Bourbaki

Theorem 2

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

Commutative Algebra, chapter V, § 3. Finitely generated algebras over a field

Let A be a finitely generated integral k-algebra, K its field of fractions and A′A' the integral closure of A in a field K' which is a finite algebraic extension of K. Then A′A' is a finitely generated A-module and a finitely generated k-algebra.

By Theorem 1 there exists a subalgebra C of A isomorphic to a polynomial algebra k[X1,…,Xn]k[X_1, \ldots, X_n] and such that A is integral over C; A′A' is obviously the integral closure of C in K' (§ 1, no. 1, Proposition 6); we may therefore confine our attention to the case where A=k[X1,…,Xn]A = k[X_1, \ldots, X_n]. Let NN be the quasi-Galois extension of KK (in an algebraic closure of KK) generated by K′K', which is a finite algebraic extension of KK (Algebra, Chapter V, § 6, no. 3, Corollary 1 to Proposition 9). It will suffice to prove that the integral closure BB of AA in NN is a finitely generated AA-module, for A′A' is a sub-AA-module of BB and AA is a Noetherian ring (Chapter III, § 2, no. 10, Corollary 2 to Theorem 2). We may therefore confine our attention to the case where K′K' is a quasi-Galois extension of KK. Then we know (Algebra, Chapter V, § 10, no. 9, Proposition 14) that K′K' is a (finite) Galois extension of a (finite) radicial extension K′′K'' of KK. If A′′A'' is the integral closure of AA in K′′K'', A′A' is the integral closure of A′′A'' in K′K' and it will suffice to prove that A′′A'' is a finitely generated AA-module and A′A' is a finitely generated A′′A''-module. Now, if it has been proved that A′′A'' is a finitely generated AA-module, it is a Noetherian domain, integrally closed by definition; the fact that AA is a finitely generated A′′A''-module will follow from § 1, no. 6, Corollary 1 to Proposition 18.

We see therefore that we may confine our attention to the case where A=k[X1,…,Xn]A = k[X_1, \ldots, X_n] and where K′K' is a finite radicial extension of K=k(X1,…,Xr)K = k(X_1, \ldots, X_r). Then K′K' is generated by a finite family of elements (yi)1≤i≤m(y_i)_{1 \leq i \leq m} and there exists a power qq of the characteristic exponent of kk such that yiq∈k(X1,…,Xr)y_i^q \in k(X_1, \ldots, X_r). Let cjc_j (1≤j≤r1 \leq j \leq r) be the coefficients of the numerators and denominators of the rational functions in X1,…,XrX_1, \ldots, X_r, equal to yiqy_i^q (1≤i≤m1 \leq i \leq m). Then K′K' is contained in the extension L=k′(X1q−1,…,Xnq−1)L = k'(X_1^{q^{-1}}, \ldots, X_n^{q^{-1}}), where k′=k(c1q−1,…,crq−1)k' = k(c_1^{q^{-1}}, \ldots, c_r^{q^{-1}}) (we are in an algebraic closure of K′K') and AA is contained in the algebraic closure B′B' of AA in L′L'. Now, k′k' is algebraic over kk and hence C′=k′[X1,…,Xn]C' = k'[X_1, \ldots, X_n] is integral over AA (§1§ 1, no. 1, Proposition 5); as k′[X1q−1,…,Xnq−1]k'[X_1^{q^{-1}}, \ldots, X_n^{q^{-1}}] is integrally closed (§1§ 1, no. 3, Corollary 2 to Proposition 13), it is seen that this ring is the integral closure of C′C' in L′L' and hence also that of AA (§1§ 1, no. 1, Proposition 6), in other words B′=k′[X1q−1,…,Xnq−1]B' = k'[X_1^{q^{-1}}, \ldots, X_n^{q^{-1}}]. Now clearly B′B' is a finitely generated C′C'-module (§1§ 1, no. 1, Proposition 4) and, as k′k' is a finite extension of kk, C′C' is a finitely generated AA-module and hence B′B' is a finitely generated AA-module; since AA is Noetherian and A⊂B′A \subset B', A′A' is a finitely generated AA-module.

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

Cited by

Cites

Transcribed from the printed volume. Not translated.