The operad $\mathcal{D}_2$ of little $2$-disks is an operad whose $n$-th space is a $K(PB_n,1)$ where $PB_n$ denotes the pure braid group on $n$-strands. Algebras over $\mathcal{D}_2$ have a categorical analogue called braided monoidal category. More precisely, there is an operad in groupoids $\mathcal{P}aB$ whose algebras in the category of categories are braided monoidal category and which is such that a levelwise application of the nerve to $\mathcal{P}aB$ yields an operad equivalent to $\mathcal{D}_2$.

The operad $\mathcal{D}_2$ has a variant called the framed little $2$-disks operad and usually denoted $f\mathcal{D}_2$. The operad $f\mathcal{D}_2$ also has the property that each of its spaces are $K(\pi,1)$'s.

My question : Is there a known framed analogue of a braided monoidal category ? More precisely, is there an operad $\mathcal{X}$ in groupoids whose algebras in the category of categories have been defined somewhere in the literature and with the property that levelwise application of the nerve to $\mathcal{X}$ yields an operad equivalent to $f\mathcal{D}_2$.

Note that the question is not about the existence of $\mathcal{X}$ (which is straightforward) but really about the existence of an $\mathcal{X}$ whose algebras have been defined somewhere.


First of all let me correct you: it is not true that the groups $PB_n$ assemble into an operad in groups.

The point is that $PB_n$ is the fundamental group of $D_2(n)$, which requires the choice of a base point. But it is impossible to choose basepoints on the spaces $D_2(n)$ simultaneously in a way that is compatible with the operad composition and the $S_n$-action. What you can do is consider the fundamental groupoids $\Pi_1(D_2(n))$. These carry natural $S_n$-actions and are compatible with composition. Since $\Pi_1(D_2(n))$ is a bit unwieldy one can try to find a "small" full subgroupoid of $\Pi_1(D_2(n))$ which is closed under the $S_n$-action and composition. This is exactly the combinatorially defined operad $PaB$ which you mention: $PaB$ is isomorphic to the suboperad generated by a single object of $\Pi_1(D_2(2))$.

Now the fundamental group of $fD_2(n)$ is $PB_n \times \mathbf Z^n$, and one can play exactly the same game. One is naturally led to an operad $PaFB$ of parenthesized framed braids which is defined in the same way as $PaB$ except the braids are framed, that is, instead of isotopy classes of embedded intervalls we are considering isotopy classes of embedded ribbons. (I hope it's clear what I mean.) The algebras over $PaFB$ are exactly balanced monoidal categories. There is in fact a standard graphical calculus for balanced monoidal categories which is based on braids whose strands are embedded ribbons.

  • $\begingroup$ Thanks, this answers my question. Also thanks for clarifying the fact that the $PB_n$ do not form an operad in groups. There is something that looks like an operadic composition but as you say there is no way to make the $S_n$ action work. It seems to me that there should still be a non-symmetric operad whose $n$-th space is $PB_n$. Of course that's not going to be a model for $\mathcal{D}_2$. $\endgroup$ Nov 12 '14 at 20:44
  • $\begingroup$ You're right, they form a nonsymmetric operad. By the way, I think there are some things you might find useful in Nathalie Wahl's Ph.D. thesis. $\endgroup$ Nov 12 '14 at 20:59

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.