Bourbaki

Proposition 8

Tag 0D8C, permanent. Cite it as [Bourbaki, Tag 0D8C].

Commutative Algebra, chapter X, § 6. Extension des scalaires dans les algèbres régulières

Let k be a field and A a k-algebra essentially of finite type. Let I denote the kernel of the homomorphism of k-algebras μ:A⊗kA→A\mu : A \otimes_k A \to A such that μ(x⊗y)=xy\mu(x \otimes y) = xy for x and y in A. The following conditions are equivalent:

(i) the k-algebra A is absolutely regular; (ii) for every regular k-algebra C, the ring A⊗kCA \otimes_k C is regular; (iii) the ring A⊗kAA \otimes_k A is regular; (iv) the ideal I of A⊗kAA \otimes_k A is completely secant (§5§ 5, No. 1, Definition 1).

Let B denote the ring A⊗kAA \otimes_k A and provide it with the structure of an A-algebra deduced from the homomorphism ρ:A→A⊗kA\rho : A \to A \otimes_k A such that ρ(x)=x⊗1\rho(x) = x \otimes 1; then μ\mu is a homomorphism of A-algebras, and induces by passing to the quotient an isomorphism of B/IB/I onto A.

(i) ⇒\Rightarrow (ii): this results from Prop. 7. (ii) ⇒\Rightarrow (iii): it is enough to apply (ii) with C=kC = k, then with C=AC = A. (iii) ⇒\Rightarrow (iv): the A-module B is free, hence faithfully flat. If the ring B is regular, A is regular (§4§ 4, No. 5, Prop. 8); the ideal I is then completely secant (§5§ 5, No. 3, Prop. 2). (iv) ⇒\Rightarrow (i): let us suppose the ideal I completely secant and first prove that A is regular. Let m be a maximal ideal of A and let ν:(A/m)⊗kA→A/m\nu : (A/m) \otimes_k A \to A/m be the homomorphism deduced from μ\mu. The maximal ideal n=Ker νn = \mathrm{Ker}\,\nu is equal to 1((A/m)⊗kA)1((A/m) \otimes_k A); applying Prop. 6 of §5§ 5, No. 6 to the A-algebra A′=A/mA' = A/m, one sees that the ideal n is completely secant in (A/m)⊗kA(A/m) \otimes_k A. Consequently (§5§ 5, No. 3, Prop. 3) the local ring ((A/m)⊗kA)n((A/m) \otimes_k A)_n is regular. Let us denote by j:A→(A/m)⊗kAj : A \to (A/m) \otimes_k A the homomorphism x↦1⊗xx \mapsto 1 \otimes x; since ν∘j\nu \circ j is the canonical homomorphism of A into A/mA/m, one has j−1(n)=mj^{-1}(n) = m. Thus j extends to a local homomorphism of local rings from AmA_m into ((A/m)⊗kA)n((A/m) \otimes_k A)_n, which makes the latter into a faithfully flat AmA_m-module. By Prop. 8 of §4§ 4, No. 5, the ring AmA_m is therefore regular. We have thus proved that A is regular.

Now let k′k' be an extension of k. The kernel of the mapping μ′:A(k′)⊗k′A(k′)→A(k′)\mu' : A_{(k')} \otimes_{k'} A_{(k')} \to A_{(k')} deduced from the multiplication of A(k′)A_{(k')} is none other than IA(k′)IA_{(k')}; it is therefore completely secant in A(k′)A_{(k')} (§5§ 5, No. 6, Prop. 6). The k′k'-algebra A(k′)A_{(k')} therefore satisfies condition (iv); by what has just been seen, it is regular, and this proves (i).

Recall (A, III, p. 133-134) that the quotient I/I2I/I^2 endowed with the A-module structure deduced from ρ\rho is denoted by Ωk(A)\Omega_k(A) and called the module of k-differentials of A. When the k-algebra A is essentially of finite type, the ring A⊗kAA \otimes_k A is noetherian, so that the A-module Ωk(A)\Omega_k(A) is finitely generated.

One denotes by dA/kd_{A/k}, or simply by d, the k-linear mapping from A into Ωk(A)\Omega_k(A) which associates with an element x of A the class of x⊗1−1⊗xx \otimes 1 - 1 \otimes x in Ωk(A)\Omega_k(A). The mapping d is a k-derivation; for every A-module M and every k-derivation D : A →\to M, there exists a unique A-linear mapping g:Ωk(A)→Mg : \Omega_k(A) \to M such that D=g∘dD = g \circ d (loc. cit., Prop. 18).

If S is a multiplicative subset of A, the canonical S−1AS^{-1}A-linear mapping (loc. cit., p. 136)

S−1Ωk(A)→Ωk(S−1A) S^{-1}\Omega_k(A) \to \Omega_k(S^{-1}A)
is bijective: in fact, it is enough to verify that, for every S−1AS^{-1}A-module M, every k-derivation D : A \to M extends in a unique way to a k-derivation D : S^{-1}A \to M; but this results from the reasoning of loc. cit., p. 123, Prop. 5.

Let k be a field and A a local k-algebra essentially of finite type. Put n = \dim(A) + \deg.\mathrm{tr}_k(\kappa_A). Let B be a k-algebra of finite type and q a prime ideal of B such that the k-algebra A is isomorphic to BqB_q (cf. No. 1, Example 2); one has n = \dim_q(B) (VIII, § 2, No. 4, Cor. 5 of Th. 3). By loc. cit., Th. 3, c) and Cor. 5, one also has

n=sup⁡deg⁡.trk(κ(p)) n = \sup \deg.\mathrm{tr}_k(\kappa(\mathfrak{p}))
where \mathfrak{p} runs through the (finite) family of minimal prime ideals of A. If A is an integral domain, n is therefore the transcendence degree of the field of fractions of A.

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

Cited by

Nothing.

Cites

Nothing.

Translated with machine help, not yet reviewed by a person.