2
$\begingroup$

When applying Block's Theorem on the structure of differentiably simple rings to Lie algebras most authors require an algebraically closed field, but I can see no reference to algebraic closure in Block's paper. What extra does the algebraic closure give? I apologise for asking such a basic (and perhaps easy) question but I've talked to a number of other algebraists and asked this on another forum without getting any answer.

$\endgroup$

1 Answer 1

4
$\begingroup$

Block's theorem does not require the base field $k$ to be algebraically closed but one has to be careful when $k$ is imperfect. Then $k$ will admit field extensions of the form $K=k(a)$ with $a\not\in k$ and $a^p\in k$. Note that $K$ will become a truncated polynomial ring in one variable over a field extension $L=k(b)$ with $b^p=a^p$. If $p>2$ then the $k$-Lie algebra $\mathfrak{g}=\mathfrak{sl}_2\otimes_k K$ is simple of dimension $3p$ (but not absolutely simple) and ${\rm Der}_k(K)$ is a nontrivial $k$-form of the Witt algebra $W(1;\underline{1})$ (such twisted forms can only exist over imperfect fields). By Block's theorem, the derivation algebra of $\mathfrak{g}$ is the semidirect product ${\rm Der}_k(K)\ltimes {\rm ad}\,\mathfrak{g}$ of dimension $4p$ over $k$ but it is not isomorphic to ${\rm Id}\otimes_k W(1;\underline{1})\ltimes ({\rm ad}\,\mathfrak{sl}_2)\otimes_k k[X]/(X^p)$ over $k$. However, such an isomorphism will exist over $L$.

$\endgroup$
1
  • $\begingroup$ Many thanks Sasha for your clear explanation. $\endgroup$ Commented Oct 13, 2015 at 12:14

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .