a,b) prime, when one orders endomorphisms by the lengths of the images of a fixed free basis (a,b) ? Writing \varphi_{u,v}\colon F_{2}\to F_{2} for the endomorphism with a\mapsto u and b\mapsto v , the natural finite model is to choose (u,v) uniformly from B_{n}\times B_{n} , where B_{n} is the ball of radius n in F_{2} . The problem was suggested by Ian Agol, who also outlined in [ 1 ] a heuristic argument for the fact that the density of non-prime endomorphisms should tend to zero. Our result is more precise than this, giving explicit lower and upper bounds of the exponential decay rate for the density of non-prime endomorphisms. Since |B_{n}|=2\cdot 3^{n}-1 , the total number of pairs in B_{n}\times B_{n} has order about \Theta(3^{2n}) . Therefore, it is natural to measure the size of exceptional sets by its logarithmic density. Let \mathcal{N}_{n}=\{(u,v)\in B_{n}\times B_{n}:\varphi_{u,v}\text{ is not prime}\} and define \delta_{\mathrm{np}}=\limsup_{n\to\infty}\frac{\log|\mathcal{N}_{n}|}{\log|B_{n}\times B_{n}|}. Because \log|B_{n}\times B_{n}|=2n\log 3+O(1), this quantity can equally be recovered from the exponential decay of probability that a random endomorphism is not prime. In particular, an upper bound of this probability of the form n^{B}3^{-\alpha n} implies \delta_{\mathrm{np}}\leq 1-\frac{\alpha}{2}. Our two main results are about lower and upper bounds for \delta_{\mathrm{np}} . Theorem A . Let B_{n} be the ball of radius n in F_{2} . There are absolute constants A,B>0 such that \mathbb{P}_{(u,v)\in B_{n}\times B_{n}}\bigl(\varphi_{u,v}\text{ is not prime}\bigr)\leq An^{B}3^{-n/14}. Consequently, we obtain \delta_{\mathrm{np}}\leq\frac{27}{28}. The proof of the upper bound has three ingredients. First, primeness is translated into the absence of a proper intermediate rank-two subgroup. Second, we restrict our attention to pairs (u,v) that satisfy a small cancellation condition and show that the generated subgroup cannot lie properly inside a rank-two overgroup whose Stallings core has large volume. Third, proper rank-two subgroups have a sharp uniform cogrowth gap: their non-backtracking spectral radius is at most \rho_{0}<9/4 , where \rho_{0} is the largest real root of \rho^{3}-\rho^{2}-\rho-3=0 . Combining this with an explicit count of small volume rank-two cores gives the bound n^{B}3^{-n/14} . Theorem B . \delta_{\mathrm{np}}\geq\frac{3}{4}. The lower bound comes from a high-entropy family of variable endomorphic images obtained by factoring through \varphi_{a^{2},b} . Thus, as a consequence of the main results, the logarithmic density of non-prime pairs is constrained by the inequality \frac{3}{4}\leq\delta_{\mathrm{np}}\leq\frac{27}{28}=0.96429\dots. It is a natural problem to determine the exact value of \delta_{\mathrm{np}} . We conjecture that the lower bound is sharp. Conjecture . \delta_{\mathrm{np}}=\frac{3}{4}. The paper is organized as follows. Section 1 collects basic facts about injective endomorphisms and rank-two overgroup lattices, Sections 2 – 4 prove the sphere estimate, Section 5 turns it into logarithmic-density bounds in the ball model, and Section 5.3 treats simple random walks. We end with a section of questions and remarks. 1. Basic observations on endomorphisms Let F_{2}=F(a,b) be the free group of rank two. For u,v\in F_{2} , write H(u,v)=\langle u,v\rangle\leq F_{2} and let \varphi_{u,v}\colon F_{2}\to F_{2} denote the endomorphism with a\mapsto u and b\mapsto v . 1.1. Prime endomorphisms and algebraic overgroups We start out by collecting some basic facts about endomorphisms of F_{2} . Lemma 1.1 . The endomorphism \varphi_{u,v} is injective if and only if u and v do not commute. Proof. If u and v commute, then H(u,v) is cyclic, so \varphi_{u,v} cannot be injective. Conversely, if u and v do not commute, then H(u,v) is a non-cyclic subgroup of the free group F_{2} generated by two elements. Hence H(u,v) is free of rank two, and one can show that the images of a and b freely generate the image of \varphi_{u,v} . Therefore \varphi_{u,v} is injective. ∎ Definition 1.2 . An endomorphism \varphi\colon F_{2}\to F_{2} is called prime if it is injective and, whenever \varphi=\alpha\circ\beta with \alpha,\beta\colon F_{2}\to F_{2} endomorphisms, at least one of \alpha or \beta is an automorphism. The following elementary criterion is the bridge between factorization of endomorphisms and subgroup theory. Lemma 1.3 . Let \varphi\colon F_{2}\to F_{2} be an endomorphism and put H=\varphi(F_{2}) . Then \varphi is prime if and only if \varphi is injective and there is no subgroup K with H<K<F_{2} and \operatorname{rk}K=2 . Proof. Suppose first that \varphi is prime. Then \varphi is injective by definition. If there were a subgroup K with H<K<F_{2} and \operatorname{rk}K=2 , choose an isomorphism \alpha\colon F_{2}\to K . Since H\leq K , the map \beta=\alpha^{-1}\circ\varphi\colon F_{2}\to F_{2} is an injective endomorphism and \varphi=\alpha\circ\beta . The subgroup K is proper in F_{2} , so \alpha is not an automorphism. Also H<K , so \beta(F_{2})<F_{2} , and hence \beta is not an automorphism. This contradicts primeness. Therefore no such subgroup K exists. Conversely, suppose that \varphi is injective and that there is no subgroup K with H<K<F_{2} and \operatorname{rk}K=2 . If \varphi=\alpha\circ\beta with \alpha,\beta\colon F_{2}\to F_{2} endomorphisms and neither an automorphism, then the injectivity of \varphi implies that \beta is injective. Moreover, \alpha(F_{2}) contains H=\varphi(F_{2}) , which has rank two because \varphi is injective. Hence \alpha(F_{2}) is non-cyclic, and therefore \alpha is injective by Lemma 1.1 . It follows that H=\varphi(F_{2})=\alpha(\beta(F_{2}))<\alpha(F_{2})<F_{2} . Since \alpha is injective, \alpha(F_{2}) is free of rank two. Thus \alpha(F_{2}) is a proper intermediate rank-two subgroup, contradiction. Therefore \varphi is prime. ∎ The following definition has been studied extensively in [ 4 ] . Definition 1.4 . Let H\leq J be free groups. We say that J is an algebraic extension of H , and write H\leq_{\operatorname{alg}}J , if H is not contained in a proper free factor of J . Lemma 1.5 . Let H\leq K\leq F_{2} , H non-cyclic and \operatorname{rk}K=2 . Then H\leq_{\operatorname{alg}}K . Proof. If H were contained in a proper free factor L of K , then \operatorname{rk}L\leq 1 because K has rank two. This is impossible since H is non-cyclic. ∎ Thus, for a rank-two subgroup H\leq F_{2} , non-primality of the corresponding injective endomorphism is equivalent to the existence of a proper rank-two algebraic extension H\leq_{\operatorname{alg}}K<F_{2} . 1.2. Rank-two factorization lattices For a finitely