Knotted solid tori in contact manifolds
John Etnyre, Youlin Li, Bülent Tosun · 2025
Read Knotted solid tori in contact manifolds by John Etnyre, Youlin Li, Bülent Tosun (2025) 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-2502-15165.txt.
Introduction
John Etnyre, Youlin Li, and Bülent Tosun study in 2025 solid tori that sit inside contact manifolds in a knotted way. For a knot type, they look at a measure called width and ask when it matches a classic invariant tied to the knot. They also ask when the width is larger than that value.
The authors give tests for both cases. They also prove that many knot types contain tori that cannot be thickened, a trait that had been seen before only in the three sphere. Readers in geometry get new tools for telling how knotted forms sit inside these spaces.
Catalog record
| CREATOR | John Etnyre, Youlin Li, Bülent Tosun |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 51 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2502-15165
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Knotted solid tori in contact manifolds John Etnyre, Youlin Li, Bülent Tosun Abstract . In this note we study solid tori in contact manifolds. Specifically, we study the width of a knot type and give criteria for when it is equal to the maximal Thurston-Bennequin invariant, and when it is larger. We also prove there are many “non-thickenable” tori in many knot types. These had previously only been observed in S 3 S^{3} . 1. Introduction Understanding solid tori in contact manifolds has been an essential tool for our understanding of contact manifolds and Legendrian and transverse knots. For example, they are essential in the study of Legendrian and transverse cables [ 12 , 13 ] — and were responsible for the construction of some of the first transversely non-simple knots — and more generally satellite knots [ 8 ] . More recently, they have appeared in the study of “Legendrian large cables” [ 34 , 38 ] and the classification of tight contact structures on some small Seifert fibered spaces [ 15 ] . The first real study of solid tori in contact manifolds happened in [ 12 ] where the property of being uniformly thick was defined. A knot type \mathcal{K} is called uniformly thick if any solid torus with core realizing the knot type \mathcal{K} can be contained in another solid torus in the same knot type that is a standard neighborhood of a maximal Thurston-Bennequin Legendrian representative of the knot. If a knot type is uniformly thick, then it is easy to study Legendrian representatives of cables [ 2 , 12 ] and more general satellites [ 8 ] of the knot. It is known that negative torus knots and the figure eight knot are uniformly thick [ 10 , 12 ] as well as many iterated cables of torus knots [ 28 , 29 ] . Positive torus knots are not uniformly thick, and it is due to this that many of their cables are not transversely simple [ 12 , 13 ] . We note that the contact structure on a standard neighborhood of a Legendrian knot is universally tight. Thus if K is uniformly thick, the contact structure on any solid tori in this knot type must be universally tight. A knot type can fail to be uniformly thick in one of two ways. First, there can be solid tori with convex boundary having dividing slope larger than the maximal Thurston-Bennequin invariant, and the second way is that there can be solid tori with dividing slope less than the maximal Thurston-Bennequin invariant, but that does not thicken to a neighborhood of a maximal Thurston-Bennequin invariant knot. These latter tori are called non-thickenable tori . To study the first failure of uniform thickness, the first author and Honda [ 12 ] defined the width of a knot type to be 1 1 1 The slope convention in this paper is the inverse of the one in [ 12 ] and other main references. The convention in this paper agrees with the convention used by topologists for specifying cables and surgery slopes. w(K)=\sup\{\text{slope of }\Gamma_{\partial S}\} where the supremum is taken over all convex solid tori S in the knot type K and where \Gamma_{T} denotes the dividing curves on a convex torus T . It is easy to see that w(K)\in[\tbb(K),\tbb(K)+1]. So in trying to verify a knot is uniformly thick one usually first tries to show that w(K)=\tbb(K) . Unfortunately, we do not know that many examples of knots where we can verify this. Our first result gives a large family of such examples. Theorem 1.1 . If K is an L -space knot and \tbb(K)=2g(K)-1 , where g(K) is the minimal genus of a Seifert surface for K , then the width of K is equal to the maximal Thurston-Bennequin invariant of K : w(K)=\tbb(K). It is conjectured [ 31 ] that for any L -space knot K we have \tbb(K)=2g(K)-1 . If true, then we know the width of any L -space knot, but we have many families of knots for which we know this is true, such as positive knots (that is a knot with a diagram in which all crossings are right-handed), the Berge knots (conjecturally the only knots with lens space surgeries), and if the equality is true for an L -space knot K then it also holds for any cable of K which is also an L -space knot [ 31 ] . We recall that by the (p,q) -cable of a knot K , denoted by K_{(p,q)} , we mean the knot type obtained by taking the isotopy class of the (p,q) -torus knot on the boundary of a tubular neighborhood of K , where we assume p denotes the number of times that the knot winds along the longitudinal direction and q the number of times along the meridional direction. We can show that having width equal to the maximal Thurston-Bennequin invariant is a common feature of cables. Theorem 1.2 . Suppose q/p<w(K) for some knot type K that is not Lagrangian slice (that is there is no Legendrian representative of K that bounds a Lagrangian disk). Then the (p,q) -cable K_{p,q} satisfies w(K_{p,q})=\tbb(K_{p,q}). The same is true if K is Lagrangian slice and if q/p\leq\min\{w(K),-1/2\} . In this theorem, we note that if K is Lagrangian slice then \tbb(K)=-1 so w(K)\in[-1,0] . Corollary 1.3 . If K is an L-space knot and \overline{tb}(K)=2g(K)-1 , then the (p,q) -cable K_{p,q} satisfies w(K_{p,q})=\overline{tb}(K_{p,q}). Proof. Since K is an L-space knot and satisfies \overline{tb}(K)=2g(K)-1 , it follows from Theorem 1.1 that w(K)=2g(K)-1 . Additionally, K is not Lagrangian slice. By Theorem 1.2 , if q/p<2g(K)-1 , then w(K_{p,q})=\tbb(K_{p,q}). Since K is an L-space knot, if q/p\geq 2g(K)-1 , then K_{p,q} is also an L-space knot [ 23 , Theorem 1.10] , and \tbb(K_{p,q})=2g(K_{p,q})-1 [ 31 , Proposition 3.6] . Thus, by Theorem 1.1 , if q/p\geq 2g(K)-1 , then w(K_{p,q})=\overline{tb}(K_{p,q}) . ∎ We now study a large family of knots where w(K)>\tbb(K) . It has long been thought that Legendrian representatives of a (p,q) -cabled knot had their Thurston-Bennequin invariant bounded above by pq , [ 31 ] ; however, in [ 38 ] Yasui showed that this was not the case. This motivated the definition of Legendrian large cables [ 34 , Definition 1.2. ] . We say a non-trivial Legendrian cable L\in\mathcal{L}(K_{p,q}) is large if tb(L)>pq . Later McCullough [ 34 ] gave many more examples of Legendrian large cables and showed that if K admits Legendrian large cables then it also has solid tori representatives on which the contact structure is virtually overtwisted. In particular, these knot types cannot be uniformly thick! Further work of Chakraborty, Min, and the first author [ 2 ] showed that we actually have w(K)>\tbb(K) for a knot type admitting Legendrian large cables. It is interesting to note that all the examples of knots K with their (p,q) -cables having Legendrian representatives with \tb>pq were Lagrangian slice, and the cables all happened to be (n,-1) -cables. It seemed likely, at the time, that this was just due to the techniques used to construct the cables. But our next theorem