Just Reviewing some elementary algebra and would just like to have my proof checked.

Show that the following conditions on a ring $R$ are equivalent:

  1. $R$ has only one maximal ideal.
  2. The nonunits of $R$ form an ideal.

Proof: Let $R$ be a ring with unique maximal ideal $\mathfrak{m}$ and let $f : R \to R/\mathfrak{m}$ be the quotient mapping. We note that $R/\mathfrak{m}$ is a field and therefore has two ideals, namely $\langle 0 \rangle$ and $\langle 1 \rangle$. For $r \in R$ not a unit, we observe that $r \in \ker f$, indeed, since ring homomorphism preserve units. Therefore the set of all non-units in $R$ lie in the kernel of $f$, which is an ideal. Moreover, this ideal does not contain any non-units, since otherwise, these would lie in the image of $f$.

For the converse, suppose the non-units form an ideal in $R$, call this ideal $I$. By definition, $I$ is contained in some maximal ideal $\mathfrak{m}$. Now suppose $R$ has two distinct maximal ideals $\mathfrak{m}$ and $\mathfrak{n}$. We may then consider two quotient mappings $$f: R \to R/\mathfrak{m}, \ \ \text{and} \ \ g : R \to R/\mathfrak{n},$$ where $R/\mathfrak{m}$ and $R/ \mathfrak{n}$ are fields. Note that $\ker(f) = I = \ker(g)$, from the argument above. The first isomorphism theorem then gives us that $\mathfrak{m} = \mathfrak{n}$.

Note: The last step I included a commutative diagram, but Math.Stack does not seem to support tikz.


  • $\begingroup$ Where in the proof of $1 \Rightarrow 2$ did you use the fact that $\mathfrak m$ is unique? $\endgroup$ – user491874 Dec 3 '17 at 23:53
  • $\begingroup$ Your $1\Rightarrow 2$ argument seems to be based on the incorrect notion that ring homomorphisms send non-units to non-units. Also, the sentence "this ideal does not contain any non-units, since otherwise, these would lie in the image of $f$" is confusing. First, I assume "non-units" is a typo for "units". More seriously, without qualification "the image of $f$" means $f[R]$ but $f$ is surjective so every element of $R/\mathfrak m$ is in $f[R]$. Perhaps the image you meant was $f[\mathfrak m]$, but in general the image of an ideal could contain a unit, so this argument is incomplete at best. $\endgroup$ – Derek Elkins Dec 4 '17 at 5:02

$1\Rightarrow 2$: Let $\mathfrak m$ be the unique maximal ideal of $R$ and $x\in R$ is nonunit. The principal ideal $xR$ is then a proper ideal, and, as a consequence of a (presumably known, based on Zorn’s lemma) theorem it is contained in a maximal ideal of $R$. Because $\mathfrak m$ is unique, that maximal ideal must be $\mathfrak m$, thus $xR\subseteq \mathfrak m$ and therefore $x\in\mathfrak m$. This proves $\mathfrak m$ is the set of all nonunits, which is therefore an ideal.

$2\Rightarrow 1$: Let $\mathfrak m$ be the ideal made up of non-units. Any proper ideal $\mathfrak n$ can only consist of non-units, so $\mathfrak n\subseteq \mathfrak m$. If $\mathfrak n$ is maximal, this implies $\mathfrak n=\mathfrak m$, i.e. $\mathfrak m$ is the unique maximal ideal.


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.