The slice-Bennequin inequality for the fractional Dehn twist coefficient
Peter Feller
TL;DR
The paper characterizes the Fractional Dehn Twist Coefficient ${\omega}$ on the braid group as the unique homogeneous quasimorphism of defect at most $1$ with ${\omega}(\Delta^2)=1$ and ${\omega}(B_{n-1})=0$, and it proves a slice-Bennequin type inequality for ${\omega}$, giving an affine linear lower bound for the smooth slice genus in terms of the braid's ${\omega}$. The approach ties ${\omega}$ to the homogenization of Upsilon and leverages cobordism arguments and Dehornoy order to relate braid invariants to 4-dimensional knot invariants. The authors also show that many other homogeneous quasimorphisms with similar normalization exist via Bestvina–Fujiwara-type constructions, while a unifying appendix places the slice-Bennequin inequality and its ${\omega}$-counterpart in a common homogenization framework. Overall, the work strengthens connections between braid group quasimorphisms, concordance, and 4-manifold topology, and provides a robust toolkit for bounding slice genus from braid data.
Abstract
We characterize the fractional Dehn twist coefficient (FDTC) on the $n$-stranded braid group as the unique homogeneous quasimorphism to the real numbers of defect at most 1 that equals 1 on the positive full twist and vanishes on the $(n-1)$-stranded braid subgroup. In a different direction, we establish that the slice-Bennequin inequality holds with the FDTC in place of the writhe. In other words, we establish an affine linear lower bound for the smooth slice genus of the closure of a braid in terms of the braid's FDTC. We also discuss connections between these two seemingly unrelated results. In the appendix we provide a unifying framework for the slice-Bennequin inequality and its counterpart for the FDTC.
