2. Primary decomposition
1. Primary s bmodules
Proposition 1
Let be a Noetherian ring and an -module. The following conditions are equivalent: (a) is reduced to a single element.
(b) and every homothety of is either injective or almost nilpotent (, no. 4). If these conditions are fulfilled and is the set of such that the homothety is almost nilpotent, then .
This follows immediately from , no. 4, Corollary to Proposition 9 and no. 1, Corollary 2 to Proposition 2.
Definition 1
Let be a Noetherian ring, an -module and a submodule of . If the module satisfies the conditions of Proposition 1, is called p-primary with respect to (or in ).
When there is no ambiguity, we shall simply say that is "p-primary" or "primary"; clearly for every submodule of containing , is p-primary in .
Definition 1 applies in particular to the case ; the submodules of are then the ideals of and hence an ideal of is called primary if has a single element or, what amounts to the same, if and every divisor of zero in the ring is nilpotent. If is p-primary, it follows from Definition 1 that is the radical (Chapter II, , no. 6) of the ideal .
Remark
Let be a p-primary submodule of an -module . If is finitely generated, there exists an integer such that by , no. 4, Proposition 9.
Examples
Example 1
If is a prime ideal of , is p-primary (, no. 1, Proposition 1).
Example 2
Let be an ideal of such that there exists a single prime ideal (necessarily maximal) containing ; then, if is an -module such that , is m-primary with respect to . For every element of contains , hence is equal to and (, no. 1, Corollary 1 to Proposition 2). In particular is an m-primary ideal in .
Example 3
Let be a maximal ideal of ; the m-primary ideals are then the ideals of for which there exists an integer such that . For if , is the only prime ideal containing (Chapter 11, , no. 1, Corollary to Proposition 1) and the conclusion follows from Example 2; conversely, if is m-primary, is the radical of and there therefore exists such that (Chapter II, , no. 6, Proposition 15).
Example 4
In a principal ideal domain , the primary ideals are and the ideals of the form , where is an extremal element and ; this follows immediately from Example 3.
Example 5
The powers of any prime ideal are not necessarily primary ideals (Exercise 1). On the other hand, there exist primary ideals which are not powers of prime ideals (Exercise 1).
Proposition 2
Let M be a module over a Noetherian ring A, p a prime ideal of A and a non-empty finite family of submodules of M which are p-primary with respect to M. Then is p-primary with respect to M.
is isomorphic to a submodule of the direct sum . Now
(§ 1, no. 1, Corollary 1 to Proposition 3). Hence (§ 1, no. 1, Proposition 3 and Corollary 1 to Proposition 2).
Proposition 3
Let A be a Noetherian ring, S a multiplicative subset of A, p a prime ideal of A, M an A-module, N a submodule of M and the canonical mapping of M to .
(i) Suppose that . If N is p-primary with respect to M, then .
(ii) Suppose that . For N to be p-primary with respect to M, it is necessary and sufficient that N be of the form , where is a sub--module of which is -primary with respect to ; then .
(i) If and N is p-primary with respect to M, then
(§ 1, no. 2, Corollary to Proposition 5) and hence (§ 1, no. 1, Corollary 1 to Proposition 2), whence .
(ii) Suppose that . If N is p-primary with respect to M, then (§ 1, no. 2, Corollary to Proposition 5) and hence the submodule of is -primary; moreover, if and are such that , then , for the homothety with ratio s in M/N is injective, whence (Chapter II, § 2, no. 4, Proposition 10). Conversely, let be a submodule of which is -primary with respect to ; let us write ; then (Chapter II, § 2, no. 4, Proposition 10) and . As the canonical mapping is injective, no prime ideal of A associated with M/N meets S (§ 1, no. 2, Proposition 6); it follows that (§ 1, no. 2, Corollary to Proposition 5), so that N is p-primary with respect to M.
2. The existence of a primary decomposition
Definition 2
Let be a Noetherian ring, an -module and a submodule of . A finite family of submodules of which are primary with respect to and such that is called a primary decomposition of in .
Example
Let us take , , for some integer . If is the decomposition of into prime factors,
By an abuse of language, the relation is called a primary decomposition of in . It amounts to the same to say that is a primary decomposition of in . If is a primary decomposition of in , the canonical mapping from to is injective. Conversely let be a submodule of and an injective homomorphism from to a finite direct sum , where each set is reduced to a single element ; let be the homomorphism obtained by taking the composition off with the projection , and let be the kernel off.; then the distinct from are primary with respect to (no. 1, Definition 1) and . Moreover, by virtue of § 1, no. 1, Proposition 3.
Theorem 1
Let be a finitely generated module over a Noetherian ring and let be a submodule of . There exists a primary decomposition of of the form
We may replace by and therefore suppose that . By § 1, no. 4, Corollary to Theorem 2, is finite; by § 1, no. 1, Proposition 4, there exists, for each , a submodule of such that and . Let us write ; for all , and hence , which implies (§ 1, no. 1, Corollary 1 to Proposition 2) and therefore proves the theorem.
3. Uniqueness properties in the primary decomposition
Definition 3
Let be a module over a Noetherian ring and a submodule of . A primary decomposition of in is called reduced if the following conditions are fulfilled:
(a) there exists no index such that
(b) if , the (i ) are distinct.
From every primary decomposition of in a reduced primary decomposition of in can be deduced as follows: let be a minimal element of the set of subsets of such that . Clearly satisfies condition (a). Then let be the set of for ; for all , let be the set of such that and let ; it follows from Proposition 2 of no. 1 that is -primary with respect to ; further and the family is therefore a reduced primary decomposition of in .
We shall see that the primary decomposition defined in the proof of Theorem 1 of no. 2 is reduced; this follows from the following proposition:
Proposition 4
Let* *be a module over a Noetherian ring, N a submodule of M, a primary decomposition of N in M and, for all , let . For this decomposition to be reduced, it is necessary and sufficient that the be distinct and belong to ; *then
(2)
(3)
If the condition of the statement is fulfilled, cannot hold, for we would deduce that (§ 1, no. 1, Corollary 2 to Proposition 3) contrary to the hypothesis; the primary decomposition of is then certainly reduced. Conversely, always holds (§ 1, no. 1, Corollary 2 to Proposition 3); on the other hand, for all , let us write ; then and if is reduced, hence is non-zero and is isomorphic to the submodule of , whence (§ 1, no. 1, Proposition 3 and Corollary 1 to Proposition 2); as , which completes the proof of the necessity of the condition in the statement and formula (2). Finally, as (§ 1, no. 1, Corollary 2 to Proposition 3); but is isomorphic to the submodule of , hence and
Corollary
Let be a Noetherian ring, an -module, a submodule of and a primary decomposition of in . Then ; for to be a reduced primary decomposition, it is necessary and sufficient that .
It follows from the remarks preceding Proposition 4 that there exists a reduced primary decomposition of in such that ; the first assertion then follows from the second and the latter is a consequence of Proposition 4.
Proposition 5
Let be a Noetherian ring, an -module, a submodule of , a reduced primary decomposition of in and, for all , let . If is a minimal element of , is equal to the saturation on with respect to (Chapter 11, § 2, no. 4) (cf. Exercise 2).
We can obviously restrict our attention to the case where , replacing if need be by . If is minimal in , the set of elements of which do not meet reduces to ; the proposition then follows from formula (3) above and § 1, no. 2, Proposition 6, the kernel of the canonical mapping , being equal to the saturation of 0 with respect to (Chapter 11, § 2, no. 4).
Remark
The prime ideals which are not minimal elements of this set are sometimes called the immersed prime ideals associated with ; if is finitely generated, for to be immersed, it is necessary and sufficient that be not an irreducible component of (Chapter II, § 4, no. 3, Corollary 2 to Proposition 14); if and are two reduced primary decompositions of in , it may be that (Exercise 24 (c)); a canonical reduced primary decomposition of in may always be defined by imposing supplementary condition on the primary submodules which appear in it (Exercise 4).
4. The localization of a primary decomposition
Given a submodule of a module over a Noetherian ring , to simplify we shall denote by , in this no., the set of reduced primary decompositions of in whose indexing set is (equipotent to ).
Proposition 6
Let be a Noetherian ring, an -module, a submodule of and . Let be a multiplicative subset of and the subset of consisting of the indices i such that . Let be the saturation of with respect to in . Then:
(i) If is an element of , the family is an element of and the family is an element of .
(ii) The mapping is a bijection of onto .
(iii) If is an element of and an element of , the family such that for and for is an element of .
(i) We know (no. 1, Proposition 3) that for is primary for and that for ; as (Chapter II, § 2, no. 4), then also . The for are distinct and their set is (, no. 2, Corollary to Proposition 5); then (Proposition 4) is a reduced primary decomposition of . Moreover, (no. 1 Proposition 3), hence and is obviously a reduced primary decomposition of in .
(ii) As , we may replace by , that is suppose that . Let be a reduced primary decomposition of in and let us write ; it follows from no. 1, Proposition 3 that is primary for () and is then a reduced primary decomposition of in by virtue of no. 3, Corollary to Proposition 4. Finally, as, for all and every submodule of which is -primary with respect to by virtue of no. 1, Proposition 3 and the hypothesis , we see that two mappings and have been defined whose compositions are the identities on and , which proves (ii).
(iii) By virtue of (i), , whence
and it follows immediately from no. 3, Corollary to Proposition 4 that this primary decomposition is reduced.
Corollary
The mappings
defined in Proposition 6 (i) are surjective.
Proposition 6 (iii) shows that the mapping is surjective and Proposition 6 (ii) then shows that the mapping
5. Rings and modules of finite length
If an -module is of finite length, we shall denote this length by or . Recall that every Artinian ring is Noetherian (Algebra, Chapter VIII, § 6, no. 5, Corollary 3 to Proposition 12) and that every finitely generated module over an Artinian ring is of finite length (loc. cit., Corollary 1 to Proposition 12). Moreover, every Artinian integral domain is a field (Algebra, Chapter VIII, § 6, no. 4, Proposition 9).
Proposition 7
Let be a finitely generated module over a Noetherian ring . The following properties are equivalent:* (a) is of finite length. (b) Every ideal is a maximal ideal of . (c) *Every ideal is a maximal ideal of .
Let be a composition series of such that, for , is isomorphic to , where is prime (, no. 4, Theorem 1). If is of finite length, so is each of the -modules , which implies that each of the rings is Artinian; but as is an integral domain, it is therefore a field, in other words is maximal; we conclude that (a) implies (b) (, no. 4, Theorem 2). Condition (b) implies (c) by , no. 3, Proposition 7. Finally, if all the ideals of are maximal, so are the (, no. 4, Theorem 2), hence the are simple -modules and is of finite length, which completes the proof.
Corollary 1
For every module of finite length over a Noetherian Zing , .
Every element of is then minimal in and the conclusion follows from , no. 3, Corollary 1 to Proposition 7.
Corollary 2
Let be a finitely generated module over a Noetherian ring and a prime ideal of . For to be a non-zero -module of finite length, it is necessary and sufficient that be a minimal element of .
By , no. 2, Corollary to Proposition 5, is the set of ideals , where runs through the set of ideals of which are contained in . On the other hand, is the unique maximal ideal of ; by Proposition 7, for to be an -module of finite length, it is necessary and sufficient that no element of be strictly contained in . On the other hand, for , it is necessary and sufficient by definition that (Chapter 11, § 4, no. 4), that is that contain an element of (, no. 3, Proposition 7). This proves the corollary.
Remark (1). Let be a finitely generated module over a Noetherian ring ; let be a composition series of such that, for , is isomorphic to , where is a prime ideal of (, no. 4, Theorem 1). If is a minimal element of , the length is equal to the number of indices such that . For the form a composition series of , and is isomorphic to and hence to if (since is minimal in the set of by , no. 4, Theorem 2) and to which is a field, if .
Proposition 8
Let be a module of finite length over a Noetherian ring .
(i) There only exists a single primary decomposition of with respect to indexed by (necessarily reduced) — let be this decomposition, where is -primary with respect to .
(ii) There exists an integer such that, for all and all , .
(iii) For all , the canonical mapping of to is surjective and its kernel is .
(iv) The canonical injection of into is bijective.
As every element is minimal in (Proposition 7), assertion (i) follows from no. 3, Proposition 5. As is finitely generated, there exists such that for all and all (no. 1, Remark); but as is a maximal ideal, is -primary with respect to (no. 1, Examples 2 and 3) and, as , it follows from (i) that necessarily for all ; whence (ii). As the , for , are relatively prime in pairs (Chapter 11, § 1, no. 2, Proposition 3), the canonical mapping is surjective (Chapter 11, § 1, no. 2, Proposition 6), whence (iv). Then and (no. 3, Proposition 4); as the elements of are maximal ideals, is the only element of which does not meet ; is therefore the kernel of the canonical mapping (, no. 2, Proposition 6). If , the homothety of with ratio is injective by virtue of the relation (no. 1, Proposition 1); since is Artinian, this homothety is bijective (Algebra, Chapter VIII, § 1, no. 2, Lemma 3). The canonical mapping is then written , where is an -homomorphism (Chapter II, § 2, no. 2, Proposition 3); as , is injective; we conclude that is surjective and bijective.
Corollary
If M is a module of finite length over a Noetherian ring A, then
This will follow from Proposition 8 (iv) if we prove that
Now, it follows from Proposition 1 of no. 1 that for all the homothety with ratio s on is injective; the homothety with ratio s on every submodule R of is therefore injective and, as R is Artinian, it is bijective (Algebra, Chapter VIII, § 2, no. 2, Lemma 3); we conclude that the sub-A-modules of are the images under the bijection of the sub-A,-modules of M, (Chapter 11, § 2, no. 3), whence our assertion.
Proposition 9
Let A be a Noetherian ring. The following conditions are equivalent :* (a) A is Artinian. (b) All the prime ideals in A are maximal ideals. (c) *All the elements of Ass(A) are maximal ideals.
If these conditions are fulfilled, A has only a finite number of prime ideals, which are all maximal and associated with the A-module A; further, A is a semi-local ring and its Jacobson radical is nilpotent.
To say that A is Artinian is equivalent to saying that A is an A-module of finite length; hence (a) and (c) are equivalent by Proposition 7. Clearly (b) implies (c). Finally, (a) implies (b) since every Artinian integral domain is a field. The properties (a), (b) and (c) are therefore equivalent.
Suppose they hold. As every prime ideal of A belongs to Supp(A) and every element of Supp(A) contains an element of Ass(A) ( 1, no. 3, Proposition 7), it follows from (c) that Ass(A) is the set of all prime ideals of A; then A has only a finite number of prime ideals, all of them maximal and associated with the A-module A. This obviously implies that A is semi-local; finally, we know that the Jacobson radical of an Artinian ring is nilpotent (Algebra, Chapter VIII, § 6, no. 4, Theorem 3).
Remark (2) The conditions of Proposition 9 for a Noetherian ring A imply that the spectrum of A is finite and discrete, every point of Spec(A) being therefore closed (Chapter 11, § 4, no. 3, Corollary 6 to Proposition 11). Conversely, for a Noetherian ring A, to say that every point of Spec(A) is closed means that every prime ideal of A is maximal (loc. cit..) and hence this condition is equivalent to that of Proposition 9.
Corollary 1
Every Artinian ring A is isomorphic to the direct composition of a finite family of Artinian local rings.
It follows from Proposition 9 and Proposition 8 (iii) and (iv) that, if is the family of maximal ideals of , the canonical mapping is bijective.
Remark (3). This corollary can also be deduced from the fact that is finite and discrete and Chapter 11, § 4, no. 3, Proposition 15.
Corollary 2
Let be a Noetherian ring and an ideal of . The following conditions are equivalent:
(a) is a semi-local ring and is a defining ideal of . (b) is a Zariski ring with the -adic topology and is Artinian.
If (a) holds, is a Zariski ring with the -adic topology (Chapter III, § 3, no. 3, Example 3); further, as by hypothesis contains a power of the Jacobson radical of , every prime ideal of which contains also contains (Chapter II, § 1, no. 1, Proposition 1); it is therefore maximal, since is a finite intersection of maximal ideals (\emph{loc. cit.}); Proposition 9 then shows that is Artinian. Conversely, if (b) holds, every maximal ideal of contains the Jacobson radical of and hence contains (Chapter III, § 3, no. 3, Proposition 6); as is Artinian, the ideals are finite in number (Proposition 9) and hence has only a finite number of maximal ideals, which implies that it is semi-local.
Corollary 3
Let be two rings and a homomorphism from to . Suppose that is semi-local and Noetherian and that is a finitely generated -module. Then the ring is semi-local and Noetherian; if is a defining ideal of , is a defining ideal of .
We know that is a Zariski ring with the -adic topology (Chapter III, § 3, no. 3, Proposition 7). As is Artinian (Corollary 2) and is a finitely generated -module, is an Artinian ring, hence is semi-local and is a defining ideal of (Corollary 2).
Corollary 4
Let be a complete semi-local Noetherian ring, a defining ideal of , a finitely generated -module and a decreasing sequence of submodules of such that . Then, for all , there exists such that .
As is a Zariski ring, is Hausdorff and the are closed under the -adic topology. On the other hand, is complete (Chapter III, § 2, no. 12, Corollary 1 to Proposition 16). Finally, is a finitely generated module over the ring , which is Artinian (Corollary 2); then is an Artinian -module and hence an Artinian -module. The corollary then follows from Chapter III, § 2, no. 7, Proposition 8.
Corollary 5
In a complete semi-local Noetherian ring every decreasing sequence ideals whose intersection is 0 is a filter base which converges to 0.
It is sufficient to apply Corollary 4 to the A-module .
Proposition 10
Let be a Noetherian ring and the prime ideals associated with the A-module , where for .
(i) The set is the set of elements which are not divisors of 0 in .
(ii) If all the are minimal elements of , the total ring of fractions of is Artinian.
(iii) If the ring is reduced, all the are minimal elements of (and therefore are the minimal elements of ) and each of the is a field; for each index , the canonical homomorphism (Chapter II, § 2, no. 1, Corollary 1 to Proposition 2) is surjective and its kernel is ; finally the canonical homomorphism from to is bijective.
The fact that is the set of elements which are not divisors of 0 in has already been seen (§ 1, no. 1, Corollary 3 to Proposition 2). The prime ideals of are of the form , where is a prime ideal of contained in (Chapter 11, § 2, no. 5, Proposition 10) that is contained in one of the (Chapter II, § 1, no. 1, Proposition 2). If is a minimal element of , it is a minimal element of (§ 1, no. 3, Corollary to Proposition 7); if each of the is a minimal element of , we then see that the prime ideals of are the and they are therefore all maximal, which proves that is Artinian (Proposition 9).
Suppose finally that the ring is reduced. Then (§ 1, no. 3, Corollary 2 to Proposition 7). We deduce that is a reduced primary decomposition of the ideal (no. 3, Corollary to Proposition 4); in particular, none of the can contain a of index and therefore the are all minimal elements of . The ring is then Artinian by (ii). The are prime ideals associated with the -module (§ 1, no. 2, Corollary to Proposition 5) and (Chapter II, § 2, no. 4); as the are distinct, is a reduced primary decomposition of in (no. 3, Corollary to Proposition 4). Proposition 8 then shows that the canonical homomorphismg,: is surjective and has kernel and the canonical homomorphism is bijective. We know moreover that the canonical homomorphism , is composed of g, and an isomorphism , (Chapter II, § 2, no. 3, Proposition 7). Finally, it follows from Proposition 8 that is isomorphic to and hence is a field since is a maximal ideal.
6. Primary decomposition and extension of scalars
In this no., A and B will denote two rings and we shall consider a ring homomorphism which makes into an A-algebra; recall that, for every B-module F, is the commutative group F with the A-module structure defined by for all .
Lemma 1
Let A be a Noetherian ring, a prime ideal of A, E an A-module whose annihilator contains a power of and such that and F a B-module such that is a flat A-module. The condition then implies .
If n is such that , then , whence , which implies and therefore since is prime. Moreover, if , the homothety h with ratio a on E is injective (, no. 1, Corollary 2 to Proposition 2); as is the homothety with ratio on and is flat, is injective (Chapter I, § 2, no. 2, Definition 1); this proves that , whence .
Theorem 2
Let be a ring homomorphism, E an A-module and F a B-module such that is a flat A-module. Then
When A is Noetherian, the two sides of (5) are equal.
Let ; by definition there exists an exact sequence
Since F is a flat A-module we derive an exact sequence
whence , which proves the inclusion (5).
Suppose now that A is Noetherian and let us prove the opposite inclusion.
We proceed in stages:
(i) Suppose first that E is a finitely generated A-module and that Ass_A(E) is reduced to a single element . By § 1, no. 4, Theorem 1 there exists a composition series of E such that is isomorphic to , where is a prime ideal of A; moreover (, no. 4, Theorem 2 and no. 3, Proposition 7) all the contain . As F is a flat A-module, the form a composition series of and is identified with
(ii) Suppose only that E is a finitely generated A-module. Let () be the elements of and let be a corresponding reduced primary decomposition (no. 3); E is then isomorphic to a submodule of the direct sum of the and, as F is a flat A-module, is isomorphic to a submodule of the direct sum of B-modules . We deduce (, no. 1, Proposition 3 and Corollary 1 to Proposition 3) that
(iii) General case. The B-module is the union of the submodules , where runs through the set of finitely generated submodules of the A-module E. If belongs to , then there exists a finitely generated submodule of E such that . By (ii), there exists such that ; as , this completes the proof of Theorem 2.
Corollary 1
If A is Noetherian and , then and is the only prime ideal of A such that .
Corollary 2
Suppose that and are Noetherian and that is a flat -module. Let be a prime ideal of , a -primary submodule and a prime ideal of . For to be a -primary submodule of , it is necessary and sufficient that be a 'p-primary ideal' of .
Let us apply Theorem 2 to the -module and the -module ; then and is isomorphic to and hence . To say that is -primary in therefore means that is reduced to , whence the corollary.
Remark
Suppose that and are Noetherian. Let be a prime ideal of and let ; let us write and let be the field of fractions of . Since contains , is a prime ideal of . If is the composite homomorphism , we know that is identified with the ring and with an ideal of this ring (Chapter II, § 2, no. 2, Proposition 6); as does not meet , is a prime ideal of (Chapter 11, § 2, no. 5, Proposition 11); moreover there are canonical isomorphisms between , and ; similarly is canonically identified with . This being so, under the hypotheses of Theorem 2, in order that , it is necessary and sufficient that and
Proposition 11
Suppose that and are Noetherian and that is a flat -module. Let be an -module and a submodule of such that, for every ideal , is a prime ideal of or equal to . Let be a reduced primary decomposition of in , being -primary for all . (i) If and , then . (ii) If and is prime, is *-primary* in . (iii) If is the set such that is prime, then
If , Theorem 2 applied to and shows that
Corollary
Suppose that* is prime for all . Then, if are the minimal elements of , the are minimal elements of
Examples (1) Let us take , where is a multiplicative subset of ; if is Noetherian, the hypotheses of Proposition 11 are satisfied and we recover a part of Proposition 6 of no. 4. (2) Let be a Noetherian ring, an ideal of and the Hausdorff completion of with respect to the -adic topology; then is a flat -module and Theorem 2 may be applied with ; but in general the hypotheses of Proposition 11 are not satisfied for the prime ideals of (Chapter III, § 2, Exercise 15 (b)). (3) Let be a Noetherian ring and the polynomial algebra ; is Noetherian and is a free -module and therefore flat. Also, if is a prime ideal of , is isomorphic to , which is an integral domain, and hence is prime; the hypotheses of Proposition 11 are therefore satisfied for every -module and every submodule of . (4) Let be a finitely generated algebra over a field , an extension of and the algebra over obtained by extension of scalars; and are Noetherian and is a free -module and hence Theorem 2 may be applied to . In certain cases (for example if is algebraically closed) it can be shown that for every prime ideal of , is prime or equal to ; we shall return later to this example.
Exercises
See the exercises for § 2.