Completion

Core Idea

The I-adic completion of a ring R with respect to an ideal I is the inverse limit of the quotients R/Iⁿ, capturing the "formal neighborhood" of V(I). For a local ring (R, m), the m-adic completion R-hat retains the essential algebraic properties of R while gaining powerful analytic-like tools such as Hensel's lemma, which lifts approximate factorizations to exact ones. Completion is the algebraic analogue of passing from polynomials to power series, or from rational numbers to real numbers, and is indispensable in local algebraic geometry and number theory.

Explainer

Completion is the algebraic process of formally adjoining limits of Cauchy sequences with respect to an ideal-adic topology. Given a commutative ring R and an ideal I, the I-adic topology on R has the powers Iⁿ as a basis of open neighborhoods of 0. The I-adic completion is the inverse limit R-hat = lim R/Iⁿ, whose elements are coherent sequences (r_1, r_2, r_3, ...) with r_n in R/Iⁿ and r_n ≡ r_{n+1} mod Iⁿ. The natural map R → R-hat sends each element to the sequence of its residues, and this map is injective when the intersection of all Iⁿ is zero (guaranteed by the Krull intersection theorem in the Noetherian local case).

The most important instance is the m-adic completion of a local ring (R, m). The completion R-hat is again a local ring with maximal ideal m-hat (the closure of m) and the same residue field R/m. The completion of k[x]_(x) is the formal power series ring k[[x]]; the completion of Z_(p) is the p-adic integers Z_p. These examples illustrate the general principle: completion replaces "polynomial-like" objects with "power-series-like" objects, gaining convergence properties at the cost of losing finite presentation.

The central payoff of completion is Hensel's lemma, which comes in several versions. The simplest: if f(x) is a polynomial over a complete local ring (R, m) and a in R satisfies f(a) ≡ 0 mod m with f'(a) a unit modulo m, then there exists a unique b in R with f(b) = 0 and b ≡ a mod m. The multiplicative version lifts coprime factorizations from the residue field to the complete ring. Hensel's lemma is the algebraic counterpart of Newton's method -- iterative refinement converges because completeness provides the necessary limits. It is the reason p-adic numbers are so powerful in number theory: factorization questions over Z can be reduced to factorization over the residue field F_p, then lifted to Z_p.

The Cohen structure theorem classifies complete local rings: every complete Noetherian local ring containing a field is isomorphic to a quotient k[[x_1, ..., x_n]]/I of a formal power series ring. This structure theorem has no analogue for non-complete rings and is one of the main reasons algebraic arguments often proceed by "passing to the completion." Completion is faithfully flat over the original ring, which means many properties (regularity, depth, being Cohen-Macaulay) can be checked after completion. This interplay between a ring and its completion -- reducing hard questions to the complete case where Cohen's theorem and Hensel's lemma provide powerful tools -- is a central technique in commutative algebra and algebraic geometry.