Bourbaki

Proposition 15

Tag 04NW, permanent. Cite it as [Bourbaki, Tag 04NW].

Commutative Algebra, chapter V, § 1. Notion of an integral element

Let AA be a filtered ring whose filtration is exhaustive and such that every principal ideal of AA is closed under the topology defined by the filtration. If the associated graded ring gr(A)\mathrm{gr}(A) (Chapter III, § 2, no. 3) is a completely integrally closed domain, then AA is a completely integrally closed domain.

Let (An)n∈Z(A_n)_{n \in \mathbf{Z}} be the filtration defined on AA; as ⋂n∈ZAn\bigcap_{n \in \mathbf{Z}} A_n, is the closure of the ideal (0) (Chapter III, § 2, no. 5), the hypothesis implies first that the filtration (An)(A_n) is separated and, as gr(A)\mathrm{gr}(A) is an integral domain, so then is AA (Chapter III, § 2, no. 3, Corollary to Proposition 1). Let x=b/ax = b/a be an element of the field of fractions KK of AA (a∈A,b∈Aa \in A, b \in A) for which there exists an element d≠0d \neq 0 of AA such that dxn∈Adx^n \in A for all n≥0n \geq 0. We must prove that b∈Aab \in A a and, as by hypothesis the ideal AaA a is closed, it is sufficient to show that, for all n∈Z,b∈Aa+An \in \mathbf{Z}, b \in A a + A. As the filtration of AA is exhaustive, there exists an integer q∈Zq \in \mathbf{Z} such that b∈Aa+Aqb \in A a + A_q. It will therefore suffice to prove that the relation b∈Aa+Aqb \in A a + A_q implies b∈Aa+Aq+1b \in A a + A_{q+1}.

Suppose then that b=ay+zb = a y + z where y∈A,z∈Aq+1y \in A, z \in A_{q+1}. Then by hypothesis dxn∈Adx^n \in A for all n≥0n \geq 0, whence we obtain immediately d(x−y)n∈Ad(x - y)^n \in A for all n≥0n \geq 0; in other words, dzn=antndz^n = a^n t_n where tn∈At_n \in A for all n≥0n \geq 0. We can obviously limit our attention to the case where z≠0z \neq 0. Let vv denote the order function on AA (Chapter III, § 1, no. 2) and let us write v(d)=n1, v(z)=n2≥m, v(a)=n3v(d) = n_1,\ v(z) = n_2 \geq m,\ v(a) = n_3; let d′,z′,a′d', z', a' be the respective images of d,z,ad, z, a in An1/An1+1, An2/An2+1, An3/An3+1A_{n_1}/A_{n_1+1},\ A_{n_2}/A_{n_2+1},\ A_{n_3}/A_{n_3+1}. For all n≥0, v(dzn)=n1+nn2n \geq 0,\ v(dz^n) = n_1 + nn_2 (Chapter III, § 2, no. 3, Proposition 1) and hence the canonical image in gr(A)\mathrm{gr}(A) of dzndz^n is d′z′nd'{z'}^n; similarly it is seen that the canonical image in gr(A)\mathrm{gr}(A) of antna^n t_n is of the form d′ntn′{d'}^n t'_n where tn′∈gr(A)t'_n \in \mathrm{gr}(A) and, as a′≠0a' \neq 0 we deduce that, for all n≥0, d′(z′/a′)n∈gr(A)n \geq 0,\ d'(z'/a')^n \in \mathrm{gr}(A). The hypothesis that gr(A)\mathrm{gr}(A) is completely integrally closed therefore implies the existence of an s′∈gr(A)s' \in \mathrm{gr}(A) such that z′=a′s′z' = a's'; decomposing s′s' into a sum of homogeneous elements, it is further seen (since z′z' and a′a' are homogeneous) that s′s' may be assumed to be homogeneous, that is the image of an element s∈As \in A; then v(as)=v(z)=n2v(as) = v(z) = n_2 and z≡as(modAn2+1)z \equiv as \pmod{A_{n_2+1}}; as n2≥m,afortiorin_2 \geq m, a fortiori z≡as(modAm+1)z \equiv as \pmod{A_{m+1}}, hence b≡a(y+s)(modAm+1)b \equiv a(y+s) \pmod{A_{m+1}}.

Bourbaki puts the proof directly after the statement, so what is above is the statement and its proof together.

Cited by

Nothing.

Cites

Transcribed from the printed volume. Not translated.