1
$\begingroup$

In my research I encounter some elements in a root lattice and I would like to verify that these elements are imaginary roots. Consider the root system $J_{6, 11}$ with the following Dynkin diagram: \begin{align} \circ - \circ - \circ - \circ - \circ - & \circ - \circ - \circ - \circ - \circ \\ & \ | \\ & \ \bullet \end{align} where the $\circ$'s corresponds to the simple roots $\alpha_1, \ldots, \alpha_{10}$ (from left to right) and $\bullet$ corresponds to the simple root $\alpha_{11}$.

Let $\gamma = \alpha_1+2\alpha_2+3\alpha_3+4\alpha_4+5\alpha_5+6\alpha_6+5\alpha_7+4\alpha_8+3\alpha_9+2\alpha_{10}+2\alpha_{11}$. I would like to verify that $\gamma$ is an imaginary root in $J_{6,11}$. Are there some method to do this? Thank you very much.

$\endgroup$
1
$\begingroup$

Denote $$K = \{ \alpha\in Q_+\setminus\{0\} \mid \langle \alpha,\alpha_i^\vee \rangle \leqslant 0 \text{ for all $i$ and $\operatorname{supp}(\alpha)$ is connected} \}.$$ Here $Q_+$ is the positive part of the root lattice and $\operatorname{supp}(\alpha)$ is the support of $\alpha$, that is, the subdiagram of the Dynkin diagram corresponding to the simple roots having non-zero coefficient in $\alpha$.

Then Lemma 5.3 in "Infinite dimensional Lie algebras" by V. Kac states that $K\subset \Delta_+^\mathrm{im}$ (the set of positive imaginary roots), and since $\Delta_+^\mathrm{im}$ is $W$-invariant, $WK\subseteq\Delta_+^\mathrm{im}$ (in fact, Theorem 5.4 shows that they are equal).

Now for the root $\gamma$ you mention. Using simple reflections $s_1,\ldots,s_{10}$, one can transform $\gamma$ to the following element of the root lattice: $$\gamma' = \alpha_2+2\alpha_3+3\alpha_4+4\alpha_5+5\alpha_6+4\alpha_7+3\alpha_8+2\alpha_9+\alpha_{10}+2\alpha_{11}$$ (this is the lowest height element in the $W(\langle\alpha_1,\ldots,\alpha_{10}\rangle)$-orbit of $\gamma$). Then $$\langle\gamma',\alpha_1^\vee\rangle = \langle\gamma',\alpha_{11}^\vee\rangle=-1 \quad \text{and} \quad \langle\gamma',\alpha_i^\vee\rangle=0 \quad \text{for} \quad i=2,\ldots,10,$$ so $\gamma'\in K$ and hence $\gamma$ is an imaginary root.

$\endgroup$
1
  • $\begingroup$ thank you very much! $\endgroup$ – Jianrong Li Nov 3 '20 at 14:34

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.