Adjoint group of a radical ring

From Groupprops
Revision as of 16:22, 24 August 2012 by Vipul (talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

Definition

Suppose N is an radical ring, i.e., an associative ring where, for every xN, there exists yN such that x+y+xy=0. Note that N cannot be a unital ring, because it's not possible to find a y that works for 1.

First, make N a semigroup with the operation x*y:=x+y+xy (this is the multiplicative formal group law, but unlike other formal group laws, it works even without commutativity assumptions). The adjoint group of N is the subgroup of this semigroup comprising the elements with two-sided inverses. The identity element for the adjoint group is 0N.

1 plus notation

For convenience, and to avoid confusing the elements of N with the same elements viewed as elements of the adjoint group, we denote the adjoint group element for xN as the formal expression 1+x. The advantage of this is that the group multiplication now arises formally from the multiplication in N and distributivity, i.e.:

(1+x)(1+y)=1+x+y+xy=1+(x*y)

Interpretation inside unitization

The adjoint group can be thought of concretely as the coset 1+N for N of 1 in the unitization N+Z. Alternatively, if N is an algebra over a commutative unital ring R, the adjoint group can be thought of as the coset 1+N for N of 1 in the unitization 1+N.