2
$\begingroup$

Let $k$ be an algebraically closed field of characteristic $0$ and let $\mathfrak{g}$ be a finite dimensional semisimple $k$-Lie algebra. By Whitehead's second Lemma, we know that $H^{2}(\mathfrak{g}, M)=0$ for any finite dimensional $\mathfrak{g}$-module $M$. Taking $k$ to be the trivial module, we have in particular that $H^{2}(\mathfrak{g},k):=H^{2}(\mathfrak{g})=0$.

It is also known (proved in, I think, Weibel's ``Introduction to Homological Algebra") that, for any finite dimensional $k$-Lie algebra $\mathfrak{h}$, $(\wedge^{2}\mathfrak{h})^{\mathfrak{h}}\cong H^{2}(\mathfrak{h})$, where $(\wedge^{2}\mathfrak{h})^{\mathfrak{h}}:=\{f\in \wedge^{2}\mathfrak{h}: h\cdot f=0 \forall h\in \mathfrak{h}\}$ (the invariants under the canonical action of $\mathfrak{h}$ on $\wedge^{2}\mathfrak{h}$).

EDIT: by $(V\wedge \mathfrak{g} +\mathfrak{g}\wedge V)$ I mean the subspace of skew symmetric elements of $V\otimes \mathfrak{g}+\mathfrak{g}\otimes V$.

Combining the above, we find that for our finite dimensional semisimple Lie algebra $\mathfrak{g}$, $(\wedge^{2}\mathfrak{g})^{\mathfrak{g}}=0$. In particular, $(\wedge^{2}\mathfrak{g})$ can contain no 1-dimensional $\mathfrak{g}$-submodules.

I'm interested in whether or not there is any way to generalise this: given a finite dimensional simple Lie algebra $\mathfrak{g}$ and an (irreducible, say) $\mathfrak{g}$-module $V$, can the $\mathfrak{g}$-module $(\mathfrak{g}\wedge V+V\wedge \mathfrak{g})$ ever contain a 1-dimensional $\mathfrak{g}$-submodule?

$\endgroup$
1
$\begingroup$

The way the question is formulated, it is trivial.

If $V\ne\mathfrak{g}$ ($\mathfrak{g}$ here being the adjoint module, in which case you already know everything), the corresponding sum is clearly direct, and moreover that subspace in question is isomorphic, as a $\mathfrak{g}$-module, to $\mathfrak{g}\otimes V$ (via $g\otimes v-v\otimes g\mapsto g\otimes v$). The latter is isomorphic to $\mathop{\mathrm{Hom}}(V^*,\mathfrak{g})$, and the invariant elements correspond to morphisms of $\mathfrak{g}$-modules. This implies, by Schur lemma, follows that for $V\not\simeq\mathfrak{g}^*$ there are no invariants. Finally, non-degeneracy of Killing form of course implies that $\mathfrak{g}^*\simeq\mathfrak{g}$, so $V\not\simeq\mathfrak{g}^*$ by assumption.

$\endgroup$

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.