7

A commutative ring with 1 is called semi-local if it has finitely many maximal ideals and is called local if it has only one maximal ideal. There are some algebraic charactrizations for local rings. For example a ring $R$ is local ring if and only if all elements of $R$ that are not units form an ideal if and only if either $r$ or $1-r$ is unit for all $r\in R$. Is there any such caracterizations for semi-local rings with more than one maximal ideals?

E.R
  • 3,015

2 Answers2

1

I did some thinking and came to the conclusion that an analogous statement seems most likely for the slightly better class of semiperfect rings.

Recall that a ring is a clean ring if every element is the sum of an idempotent and a unit.

Proposition: a ring is local iff it is clean and has trivial idempotents.

You see, if $x$ is not already a unit, so that $x+0$ is such a decomposition, then $(x-1)+1$ works.

I think the analogous statement would be

Proposition: a ring is semiperfect iff it is clean and has a complete orthogonal set of primitive idempotents.

If the ring is commutative, then such a ring is a finite product of local rings, and every element can be expressed as $u+e$ where $e$ is the sum of finitely many elements of the complete orthogonal set.

I could not see a version specifically for semilocal rings. They can be a bit less nice.

rschwieb
  • 160,592
0

For proving things about a semi-local ring $R$, the following local-global property can be useful:

If a polynomial $f \in R[x_1, \ldots, x_n]$ represents a unit over $R_P$ for every maximal ideal $P$ of $R$, then $f$ represents a unit over $R$.

This property holds for semi-local rings and a few others, such as the ring of all algebraic integers. A good reference for consequences of this property as well as some similar properties is the paper

As an example of how this hypothesis may be used, this class of rings has a primitive element theorem and normal basis theorem, see

Skip
  • 101
  • See also Neher, Gille, "Group schemes over LG-rings and applications to cancellation theorems and Azumaya algebras", https://arxiv.org/abs/2407.02021 – Skip Sep 04 '24 at 02:21