Communications in Algebra

ISSN: 0092-7872 (Print) 1532-4125 (Online) Journal homepage: https://www.tandfonline.com/loi/lagb20

Commutative rings with absorbing factorization

Malik Tusif Ahmed, Tiberiu Dumitrescu & M. Azeem Khadam

To cite this article: Malik Tusif Ahmed, Tiberiu Dumitrescu & M. Azeem Khadam (2020):
Commutative rings with absorbing factorization, Communications in Algebra

To link to this article: https://doi.org/10.1080/00927872.2020.1778714

Published online: 19 Jun 2020.

Submit your article to this journal

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=lagb20
COMMUNICATIONS IN ALGEBRAV
R

https://doi.org/10.1080/00927872.2020.1778714

Commutative rings with absorbing factorization

Malik Tusif Ahmeda , Tiberiu Dumitrescub, and M. Azeem Khadama
a
Abdus Salam School of Mathematical Sciences, GC University Lahore, Lahore, Pakistan; bFacultatea de
Matematica si Informatica, University of Bucharest, Bucharest, Romania

ABSTRACT ARTICLE HISTORY

We use the concept of n-absorbing ideal introduced by Badawi to study Received 24 March 2020
commutative rings having absorbing factorization. To this end, we intro- Revised 1 June 2020
duce and investigate the AF-dimension of a ring R – the minimum positive Communicated by Alfred
integer n such that R has n-absorbing factorization. We show that the Krull Geroldinger
dimension of a finite AF-dimensional ring is at most one and also present KEYWORDS
several applications. Chained ring; DVR;
factorization; n-absorbing
ideal; ZPI ring

2010 MATHEMATICS
SUBJECT
CLASSIFICATION
Primary 13A15;
Secondary 13F15

1. Introduction
Badawi introduced the concept of n-absorbing ideal in [4] which is a generalization of a prime
ideal and studied the case when n ¼ 2 in the same article. He along with Anderson later investi-
gated n-absorbing ideal in [3]. An ideal I of a ring R is an n-absorbing ideal if whenever
a1    anþ1 2 I for a1 , :::, anþ1 2 R, then there are n of the ai’s whose product is in I. In this case,
pffiffi
due to Choi and Walker [7, Theorem 1], ð I Þn  I: In a Pr€ ufer domain, every n-absorbing ideal
is a product of prime ideals, cf. [3, Theorem 5.7]. The study of n-absorbing ideals is an active
area of research and some of the recent articles are [5, 14, 22].
For m  n an n-absorbing ideal is also an m-absorbing ideal, see [3, Theorem 2.1(b)].
Therefore, in [3] the authors studied an x-invariant for a proper ideal I in a ring R, defined as
xR ðIÞ ¼ minfn j I is an n  absorbing ideal of Rg, otherwise set xR ðIÞ ¼ 1: As the authors
pointed out, xR ðIÞ measures how far I is from being a prime ideal of R. Recently, it was proved that
xR ðIÞ is finite for all r-ideals I, where r is a weak r-Noetherian ideal system on the ring R (among
others, this applies to all Mori domains), cf. [8, Proposition 3.3]. To sum up, this also shows that
studying absorbing ideals is the same as studying the x-invariants in semigroups of ideals. The
x-invariants for monoids were introduced by Geroldinger [10] and further studied by Geroldinger
and Hassler [11]. Furthermore, recall that the colength of an ideal I is equal to lR ðR=IÞ the length of
R-module R/I, see [6]. It is notable that xR ðIÞ  lR ðR=IÞ for a primary ideal I, see Remark 2.3.
Let n be a fixed positive integer. This note aims to study commutative rings whose proper
ideals have an n-absorbing factorization. Let I be a proper ideal of R. By an n-absorbing factoriza-
tion of I we mean an expression of I as a product I1    Ir of proper n-absorbing ideals. In the

CONTACT M. Azeem Khadam [email protected] Abdus Salam School of Mathematical Sciences, GC University
Lahore, 68-B New Muslim Town Lahore-Pakistan, Lahore, 54000 Pakistan.
ß 2020 Taylor & Francis Group, LLC
2 M. T. AHMED ET AL.

light of [3, Theorem 2.1(b)], it is natural to look at the minimum positive integer n such that
every ideal of R has an n-absorbing factorization; call n the AF-dimension (absorbing factorization
dimension) of R and denote it by AF-dimðRÞ: If no such n exists, set AF-dimðRÞ ¼ 1: We call R
a FAF-ring (finite absorbing factorization ring) if AF-dimðRÞ < 1: Recall that a ZPI-ring is a
ring whose proper ideals can be written as a product of prime ideals. Therefore, AF-dimðRÞ
measures, in some sense, how far R is from being a ZPI-ring.
Factorization of an ideal is an important topic of multiplicative ideal theory (for a detailed
study see the manuscript [9]). For recent work on the factorization of ideals, see [1, 2, 17–21].
Note that our investigation also generalizes the concept of TAF-rings, that is, rings in which every
proper ideal can be factored as a product of two-absorbing ideals [16].
This article is divided into four sections besides the introduction. The second section considers
a further investigation of absorbing ideals, which we not only require for upcoming sections, but
it also contributes some new results in the study of absorbing ideals. For example, we prove that
a P-primary ideal I is n-absorbing if and only if Pn  I (Corollary 2.4), and it is equivalent to I
being a prime power, provided R is a chained ring and I a regular ideal (Theorem 2.7).
In Section 3, we introduce and study the main ingredients of the article, that is, the absorbing
factorization of an ideal, the AF-dimension of a ring and FAF-rings. We produce several classes
of examples, in particular, we present the complete description for the AF-dimension of the ring
pffiffiffi
Z½ d for a square-free integer d (Corollary 3.9).
In Section 4, we prove that the Krull dimension of a FAF-ring is at most one (Theorem 4.1).
Further, we give a structure theorem for a FAF-ring (Theorem 4.2). In the end, we prove that the
polynomial rings which are FAF are ZPI (Corollary 4.4). In the last section, we present a few
applications of our work.
We assume throughout this article that all rings are commutative and unitary. Our termin-
ology and notation follow that of Gilmer [12], or Kaplansky [15].

2. A few results on n-absorbing ideals

In this section, we present a few properties of n-absorbing ideals which are not only useful for
upcoming sections, but also contribute to the study of n-absorbing ideals as initiated in [4] (for
n ¼ 2) and [3]. In particular, we investigate being an n-absorbing ideal property in chained rings.
For the sake of completeness, we begin with the definition.
Definition 2.1. Let n be a positive integer. An ideal I of a ring R is said to be n-absorbing if
whenever a1    anþ1 2 I for a1 , :::, anþ1 2 R, then there are n of the ai’s whose product is in I.
pffiffi
For an n-absorbing ideal I of R, the well-known conjecture ð I Þn  I is recently proved in
[7]. By combining this with [3, Theorem 3.1], we obtain the following proposition.
Proposition 2.2. Let P be a prime ideal of a ring R and I a P-primary ideal of R. Then, I is an
n-absorbing ideal of R if and only if Pn  I:

Remark 2.3. If I is an ideal of a ring R and lR ðR=IÞ ¼ n, we say that I has colength n. Note that
for a P-primary ideal I, xðIÞ is at most the colength of I. Indeed, if lR ðR=IÞ ¼ n then by [6,
Proposition 2.2], Pn RP  IRP and hence Pn  I: Now apply the previous Proposition 2.2.
For any positive integer n, it is well-known that the nth symbolic power PðnÞ ¼ Pn RP \ R of a
prime ideal P of R is P-primary and Pn  PðnÞ :
Further, recall that a prime ideal P of a ring R is called divided if P  xR for every x 2 R n P:
Corollary 2.4. Let R be a ring and I, P nonzero ideals of R with P prime.
 COMMUNICATIONS IN ALGEBRAV
R
3

(a) PðnÞ is an n-absorbing ideal. In particular, for each maximal ideal M of R, Mn is an n-absorbing
ideal.
pffiffi
(b) If I ¼ P and P is divided, then I is n-absorbing if and only if I is P-primary and Pn  I:

Proof. (a) and “only if” part of (b) follows from [3, Theorem 3.1], and “if” part of (b) follows
from [3, Theorem 3.2]. The if part of (b) also follows from Proposition 2.2. w

A chained ring is a ring whose every two ideals are comparable under inclusion. Obviously, a
chained domain is a valuation domain. Note that in a chained ring every radical ideal is prime.
Lemma 2.5. If R is a chained ring, then its n-absorbing ideals are primary ideals Pk with P prime
ideal and k  n:
pffiffi
Proof. Let I be an n-absorbing ideal of R. As R is chained, I ¼ P is a prime ideal. By Corollary
2.4, I is P-primary and Pn  I: If P ¼ P2 then I ¼ P. Suppose that P 6¼ P2 and choose k minimal
such that Pk  I: Then, there is an element y of Pk1 not in I. Hence I  yR, and I ¼ yJ for
some ideal J of R. Since I is P-primary and y 62 I, it follows that J  P: Hence I ¼ yJ  yP  Pk ,
as we desired. w

An element in a ring R is called regular if it is not a zero-divisor, and I is called a regular ideal
of R if it contains regular elements.
Lemma 2.6. If P is a proper regular prime ideal of a chained ring R, then the product of two
P-primary ideals is also a P-primary ideal.
pffiffiffiffiffiffiffiffiffiffiffi
Proof. Assume that Q1 and Q2 are two P-primary ideals of R. Clearly Q1 Q2 ¼ P: Suppose x, y
are two elements of R with xy 2 Q1 Q2 and x 62 P: Hence Q1  xR, and Q1 ¼ xJ for some ideal J
of R. Since Q1 is P-primary and x 62 P, it follows that J ¼ Q1, and hence xy 2 xQ1 Q2 : As P is a
regular ideal and P  xR, therefore x is a regular element. Consequently, y 2 Q1 Q2 , as
we desired. w

The previous lemma is no longer true if the regular condition is dropped. Indeed, [12,
Exercise 4, page 200] provides the chained ring R, the idealization of the Zð2Þ -module Z21 , where
P ¼ Z21 is a prime ideal but P2 ¼ 0 is not P-primary (note that Pð2Þ ¼ P).
The following is the main result of this section.
pffiffi
Theorem 2.7. Let R be a chained ring and I a proper regular ideal of R with P ¼ I a prime ideal.
Then the following assertions are equivalent.
(a) I is an n-absorbing ideal of R.
(b) I is a P-primary ideal of R such that Pn  I:
(c) I 2 fP, P2 , :::, Pn g:

Proof. ðaÞ () ðbÞ follows by Corollary 2.4. For ðaÞ ) ðcÞ, see Lemma 2.5 and ðcÞ ) ðbÞ follows
from Lemma 2.6. w

Remark 2.8. The previous result extends [4, Proposition 3.10] from two-absorbing ideals in
domain case to n-absorbing ideals in rings with zero-divisors. Note that ðaÞ ) ðbÞ ) ðcÞ is true
for any proper ideal I, but for ðcÞ ) ðaÞ we need I to be a regular ideal.
We conclude this section with the following consequence of the previous Theorem 2.7.
pffiffi
Corollary 2.9. Let R be a valuation domain and I a proper ideal of R with P ¼ I a prime ideal.
Then the following assertions are equivalent.
4 M. T. AHMED ET AL.

(a) I is an n-absorbing ideal of R.

(b) I is a P-primary ideal of R such that Pn  I:
(c) I 2 fP, P2 , :::, Pn g:

3. The AF-dimension of a ring

Let n be a fixed positive integer. A ring is said to have an n-absorbing factorization if it’s every
proper ideal is a product of n-absorbing ideals. This is, in fact, a generalization of the two-
absorbing factorization rings as introduced in [16]. Recall that for m  n an n-absorbing ideal is
also an m-absorbing ideal, see [3, Theorem 2.1(b)]. Therefore, it is natural to look at the min-
imum n such that our ring has n-absorbing factorization. Formally:
Definition 3.1. Let R be a ring. The AF-dimðRÞ – the AF-dimension of R is defined as the
minimum positive integer n such that every ideal of R has an n-absorbing factorization. If no
such n exists, set AF-dimðRÞ ¼ 1: We call R a FAF-ring (finite absorbing factorization ring)
if AF-dimðRÞ < 1:
Before going into examples, we have the following observation.
Remark 3.2. Let AF-dimðRÞ  n and I be an M-primary ideal of R for some maximal ideal M of
R. If I 6 M2 , then I is an n-absorbing ideal of R and hence by Proposition 2.2, Mn  I:
We have a couple of examples. Recall that a ZPI-ring is a ring whose proper ideals can be
written as a product of prime ideals.
Example 3.3. (1) R is a ZPI-ring if and only if AF-dimðRÞ ¼ 1: Indeed, by definition, every prime
ideal is 1-absorbing. Q
(2) The AF-dimension of the ring R ¼ 1 i¼1 Z2 is not finite. Indeed, the ideal I ¼ fðxi Þ 2
R j x2i1 ¼ 0 for all i 2 Ng has no n-absorbing factorization for all n  1: By [3, Example 2.3], I
is not an n-absorbing ideal for any n  1: Further, if for some n, I ¼ I1    Ir where each Ii is an
n-absorbing ideal, then I ¼ I1    Ir ¼ I1 \    \ Ir since R is a von Neumann regular ring. Hence
I is an nr-absorbing ideal by [3, Theorem 2.1], a contradiction.
Recall that a ring R is said to be special primary if MaxR ¼ fMg and it’s every proper ideal is
a power of M. Note that local ZPI rings are special primary. Combining this fact with Lemma
2.5, we obtain the following result.
Proposition 3.4. A chained ring R is a FAF-ring if and only if R is a local ZPI-ring, that is, a spe-
cial primary ring.
The next result says that the AF-dimension of a factor (resp. fraction) ring is bounded above
by the AF-dimension of the ring.
Proposition 3.5. Let R be a FAF-ring and T a factor or a fraction ring of R. Then AF-dimðTÞ 
AF-dimðRÞ:

Proof. Let R be a ring and I  J ideals of R such that J ¼ J1 J2    Jr with each Ji is an n-absorbing
ideal of R. Then J=I ¼ ðJ1 =IÞðJ2 =IÞ    ðJr =IÞ, where each Ji =I is also an n-absorbing ideal of T ¼
R=I, cf. [3, Corollary 4.3]. The fraction ring assertion is a consequence of the fact that localiza-
tion of an n-absorbing factorization is still an n-absorbing factorization, cf. [3, Theorem 4.1]. w

Proposition 3.6. Let R1 , :::, Rk be FAF-rings and R their direct product. Then AF-dimðRÞ ¼
supi fAF-dimðRi Þg:
 COMMUNICATIONS IN ALGEBRAV
R
5

Proof. It suffices to prove our assertion for two rings, say A and B. Note that an ideal H of A is
n-absorbing if and only if H B is an n-absorbing ideal of A B, because ðA BÞ=ðH BÞ ¼
A=H: Therefore, if I ¼ I1 I2    Ir (resp. J ¼ J1 J2    Js ) is an n-absorbing factorization in A (resp.
B), then I J ¼ ðI1 BÞ    ðIr BÞðA J1 Þ    ðA Js Þ is an n-absorbing factorization in A B. w
Denote by MinðIÞ the set of minimal prime ideals over an ideal I.
Proposition 3.7. Let R be a FAF-ring and I a proper ideal. Then MinðIÞ is finite.

Proof. If I ¼ I1    Ir is an n-absorbing factorization of I, then MinðIÞ  [rj¼1 MinðIj Þ and each

MinðIj Þ is finite by [3, Theorem 2.5]. w

Proposition 3.8. Let R be a finite ring of order m such that pnþ2 - m for each prime p. Then
AF-dimðRÞ  n. Moreover, AF-dimðZpnþ1 ½X=ðX 2 , pXÞÞ ¼ n þ 1:

Proof. Combining Proposition 3.6 with the well-known fact that each finite ring is a product of
local rings of prime power order, we may assume that (R, M) is a local ring of order m ¼ pk with
k  n þ 1: Then M nþ1 ¼ ð0Þ: Also, if I is a nonzero proper ideal of R, then M n  I, so I is an n-
absorbing ideal, cf. Proposition 2.2. For the “Moreover” part, denote by x the image of X in A ¼
pffiffiffiffiffiffi
Zpnþ1 ½X=ðX 2 , pXÞ: Then xA ¼ ðp, xÞA, the unique prime ideal of A. Further, neither xA is con-
tained in ððp, xÞAÞ2 ¼ ðp2 ÞA nor ððp, xÞAÞn ¼ ðpn ÞA is contained in xA. This shows that xA is nei-
ther a product of at least two n-absorbing ideals nor an n-absorbing ideal itself, cf. Proposition
2.2. That is, the AF-dimension of A is at least n þ 1. The reverse inequality follows from the pre-
vious part. w

Note that the above result also extends [16, Proposition 2.6].
For a square-free integer d, the following result gives the complete description of AF-dimension
pffiffiffi
of Z½ d:
Corollary 3.9. Let d 2 Z  f0, 1g be a square-free integer d
1 mod 4. Then the following asser-
tions are equivalent.
(a) d
5 mod
 p 8: 
ffiffiffi
(b) AF-dim Z½ d ¼ 2:
pffiffiffi
(c) Z½ d is a FAF-domain.

Proof. ðaÞ () ðbÞ follows by combining [16, Corollary 4.11] with the well-known fact that
pffiffiffi
Z½ d is Dedekind if and only if d
2, 3 mod 4, and ðbÞ ) ðcÞ is trivial. For  ðcÞ ) ðaÞ, we
pffiffiffi 
assume that d
1 mod 8: Fix some n  1: It suffices to show that AF-dim Z½ d  n: By
pffiffiffi
Hensel’s Lemma q2
d mod 2n for some integer q, see [13, page 71]. Then sending d into
pffiffiffi
x þ q, we map Z½ d onto A ¼ Z2n ½X=ðX 2 , 2XÞ, where x is the image of X in A. By Propositions
pffiffiffi  pffiffiffi 
3.5 and 3.8, AF-dim Z½ d  n: As n is arbitrary, AF-dim Z½ d ¼ 1: w

4. Krull dimension of an FAF-ring

We begin with the main result of the article.
Theorem 4.1. Let R be a FAF-ring. Then RM is a zero-dimensional ring or a one-dimensional
domain for each M 2 MaxðRÞ. In particular, the Krull dimension of R is at most one.
6 M. T. AHMED ET AL.

Proof. Let us suppose that AF-dimðRÞ ¼ n: By Proposition 3.5, we may assume that R is local
with nonzero maximal ideal M. It suffices to show that M is the only nonzero prime ideal.
Suppose on the contrary that R has a nonzero prime ideal P 6¼ M: Consider y 2 M  P and
hence, shrinking M, we may assume that M is minimal over (P, y). First, we show that M 6¼ M 2 :
Assume to the contrary that M ¼ M 2 : Then M is the only n-absorbing ideal of R with radical M.
As AF-dimðRÞ ¼ n, y 2 M  P and M is minimal over (P, y), it follows that ðP, y2 Þ ¼ M: We get
ðP, yÞ ¼ M ¼ ðP, y2 Þ and hence yð1  yaÞ 2 P for some a 2 R, which leads to a contradiction.
Therefore, we can pick x 2 M  M 2 : Now, suppose for the moment that the following assertion
is true.
ðÞ Every prime ideal Q  M is contained in Mn.
pffiffiffiffiffiffiffi
By ðÞ we have xR ¼ M, so M n  xR, cf. Remark 3.2. Let x 2 P  f0g and xR ¼ I1    Ir
be an n-absorbing factorization. As P is prime, it contains some Ii ¼: I: By ðÞ we have I  P 
Mn  xR: Since P is prime and x 62 P, we get P ¼ xn P: Then I ¼ xn J for some proper ideal J.
Since xn 62 I (as x 62 P) and I is an n-absorbing ideal of R we get xn1 J  I, so I  xI, and
hence I ¼ xI: Combining this fact with xR ¼ I1    Ir , we get xR ¼ xxR, so xð1  xpÞ ¼ 0 for
some p 2 R: Since 1  xp is a unit, we get x ¼ 0 which is a contradiction.

Proof of ðÞ: Assume, to the contrary, that there exists a prime ideal Q  M such that Q 6 Mn :
2
Pick an element z 2 M  Q: As AF-dimðRÞ ¼ n and Q 6 Mn , it follows that ðQ, zn Þ ¼ I1    Is
2
where each Ij is an n-absorbing ideal and s < n. Then zn 2 Ij for each j 2 f1, :::, sg: So zn 2 Ij
2
because each Ij is an n-absorbing ideal and therefore zns 2 ðQ, zn Þ: Note that ns < n2 and hence-
forth, modding out by Q, we are led to a contradiction. w

The following is a structure theorem for the FAF-rings.

Theorem 4.2. Any FAF-ring is a finite direct product of one-dimensional domains and zero-
dimensional local rings with nilpotent maximal ideal. In particular, a FAF-ring of Krull dimen-
sion one having unique height-zero prime ideal is a domain.

Proof. We adapt the proof of [12, Theorem 46.11]. Let AF-dimðRÞ ¼ n and therefore ð0Þ ¼
pffiffiffi
I1    Ir be an n-absorbing factorization of (0). Note that ð Ii Þn  Ii for every i ¼ 1, :::, r by [7,
Theorem 1], and hence each Ii contains a product of prime ideals by [3, Theorem 2.5]. We get
ð0Þ ¼ P1c1    Pscs where P1 , :::, Ps are (the) distinct height-zero prime ideals of R. By Theorem 4.1,
the ideals P1 , :::, Ps are pairwise co-maximal, and so by Chinese Remainder Theorem we have R ¼
R=P1c1    R=Pscs : Set P ¼ Pi and c ¼ ci for some i 2 f1, :::, sg: If P is maximal, then R=Pc is local
and zero-dimensional. If P is not maximal, then by Theorem 4.1 we have PRM ¼ ð0Þ for every max-
imal ideal M containing P. Hence Pc ¼ P and therefore R=Pc is a one-dimensional domain. w

Recall that a domain A has finite character if every x 2 A  f0g belongs to only finitely many
maximal ideals of A.
Theorem 4.3. For a one-dimensional domain A, the following are equivalent.
(a) A is a FAF-domain.
(b) A has finite character and there is some positive integer d such that AF-dimðAM Þ  d for
each M 2 MaxðAÞ:

Proof. ðaÞ ) ðbÞ follows from Propositions 3.5 and 3.7. For ðbÞ ) ðaÞ, note that every nonzero
ideal I of A is a finite product of primary ideals. Indeed, we can locally check that I ¼
Qn
i¼1 ðIAMi \ AÞ where VðIÞ ¼ fM1 , :::, Mn g (alternatively, we can invoke Jaffard’s Theorem [9,
Theorem 2.1.5], because A is h-local). So it suffices to show that every nonzero M-primary ideal I
 COMMUNICATIONS IN ALGEBRAV
R
7

has a d-absorbing factorization. Set B ¼ AM, N ¼ MB and note that every proper nonzero ideal
of B is N-primary. By our hypothesis, IB ¼ J1    Jk for some d-absorbing N-primary ideals Ji of B;
so each Ji contains Nd. Then Ii ¼ Ji \ A is an M-primary ideal of A which contains Md, so Ii is a
d-absorbing ideal. We check locally that I ¼ I1    Ik : This completes the proof. w

Corollary 4.4. For a ring R, the following assertions are equivalent.

(a) R½X is a FAF-ring.
(b) R is a von Neumann regular FAF-ring.
(c) R is a finite direct product of fields.
(d) R½X is a ZPI-ring.

Proof. It suffices to prove ðaÞ ) ðcÞ and ðbÞ ) ðcÞ: For ðaÞ ) ðcÞ, note that the polynomial ring
over a direct product is a direct product of polynomial rings. Therefore, by using Theorem 4.2,
we may assume that R (and hence also R½X) has a unique height-zero prime ideal. Hence, again
by applying Theorem 4.2, R½X is a one-dimensional domain and so R is a field. For ðbÞ ) ðcÞ,
by Proposition 3.7, R is a semilocal von Neumann regular ring, and hence by Chinese Remainder
Theorem a finite direct product of fields. w

5. Noetherian FAF-domains
We begin with the following lemma.
Lemma 5.1. Let (A, M) be a local domain having a discrete rank-one valuation overring (B, pB)
with a nonzero conductor ðA : BÞ such that pB \ A ¼ M. Then A is a FAF-ring.
Proof. Let I be a nonzero proper ideal of A and ðIÞ denote its p-adic value, that is, IB ¼ pðIÞ B:
Set m ¼ ðMÞ and n ¼ ðA : BÞ: Note that if J is an ideal of A with ðJÞ  ðIÞ þ n, then J  I:
If ðIÞ  2n, then I is a product of smaller p-adic value ideals, e.g., I ¼ pn ðI : pn Þ: Suppose
that ðIÞ  2n  1 and set k ¼ dð3n  1Þ=me, where de means ceiling function. Then ðM k Þ 
3n  1  ðIÞ þ n, so Mk  I, hence I is k-absorbing. Thus, AF-dimðAÞ  dð3n  1Þ=me: w

We have another lemma.

Lemma 5.2. Let (A, M) be a local domain and B an overring of A with a nonzero conductor
ðA : BÞ. Assume that B has two distinct maximal ideals P, Q such that P \ A ¼ Q \ A ¼ M and
\n1 Pn ¼ \n1 Qn ¼ ð0Þ. Then A is not a FAF-ring.

Proof. Let c 2 ðA : BÞ  f0g and z 2 Q  P: Then c 2 M and c 62 Pk for some k. Suppose that A
has AF-dimension d. Fix an n  1: Then Aðczn Þ ¼ I1    Ir with each Ii a d-absorbing ideal, so
Ii  M d : Therefore, r < k (since c 62 Pk and z 62 P), so Mkd  I1    Ir  Qn : Since \Qn ¼ ð0Þ, we
get Mkd ¼ ð0Þ which is a contradiction. w

From Lemmas 5.1 and 5.2, we get the following corollary.

Corollary 5.3. Let (A, M) be a Noetherian local one-dimensional domain with a nonzero conductor
ðA : A0 Þ, where A0 is the integral closure of A. Then A is a FAF-domain if and only if A0 is local.
We state the main result of this section.
Theorem 5.4. Let A be a Noetherian one-dimensional domain with a nonzero conductor ðA : A0 Þ,
where A0 is the integral closure of A. The following assertions are equivalent.
(a) A is a FAF-domain.
(b) AM is a FAF-domain for each M 2 MaxðAÞ:
8 M. T. AHMED ET AL.

(c) The spectral map SpecðA0 Þ ! SpecðAÞ is bijective.

Proof. ðaÞ () ðbÞ: If M is a maximal ideal not containing ðA : A0 Þ, then AM is a discrete rank-
one valuation domain. Let M1 , :::, Mn be the maximal ideals containing ðA : A0 Þ: We apply
Theorem 4.3 with d equal to the maximum of AF-dimðAMi Þ, i ¼ 1, :::, n:
ðbÞ () ðcÞ: Note that condition (c) can be tested locally, because the integral closure of AM is
A0M for each M 2 MaxðAÞ: So we may assume that A is local. Then A0 is a PID, so Corollary
5.3 applies. w

Examples 5.5. (1) AF-dimðZ½2iÞ ¼ 3: Indeed, all localizations of Z½2i at its maximal ideals are
discrete rank-one valuation domains except A ¼ Z½2iM where M ¼ ð2, 2iÞ: Applying the proof
of Lemma 5.1 with B ¼ Z½ið1þiÞ , p ¼ 1 þ i, n ¼ 2, m ¼ 2, we get that AF-dimðAÞ  d5=2e ¼ 3:
Hence AF-dimðZ½2iÞ  3 (see the proof of Theorem 5.4), and therefore AF-dimðZ½2iÞ ¼ 3
because ð2 þ 2iÞ has no 2-absorbing factorization.
pffiffiffi pffiffiffi
(2) A ¼ Z½ 3 4 is a FAF-ring. Indeed, its integral closure is A0 ¼ Z½ 3 2 and A  A0 is a root
extension (i.e., z2 2 A for each z 2 A0 ), so the map SpecðA0 Þ ! SpecðAÞ is bijective. Apply
Theorem 5.4.
pffiffiffiffiffi pffiffiffiffiffi
(3) Z½ 3 10 is not a FAF-ring. Indeed, the integral closure of Z½ 3 10 is Z½t with t ¼
p3 ffiffiffiffi p3 ffiffiffiffiffi pffiffiffiffiffi
1þ 10þ 100
: Further, ð3, tÞ and ð3, t  1Þ are two distinct primes lying over ð3, 1  3 10Þ in
p ffiffiffiffiffi
3
Z½ 3 10: Now apply Theorem 5.4.
(4) Let K be a field. Consider the Noetherian one-dimensional domains A ¼ K þ XðX 
1ÞK½X and B ¼ K þ X n K½X for some n  2: Their integral closure is K½X: Consider the spectral
maps SpecðK½XÞ ! SpecðAÞ and SpecðK½XÞ ! SpecðBÞ: Since only the second one is bijective,
we get that B is a FAF-domain while A is not.

Acknowledgments
The authors are grateful to the referee for useful comments and suggestions.

Funding
First- and third-named authors are supported by ASSMS GC University Lahore under postdoctoral fellowships.

ORCID
Malik Tusif Ahmed http://orcid.org/0000-0001-5177-184X
M. Azeem Khadam http://orcid.org/0000-0002-2917-6275

References
[1] Ahmed, M. T., Dumitrescu, T. (2017). SP-rings with zero-divisors. Commun. Algebra 45(10):4435–4443.
DOI: 10.1080/00927872.2016.1267184.
[2] Ahmed, M. T., Dumitrescu, T. (2018). Domains with invertible-radical factorization. Bull. Korean Math.
Soc. 55:809–818.
[3] Anderson, D. F., Badawi, A. (2011). On n-absorbing ideals of commutative rings. Commun. Algebra 39(5):
1646–1672. DOI: 10.1080/00927871003738998.
[4] Badawi, A. (2007). On 2-absorbing ideals of commutative rings. Bull. Austral. Math. Soc. 75(3):417–429.
DOI: 10.1017/S0004972700039344.
 COMMUNICATIONS IN ALGEBRAV
R
9

[5] Badawi, A., Issoual, M., Mahdou, N. (2019). On n-absorbing ideals and (m, n)-closed ideals in trivial ring
extensions of commutative rings. J. Algebra Appl. 18(7):1950123. DOI: 10.1142/S0219498819501238.
[6] Barucci, V., Fr€ oberg, R. (2001). On the number of ideals of finite colength. In Geometric and
Combinatorial Aspects of Commutative Algebra (Messina, 1999). Vol. 217 of Lecture Notes in Pure and
Applied Mathematics. New York: Dekker, pp. 11–19.
[7] Choi, S. H., Walker, A. (2020). The radical of an n-absorbing ideal. J. Commut. Algebra 12(2):171–177.
https://projecteuclid.org/euclid.jca/1507276946 DOI: 10.1216/jca.2020.12.171.
[8] Fan, Y., Geroldinger, A., Kainrath, F., Tringali, S. (2017). Arithmetic of commutative semigroups with a
focus on semigroups of ideals and modules. J. Algebra Appl. 16(12):1750234–1750242. DOI: 10.1142/
S0219498817502346.
[9] Fontana, M., Houston, E., Lucas, T. (2012). Factoring Ideals in Integral Domains. New York: Springer.
[10] Geroldinger, A. (1997). Chains of factorizations in weakly Krull domains. Colloq. Math. 72(1):53–81. DOI:
10.4064/cm-72-1-53-81.
[11] Geroldinger, A., Hassler, W. (2008). Local tameness of v-Noetherian monoids. J. Pure Appl. Algebra 212(6):
1509–1524. DOI: 10.1016/j.jpaa.2007.10.020.
[12] Gilmer, R. (1972). Multiplicative Ideal Theory. New York: Dekker.
[13] Gouv^ea, F. Q. (1997). Gouv^ea p-adic Numbers: An Introduction. Berlin: Universitext, Springer.
[14] Issoual, M., Mahdou, N., Moutui, M. A. S. (2019). On n-absorbing and strongly n-absorbing ideals of amal-
gamation. J. Algebra Appl. DOI: 10.1142/S0219498820501996.
[15] Kaplansky, I. (1974). Commutative Rings. Chicago: The University of Chicago Press.
[16] Mukhtar, M., Ahmed, M. T., Dumitrescu, T. (2018). Commutative rings with two-absorbing factorization.
Commun. Algebra 46(3):970–978. DOI: 10.1080/00927872.2017.1332202.
[17] McGovern, W. W. (2007). Bezout SP-domains. Commun. Algebra 35(5):1777–1781. DOI: 10.1080/
00927870601169291.
[18] Olberding, B. (2000). Factorization into prime and invertible ideals. J. London Math. Soc. 62(2):336–344.
DOI: 10.1112/S0024610700001319.
[19] Olberding, B. (2005). Factorization into radical ideals. In Arithmetical Properties of Commutative Rings and
Monoids, Vol. 241. Boca Raton: Chapman & Hall/CRC, pp. 363–377.
[20] Olberding, B. (2009). Factorization into prime and invertible ideals II. J. London Math. Soc. 80(1):155–170.
DOI: 10.1112/jlms/jdp017.
[21] Olberding, B., Reinhart, A. (2019). Radical factorization in commutative rings, monoids and multiplicative
lattices. Algebra Univers. 80(2):24–29. DOI: 10.1007/s00012-019-0597-1.
[22] Payrovi, S., Babaei, S. (2013). On the 2-absorbing ideals in commutative rings. Bull. Malays. Math. Sci. Soc.
36(2):895–900.