We have that a commutative ring $R$ with $1$ is called local if $R − R^×$ is an ideal of $R$. I have to proof the following:

$R$ is local $\iff$ $R$ has exactly one maximal ideal.

We have that every ideal $I$ $\subseteq$ $R$ is contained in a unique maximal ideal. $R$ is local so that the only maximal ideal is $M$. So, $I$ $\subseteq$ $M$. The homomorphism $\phi$: $R$ $\rightarrow$ $R/I$ tells me that the ideals of $R/I$ are all contained in $\phi$(M). $\phi$(M) is a maximal ideal in $R/I$ so that it is also the unique maximal ideal in $R/I$ since any other maximal ideal is contained in it.

So, $R/I$ has a unique maximal ideal $\phi$(M) and hence it is local.

I am not sure if this is a complete proof of the above statement. Any help would be grateful.


Assume $R$ is local and let $M=R\setminus R^\times$. By assumption $M$ is an ideal. It is also maximal because any ideal which properly contains it must contain invertible elements and hence it is all $R$. Now let $I\subseteq R$ be any proper ideal. Then all the elements of $I$ must be non invertible and hence $I\subseteq M$. So this shows any proper ideal is contained in $M$, hence it must be the only maximal ideal of $R$.

Second direction: suppose there is exactly one maximal ideal and let's call it $M$. We will show that $M=R\setminus R^x$. Obviously $M\subseteq R\setminus R^x$. Now suppose there is a non invertible element $y\in R\setminus R^x$ such that $y\notin M$. Then define:

$I=\{ry: r\in R\}$

It is easy to see that $I$ is an ideal of $R$. It is also a proper ideal because $1\notin I$. (because $y$ is not invertible). Then $I$ is contained in some maximal ideal. Since by assumption $M$ is the only maximal ideal we must have $I\subseteq M$. But this means $y\in M$, a contradiction. Hence we must have $M=R\setminus R^x$, so $R$ is indeed local.

  • $\begingroup$ We don't know yet that $M=R\setminus R^x$. In this direction of the proof I defined $M$ to be the only maximal ideal of $R$. The equality $M=R\setminus R^x$ is what we are trying to prove. $\endgroup$ – Mark May 26 '20 at 9:37

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.