2
$\begingroup$

Let $\mathfrak{g}$ be a semisimple Lie algebra and $U(\mathfrak{g})$ its universal enveloping algebra. Then $U(\mathfrak{g})$ is a hopf algebra. Is $\Delta(1) = 1 \otimes 1$ or $\Delta(1) = 1 \otimes 1 + 1 \otimes 1$? Here $1$ is the identity in $U(\mathfrak{g})$. Thank you very much.

$\endgroup$
  • 1
    $\begingroup$ By definition of a Hopf algebra, the comultiplication map is an algebra homomorphism, so $\Delta(1)=1\otimes 1$. This is not to be confused with the comultiplication of Frobenius algebras, which are not algebra homomorphisms. $\endgroup$ – Hanno Oct 26 '14 at 14:27
  • $\begingroup$ @Hanno, thank you very much. Yes, I think we have $\Delta(1) = 1 \otimes 1$ since $1 \otimes 1$ is the identity of $U(\mathfrak{g}) \otimes U(\mathfrak{g})$ and $\Delta$ is a homomorphism of algebras. $\endgroup$ – LJR Oct 30 '14 at 7:15
3
$\begingroup$

One of the axioms of a bialgebra $B$ (in particular a Hopf algebra) is:

axiom

(taken from Wikipedia: in their notation $K$ is the base field and $B$ is the bialgebra).

Here $\eta$ is the unit and $\Delta$ is coproduct. In plain words, $\eta(\lambda \in K) = \lambda 1_B$. By the commutativity of the diagram, $$\Delta(1_B) = \Delta(\eta(1_K)) = (\eta \otimes \eta)(1_K \otimes 1_K) = 1_B \otimes 1_B.$$

As Hanno says you can interpret that by remembering that $\Delta$ is a algebra homomorphism in a bialgebra, so it has to preserve the unit.

The equation $\Delta(x) = 1 \otimes x + x \otimes 1$ is only valid for the elements of $\mathfrak{g}$ in $U(\mathfrak{g})$. Such an element is called a primitive element. You should be very careful when computing the coproduct in $U(\mathfrak{g})$, because it's tempting to believe that since $\Delta(x) = 1 \otimes x + x \otimes 1$ on generators, the same holds true for every element. That's not the case. The rule to use is again that $\Delta$ is a algebra homomorphism, so for example if $x, y \in \mathfrak{g}$, $$\Delta(xy) = \Delta(x) \Delta(y) = (1 \otimes x + x \otimes 1) (1 \otimes y + y \otimes 1) = 1 \otimes xy + x \otimes y + y \otimes x + xy \otimes 1.$$ And a similar pattern holds true for products with more terms.

$\endgroup$
-1
$\begingroup$

On my master thesis two year ago, i think i used $\Delta(1) = 1 \otimes 1 + 1 \otimes 1$. By a fast googling, i found the same on wiki.

I hope that helps.

$\endgroup$
  • 1
    $\begingroup$ The equality $\Delta(X) = X\otimes 1 + 1\otimes X$ holds for $X\in {\mathfrak g}\subset{\mathscr U}{\mathfrak g}$, but not for the unit $1$ of ${\mathscr U}{\mathfrak g}$. As said in the comments, the unit has to be preserved under comultiplication, and the unit of ${\mathscr U}{\mathfrak g}\otimes{\mathscr U}{\mathfrak g}$ is $1\otimes 1$. $\endgroup$ – Hanno Oct 30 '14 at 8:00
  • 1
    $\begingroup$ Hi Giannis. Nice to see you here! Btw: I am sure you have used the correct expression $\Delta(1) = 1 \otimes 1 $ in your master thesis. (I verified this by looking at the bottom of p. 40 of your thesis as this is posted in the arXiv) ;) $\endgroup$ – KonKan Jan 7 '17 at 20:57

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.