5. Flatness properties of filtered modules
1. Ideally Hausdorff modules
Definition 1
Let A be a commutative ring and an ideal of A. An A-module M is called ideally Hausdorff with respect to 3 (or simply ideally Hausdorff if there is no ambiguity) if, for every finitely generated ideal of A, the A-module is Hausdorff with the 3-adic topology.
Putting in this definition, we have already seen that M is necessarily Hausdorff with the 3-adic topology.
Examples
Example 1
If A is Noetherian and is contained in the Jacobson radical of A (in other words if A is a Zariski ring with the 3-adic topology), every finitely generated A-module is ideally Hausdorff (§3, no. 3, Proposition 6).
Example 2
Every direct sum of ideally Hausdorff modules is an ideally Hausdorff module, by virtue of the relations
Example 3
If an A-module M is flat and Hausdorff with the -adic topology it is ideally Hausdorff, for is then identified with a submodule of M and the -adic topology on is finer than the topology induced on by the -adic topology on M, which is Hausdorff by hypothesis.
2. Statement of the flatness criterion
Let A be a ring, a two-sided ideal of A, M a left-module and gr(A) and gr(M) the graded ring and graded gr(A)-module associated respectively with the ring A and with the module M with the -adic filtrations (§ 2, no. 3). We have seen (loc. cit.) that for every integer there is a surjective -module homomorphism
Theorem 1
Let A be a commutative ring, an ideal of A and M an A-module. Consider the following properties: (i) M is afat A-module. (ii) for every A-module N annihilated by . (iii) is ajlat -module and the canonical mapping is bijective (the latter condition being equivalent to by virtue of the relation and the exact sequence
Remark
If is a field (as often happens in applications) the condition "M/ is a flat -module" holds automatically for every A-module M, which simplifies the statement of properties (iii) and (iv); moreover, in this case, property (v) is equivalent to saying that is a free -module for every integer (Chapter 11, § 3, no. 2, Corollary 2 to Proposition 5).
3. Proof of the flatness criterion
(A) The implications (i) (ii) (iii)
The implication (i) (ii) is immediate (Chapter I, § 4). The equivalence (ii) (iii) is a special case of Chapter I, § 4, Proposition 2 applied to , taking account of the fact that being given an -module structure on is equivalent to being given an -module structure under which is annihilated by 3.
Remark (1). Condition (ii) is also equivalent to the following:
(ii') for every *-module* annihilated by a power of 3.
Clearly (ii') implies (ii). Conversely, if (ii) holds, then in particular for all ; from the exact sequence
It follows from this that if 9 is nilpotent, (ii) implies (i), for (ii') then means that for every *-module* and hence that is flat (Chapter I, § 4).
(B) Let us prove the following proposition:
Proposition 1
Let* be a commutative ring, 3 an ideal of and an *-module*. The following conditions are equivalent: (a) For all , . (b) For all , the canonical homomorphism
Moreover these conditions imply : (c) The canonical homomorphism is bijective. Conversely, if is nilpotent, (c) implies (a) and (b).
The equivalence of (a) and (b) follows from the exact sequence
Consider next the diagram
where we note that is canonically identified with . This diagram is commutative by definition of and its rows are exact. If (b) holds, , and are bijective and so therefore is by definition of cokernel, hence (b) implies (c). Conversely, assuming that is nilpotent, let us show that (c) implies (b); we shall argue by descending induction on , since for sufficiently large. Suppose then that in diagram (1), and are bijective; then so is by virtue of Chapter I, § 1, no. 4, Corollary 1 to Proposition 2.
(C) The implication (ii) (iv)
If (ii) holds, so does (ii') by Remark 1; Proposition 1 then shows that is an isomorphism. On the other hand, we already know that (ii) implies (iii) and hence is a flat -module, which completes the proof that (ii) implies (iv).
Remark (2). Proposition 1 shows that, if is nilpotent, (iv) implies (iii); taking account of Remark 1, we have therefore proved in this case that (i), (ii), (iii) and (iv) are equivalent.
(D) The equivalence (iv) (v)
For all , has a canonical -module structure. If it is filtered by the -adic filtration, it is immediate that if and if . For all , let , , ; let (iv), (resp. ) denote the assertion derived from (iv) (resp. (v)) by replacing by . It follows from what has just been said that (iv) is equivalent to "for all , " and obviously (v) is equivalent to "for all , ". Then it will suffice to establish the equivalence for all or also to show that (iv) (v) when is nilpotent. Now (Remark 2) we have seen that in that case (iv) is equivalent to (i). As is isomorphic to , (i) implies (v) (Chapter I, § 2, no. 7, Corollary 2 to Proposition 8); moreover clearly (v) then implies (i). We have therefore shown the equivalence (iv) (v) in all cases and also that of all the properties of the theorem in the case where is nilpotent.
(E) The implication (v) (i) when is Noetherian and ideally Hausdorff
It is sufficient to prove that for every ideal of the canonical mapping is injective (Chapter I, § 2, no. 3, Proposition 1). Let ; is Hausdorff with the -adic topology, it suffices to verify that, for every integer , . Let be the canonical injection; it suffices to show that ; for if , and , the image under , of the element of is the element of and hence . By virtue of Krull's Theorem (§ 3, no. 2, Theorem 2), there exists an integer such that ; if is the canonical injection, it will then be sufficient to show that . Now, denoting by and the canonical mappings, there is a commutative diagram
in which the first row is exact. It suffices to prove that and, as by hypothesis, it will suffice to verify that the mapping , is injective. Now, it may also be written (Algebra, Chapter II, § 3, no. 6, Corollary 3 to Proposition 6)
and, as is injective and, by (v), is a flat -module, this completes the proof.
4. Applications
Proposition 2
Let be a commutative ring, an ideal of and a commutative Noetherian -algebra such that is contained in the Jacobson radical of . Then every finitely generated -module is an ideally Hausdorff -module with respect to .
We shall see more generally that for every finitely generated -module , is Hausdorff with the -adic topology. For is a finitely generated -module and the -module is canonically identified with by virtue of the associativity of the tensor product. Let be the Jacobson radical of ; as is contained in , the -adic topology on is therefore identified with a finer topology than the -adic topology on ; but this latter topology is Hausdorff since is a finitely generated -module (no. 1, Example 1), whence the conclusion.
Proposition 3
Let be a commutative ring, a commutative -algebra, an ideal of and a -module. Suppose that is a Noetherian ring and a flat -module and that is ideally Hausdorff with respect to . The following conditions are equivalent: (a) is a flat -module. (b) is a flat -module and is a flat -module. If further the canonical homomorphism is bijective, conditions (a) and (b) are also equivalent to: (c) is a flat -module.
Condition (a) implies (b) by Chapter I, §2, no. 7, Corollaries 2 and 3 to Proposition 8 and the fact that is isomorphic to . Suppose condition (b) holds; to show that is a flat -module, we shall apply Theorem 1 of no. 2 with replaced by and by . It will therefore be sufficient to show that the canonical mapping is injective. Let be the canonical mapping and the canonical isomorphism is the canonical mapping , as is easily verified. Now is an isomorphism since is a flat -module, whilst is an isomorphism because is flat over is then an isomorphism.
Let be the canonical homomorphism; the -module structure on derived by means of from its -module structure is isomorphic to that on . Then it follows that, if is a flat -module, is a flat -module and hence also a flat -module if is an isomorphism; we have thus proved that (c) (b) in that case.
Corollary
Let be a commutative Noetherian ring, an ideal of , the Hausdorff completion of with respect to the -adic topology and an ideally Hausdorff -module with respect to . For to be a flat -module, it is necessary and sufficient that be a flat -module.
We know in fact that is a Noetherian ring (, no. 4, Proposition 8) and a flat -module (, no. 4, Theorem 3), that (, no. 12, Proposition 16) and that the canonical homomorphism is bijective (, no. 12, Proposition 15); Proposition 3 can therefore be applied.
Proposition 4
Let and be two commutative Noetherian rings, a ring homomorphism, an ideal of and an ideal of containing and contained in the Jacobson radical of . Let be the Hausdorff completion of with respect to the -adic topology and the Hausdorff completion of with respect to the -adic topology; is continuous with these topologies and therefore makes into an -algebra. Let be a finitely generated -module and its Hausdorff completion with respect to the -adic topology; the following properties are equivalent: (a) is a flat -module. (b) is a flat -module. (c) is a flat -module.
As with the -adic topology is a Zariski ring, is a faithfully flat -module (, no. 5, Proposition 9) and is canonically isomorphic to (, no. 4, Theorem 3); it is immediately verified that this canonical isomorphism is an isomorphism of the -module structure on onto the -module structure on derived from that on . Applying Proposition 4 of Chapter I, , no. 2 with replaced by , by , by , by , we see that for to be a flat -module, it is necessary and sufficient that be a flat -module. Moreover, is a finitely generated -module and is contained in and hence in the Jacobson radical of (, no. 4, Proposition 8); therefore is an ideally Hausdorff -module with respect to (Proposition 2). Conditions (b) and (c) are therefore equivalent by the Corollary to Proposition 3.
Exercises
See the exercises for § 5.