4
$\begingroup$

Let $g$ be a Lie algebra with enveloping algebra $U(g)$ and $Z(g)$ the center of $U(g)$.

Question 1: When is $Z(g)$ noetherian?

Question 2: When is $U(g)$ a finitely generated $Z(g)$-module? Is this true when $g$ is finite dimensional?

$\endgroup$

1 Answer 1

3
$\begingroup$

For Question 2, I think the answer is "very rarely" in characteristic 0 (unless $\mathfrak{g}$ is abelian).

For example, let $\mathfrak{g} = \langle x, y, z\rangle_{\mathbb{C}}$ with relation $[x,y] = z$ and $[x,z] = [y,z] = 0$. It follows that $Z(\mathfrak{g}) = \mathbb{C}[z]$. Indeed, let $$ a = \sum_{m, n, l\in \mathbb{N}} c_{m, n, l}x^my^nz^l\in Z(\mathfrak{g}),\quad c_{m,n,l}\in \mathbb{C}. $$ Then we have $$ [x,a] = \sum_{\substack{m, n, l\in \mathbb{N} \\ n \ge 1}}n c_{m, n, l}x^my^{n-1}z^{l+1} $$ so $c_{m,n,l} = 0$ unless $n = 0$. Similarly, $c_{m,n,l} = 0$ unless $m = 0$.

However, in characteristic $p > 0$, for an arbitrary Lie algebra $\mathfrak{g}$ we have $$ U(\mathfrak{g}^p) = \mathbb{k}[x^p\;;\;x\in \mathfrak{g}]\subset Z(\mathfrak{g}) $$ so $U(\mathfrak{g})$ is a finite $Z(\mathfrak{g})$-module provided that $\mathfrak{g}$ is finite dimensional.

$\endgroup$
6
  • $\begingroup$ Thanks, how is the situation when $g$ is semisimple in characteristic 0? $\endgroup$
    – Mare
    Nov 7, 2020 at 8:06
  • 1
    $\begingroup$ en.wikipedia.org/wiki/Harish-Chandra_isomorphism $\endgroup$
    – Wille Liu
    Nov 7, 2020 at 8:07
  • $\begingroup$ Thanks, so it is module-finite when g is semisimple and the characteristic is 0?! (it seems wikipedia doesnt say it explicitly) $\endgroup$
    – Mare
    Nov 7, 2020 at 8:09
  • $\begingroup$ Ah no, sorry. I said something idiot. In the semisimple case the centre $\mathfrak{g}$ is just a ring of symmetric polynomial of dimension equal to the rank of $\mathfrak{g}$, which is smaller than the dimension of $\mathfrak{g}$. It's not module finite. $\endgroup$
    – Wille Liu
    Nov 7, 2020 at 11:03
  • $\begingroup$ Thanks, is the universal enveloping algebra for a semisimple g module-finite over some other (canonical) commutative noetherian ring? $\endgroup$
    – Mare
    Nov 7, 2020 at 13:29

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.