Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

I can't find an example of a countable injective module over a non-Noetherian ring.

share|improve this question
And, presumably, you would like to know one? (I suspect you weren't just posting a status update... have you considered actually asking the question, then?) –  Arturo Magidin Feb 6 '11 at 22:11
In other words: This is not twitter :) –  Mariano Suárez-Alvarez Feb 6 '11 at 22:45
Yeah, I would like to know one –  Wesley Farrel Feb 6 '11 at 22:46
Then perhaps you might consider editing your question so as to make it a question, not a status update. –  Arturo Magidin Feb 7 '11 at 0:45

2 Answers 2

up vote 2 down vote accepted

Another type of example: Let A be any non-Noetherian ring, let B be the field of 2 elements, and let C be the field of rational numbers. Then R = A×B×C is a non-Noetherian ring, its module 0×B×0 is injective and finite, and its module 0×0×C is injective and countably infinite. Injective modules are at least somewhat local ideas, so even if the ring is bad somewhere, it doesn't mean all the injectives are bad.

share|improve this answer

Let $R$ be a countable non-Noetherian integral domain, e.g. a polynomial ring in countably infinitely many indeterminates over $\mathbb{Z}$ (or, more appealingly to a number theorist, the ring $\overline{\mathbb{Z}}$ of all algebraic integers).

Then the fraction field $K$ of $R$ is a countable, injective $R$-module. In fact it is the injective hull of the $R$-module $R$: see this wikipedia article on injective hulls.

In general, I confess that I have not thought as much as I should on injective hulls -- I haven't even gone through the proof that they always exist! -- but I would have to think that if $M$ is any infinite $R$-module, then its injective hull has the same cardinality as $M$. This would give many examples.

share|improve this answer
I'm not much into the infinite, but apparently Eklof showed that even if you assume R is commutative (associative, unital) and card(M) ≥ card(R), then card(E(M)) need not equal card(M). In general though, I think you cannot expect to do better than a bound depending both on card(M) and the number of generators of the ideals of R. If you need to invert a continuum of independent generators, even a countable module is going to get big. If R is commutative and noetherian, then things are much better. –  Jack Schmidt Feb 7 '11 at 0:25
@Jack: thanks for the comment. I'll keep that in mind for whenever I get around to seriously thinking about injective modules. –  Pete L. Clark Feb 7 '11 at 1:32

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.