Comultiplicativity of the Ozsváth-Szabó contact invariant (Q935871)

A pair \((S,g)\), where \(S\) is a compact, oriented surface with nonempty boundary and \(g\) is an element of the mapping class group of \(S\), determines a \(3\)-manifold \(Y_g\) with open book decomposition with page \(S\) and monodromy \(g\). Such open books on \(3\)-manifolds in turn determine contact structures \(\xi_{(S,g)}\). A key problem is to fully understand the relationship between contact topological properties of \(\xi_{(S,g)}\) and mapping class group theoretic properties of \(g\). The paper under review uses the Ozsváth-Szabó contact invariant and a comultiplication map \(\tilde{\mu} : \widehat{HF}(-Y_{hg}) \rightarrow \widehat{HF}(-Y_g) \otimes \widehat{HF}(-Y_h)\), proving that the contact invariant is natural with respect to this comultiplication, to shed some new light on this problem. The main corollary is that if the contact invariant for \(\xi_{(S,g)}\) is nonzero and the contact invariant for \(\xi_{(S,h)}\) is nonzero then the invariant for \(\xi_{(S,gh)}\) is nonzero. Since it is known that contact structures with nonzero contact invariant are tight and that strongly symplectically fillable contact structures have nonzero contact invariant, we see that if \(\xi_{(S,g)}\) and \(\xi_{(S,h)}\) are strongly symplectically fillable then \(\xi_{(S,gh)}\) must be tight. This paper also generalizes the comultiplication to include \(HF^+\), which yields obstructions to certain manifolds being L-spaces and to certain contact structures being supported by planar open books.