1
$\begingroup$

I'm hoping to learn of a criterion for the existence of a bi-invariant one form on a Lie group $G$. I'm looking for a reference that there are no such one forms son $SU(n)$ (as long as this is in fact true!).

Is there is a particularly nice proof in the case of $SU(n)$ that would also be great!

$\endgroup$
1
  • 3
    $\begingroup$ Given a compact connected Lie group $G$ with Lie algebra $\mathfrak g$, the space of bi-invariant one forms is isomorphic to the space of $Ad(G)$-invariant vectors in the dual of $\mathfrak g$; for s semi-simple connected compact group $G$, such vectors do not exist; if $G$ is a torus, then they do. $\endgroup$ Aug 21, 2014 at 3:11

1 Answer 1

3
$\begingroup$

Use conjugation to rotate the one-form at the unit. Use simplicity of $SU(n)$ to show that one can rotate the one-form at the unit in any direction. That, combine with bi-invariance, would imply that the one-form is zero.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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