I am searching for two infinite dimensional algebras such that the center of their tensor product is bigger than the tensor product of their centers. Who knows of such examples? Thanks a lot.


The following result tells us we can't work over $k$-algebras; as suggested in the comments, this doesn't preclude an example where the algebra is over a commutative ring instead.

Result: Let $A$, $B$ be associative $k$-algebras. Then $Z(A \otimes_k B) = Z(A) \otimes_k Z(B)$.

Proof: Let $z = \sum_{i=1}^n a_i \otimes b_i$ be an element of $Z(A \otimes_k B)$, and assume wlog that the $b_i$ are $k$-linearly independent. Since $z$ is central, it must commute with all elements of the form $a \otimes 1$, $a \in A$. Therefore \begin{equation*}0 = z(a\otimes 1) - (a\otimes 1)z = \sum_{i =1}^n (a_ia - aa_i) \otimes b_i\end{equation*} and this holds iff $a_i \in Z(A)$ for all $i$, since $a \in A$ was arbitrary and the $b_i$ are linearly independent.

We can assume that the $a_i$ are linearly independent in $Z(A)$. Since $z$ must also commute with all elements of the form $1\otimes b$, $b \in B$, we get that $b_i \in Z(B)$ for all $i$ as well. Thus $Z(A\otimes_kB) \subseteq Z(A) \otimes_k Z(B)$, and the reverse inclusion is clear.

Let's try to use the fact this fails over $R$-algebras to construct an example. There are probably simpler ones, but here's what I've come up with.

Let $R = k[x]$, the polynomial ring in one variable. Define two Ore extensions:

\begin{equation*} A = k[x^{\pm1},u_1][u_2;\alpha], \quad B = k[x,y,t_1][t_2;\beta] \end{equation*} where $\alpha: x \mapsto x, u_1 \mapsto qu_1$, $\beta: x\mapsto x, y\mapsto y, t_1 \mapsto qt_1$ and $q \in k^{\times}$ is not a root of unity. In other words, $A$ and $B$ are both nearly polynomial or Laurent polynomial, but we've enforced the relations $u_2u_1 = qu_1u_2$ and $t_2t_1 = qt_1t_2$. $Z(A) = k[x^{\pm1}]$, $Z(B) = k[x,y]$ and we're viewing both of them as algebras over $R = k[x]$.

Now define $z = u_1 \otimes y - u_1x^{-1} \otimes xy$, which is not in $Z(A) \otimes_R Z(B)$ since neither $u_1$ nor $u_1x^{-1}$ are in $Z(A)$. However,

\begin{eqnarray*} z(a\otimes b) - (a\otimes b)z &=& u_1a \otimes yb - u_1x^{-1}a \otimes xyb - au_1\otimes by + au_1x^{-1}\otimes xyb \\ &=& u_1a \otimes yb - u_1a \otimes yb - au_1 \otimes by + au_1 \otimes yb \\ &=& 0 \end{eqnarray*} for all $a \in A$, $b \in B$, using the centrality of $x^{-1}$ in $A$ and the fact that the tensor product is over $k[x]$. Therefore $z \in Z(A \otimes_RB)$.

(I suppose you could just take $B = k[x,y]$ or even $B = k[x]$ if you prefer, we only really need one ring to be noncommutative for this to work.)

  • 5
    $\begingroup$ I think that the key point is that you assume that the algebra is over a field instead of a commutative ring. $\endgroup$ – Maximiliano Valle Jul 24 '13 at 11:08
  • $\begingroup$ Good point; I'm so used to "algebra" meaning $k$-algebra I completely forgot about $R$-algebras. I've updated my post with an $R$-algebra example that hopefully works. $\endgroup$ – eithil Jul 25 '13 at 11:02
  • 1
    $\begingroup$ Isn't $z=0$? The tensor product is over $k[x]$, so $u_1 x^{-1}\otimes xy$ is just $u_1 \otimes y$. I got doubtful because your Argument would in fact work for $B=k[x,y]$, and I'm sure that can't be. $\endgroup$ – Ben Apr 17 '14 at 11:17
  • $\begingroup$ I think @Ben is correct, and this example does not work. Usually I wouldn't write a comment just to say I agree with someone, but given the number of upvotes for this answer it can't hurt to attract some attention to its incorrectness. $\endgroup$ – Captain Lama Feb 7 at 14:14

It seems to me that a simple example is as follows: Let $R = \mathbb{Z}$, let $A$ be the noncommutative $\mathbb{Z}$-algebra on two generators $x$, $y$, obeying $xy=yx+2$ and let $B = \mathbb{Z}/2 \mathbb{Z}$.

Note that $A$ is basically the Weyl algebra and its center is $\mathbb{Z}$. Here is a more detailed argument: Let $W = \mathbb{Q}\langle x,y \rangle / (xy-yx-2)$; $W$ is isomorphic to the Weyl algebra. There is an obvious map $A \to W$. We claim that it is injective. Proof: By a PBW like argument, $x^i y^j$ spans $A$ over $\mathbb{Z}$, and these monomials are linearly independent in $W$. $\square$ So $x^i y^j$ is a $\mathbb{Z}$-basis of $A$, and we can compute in it in the usual way to see that $\mathbb{Z} = Z(A)$.

So $Z(A) \otimes_R Z(B)$ is $\mathbb{F}_2$, but $A \otimes_R B$ is $\mathbb{F}_2[x,y]$.

  • $\begingroup$ Is it obvious that the argument about reducing the center to scalars works without invertibility of $[x,y]$? $\endgroup$ – მამუკა ჯიბლაძე Feb 20 at 16:34
  • 1
    $\begingroup$ @მამუკაჯიბლაძე You can decide whether or not it is obvious, I added the details. $\endgroup$ – David E Speyer Feb 20 at 17:14
  • $\begingroup$ Thanks! I actually meant to ask about the part that you call "compute in the usual way" but this is also straightforward: $[x,-]$ acts on the PBW basis as $2\frac\partial{\partial y}$ and $[-,y]$ as $2\frac\partial{\partial x}$... $\endgroup$ – მამუკა ჯიბლაძე Feb 20 at 17:50

I needed an example myself with one of the algebras being commutative (and I'm sure you can't take $k[x,y]$ in the other answer). In case someone is still interested in this question, here it goes.

We let $A = \mathbb{Z}[e_{ij},n_{ij}]/(e_{ij}n_{ij})$ for all positive integers with $i<j$ and define $S$ to be the commutative (flat) $A$-algebra $A[n_{ij}^{-1}]$ (the localization at all $n_{ij}$, $i<j$). Also let $B$ be the quotient of $A\langle x_{i}\rangle_{i\geq 1}$ by the two sided ideal generated by $[x_i,x_j]-e_{ij}$ for all $i<j$. Now I claim that $S\otimes_AZ(B)\not=Z(S\otimes_AB)$. In fact, since localising at all the $n_{ij}$ kills all the $e_{ij}$, the $A$-algebra $S\otimes_AB$ is commutative, hence $Z(S\otimes_AB) = S\otimes_AB$. However, $1\otimes x_1\in S\otimes_AB$ can't be an element of $S\otimes_AZ(B)$, because this would require $x_1$ to be divided by each $n_{1j}$, $j\geq 2$, and this is absurd.

This example involves $B$ being not finitely generated, but as long as we require $S$ to be flat, any counterexample has to be of this kind, because if $S$ is flat and $B$ is finitely generated as $A$-algebra (or without this finiteness assumption on $B$ if $S$ acts sufficiently nicely on $B$), then $S\otimes_A Z(B) = Z(S\otimes_A B)$.


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.