0
$\begingroup$

Let $A \subseteq B$ be commutative noetherian domains. Of course, if $M$ is an $A$-module, then $M \otimes_A B$ is a $B$-module. I am curious to know if there exist additional conditions on $A$ and $B$, such that every $B$-module $N$ is necessarily of the form $M \otimes_A B$ for some $A$-module $M$.

I do not mind to assume one or more of the following additional conditions: $A$ is a UFD (but I do not want to assume that $B$ is a UFD). $A$ is regular. $B$ is a complete intersection ring (but I do not want to assume that $B$ is regular). $B$ is a faithfully flat $A$-module. $B$ is a free $A$-module.

I once ran into a paper (unfortunately I cannot find it now) which calls such $N$ extendable (maybe that paper answers my curiosity?).

Improve this question
$\endgroup$
4
  • 1
    $\begingroup$ A module of the form $M\otimes B$ is sometimes called "relative free" (and a direct summand of such a module is called "relative projective"). You're asking for coniditions when every module is relative free. If $A$ is a field, then this means that every $B$-module is free which is more or less equivalent to being a principle ideal domain. $\endgroup$
    – Johannes Hahn
    May 10, 2015 at 16:58
  • 1
    $\begingroup$ @JohannesHahn: what do you mean by "more or less equivalent"? If $B$ is a nonzero ring that is not a field and $I$ is a maximal ideal of $B$ then $B/I$ is a $B$-module that is not free. $\endgroup$
    – grghxy
    May 10, 2015 at 17:04
  • $\begingroup$ Thank you both. Please can you recommend books/papers dealing with relative free modules? $\endgroup$
    – user237522
    May 10, 2015 at 18:08
  • $\begingroup$ Hochschild, Relative homological algebra, Trans. Amer. Math. Soc. 1956. $\endgroup$
    – Vinteuil
    May 11, 2015 at 8:55

1 Answer 1

Reset to default
1
$\begingroup$

An example: if $B$ is the henselization of a local ring $A$, then for any finite type $B$-module $N$ there exists a finite type $A$-module $M$ such that $N$ is a direct summand of $M\otimes_AB$. You can find this result and some similar others in http://arxiv.org/pdf/0707.4197v3.pdf

I think there are more recent papers on this question maybe by some of the same authors, but I cannot remmenber more precisely.

Improve this answer
$\endgroup$
1
  • $\begingroup$ Great! I think the results in the above mentioned paper are very close to what I was looking for. $\endgroup$
    – user237522
    May 11, 2015 at 13:56

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.