Some results on MP-injectivity and MGP-injectivity of rings and modules

We study MP-injective rings and MGP-injective rings satisfying some additional conditions. Using the concepts of MP-injectivity and MGP-injectivity of rings and modules, we present some new characterizations of QF-rings, semisimple Artinian rings, strongly regular rings, and simple Artinian rings.

Gespeichert in:
Bibliographische Detailangaben
Datum:2011
Hauptverfasser: Zhanmin, Zhu, Чжанмін, Чжу
Format: Artikel
Sprache:Englisch
Veröffentlicht: Institute of Mathematics, NAS of Ukraine 2011
Online Zugang:https://umj.imath.kiev.ua/index.php/umj/article/view/2815
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Назва журналу:Ukrains’kyi Matematychnyi Zhurnal
Завантажити файл: Pdf

Institution

Ukrains’kyi Matematychnyi Zhurnal
_version_ 1860508794075742208
author Zhanmin, Zhu
Чжанмін, Чжу
author_facet Zhanmin, Zhu
Чжанмін, Чжу
author_sort Zhanmin, Zhu
baseUrl_str https://umj.imath.kiev.ua/index.php/umj/oai
collection OJS
datestamp_date 2020-03-18T19:37:09Z
description We study MP-injective rings and MGP-injective rings satisfying some additional conditions. Using the concepts of MP-injectivity and MGP-injectivity of rings and modules, we present some new characterizations of QF-rings, semisimple Artinian rings, strongly regular rings, and simple Artinian rings.
first_indexed 2026-03-24T02:30:52Z
format Article
fulltext К О Р О Т К I П О В I Д О М Л Е Н Н Я UDC 512.5 Zhanmin Zhu (Jiaxing Univ., China) SOME RESULTS ON MP-INJECTIVITY AND MGP-INJECTIVITY OF RINGS AND MODULES ДЕЯКI РЕЗУЛЬТАТИ ПРО MP-IН’ЄКТИВНIСТЬ ТА MGP-IН’ЄКТИВНIСТЬ КIЛЕЦЬ ТА МОДУЛIВ We study MP-injective rings and MGP-injective rings satisfying some additional conditions. Using the concepts of MP-injectivity and MGP-injectivity of rings and modules, we present some new characterizations of QF- rings, semisimple Artinian rings, strongly regular rings, and simple Artinian rings. Вивчаються MP-iн’єктивнi та MGP-iн’єктивнi кiльця, що задовольняють деякi додатковi умови. Iз засто- суванням понять MP-iн’єктивностi та MGP-iн’єктивностi кiлець та модулiв наведено новi характеризацiї QF-кiлець, напiвпростих кiлець Артiна, сильно регулярних кiлець та простих кiлець Артiна. 1. Introduction. Throughout this article, R is an associative ring with an identity. For a subset X of R, the right and left annihilators of X are denoted by r(X) and l(X), respectively. To facilitate, r(a) is called a special right annihilator of R for each a ∈ R. The Jacobson radical of R is denoted by J = J(R), the right singular ideal of R is denoted by Zr = Z(RR). The right socle of R is denoted by Sr = Soc(RR). Let M be an R-module and N be a submodule of M, following [1], we write N ⊆ess M to indicate thatN is an essential submodule ofM. Concepts which have not been explained can be found in [1] and [2]. Recall that a ring R is right P-injective [3] if every R-homomorphism from a princi- pal right ideal of R to R extends to an endomorphism of R. A ring R is right generalized principally injective (briefly right GP-injective) [4] if, for any 0 6= a ∈ R, there exists a positive integer n such that an 6= 0 and any right R-homomorphism from anR to R extends to an endomorphism of R. GP-injective rings are studied in papers [4 – 8]. In [8], GP-injective rings are called YJ-injective rings. In [2], the concepts of right P-injective rings and right GP-injective rings are gen- eralized to right MP-injective rings and right MGP-injective rings, respectively, and some interesting results on these rings are obtained. Following [2], a right R-module N is MP-injective if, for every R-monomorphism from a principal right ideal of R to N extends to a homomorphism of R to N, the ring R is right MP-injective if RR is MP-injective; a right R-module N is MGP-injective if, for any 0 6= a ∈ R, there exists a positive integer n such that an 6= 0 and any R-monomorphism from anR to N extends to a homomorphism of R to N, the ring R is right MGP-injective if RR is MGP-injective. In this paper, we shall study some new properties of MP-injective rings and MGP-injective rings, and give some new characterizations of QF-rings rings, semisimple artinian rings, von Neumann regular rings, strongly regular rings and simple artinian rings by MP-injectivity and MGP-injectivity of rings and modules. 2. Results. Recall that a ring R is QF if it is right or left self-injective and right or left artinian, a ring R is semiregular if R/J(R) is von Neumann regular and idempotents can be lifted modulo J(R), a ring R is right CF if every cyclic right R-module embeds in c© ZHANMIN ZHU, 2011 1426 ISSN 1027-3190. Укр. мат. журн., 2011, т. 63, № 10 SOME RESULTS ON MP-INJECTIVITY AND MGP-INJECTIVITY . . . 1427 a free module, a ring R is right mininjective if every R-homomorphism from a minimal right ideal of R to R extends to an endomorphism of R. These concepts can be found in [1]. It is well known that right CF-rings are left P-injective [1] (Lemma 7.2 (1)); and a ring R is QF if and only if R is right artinian and right and left mininjective [9] (Corollary 4.8). According to [10], a ring R is right 2-simple injective if every R- homomorphism from a 2-generated right ideal of R to R with simple image extends to an endomorphism of R. Theorem 2.1. Let R be a right MGP-injective ring. Then the following statements are equivalent: (1) R is a QF-ring; (2) R is a right 2-simple injective ring with ACC on right annihilators; (3) R is right CF-ring and the ascending chain r(a1) ⊆ r(a2a1) ⊆ r(a3a2a1) ⊆ . . . terminates for every sequence {a1, a2, . . .} ⊆ R; (4) R is a semiregular right CF-ring. Proof. (1)⇒ (2). Since a QF-ring is right self-injective and right noetherian, so (1) implies (2). (2)⇒ (1). Suppose (2) holds. Then since R is a right MGP-injective ring with ACC on right annihilators, by [2] (Corollary 3.12(1)), R is semiprimary. Noting that R is right 2-simple injective, by [10] (Theorem 17(17)), R is a QF-ring. (1)⇒ (3). Assume (1). Then since every injective module over a QF-ring is projec- tive, so every right R-module embeds in a free module, and hence R is a right CF-ring. Note that a QF-ring is right noetherian, the last assertion of (3) is clear. (3)⇒ (4). By [2] (Theorem 3.11), R is right perfect, so that it is semiregular. (4) ⇒ (1). Note that right MGP-injectivity implies that J(R) = Zr by [2] (The- orem 3.4(2)), so R is right artinian by [11] (Corollary 2.9). Since R is right and left mininjective, by [9] (Corollary 4.8), R is QF. Theorem 2.1 is proved. Corollary 2.1 ([12], Corollary 3). The following statements are equivalent for a ring R: (1) R is a QF-ring; (2) R is a right 2-injective ring with ACC on right annihilators. Lemma 2.1. Let R be a left noetherian ring. If I is an ideal of R and r(I) ⊆ess RR, then I is nilpotent. Proof. Since R is left noetherian and r(Ii) is an ideal for each positive integer i, there exists k ≥ 1 such that r(Ik) = r(Ik+1) = . . . . If I is not nilpotent, choose l(x) maximal in {l(y) | Iky 6= 0}. Then I2kx 6= 0 because r(I2k) = r(Ik), so there exists a ∈ Ik such that Ikax 6= 0. Since r(I) ⊆ r(Ik) and r(I) ⊆ess RR, we have that r(Ik) ⊆ess RR. Thus axR ∩ r(Ik) 6= 0, say 0 6= axb ∈ r(Ik), then, Ikxb 6= 0 and Ika ⊆ l(xb) but Ika * l(x), which contradicts the maximality of l(x). Therefore I is nilpotent. Lemma 2.1 is proved. Theorem 2.2. Let R be a left noetherian right MGP-injective ring. Then: (1) r(J) ⊆ess RR; (2) J is nilpotent; (3) r(J) ⊆ess RR; (4) lr(J) = J. ISSN 1027-3190. Укр. мат. журн., 2011, т. 63, № 10 1428 ZHANMIN ZHU Proof. (1). Let 0 6= x ∈ R. Since R is left noetherian, the non-empty set F = = {l((xa)k) | a ∈ R, k > 0 such that (xa)k 6= 0} has a maximal element, say l((xy)n). We claim that J(xy)n = 0. If not, then there exists t ∈ J such that t(xy)n 6= 0. Since R is right MGP-injective, there exists a positive integer m such that (t(xy)n)m 6= 0 and b ∈ R(t(xy)n)m for every b ∈ R with r((t(xy)n)m) = r(b). Write (t(xy)n)m = = s(xy)n, where s = (t(xy)n)m−1t ∈ J. We proceed with the following two cases. Case 1: r((xy)n) = r(s(xy)n). Then (xy)n = cs(xy)n, i.e., (1 − cs)(xy)n = 0. Since s ∈ J, 1− cs is invertible. So we have (xy)n = 0. This is a contradiction. Case 2: r((xy)n) 6= r(s(xy)n). Then there exists u ∈ r(s(xy)n) but u /∈ r((xy)n). Thus, s(xy)nu = 0 and (xy)nu 6= 0. This shows that s ∈ l((xy)nu) and l((xy)nu) ∈ F. Noting that s /∈ l((xy)n), so the inclusion l((xy)n) ⊂ l((xy)nu) is strict. This contracts the maximality of l((xy)n) in F. Thus, J(xy)n = 0, and so 0 6= (xy)n ∈ xR ∩ r(J), proving (1). (2). By (1) and Lemma 2.1. (3). If 0 6= c ∈ R, we must show that Rc ∩ r(J) 6= 0. This is clear if Jc = 0. Otherwise, since J is nilpotent by (2), there exists m ≥ 1 such that Jmc 6= 0 but Jm+1c = 0. Then 0 6= Jmc ⊆ Rc ∩ r(J), as required. (4). By (1) and [2] (Theorem 3.4), lr(J) ⊆ Zr = J, so that lr(J) = J. Theorem 2.2 is proved. Theorem 2.3. Let R be a left noetherian right MGP-injective ring. Then the fol- lowing statements are equivalent: (1) R is right Kasch; (2) R is left C2; (3) R is left GC2; (4) R is semilocal; (5) R is left artinian; (6) the ascending chain r(a1) ⊆ r(a2a1) ⊆ r(a3a2a1) ⊆ . . . terminates for every sequence {a1, a2, . . .} ⊆ R. Proof. (1)⇒ (2). By [1] (Proposition 1.46). (2)⇒ (3); and (5)⇒ (6) are obvious. (3) ⇒ (4). Since left noetherian ring is left finite dimensional, and left finite dimen- sional left GC2 ring is semilocal [13] (Lemma 1.1), so (4) follows from (3). (4) ⇒ (5). Since R is left noetherian right MGP-injective, by Theorem 2.2(2), J is nilpotent. And so R is left noetherian and semiprimary by hypothesis, as required. (5) ⇒ (1). Assume (5). Then R is semiperfect right mininjective ring and Sr ⊆ess RR. So that R is a right minfull ring. By [1] (Theorem 3.12), R is right Kasch. (6)⇒ (4). By [2] (Theorem 3.11). Theorem 2.3 is proved. Corollary 2.2. Let R be a left noetherian right MGP-injective right finite dimen- sional ring. Then R is left artinian. Proof. Since R is right MGP-injective, by [2] (Theorem 3.4(1)), R is right GC2. But right finite dimensional right GC2 ring is semilocal, so R is left artinian by Theorem 2.3. Corollary 2.3. The following statements are equivalent for a ring R: (1) R is a QF-ring; (2) R is left artinian and right 2-injective; ISSN 1027-3190. Укр. мат. журн., 2011, т. 63, № 10 SOME RESULTS ON MP-INJECTIVITY AND MGP-INJECTIVITY . . . 1429 (3) R is left noetherian right 2-injective and right Kasch; (4) R is a left noetherian right 2-injective semilocal ring; (5) R is left noetherian right 2-injective and left C2; (6) R is left noetherian right 2-injective and left GC2; (7) R is left noetherian right 2-injective and the ascending chain r(a1) ⊆ r(a2a1) ⊆ ⊆ r(a3a2a1) ⊆ . . . terminates for every sequence {a1, a2, . . .} ⊆ R; (8) R is left noetherian right 2-injective and right finite dimensional. Proof. By Theorem 2.3 (2) through (7) are equivalent. (1)⇒ (8) is clear. (8)⇒ (2) by Corollary 2.2. (2) ⇒ (1) by [10] (Theorem 17). Lemma 2.2. Let M be a right R-module and NR ⊆ess MR. Then (N : x) ⊆ess RR for all x ∈M, where (N : x) = {a ∈ R | xa ∈ N}. Proof. Let x ∈ M. For each 0 6= a ∈ R, if xa = 0, then a ∈ (N : x), thus 0 6= aR = (N : x) ∩ aR. If xa 6= 0, then since N ⊆ess M, N ∩ xaR 6= 0, so that there exists 0 6= xar ∈ N, and thus 0 6= ar ∈ (N : x) ∩ aR. Hence, (N : x) ⊆ess RR. Lemma 2.2 is proved. Theorem 2.4. The following conditions are equivalent for a ring R : (1) R is a semisimple artinian ring; (2) R is right Kasch and every simple right R-module is MGP-injective; (3) R is right Kasch and every simple right R-module is mininjective. Proof. It is obvious that (1)⇒ (2)⇒ (3). (3) ⇒ (1). For any right R-module A, let E(A) be the injective hull of A. If A 6= 6= E(A), then there exists x ∈ E(A) − A. By Lemma 2.2, we have (A : x) ⊆ess RR. Clearly, (A : x) 6= R. Thus there exists a maximal right ideal M of R such that (A : x) ⊆ M. Clearly, M ⊆ess RR. Since R is right Kasch, there exists 0 6= a ∈ R such that M = r(a). Now we define f : aR → R/r(a); ay 7→ y + r(a), then f is a right R-homomorphism. Since aR is a minimal right ideal and R/r(a) is a simple right R-module, by hypothesis, there is b ∈ R such that 1 + r(a) = f(a) = ba+ r(a), which yields that 1 − ba ∈ r(a), and so a = aba. Let e = ba, then 0 6= e = e2. It follows that M = r(e) = (1 − e)R, and then M ∩ eR = 0 but eR 6= 0, which contradicts that M ⊆ess RR. Hence, A = E(A), i.e., A is injective. Therefore R is a semisimple artinian ring. Theorem 2.4 is proved. The following Lemma 2.3 (1) and (2) are well-known results, we give their proof here for completeness. Lemma 2.3. Let R be a prime ring, then: (1) if I is a nonzero ideal of R, then I is essential in R both as a left ideal and as a right ideal; (2) if R is a semisimple artinian ring, then it is a simple artinian ring; (3) if R satisfies the ascending chain condition for special right annihilators, then Zr = 0. Proof. (1). If K is a right ideal of R satisfies K ∩ I = 0, then KI ⊆ K ∩ I = 0. Since R is a prime ring, K = 0, and so I is an essential right ideal of R. Similarly, I is an essential left ideal of R. (2). Let I be a nonzero ideal of R. Then since R is a semisimple artinian ring, there exists a right ideal T of R such that I ⊕ T = R. By (1), T = 0, and thus I = R. This proves that R is a simple artinian ring. ISSN 1027-3190. Укр. мат. журн., 2011, т. 63, № 10 1430 ZHANMIN ZHU (3). Since R satisfies the ascending chain conditions for special right annihilators, the set {r(x) | 0 6= x ∈ R} has a maximal element r(a). If Zr 6= 0, then aZra 6= 0 because R is a prime ring (otherwise, if aZra = 0, then aZr(aR) = 0, and so aZr = 0, i.e., (Ra)Zr = 0, which implies that Zr = 0, a contradiction). Thus there is b ∈ Zr such that aba 6= 0. It follows from the maximality of r(a) that r(a) = r(aba). Since aba ∈ Zr, we have r(a) = r(aba) ⊆ess RR, and whence r(a) ∩ baR 6= 0. So that there exists c ∈ R such that bac 6= 0 and abac = 0, which implies that c ∈ r(aba) = r(a). Thus ac = 0, and then bac = 0 which contradicts bac 6= 0. Therefore Zr = 0. Lemma 2.3 is proved. Theorem 2.5. The following statements are equivalent for a ring R: (1) R is a simple artinian ring; (2) R is a right MGP-injective prime ring such that the ascending chain r(a1) ⊆ ⊆ r(a2a1) ⊆ r(a3a2a1) ⊆ . . . terminates for every sequence {a1, a2, . . .} ⊆ R; (3) R is a prime ring such that 0 6= Sr is MP-injective, and R satisfies the ascending chain condition for special right annihilators. Proof. It is obvious that (1)⇒ (2) and (3). (2)⇒ (1). By [2] (Theorem 3.17) and Lemma 2.3(2). (3) ⇒ (1). We first prove that R is a semisimple artinian ring. If not, then Sr 6= R. Since R satisfies the ascending chain conditions for special right annihilators, the set {r(x) | x ∈ R − Sr} has a maximal element r(a). By Lemma 2.3(3), there exists a nonzero right ideal T of R such that r(a)⊕T ⊆ess RR. By Lemma 2.3(1), T ∩Sr 6= 0, so that there exists 0 6= b ∈ T ∩ Sr. Now we define f : abR → Sr; abx 7→ bx, then f is a right R-monomorphism. Since Sr is MP-injective, then there is y ∈ Sr such that b = f(ab) = yab, which implies that (a−aya)b = 0, i.e., b ∈ r(a−aya). Since a /∈ Sr and y ∈ Sr, a− aya /∈ Sr. By the maximality of r(a), we have r(a) = r(a− aya). It follows that ab = 0, and so b = yab = 0, which contradicts b 6= 0. Therefore, Sr = R, i.e., R is a semisimple artinian ring. Since R is a prime ring, by Lemma 2.3(2), R is a simple artinian ring. Theorem 2.5 is proved. Recall that a ring R is a right SF ring if every simple right R-module is flat, a ring R is a right quasi-duo ring if every maximal right ideal of R is an ideal, a ring R is a quasi-duo ring if it is left or right quasi-duo. These concepts can be found in [14]. Proposition 2.1. If every maximal essential right ideal of R is MP-injective, then R is a right SF ring. Proof. Let S be a simple right R-module, then there exists a maximal right ideal M of R such that S ∼= R/M. If M is essential right ideal, then by hypothesis, M is MP -injective. So for any a ∈ R, if y = xa ∈ Ra ∩M, then since the inclusion mapping yR → M extends to a right R-homomorphism f : R → M, so that y = = f(y) = f(1)y = (f(1)x)a ∈ Ma. Hence Ra ∩M = Ma, this shows that M is a pure submodule of R, and therefore R/M is flat. Proposition 2.1 is proved. Definition 2.1. Let R be a ring. A right R-module N is WMGP-injective if, for any a ∈ R, there exists a positive integer n such that any R-monomorphism from anR to N extends to a homomorphism of R to N. The ring R is right WMGP-injective if RR is WMGP-injective. ISSN 1027-3190. Укр. мат. журн., 2011, т. 63, № 10 SOME RESULTS ON MP-INJECTIVITY AND MGP-INJECTIVITY . . . 1431 Example 2.1. Let R = {[ a v 0 a ]∣∣∣∣ a ∈ F, v ∈ V} be the trivial extension of the field F by the two-dimensional vector space V over F. Then R is a commutative WMGP- injective ring that is not MGP-injective. Proof. Let V = uF ⊕ wF. For any x ∈ R, write x = [ a v 0 a ] . If a 6= 0, then x is invertible, so xR = R, and thus any R-homomorphism from xR to R extends to an endomorphism of R. If a = 0, then x2 = 0, and so any R-homomorphism from x2R to R extends to an endomorphism of R. Hence, R is WMGP-injective. Let x0 = [ 0 u 0 0 ] , y0 = [ 0 w 0 0 ] , then x02 = 0, r(x0) = r(y0) = [ 0 V 0 0 ] , but Rx0 = [ 0 Fu 0 0 ] and Ry0 = [ 0 Fw 0 0 ] . So Ry0 * Rx0. This shows that the R-monomorphism from x0R to R via x0r 7→ y0r can not be extended to an endomorphism of R. whence R is not MGP-injective. Proposition 2.1 is proved. Next, we give some new characterizations of strongly regular rings. Theorem 2.6. The following conditions are equivalent for a ring R : (1) R is a strongly regular ring; (2) every maximal right ideal of R is MGP-injective and l(a) is an ideal for each a ∈ R; (3) R is a reduced ring and every maximal essential right ideal of R is MGP- injective; (4) R is a reduced ring and every maximal essential right ideal of R is WMGP- injective or a right annihilator; (5) R is a quasi-duo ring, and every maximal essential right ideal of R is MP- injective. Proof. (1)⇒ (2). Since R is a strongly regular ring, by [15] (Proposition 12.3), R is von Neumann regular and every left ideal is two-sided, so (2) holds. (2) ⇒ (3). We need only to prove that R is reduced. Let a ∈ R with a2 = 0, we claim that a = 0. Otherwise, if a 6= 0, then a ∈ l(a) 6= R. By (2), l(a) is an ideal, so there exists a maximal right ideal M such that l(a) ⊆ M. Since M is MGP-injective, the inclusion mapping aR → M extends to a homomorphism from R to M, and so there exists b ∈ M such that a = ba. Thus 1 − b ∈ l(a) ⊆ M, and then 1 ∈ M, a contradiction. Therefore a = 0, and hence R is reduced. (3) ⇒ (4). Since right MGP-injective module is right WMGP-injective, so (3) implies (4). (4)⇒ (1). For any a ∈ R, we claim that aR+r(a) = R. In fact, if aR+r(a) 6= R, then there exists a maximal right ideal M of R such that aR + r(a) ⊆ M. We claim that M is an essential right ideal. Otherwise, there there exists 0 6= b ∈ R such that bR ∩M = 0. Since M is a maximal right ideal, bR ⊕M = R, and so M = eR for some e2 = e ∈ R. Clearly, a = ea, i.e., 1− e ∈ l(a). Since aR is reduced, l(a) ⊆ r(a), so that 1−e ∈ r(a) ⊆M, and hence 1 ∈M, a contradiction. Therefore M is a maximal essential right ideal. By hypothesis, M is WMGP-injective or a right annihilator. Case 1: If M is WMGP-injective. Then there exists a positive integer n such that any R-monomorphism from anR to M extends to a homomorphism of R to M. Now ISSN 1027-3190. Укр. мат. журн., 2011, т. 63, № 10 1432 ZHANMIN ZHU we define f : anR → M by f(anx) = ax, where x ∈ R, noting that R is reduced, by [2] (Lemma 3.20), f is well-defined, and f is a right R-homomorphism, and so there exist u ∈ M such that a = uan. Thus, 1 − uan−1 ∈ l(a) ⊆ r(a) ⊆ M, it follows that 1 ∈M, a contradiction. Case 2: If M is a right annihilator. Then there exists 0 6= c ∈ R such that M = = r(c). Thus, c ∈ lr(c) = l(M) ⊆ l(a) ⊆ r(a) ⊆M = r(c), so that c2 = 0. Since R is reduced, c = 0, a contradiction too. Therefore, these contradictions show that aR + r(a) = R. Write 1 = as+ t, where s ∈ R, t ∈ r(a), then a = a2s+ at = a2s. Consequently, R is strongly regular. (5)⇒ (1). By Proposition 2.1 and [14] (Theorem 4.10). Theorem 2.6 is proved. Theorem 2.7. If R is a right MGP-injective ring, then it is a classical quotient ring, and so every right (left) R-module is divisible. Proof. Let l(a) = r(a) = 0. Then l(ak) = r(ak) = 0 for every positive integer k. By the right MGP -injectivity of R, there exists a positive integer n such that b ∈ Ran for every b ∈ R with r(an) = r(b), in particular, 1 = can for some c ∈ R. Thus, ancan = an, noticing that l(an) = r(an) = 0, we have anc = can = 1. Hence R is a classical quotient ring, and so every right (left) R-module is divisible. Theorem 2.7 is proved. Proposition 2.2. If every maximal essential right ideal of R is WMGP-injective or a right annihilator, then R is a classical quotient ring. Proof. Let a be a nonzero divisor of R, i.e., l(a) = r(a) = 0. Then there exists a right ideal K such that aR⊕K ⊆ess RR. We claim that aR⊕K = R. If not, then there exists a maximal right ideal M such that aR ⊕K ⊆ M, and so M is WMGP-injective or a right annihilator. If M is WMGP-injective, then there exists a positive integer n such that every monomorphism from anR to M extends to a homomorphism of R to M. Now define f : anR → M by f(anx) = ax, where x ∈ R, then f is well defined as a is a nonzero divisor, and so a = f(an) = ban for some b ∈ M. This follows that 1− ban−1 ∈ l(a) = 0, and then 1 ∈M, a contradiction. If M is a right annihilator, then since M is a maximal right ideal, there exists 0 6= t ∈ R such that M = r(t). Hence, t ∈ lr(t) = l(M) ⊆ l(a) = 0, i.e., t = 0, a contradiction too. Thus, aR ⊕ K = R. Write aR = eR, where e2 = e, then a = ea and e = ac for some c ∈ R, and so a = aca. Noting that a is a nonzero divisor, we have ac = ca = 1. This shows that R is a classical quotient ring. Proposition 2.2 is proved. At the end of this paper, we give an important property of semiprime right MGP- injective rings. Proposition 2.3. If R is a semiprime right MGP-injective ring, then R contains a unique largest reduced ideal I, and I = rl(I) = lr(I), Z(RI) = Z(IR) = 0. Proof. Let I = ∑ α∈A Iα be the sum of all reduced ideals Iα of R. It may be assumed that I 6= 0. We prove that rl(I) is reduced. Otherwise, then there exists 0 6= 6= x ∈ rl(I) such that x2 = 0. Case 1: xR ∩ Iα = 0 for all α ∈ A. Then xRIα ⊆ xR ∩ Iα = 0 for all α ∈ A, and so xRI = 0, xR ⊆ l(I). It follows that xRx = 0. But R is semiprime, x = 0, a contradiction. ISSN 1027-3190. Укр. мат. журн., 2011, т. 63, № 10 SOME RESULTS ON MP-INJECTIVITY AND MGP-INJECTIVITY . . . 1433 Case 2: There is i ∈ A such that xR ∩ Ii 6= 0. Take 0 6= a ∈ xR ∩ Ii, then aR is reduced. For any y ∈ r(a2), since (aya)2 = ay(a2y)a = 0 and aya ∈ aR, we have aya = 0, and then (ay)2 = (aya)y = 0, which implies that ay = 0. Hence, r(a2) = r(a). By the proof of [2] (Lemma 3.20), we have that r(ak) = r(a) for every positive integer k. If a2 = 0, then a = 0, a contradiction. If a2 6= 0. Since R is right MGP-injective, by [2] (Theorem 3.2), there exists a positive integer n such that a2n 6= 0 and a = ba2n for some b ∈ R. Write c = ba2n−2, then a = ca2. It is easy to see that (a − aca)2 = 0, a − aca ∈ aR, so a = aca. Let e = ac, then e2 = e, a = ea, e ∈ aR ⊆ xR. Thus, there exists d ∈ R such that e = xd, (ex)2 = ex2dx = 0. But ex ∈ aR, so ex = 0, and whence e = e2 = exd = 0, this follows that a = 0, a contradiction too. Therefore, rl(I) is reduced. Noting that rl(I) is an ideal and I ⊆ rl(I), we have I = rl(I), and so I is the unique largest reduced ideal. Since R is semiprime, it is easy to see that r(K) = l(K) for every ideal K of R. Noting that I and l(I) are ideals, we have lr(I) = ll(I) = rl(I) = I. It is obvious that Z(IR) = I ∩ Zr. Assume that I ∩ Zr 6= 0, then there exists 0 6= y ∈ I ∩ Zr. Since r(y) is an essential right ideal, r(y) ∩ yR 6= 0, and so there is 0 6= yz ∈ r(y). Thus, y2z = 0, (yzy)2 = yz(y2z)y = 0. But yzy ∈ I and I is reduced, so yzy = 0, (yz)2 = (yzy)z = 0, yz ∈ I, and hence yz = 0, which contradicts yz 6= 0. Consequently, Z(IR) = 0. Similarly, Z(RI) = 0. Proposition 2.3 is proved. 1. Nicholson W. K., Yousif M. F. Quasi-Frobenius rings. – Cambridge: Cambridge Univ. Press, 2003. 2. Zhu Z. M. MP-injective rings and MGP-injective rings // Indian J. Pure and Appl. Math. – 2010. – 41. – P. 627 – 645. 3. Nicholson W. K., Yousif M. F. Principally injective rings // J. Algebra. – 1995. – 174. – P. 77 – 93. 4. Nam S. B., Kim N. K., Kim J. Y. On simple GP-injective modules // Communs Algebra. – 1995. – 23. – P. 5437 – 5444. 5. Chen J. L., Ding N. Q. On general principally injective rings // Communs Algebra. – 1999. – 27. – P. 2097 – 2116. 6. Chen J. L., Ding N. Q. On regularity of rings // Algebra Colloq. – 2001. – 8. – P. 267 – 274. 7. Chen J. L., Zhou Y. Q., Zhu Z. M. GP-injective rings need not be P-injective // Communs Algebra. – 2005. – 33. – P. 2395 – 2402. 8. Yue Chi Ming R. On regular rings and self-injective rings. II // Glas. mat. – 1983. – 18. – P. 221 – 229. 9. Nicholson W. K., Yousif M. F. Mininjective rings // J. Algebra. – 1997. – 187. – P. 548 – 578. 10. Zhu Z. M., Chen J. L. 2-Simple injective rings // Int. J. Algebra. – 2010. – 4. – P. 25 – 37. 11. Chen J. L., Li W. X. On artiness of right CF rings // Communs Algebra. – 2004. – 32. – P. 4485 – 4494. 12. Rutter E. A. Rings with the principle extension property // Communs Algebra. – 1975. – 3. – P. 203 – 212. 13. Zhou Y. Q. Rings in which certain right ideals are direct summands of annihilators // J. Austral. Math. Soc. – 73. – 2002. – P. 335 – 346. 14. Rege M. B. On von Neumann regular rings and SF-rings // Math. Jap. – 1986. – 31. – P. 927 – 936. 15. Stenström B. Rings of quotients. – Berlin etc.: Springer, 1975. Received 14.03.11 ISSN 1027-3190. Укр. мат. журн., 2011, т. 63, № 10
id umjimathkievua-article-2815
institution Ukrains’kyi Matematychnyi Zhurnal
keywords_txt_mv keywords
language English
last_indexed 2026-03-24T02:30:52Z
publishDate 2011
publisher Institute of Mathematics, NAS of Ukraine
record_format ojs
resource_txt_mv umjimathkievua/1f/60d3fdf5d4cd2140da771b32feec291f.pdf
spelling umjimathkievua-article-28152020-03-18T19:37:09Z Some results on MP-injectivity and MGP-injectivity of rings and modules Деякi результати про MP-iн’єктивнiсть та MGP-iн’єктивнiсть кiлець та модулiв Zhanmin, Zhu Чжанмін, Чжу We study MP-injective rings and MGP-injective rings satisfying some additional conditions. Using the concepts of MP-injectivity and MGP-injectivity of rings and modules, we present some new characterizations of QF-rings, semisimple Artinian rings, strongly regular rings, and simple Artinian rings. Вивчаються MP-iн’єктивнi та MGP-iн’єктивнi кiльця, що задовольняють деякi додатковi умови. Iз застосуванням понять MP-iн’єктивностi та MGP-iн’єктивностi кiлець та модулiв наведено новi характеризацiї QF-кiлець, напiвпростих кiлець Артiна, сильно регулярних кiлець та простих кiлець Артiна. Institute of Mathematics, NAS of Ukraine 2011-10-25 Article Article application/pdf https://umj.imath.kiev.ua/index.php/umj/article/view/2815 Ukrains’kyi Matematychnyi Zhurnal; Vol. 63 No. 10 (2011); 1426-1433 Український математичний журнал; Том 63 № 10 (2011); 1426-1433 1027-3190 en https://umj.imath.kiev.ua/index.php/umj/article/view/2815/2383 https://umj.imath.kiev.ua/index.php/umj/article/view/2815/2384 Copyright (c) 2011 Zhanmin Zhu
spellingShingle Zhanmin, Zhu
Чжанмін, Чжу
Some results on MP-injectivity and MGP-injectivity of rings and modules
title Some results on MP-injectivity and MGP-injectivity of rings and modules
title_alt Деякi результати про MP-iн’єктивнiсть та MGP-iн’єктивнiсть кiлець та модулiв
title_full Some results on MP-injectivity and MGP-injectivity of rings and modules
title_fullStr Some results on MP-injectivity and MGP-injectivity of rings and modules
title_full_unstemmed Some results on MP-injectivity and MGP-injectivity of rings and modules
title_short Some results on MP-injectivity and MGP-injectivity of rings and modules
title_sort some results on mp-injectivity and mgp-injectivity of rings and modules
url https://umj.imath.kiev.ua/index.php/umj/article/view/2815
work_keys_str_mv AT zhanminzhu someresultsonmpinjectivityandmgpinjectivityofringsandmodules
AT čžanmínčžu someresultsonmpinjectivityandmgpinjectivityofringsandmodules
AT zhanminzhu deâkirezulʹtatiprompinêktivnistʹtamgpinêktivnistʹkilecʹtamoduliv
AT čžanmínčžu deâkirezulʹtatiprompinêktivnistʹtamgpinêktivnistʹkilecʹtamoduliv