A Temperley-Lieb diagram is a crossingless matching of $2n$ points. We think of this matching as living in a rectangle, with $n$ points on top and the other $n$ on the bottom. To $n$ points we can assign $V^{\otimes n}$, where $V$ is a special $2$-dimensional representation of $U_q(sl_2)$. To every Temperley-Lieb diagram, one assigns a map $V^{\otimes n} \to V^{\otimes n}$. This map is built from the "cups" and "caps" in the diagram, which correspond to evaluation and coevaluation maps. I think this is essentially where WRT invariants come from.

My question is as follows: suppose we have a $\mathbb{Z}[q,q^{-1}]$-linear combination of Temperley-Lieb diagrams, such that the induced endomorphism of $V^{\otimes n}$ is an isomorphism. Does it follow that the original linear combination is the scaled identity diagram, which consists of $n$ vertical strands? In other words, are the isomorphisms of $V^{\otimes n}$ which are induced by Temperley-Lieb diagrams exactly those which are scaled identity?


Absolutely not. The representation $V^{\otimes n}$ is not irreducible and the Temperley-Lieb algebra is $\mathrm{End}_{U_q(\mathfrak{sl}_2)}(V^{\otimes n})$. So, at the very least, one can rescale the irreducible factors independently.

I highly doubt that the representation is multiplicity-free, so the TL-algebra is probably quite a bit larger than that.

  • $\begingroup$ Interesting, thank you. I didn't know that TL is the full endomorphism ring. $\endgroup$ – Ross Apr 18 at 13:31
  • 1
    $\begingroup$ @Ross $U_q(\mathfrak{sl}_2)$-invariant endomorphisms, I think $\endgroup$ – Jules Lamers Apr 18 at 21:12

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.