Jump to ContentJump to Main Navigation
Show Summary Details
More options …

Open Mathematics

formerly Central European Journal of Mathematics

Editor-in-Chief: Vespri, Vincenzo / Marano, Salvatore Angelo


IMPACT FACTOR 2018: 0.726
5-year IMPACT FACTOR: 0.869

CiteScore 2018: 0.90

SCImago Journal Rank (SJR) 2018: 0.323
Source Normalized Impact per Paper (SNIP) 2018: 0.821

Mathematical Citation Quotient (MCQ) 2018: 0.34

ICV 2018: 152.31

Open Access
Online
ISSN
2391-5455
See all formats and pricing
More options …
Volume 15, Issue 1

Issues

Volume 13 (2015)

Rings in which elements are the sum of a nilpotent and a root of a fixed polynomial that commute

Ali H. Handam / Hani A. Khashan
Published Online: 2017-04-17 | DOI: https://doi.org/10.1515/math-2017-0031

Abstract

An element in a ring R with identity is said to be strongly nil clean if it is the sum of an idempotent and a nilpotent that commute, R is said to be strongly nil clean if every element of R is strongly nil clean. Let C(R) be the center of a ring R and g(x) be a fixed polynomial in C(R)[x]. Then R is said to be strongly g(x)-nil clean if every element in R is a sum of a nilpotent and a root of g(x) that commute. In this paper, we give some relations between strongly nil clean rings and strongly g(x)-nil clean rings. Various basic properties of strongly g(x) -nil cleans are proved and many examples are given.

Keywords: Nil clean ring; Strogly nil clean ring; g(x)-nil clean ring; Strongly g(x)-nil clean ring

MSC 2010: 16N40; 16U99

1 Introduction

Let R be an associative ring with identity. Following [10], we define an element r of a ring R to be clean if there is an idempotent eR and a unit uR such that r = u + e. A clean ring is defined to be one in which every element is clean. Clean rings were first introduced by Nicholson [10] as a class of exchange rings. In [11], an element r of a ring R is defined to be strongly clean provided that there exist an idempotent eR and a unit uR such that r = e + u and eu = ue. Analogously, a strongly clean ring is one in which every element is strongly clean. In [7], Diesl introduced the concept of strongly nil cleanness. An element rR is strongly nil clean provided that there exist an idempotent eR and a nilpotent element nR such that r = n + e and ne = en. Recently, the concept of nil clean rings was generalized to g(x)-nil clean rings (see Khashan and Handam [9]). We refer the reader to [1, 46, 14] for a survey on nil clean rings.

Let C(R) denote the center of a ring R and g(x) be a polynomial in C(R)[x]. Following Camillo and Simón [2], R is called g(x)-clean if for each rR, there exist a unit uR and sR such that r = s + u and g(s) = 0. Of course (x2x)-clean rings are clean rings.

Nicholson and Zhou [12] proved that if g(x) ∈ (xa)(xb)C(R)[x] where a, bC(R) are such that b and ba are both units in R and RM is a semisimple left R-module, then End(RM) is g(x)-clean. In 2008, Fan and Yang [8], studied more properties of g(x)-clean rings. Among many conclusions, they proved that if g(x) ∈ (xa)(xb)C(R)[x] where a, bC(R) with ba a unit in R, then R is a clean ring if and only if R is (xa)(xb)-clean. Recently, Khashan and Handam [9] completely determined the relation between nil clean rings and g(x)-nil clean rings.

In this paper, we define and study strongly g(x)-nil clean rings. For a ring R and g(x) ∈ C(R)[x], We say that rR is strongly g(x)-nil clean provided that there exist a nilpotent bR and sR such that r = s + b, g(s) = 0 and sb = bs. Moreover, R is called strongly g(x)-nil clean if every element in R is strongly g(x)-nil clean.

In section 2, we define the concept of strongly g(x)-nil clean and some general properties of strongly g(x)-nil clean rings are given. In section 3, we determine the relation between strongly nil clean rings and strongly g(x)-nil clean rings and we give many characterizations for a strongly nil clean ring R in terms of some strongly g(x)-nil clean rings.

Throughout this article, U(R), Id(R) and N(R) will denote, respectively, the group of units, the set of idempotents and the set of nilpotent elements in R. We write Mn(R) and Tn(R) for the rings of all n × n matrices and all upper triangular matrices over R, respectively. The ring of integers modulo n is denoted by ℤn.

2 Strongly g(x)-nil clean rings

We study in this section the basic properties of strongly g(x)-nil clean rings, and some examples are given.

Definition 2.1

Let R be a ring and g(x) be a fixed polynomial in C(R)[x]. An element rR is called strongly g(x)-nil clean if r = n + s where nN(R), g(s) = 0 and ns = sn. A ring R is called strongly g(x)-nil clean if each of its elements is strongly g(x)-nil clean.

Clearly, every strongly g(x)-nil clean ring is g(x)-nil clean. Moreover, the two concepts coincide in the commutative case. However, the converse need not be true in non commutative rings. For example, one can see that the ring M2(ℤ2) is (x4 + x2)-nil clean which is not strongly (x4 + x2)-nil clean.

Proposition 2.2

Every strongly g(x)-nil clean ring is strongly g(x)-clean.

Proof

Suppose R is a strongly g(x)-nil clean ring and let rR. Write r − 1 = n + s where nN(R), g(s) = 0 and ns = sn. Then r = (n + 1) + s where n + 1 ∈ U(R) and (n + 1)s = s(n + 1) . Hence, R is strongly g(x)-clean.     □

The converse of Proposition 2.2 is not true in general. For example, simple computations show that the non strongly (x4 + x2)-nil clean ring M2(ℤ2) is strongly (x4 + x2)-clean.

For any two rings R and S, consider the ring homomorphism η : C(R) → C(S) with η(1R) = 1S. If g(x)= i=0naixiC(R)[x], we let gη(x):=i=0nη(ai)xiC(S)[x]. In particular, if g(x) ∈ ℤ[x], then gη(x) = g(x).

Next, we give some properties of the class of strongly g(x)-nil clean rings. The proof of the following proposition is straightforward.

Proposition 2.3

Let R and S be two rings, η : RS be a ring epimorphism and g(x) ∈ C(R)[x]. If R is strongly g(x)-nil clean, then S is strongly gη(x)-nil clean.

Corollary 2.4

Let R be a ring.

  1. If R is a strongly g(x)-nil clean ring and I is an ideal of R, then R/I is strongly gη(x)-nil clean where η : RR/I is defined by η(r) = r + I.

  2. If the formal power series ring R[[t]] is strongly g(x)-nil clean, then R is strongly gη(x)-nil clean where

    η : R[[t]] → R is defined by η(i=0aiti)=a0.

Corollary 2.5

Let R be a ring and 1 < n ∈ ℕ. If Tn(R) is strongly g(x)-nil clean, then R is strongly gη(x)-nil clean.

Proof

Define η : Tn(R) → R by η(A) = a11 where A = (aij) ∈ Tn(R). Then clearly η is a ring epimorphism and R is strongly gη(x)-nil clean.     □

Proposition 2.6

Let R1, R2, Rk be rings and g(x) ∈ ℤ[x]. Then R=i=1kRi is strongly g(x)-nil clean if and only if Ri is strongly g(x)-nil clean for all i ∈ {1, 2, …, k}.

Proof

⇒) : For each i ∈ {1, 2, …, k}, Ri is a homomorphic image of i=1kRi under the projection homomorphism. Hence, Ri is strongly g(x)-nil clean by Proposition 2.3.

⇐) : Let r=(x1,x2,...,xk)i=1kRi. For each i, write xi = ni + si where niN(Ri), g(si) = 0 and nisi = sini. Let n = (n1, n2, nk) and s = (s1, s2, sk). Then r = n + s where clearly nN(R), g(s) = 0 and ns = sn. Therefore, R is strongly g(x)-nil clean.     □

Proposition 2.7

If R is a commutative ring, then the ring of polynomials R[t] is not strongly nil clean (and hence not strongly (x2x)-nil clean).

Proof

It is well known that N(R[t]) = {a0 + a1t + … + aktk : aiN(R)} and Id(R[t]) = Id(R). Suppose t = (a0 + a1t + … + aktk) + e where aiN(R) and e2 = e. Then a1 = 1 ∈ N(R) which is a contradiction. Therefore, R[t] is not strongly nil clean.     □

The converse of (1) in Corollary 2.4 is not true in general. For example, the field ℤ2 ≃ ℤ2[x]/〈x〉 is strongly (x2x)-nil clean but ℤ2[x] is not strongly (x2x)-nil clean by Proposition 2.7.

Let A and B be rings, BMA be a bimodule and consider the formal triangular matrix T=A0MB. If g(x)= a00m0b0+a10m1b1x+...+an0mnbnxnC(T)[x], then clearly gA(x) = a0 + a1x + … + anxnC(A)[x] and gB(x) = b0 + b1x + … + bnxnC(B)[x].

Proposition 2.8

Let A and B be rings and let BMA be a bimodule. If T=A0MB is strongly g(x)-nil clean for some g(x) ∈ C(T)[x], then A is strongly gA(x)-nil clean and B is strongly gB(x)-nil clean.

Proof

Suppose that T=A0MB is strongly g(x)-nil clean. For every aA, bB, mM, write a0mb=n10n2n3+s10s2s3wheren10n2n3N(T),g(s10s2s3)=0andn10n2n3s10s2s3=s10s2s3n10n2n3. Then a = n1 + s1 and b = n3 + s3. It is easy to see that n1N(A), n3N(B), gA(s1) = gB(s3) = 0, n1s1 = s1n1 and n3s3 = s3n3. Therefore, A is a strongly gA(x)-nil clean and B is a strongly gB(x)-nil clean.     □

3 Strongly (x2 + cx + d)-nil clean rings

In this section, we consider some types of strongly (x2 + cx + d)-nil clean rings and we will investigate the relation between strongly g(x)-nil clean rings and strongly nil clean rings.

Theorem 3.1

Let R be a ring and a, bC(R). Then R is strongly nil clean with baN(R) if and only if R is strongly (x − (a + 1))(xb)-nil clean.

Proof

⇒) : Suppose R is strongly nil clean with baN(R) and let rR. Then r(a+1)(ba)1=e+n, where e2 = e, nN(R) and en = ne. Therefore, r = [e((ba) − 1) + (a + 1)] + n((ba) − 1). Write s = [e((ba) − 1) + (a + 1)] and v = n((ba) − 1). Then r = s + v where clearly vN(R) and sv = vs. Moreover, s is a root of (x − (a + 1))(xb) since [e((ba)1)+(a+1)(a+1)][e((ba)1)+(a+1)b]=e2((ba)1)2e((ba)1)((ba)1)=0

Hence, R is strongly (x − (a + 1))(xb)-nil clean.

⇐) : Suppose R is a strongly (x − (a + 1))(xb)-nil clean. Write a = s + n where nN(R), (s − (a + 1))(sb) = 0 and sn = ns. Then, saN(R) and so s − (a + 1) ∈ U(R). Therefore, s = b and baN(R). Now, let rR. By the assumption, we can write r((ba) − 1) + (a + 1) = s + n where nN(R), s is a root of (x − (a + 1))(xb) and sn = ns. Thus, r = t + m where t = (s − (a + 1))((ba) − 1)−1 and m = n((ba) − 1)− 1. Since a, bC(R), it is clear that mN(R) and tm = mt. Moreover, t2=((s(a+1))((ba)1)1)2=(s(a+1))(sb+b(a+1))((ba)1)2=(s(a+1))(sb)+(s(a+1))(b(a+1))((ba)1)2=(s(a+1))((ba)1)1=t

Therefore, R is strongly nil clean.     □

Lemma 3.2

([7]). If a ring R is nil clean, then 2 is a (central) nilpotent element in R.

As 2 is a central nilpotent in any (strongly) nil clean ring R, then 2nN(R) for any integer n. On the other hand, a non zero strongly g(x)-nil clean ring can’t have a nilpotent of the form 2n + 1 since otherwise 1 ∈ N(R) which is a contradiction. Next, we give some special cases of Theorem 3.1.

Corollary 3.3

Let R be a ring and n ∈ ℤ. For any bC(R), the following conditions are equivalent

  1. R is strongly nil clean.

  2. R is strongly (x2 − (2b + 1)x + b(b + l))-nil clean.

  3. R is strongly (x2 − (2b + 1 − 2n)x + (b2 + b(1 − 2n))-nil clean.

  4. R is strongly (x2 − (2b + 1 + 2n) x + (b2 + b(1 + 2n))-nil clean.

Proof

In Theorem 3.1, we take a = b to get (1) ⇔ (2), a = b − 2n to get (1) ⇔(3) and a = b + 2n to get (1) ⇔(4).     □

For example, strongly nil clean, strongly (x2 − 3x + 2)-nil clean, strongly (x2 − 5x + 6)-nil clean, strongly (x2 + x)-nil clean, strongly (x2 + 3x)-nil clean and strongly (x2 − 3x)-nil clean rings are all equivalent.

Remark 3.4

The equivalence in Theorem 3.1 of strongly nil clean rings and (x − (a + 1))(xb)-nil clean rings is a global property. That is, it holds for a ring R but it may fail for a single element. For example, 1 ∈ ℤ10 is (strongly) nil clean but it is not (strongly) (x2 + x)-nil clean in10.

Let R be a strongly g(x)-nil clean ring and suppose 0 = s + n where nN(R), g(s) = 0 and ns = sn. If s is a unit, then clearly R = {0}. In particular, if g(x) = xn − 1 or g(x) = xn + 1, then the only strongly g(x)-nil clean ring is the zero ring. Therefore, we give the following definition.

Definition 3.5

A ring R is called strongly g(x)-nil* clean if for every 0 ≠ rR, r = s + n where nN(R), g(s) = 0 and ns = sn.

It is clear that every strongly g(x)-nil clean ring is strongly g(x)-nil*clean. Moreover, the converse is true if there is a non unit root s of g(x) such that 0 = s + b for some bN(R). In particular, strongly g(x)-nil clean rings and strongly g(x)-nil* clean rings coincide if all roots of g(x) are non unit. However, there are examples of strongly g(x)-nil*clean rings which are not strongly g(x)-nil clean.

Example 3.6

For any prime integer p, the fieldp is strongly (xp − 1 − 1)-nil* clean which is not strongly (xp − 1 − 1)-nil clean. In fact by Fermat’s Theorem, any non zero element inp is a root of xp − 1−1 and sop is strongly (xp − 1−1)-nil* clean. On the other hand, the element 0 can’t be written as a sum of a nilpotent and a root of xp − 1 − 1. Therefore, ℤp is not (xp − 1 − 1)-nil clean.

In general, we prove in the next example that for any polynomial g(x) ∈ C(R)[x] with a unit constant term, R is not strongly g(x)-nil clean.

Example 3.7

Let g(x) = xn + an − 1xn − 1 + … + a1x + a0C(R)[x] where R is a non zero ring and n ∈ ℕ. If a0U(R), then R is not strongly g(x)-nil clean. In particular, R is not strongly (xn ± 1)-nil clean.

Proof

Suppose R is strongly g(x)-nil clean. If we write 0 = s + r where rN(R), g(s) = 0 and rs = sr, then s(sn − 1 + an − 1sn−2 + … + a1) = −a0U(R) and so sU(R) . But, also s = −rN(R), so 1 = 0 and R = {0}, a contradiction.     □

Next, for any n ∈ ℕ, we give in the following proposition a characterization of strongly (xn − 1)-nil*clean rings. The proof is trivial.

Proposition 3.8

Let R be a ring and n ∈ ℕ. Then R is strongly (xn − 1)-nil* clean if and only if for every 0 ≠ rR, r = v + b where bN(R), vn = 1 and vb = bv.

Since in any ring, the sum of a unit and a nilpotent that commute is again a unit, then we conclude by Proposition 3.8 that any strongly (xn − 1)-nil* clean ring is a division ring.

Proposition 3.9

Let R be a ring and 2 ≤ n ∈ ℕ. If R is strongly (xn − 1 − 1)-nil* clean, then R is strongly (xnx)- nil clean. Moreover, the converse is true if for every 0 ≠ aR, (a + 1)R (or\ R(a + 1)) does not contain non trivial idempotents.

Proof

Suppose R is strongly (xn − 1 − 1)-nil*clean. If r = 0, then clearly r is a strongly (xnx)-nil clean element. Suppose 0 ≠ rR. Then r = v + a where aN(R) and vn − 1 = 1 and so v is a root of xnx. Therefore, R is strongly (xnx)-nil clean. Now, suppose that for every aR, (a + 1)R contains no non trivial idempotents and R is strongly (xnx)-nil clean. Let 0 ≠ rR and write r = s + a where aN(R), sn = s and sa = as. Then rsn − 1 = s + asn − 1 and so r (1 − sn − 1) = a(1 − sn − 1). If we set y = 1 + a, then yU(R) and (r + 1)(1 − sn − 1) = (a + 1)(1 − sn − 1) = y (1 − sn − 1). Now, clearly y(1 − sn − 1)y− 1 is an idempotent with y(1 − sn − 1)y− 1 = (r + 1)(1 − sn − 1)y− 1 ∈ (r + 1)R. By assumption, y(1 − sn − 1)y− 1 = 0 and so 1 − sn − 1 = 0. Therefore, s is a root of xn − 1 − 1 and R is strongly (xn − 1 − 1)-nil*clean.     □

For a ring R and n ∈ ℕ, let Un (R) denotes the set of elements in R that can be written as a sum of no more than n units.

Proposition 3.10

Let R be a ring and g(x) = xn + an−1xn − 1 + … + a1x + a0C(R)[x] where n ∈ ℕ and a0U(R). If R is strongly g(x)-nil* clean, then R = U2(R). In particular if R is strongly (xn − 2 + xn − 3 + … + x + 1)-nil* clean, then R = U2(R) is strongly (xnx)-nil clean.

Proof

Let 1 ≠ rR and write r − 1 = s + b where bN(R), sn + an − 1 sn − 1 + … + a1s + a0 = 0 and sb = bs. Then r = s + (b + 1) where b + 1 ∈ U(R). Since also, s(sn − 1 + an − 1 sn − 2 + … +a1) = −a0U(R), then sU(R). Thus, rU2(R). Since also 1 ∈ U2(R), then R = U2(R). Now, suppose R is strongly (xn − 2 + xn − 3 + … + x + 1)-nil* clean. Then R = U2(R) by taking a0=1 ∈ U(R). If r = 0, then r is clearly a strongly (xnx)-nil clean element. Let 0 ≠ rR and write r = s + b where bN(R), sn − 2 + sn − 3 + … + s + 1 = 0 and sb = bs. Then sns = s(s − 1)(sn − 2 + sn − 3 + … + s + 1) = 0 and so R is strongly (xnx)-nil clean.     □

By choosing n = 4 in Proposition 3.10, we conclude that if R is strongly (x2 + x + 1)-nil*clean, then R = U2(R) is strongly (x4x)-nil clean. Also, we have R = U2(R) if R is strongly (xn ± 1)-nil*clean.

Lemma 3.11

Let R be a Boolean ring and G any cyclic group of order p (prime). Then all elements of the group ring RG are roots of g(x) = x2p−1x.

By taking for example p = 3 in Lemma 3.11, we see that the group ring ℤ2(C3) is (strongly) (x4x)-nil clean which is not strongly nil clean. Moreover, the field ℤ3 is (strongly) (x3x)-nil clean which is not strongly nil clean. Therefore, strongly (xnx)-nil cleanness of a ring R does not imply strongly nil cleanness of R whether n is odd or even.

In the following proposition, we can see that (xnx)-nil clean and (xn + x)-nil clean rings coincide for any even integer n. The proof can be easily verified.

Proposition 3.12

Let R be a ring, n ∈ ℕ and a, bR. Then R is strongly (ax2nbx)-nil clean if and only if R is strongly (ax2n + bx)-nil clean.

By Proposition 3.12 and Lemma 3.11, we conclude that ℤ2(Cp) is also (x2p−1 + x)-nil clean for any prime p. However, one can see that Proposition 3.12 need not be true for odd degrees. For example, the field ℤ3 is strongly (x3x)-nil clean which is not strongly (x3 + x)-nil clean.

Recall that an element p in a ring R is called periodic if pk = pm for some distinct positive integers k and m. Equivalently, p is periodic if p is a root of g(x) = x2nxn for some positive integer n. R is called periodic if every element in R is periodic. In the following Lemma, we can see a sufficient condition for a ring to be periodic.

Lemma 3.13

([3]). Let R be a ring. If for each xR, there is n ∈ ℕ and p(x) ∈ ℤ[x] such that xn = xn+1p(x), then R is periodic.

In the following Theorem, periodic rings are characterized as strongly (xnx)-nil clean rings for some n ∈ ℕ. This Theorem is essentially proved in [13]. However, for the sake of completeness, we shall give the proof.

Theorem 3.14

The following are equivalent for a ring R.

  1. R is periodic.

  2. R is a strongly (xnx)-nil clean rings for some n ∈ ℕ.

  3. For every rR, there exists k ∈ ℕ and a ∈ 〈rsuch that rk = rk+1a.

Proof

(1) ⇒ (2) : Suppose R is periodic and let rR. Then there exists k ∈ ℕ such that r2k = rk. Let s = rk+1 and b = rrk+1. Then r = s + b where clearly s is a root of g(x) = xk+1x, bk = 0 and sb = bs. Therefore, R is a strongly (xk+1x)-nil clean.

(2)⇒ (3) : Suppose R is strongly (xnx)-nil clean for some n ∈ ℕ. Let rR and write r = s + b where sns = 0, bN(R) and sb = bs. Now, rrn = (s + b) − (s + b)n = b + bt where t ∈ 〈s, b〉 and so rrnN(R). By simple computations, there exists k ∈ ℕ such that rk = rk+1 a where a ∈ 〈r〉.

(3) ⇒ (1) : Lemma 3.13. □

References

  • [1]

    Breaz S., Călugăreanu G., Danchev P., Micu T., Nil-clean matrix rings, Linear Algebra and its Applications, 2013, 439(10), 3115–3119. Web of ScienceCrossrefGoogle Scholar

  • [2]

    Camillo V.P., Simón J.J., The Nicholson-Varadarajan theorem on clean linear transformations, Glasgow Math. J., 2002, 44, 365-369.CrossrefGoogle Scholar

  • [3]

    Chacron M., On a theorem of Herstein, Cand. J. Math., 1969, 21, 1348-1353.CrossrefGoogle Scholar

  • [4]

    Chen H., Strongly nil clean matrices over R[x]/(x2 – 1), Bull. Korean Math. Soc, 2012, 49(3), 589-599. CrossrefGoogle Scholar

  • [5]

    Chen H., On Strongly Nil Clean Matrices, Communications in Algebra, 2013, 41(3), 1074-1086.Web of ScienceCrossrefGoogle Scholar

  • [6]

    Diesl A.J., Classes of Strongly Clean Rings, PhD thesis, University of California, Berkeley, 2006. Google Scholar

  • [7]

    Diesl A.J., Nil clean rings, Journal of Algebra, 2013, 383, 197-211. CrossrefGoogle Scholar

  • [8]

    Fan L., Yang X., On rings whose elements are the sum of a unit and a root of a fixed polynomial, Communications in Algebra, 2008, 36, 269-278. CrossrefWeb of ScienceGoogle Scholar

  • [9]

    Khashan H.A., Handam A.H, g(x)- nil clean rings, Scienticae Mathematicae Japonicae, 2016, 2, 145-154. Google Scholar

  • [10]

    Nicholson W.K., Lifting idempotents and exchange rings, Transactions of the American Mathematical Society, 1977, 229, 269–278. CrossrefGoogle Scholar

  • [11]

    Nicholson W.K., Strongly clean rings and Fitting’s lemma, Communications in Algebra, 1999, 27, 3583–3592. CrossrefGoogle Scholar

  • [12]

    Nicholson W.K., Zhou ., Endomorphisms that are the sum of a unit and a root of a fixed polynomial, Canad. Math. Bull., 2006, 49, 265-269. CrossrefGoogle Scholar

  • [13]

    Ôhori M., On strongly π-regular rings and periodic rings, Mathematical Journal of Okayama University, 1985, 27(1), 49-52. Google Scholar

  • [14]

    Sahebi S., Jahandar M., A note on uniquely (nil) clean ring, Journal of Linear and Topological Algebra, 2012, 1(2), 65-67. Google Scholar

About the article

Received: 2016-01-12

Accepted: 2017-02-02

Published Online: 2017-04-17


Citation Information: Open Mathematics, Volume 15, Issue 1, Pages 420–426, ISSN (Online) 2391-5455, DOI: https://doi.org/10.1515/math-2017-0031.

Export Citation

© 2017 Handam and Khashan. This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License. BY-NC-ND 3.0

Comments (0)

Please log in or register to comment.
Log in