0
$\begingroup$

Write the 4-ball as $\mathbb{D}^4=\mathbb{D}^2\times \mathbb{D}^2$. Then its boundary $\mathbb{S}^3\simeq \mathbb{S}^1\times \mathbb{D}^2\cup \mathbb{D}^2\times \mathbb{S}^1$. We will use implicitely this homeomorphism.

Consider the unknot $K=\mathbb{S}^1\times\{0\}$ in the boundary $\partial \mathbb{D}^4\simeq\mathbb{S}^3$ of the four ball.

Notice that the disk $D = \mathbb{D}^2\times \{0\}$ is a smooth slice disk for $K$.

Is $D$ boundary parallel? I.e. is it obtained by pushing an unknotting disk $D'\subset \mathbb{S}^3,$ $\partial D' = K$ inside $int(\mathbb{D}^4)?$

A possible approach: take as $D'\subset \partial \mathbb{D}^4$ the PL disk obtained by gluing the annulus $\mathbb{S}^1\times ([-1,1]\times \{0\})\in \mathbb{S}^1\times \mathbb{D}^2\subset\partial \mathbb{D}^4$ to the disk $\mathbb{D}^2\times (\{-1\}\times \{0\})\in \mathbb{D}^2\times \mathbb{S}^1\subset\partial \mathbb{D}^4$. Then gluing $D'$ to $D$ we get an embedded 2-sphere. If we manage to prove that this sphere is unknotted in $\mathbb{D}^4$, i.e. it bounds a 3-ball then we can use the latter to push $D'$ to $D$.

Relevance of this problem In studying Kirby calculus, one finds often the claim that when you attach a $2$-handle, the cocore of a $2$-handle is an unknotted $2$-disk, i.e. boundary parallel.
The above problem is a possible way to prove this.

$\endgroup$
2
  • 1
    $\begingroup$ I don't understand the -1. I've posted this question on MSE before trying MO, in 2 days I've got +3 votes and no hints. It is surely something well known to people in the field but that's not a good reason for a down-vote. $\endgroup$ Sep 18, 2020 at 12:02
  • 4
    $\begingroup$ There's a continuous family of disks $S^1\times[0,t] \cup (D^2 x \{t\})$ with boundary $K$. At $t=0$ you have $D^2 \times \{0\}$, and at $t=1$ the whole disk lies in $S^3$. $\endgroup$ Sep 18, 2020 at 17:53

0

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.