I'm trying to understand the computation of the ribbon element of $\operatorname{U_q}(\mathfrak{sl}_2)$. I've laid out an argument by Andre Henriques below. I don't understand how the conclusion (bold sentence in last paragraph) is arrived at. The result $q^{\frac{1}{2}i^2+i}$ comes from what I call equation '***' directly above that sentence in bold. However, it seems to only be true for the case $n=1$. My question is, what allows us to forget about the other values of $n$? Here is the argument:

According to a computation by Andre Henriques, the $R$-matrix for quantum group $\operatorname{U_q}(\mathfrak{sl}_2)$ is $$\sum_{n,m}\frac{1}{n!}(\frac{1}{2}\operatorname{log}(q))^nq^{m \choose 2}\frac{1}{[m]_{q}!}H^nE^m\otimes H^nF^m$$

With the $R$-matrix in hand we proceed with computing the ribbon element. The ribbon element is defined as $\nu=m(1\otimes \mu^{-1})R=m(1\otimes \mu)\bar{R}$. Here, Turaev tells us that $\mu=\prod K^{2n_i}_i$ where $n_i\in \mathbb{Z}$ come from a particular linear combination of simple roots $\alpha_i$ satisfying $2\rho=\sum n_i \alpha_i$. Here, $\rho$ is the element a weight space that satisfies $\langle 2\rho, \alpha_i\rangle=\langle \alpha_i, \alpha_i \rangle$. For the case $\operatorname{U_q}(\mathfrak{sl}_2)$ this just means that $\mu = K$ since the root system for $\operatorname{U_q}(\mathfrak{sl}_2)$ is one dimensional.

Look again at the definition of the ribbon element $\nu\in \operatorname{U_q}(\mathfrak{sl}_2)$. The definition is telling us that $\nu$ is an element that behaves the same for different compositions of $R$, $R^{-1}$, $\mu$, $\mu^{-1}$ and $m$. So let's do just that, compose both ways and see what falls out when we let $\nu$ act on a representation. The $R$ matrix for $\operatorname{U_q}(\mathfrak{sl}_2)$ is $$ \sum_{n,m}\frac{1}{n!}(\frac{1}{2}\operatorname{log}(q))^nq^{m \choose 2}\frac{1}{[m]_{q}!}H^nE^m\otimes H^nF^m $$ so $$ \nu=m(1\otimes \mu^{-1})R=\sum_{n,m}\frac{1}{n!}(\frac{1}{2}\operatorname{log}(q))^nq^{m \choose 2}\frac{1}{[m]_{q}!}H^nE^m K^{-1}H^nF^m\quad\quad \text{*} $$ and $$ \nu=m(1\otimes \mu)\bar{R}=\sum_{n,m}\frac{1}{n!}(\frac{1}{2}\operatorname{log}(q))^nq^{m \choose 2}\frac{1}{[m]_{q}!}H^nF^m KH^nE^m\quad\quad \text{**} $$ Now we are interested in how $\nu$ acts on a category of representations. In particular, we want to check that $*$ and $**$ above act in the same way on each representation. Note that for an irreducible representation a central element should act by a scalar. Fix a representation $W_i$ and take the lowest weight vector $w_{-i}$ of that representation. For $\nu=m(1\otimes \mu^{-1})R$ we get \begin{align*} \nu(w_{-i})&=\big (\sum_{n,m}\frac{1}{n!}(\frac{1}{2}\operatorname{log}(q))^nq^{m \choose 2}\frac{1}{[m]_{q}!}H^nE^m K^{-1}H^nF^m \big ) w_{-i}\\ &=\big (\sum_{n}\frac{1}{n!}(\frac{1}{2}\operatorname{log}(q))^nH^nK^{-1}H^n\big ) w_{-i}\quad\quad \text{since $F$ on the lowest weight vector is 0}\\ &= \big (\sum_{n}\frac{1}{n!}(\frac{1}{2}\operatorname{log}(q))^nH^{2n}K^{-1}\big ) w_{-i}\quad\quad \text{***} \end{align*} In other words, $\nu$ acts on each representation of $\operatorname{U_q}(\mathfrak{sl}_2)$ by multiplication of the scalar $q^{\frac{1}{2}i^2 +i}$. Taking the highest weight vector $w_i\in W_i$ and acting on it by $\nu=m(1\otimes \mu)\bar{R}$ gives the same scalar action $q^{\frac{1}{2}i^2+i}$ on each representation as we were hoping.


Your Answer

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

Browse other questions tagged or ask your own question.