0
$\begingroup$

Definition. Let $\varphi : R \to S$ be a ring map. Let $R \to R'$ be any ring map. The base change of $\varphi$ by $R \to R'$ is the ring map $\rho: R' \to S \otimes_R R'$.

Now, in the above definition, $S$ and $S\otimes_R R'$ may be considered as an $R$- and $R'$-modules, respectively. Is it true that if $S$ is finitely generated (over $R$), then so is $S \otimes_R R'$ (but over $R'$)? If the answer is affirmative, how does one see this?

Update:

Alright, let me make this question more specific. Let $S$ be generated over $R$ by $m_1,\dots,m_n$. By the definition of the $R$-module structure on $S$, this means that any element $s\in S$ may be written as $s=\sum_{i=1}^nm_i\varphi(r_i)$ for some $r_i\in R$.

Now the claim is that the $n$ elements $m_1\otimes 1,\dots, m_n \otimes 1$ generate the module $S\otimes_R R'$ over $R'$. To establish this, we need to show that every element $\gamma \in S\otimes_R R'$ may be written as $\gamma =\sum_{i=1}^n \rho(r_i') \cdot m_i\otimes 1$ for some $r_i'\in R'$.

My attempt:

As is known, every $\gamma \in S\otimes_R R'$ may be written as $\gamma=\sum_{i=1}^k\rho(\widehat{r_i'})\cdot s_i\otimes r_i'$ for some $s_i\in S$ and $r_i'\in R$. Further, since $S$ is finitely generated over $R$, we have $s_j=\sum_{i=1}^n\varphi(r_{ij})\cdot m_i$ for some $r_{ij}\in R$ and for every $j=1,\dots,k.$ On substituting these $s_j$ in the above expression of $\gamma$, we obtain $$\gamma=\left[\rho(\widehat{r_1'})\sum_{i=1}^n\varphi(r_{i1})\cdot m_i\otimes r_1'\right]+\dots+\left[\rho(\widehat{r_k'})\sum_{i=1}^n\varphi(r_{ik})\cdot m_i\otimes r_k'\right]=\\=\sum_{j=1}^k\rho(\widehat{r_j'})\sum_{i=1}^n\varphi(r_{ij})\cdot m_i\otimes r_j'.$$

How do I proceed?

$\endgroup$
  • $\begingroup$ Finitely generated over what base? $\endgroup$ – Hanno Nov 30 '16 at 9:30
  • $\begingroup$ I mean if $S$ is a finitely generated $R$-module, does it follow that $S\otimes_R R'$ is a finitely generated $R'$-module? $\endgroup$ – Cary Nov 30 '16 at 9:34
  • $\begingroup$ Let $s_1, \dots , s_n$ be generators of $S$. Can you see that $s_1 \otimes 1 , \dots , s_n \otimes 1$ generate $S \otimes_R R'$? $\endgroup$ – Crostul Nov 30 '16 at 9:40
  • $\begingroup$ I am confused: You give a link to the definition and only a few lines after this definition, there is a proof that the answer is affirmative. How does this fit together? $\endgroup$ – MooS Nov 30 '16 at 10:17
  • $\begingroup$ @Hanno, Crostul: thanks for your comments, I updated the question. MooS: oops, indeed, I never noticed it. Anyway, it is based on a few lemmas, and the purpose of the updated question is to establish a more direct proof (if possible). $\endgroup$ – Cary Nov 30 '16 at 10:30
2
$\begingroup$

Saying that $S$ is finitely generated as $R$-module is the same as saying there is a surjective homomorphism of $R$-modules $f\colon R^n\to S$, for some positive integer $n$.

Then the homomorphism $f\otimes_R\mathit{id}_{R'}\colon R^n\otimes_RR'\to S\otimes_RR'$ is surjective too. Since $R^n\otimes_RR'\cong(R')^n$, you're done.

$\endgroup$

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.