Research article

Regular local hyperrings and hyperdomains

  • Received: 14 July 2022 Revised: 09 September 2022 Accepted: 14 September 2022 Published: 26 September 2022
  • MSC : 20N20, 13E05

  • This paper falls in the area of hypercompositional algebra. In particular it focuses on the class of Krasner hyperrings and it studies the regular local hyperrings. These are Krasner hyperrings R with a unique maximal hyperideal M having the dimension equal to the dimension of the vectorial hyperspace MM2. The aim of the paper is to show that any regular local hyperring is a hyperdomain. For proving this, we make use of the relationship existing between the dimension of the vectorial hyperspaces related to the hyperring R and to the quotient hyperring ¯R=Ra, where a is an element in MM2, and of the regularity of ¯R.

    Citation: Hashem Bordbar, Sanja Jančič-Rašovič, Irina Cristea. Regular local hyperrings and hyperdomains[J]. AIMS Mathematics, 2022, 7(12): 20767-20780. doi: 10.3934/math.20221138

    Related Papers:

    [1] Bin Zhou, Xiujuan Ma, Fuxiang Ma, Shujie Gao . Robustness analysis of random hyper-networks based on the internal structure of hyper-edges. AIMS Mathematics, 2023, 8(2): 4814-4829. doi: 10.3934/math.2023239
    [2] Qi Xiao, Jin Zhong . Characterizations and properties of hyper-dual Moore-Penrose generalized inverse. AIMS Mathematics, 2024, 9(12): 35125-35150. doi: 10.3934/math.20241670
    [3] Shovan Dogra, Madhumangal Pal, Qin Xin . Picture fuzzy sub-hyperspace of a hyper vector space and its application in decision making problem. AIMS Mathematics, 2022, 7(7): 13361-13382. doi: 10.3934/math.2022738
    [4] Mohammad Hamidi, Irina Cristea . Hyperideal-based zero-divisor graph of the general hyperring Zn. AIMS Mathematics, 2024, 9(6): 15891-15910. doi: 10.3934/math.2024768
    [5] Abdelaziz Alsubie, Anas Al-Masarwah . MBJ-neutrosophic hyper BCK-ideals in hyper BCK-algebras. AIMS Mathematics, 2021, 6(6): 6107-6121. doi: 10.3934/math.2021358
    [6] Shahbaz Ali, Muhammad Khalid Mahmmod, Raúl M. Falcón . A paradigmatic approach to investigate restricted hyper totient graphs. AIMS Mathematics, 2021, 6(4): 3761-3771. doi: 10.3934/math.2021223
    [7] Yamin Sayyari, Mehdi Dehghanian, Choonkil Park, Jung Rye Lee . Stability of hyper homomorphisms and hyper derivations in complex Banach algebras. AIMS Mathematics, 2022, 7(6): 10700-10710. doi: 10.3934/math.2022597
    [8] Faik Babadağ, Ali Atasoy . On hyper-dual vectors and angles with Pell, Pell-Lucas numbers. AIMS Mathematics, 2024, 9(11): 30655-30666. doi: 10.3934/math.20241480
    [9] Jia Chen, Renato De Leone . A survival tree for interval-censored failure time data. AIMS Mathematics, 2022, 7(10): 18099-18126. doi: 10.3934/math.2022996
    [10] Narjes Alabkary . Hyper-instability of Banach algebras. AIMS Mathematics, 2024, 9(6): 14012-14025. doi: 10.3934/math.2024681
  • This paper falls in the area of hypercompositional algebra. In particular it focuses on the class of Krasner hyperrings and it studies the regular local hyperrings. These are Krasner hyperrings R with a unique maximal hyperideal M having the dimension equal to the dimension of the vectorial hyperspace MM2. The aim of the paper is to show that any regular local hyperring is a hyperdomain. For proving this, we make use of the relationship existing between the dimension of the vectorial hyperspaces related to the hyperring R and to the quotient hyperring ¯R=Ra, where a is an element in MM2, and of the regularity of ¯R.



    In classical algebra, it is well known that the Krull dimension d of any Noetherian local ring R with maximal ideal M is less than or equal to the dimension of MM2. If the equality holds, the ring is called regular. One of the crucial results in the theory of regular rings states that any regular local ring is an integral domain. If the dimension of R is zero, then its maximal ideal must be trivial and by consequence R is a field. The same problem has been started to be investigated by Bordbar and Cristea [5] in the theory of Krasner hyperrings. These are ring-like hypercompositional structures, defined by Krasner in 1956, having the additive part a canonical hypergroup (this is a natural generalization of the abelian group) and the multiplicative one a semigroup with a bilaterally absorbing element 0, such that the multiplication distributes from both sides over the addition [11]. Inspired by the properties of the hyperpolynomials (they are defined in the same manner as polynomials, but with coefficients in a Krasner hyperring), Mittas [16] noticed that the multiplication of hyperpolynomials is a hyperoperation and in 1973 he introduced a new class of hyperrings, called superrings. Their additive part remains a canonical hypergroup, while the multiplicative one becomes a semihypergroup with a bilateral absorbing element and the multiplication is distributive on both sides with respect to the addition. Notice that this structure was then called by Ameri et al. [1] a strongly distributive superring, which is in fact an additive-multiplicative hyperring in the sense of Davvaz and Musavi [9]. The definition of the distributivity created some confusions in the terminology, as explained in [10], that we will also clarify in the next section of this paper. If we relax the conditions on the additive structure and consider it as a hypergroup, while the multiplicative part remains a semihypergroup, we obtain the notion of (strong) general hyperring, introduced by Vougiouklis [21]. The last type of hyperring is the multiplicative one, defined by Rota [19], where the addition is a binary operation and the multiplication is a hyperoperation.

    Inspired by similar investigations on classical algebra, Bordbar and Cristea have started the study of regular local hyperrings [5], by defining the regular parameter elements in a commutative local hyperring. They have shown that the dimension of a local hyperring R with maximal hyperideal M is equal to the height of M. Since RM is a hyperfield, the quotient hypermodule MM2 becomes a vectorial hyperspace, having the dimension greater than or equal to the dimension of the local hyperring R. In the case of equality, the hyperring is called regular. In this paper we continue to investigate on the properties of regular local hyperrings, with the aim to prove that any regular local hyperring is a hyperdomain, i.e., a commutative hyperring without divisors of zero.

    The structure of this paper is as follows. In the preliminary section we gather the basic notions related to Krasner hyperrings, dimension of a Krasner hyperring, height of hyperideals, and hypermodules. After shortly recalling the definitions of a regular local hyperring and regular parameter elements, Section 3 focuses on the study of the relationship existing between the dimensions of the vectorial hyperspaces MM2 and ¯M¯M2 related to a local Krasner hyperring R with maximal hyperideal M, where ¯M=M<a> and aMM2. First, we prove that, if R is a regular local hyperring, then ¯R=R<a> is regular local, too. Then we prove several properties concerning the dimension of R, primary hyperideals, and the dimension of the vectorial hyperspace related to R and ¯R. These help us to state and prove the main result of this article: every regular local hyperring is a hyperdomain (see Theorem 3.13). The article ends with a non-trivial example of a regular local hyperring. The conclusive section summarizes the findings of this article and their impact on the existing theory, as well as some new lines of research that this study could open.

    This section collects the main results about the Krasner hyperrings, that will help the reader to better understand the topic. For a detailed description of the theory of Krasner hyperrings, we suggest to consult the works [8,12,13,14,15,18].

    A (Krasner) hyperring is a hypercompositional structure (R,+,) where

    (1) (R,+) is a canonical hypergroup, i.e.,

    (a) (a,bRa+bR),

    (b) (a,b,cR)(a+(b+c)=(a+b)+c),

    (c) (a,bR)(a+b=b+a),

    (d) (0R)(aR)(a+0={a}),

    (e) (aR)(aR)(0a+xx=a),

    (f) (a,b,cR)(ca+bac+(b)).

    (2) (R,) is a semigroup with a bilaterally absorbing element 0, i.e.,

    (a) (a,bRabR),

    (b) (a,b,cR)(a(bc)=(ab)c),

    (c) (aR)(0a=a0=0).

    (3) The multiplication strongly distributes bilaterally over the addition:

    (a,b,cR)(a(b+c)=ab+ac and (b+c)a=ba+ca).

    If moreover (R,) is commutative, i.e.,

    (4) (a,bR)(ab=ba), then the hyperring is called commutative.

    Finally, if (R,) is a monoid, i.e.,

    (5) (1R)(aR)(a1=a=1a), we say that R is with a unit element.

    (6) A hyperring with a multiplicative identity 1, where every nonzero element is invertible, is called a hyperfield.

    Throughout this paper, R denotes a Krasner hyperring, called here by short, hyperring, unless otherwise stated.

    A nonempty set I of a hyperring R is called a right hyperideal if, for all a,bI and rR, we have abI and arI. Similarly, a left hyperideal is defined dually and a set I is called a hyperideal if it is a left and right hyperideal of R. A proper hyperideal M of a hyperring R is called maximal if M itself and R are the only hyperideals of R containing M. A hyperideal P of a hyperring R is called prime if, for every pair of elements a and b of R, the fact that abP implies either aP or bP. Moreover, a nonzero hyperring R having exactly one maximal hyperideal is called a local hyperring. A prime hyperideal P of R is called a minimal prime hyperideal over a hyperideal I of R if it is minimal (with respect to the inclusion) among all prime hyperideals of R containing I. A prime hyperideal P is called a minimal prime hyperideal if it is a minimal prime hyperideal over the zero hyperideal of R.

    A hyperring R is called Noetherian if it satisfies the ascending chain condition on the hyperideals of R, in the sense that, for every ascending chain of hyperideals I1I2I3, there exists NN such that In=IN, for every natural number nN.

    The next result fixes the form of the hyperideals of a quotient hyperring.

    Proposition 2.1. Let R be a commutative hyperring and I be a hyperideal of R. For any hyperideal J of R such that IJ, the set JI is a hyperideal of RI. Moreover, for any arbitrary element rR, we have

    r+IJIrJ.

    Let us recall now some results introduced and studied in [4,6,7] related to the dimension of a hyperring. Let P be a prime hyperideal of a commutative hyperring R. The height of P, denoted by htRP, is defined to be the supremum of the lengths of all chains

    P0P1Pn

    of prime hyperideals of R, for which Pn=P, if this supremum exists, and it is , otherwise. We say that the dimension of a hyperring R, denoted by dimR, is the supremum of the lengths of all chains of prime hyperideals of R. If R is a local hyperring with maximal hyperideal M, then dimR=htRM.

    Proposition 2.2. Let R be a commutative Noetherian hyperring and I a proper hyperideal of R generated by n elements. Then htRIn.

    On a canonical hypergroup, endowed with an external operation using elements from a Krasner hyperring, a structure of Krasner hypermodule [2,3,13] can be defined as follows.

    Let R be a hyperring with unit element 1. An R-hypermodule M is a canonical hypergroup (M,+) together with a map R×MM defined by

    (a,m)am=amM (2.1)

    such that for all a,bR and m1,m2M we have

    (1) (a+b)m1=am1+bm1.

    (2) a(m1+m2)=am1+am2.

    (3) (ab)m1=a(bm1).

    (4) a0M=0Rm1=0M.

    (5) 1m1=m1, where 0M denotes the zero element of the canonical hypergroup (M,+) and 0R the zero element of the additive part of R, which is again a canonical hypergroup. Besides, if R is a hyperfield, then M is called a vectorial hyperspace.

    This definition was given by Stratigopoulos [20] and later, Mittas [17] relaxed the first axiom, without changing the name of the structure, asking only the inclusion to be satisfied, i.e., (a+b)m1am1+bm1, in order to state that the product Fn of n-copies of a hyperfield F is an F-vectorial hyperspace with the canonical addition and multiplication by scalars of the n-tuples of elements of F. Recently, this structure has been re-named a weak hypermodule [15] (to not be confused with the Hv-modules, where the weak distributivity [22] is involved: x(y+z)xy+xz and (x+y)zxz+yz, for any three elements x,y,z). In our paper we will use the initial definition given by Statigopoulous and call the structure a Krasner hypermodule, or simply a hypermodule.

    Let V be a vectorial hyperspace over a hyperfield F. A set A of elements of V is called:

    (1) Linearly independent if for every finite set of vectors {v1,v2,,vn}A, 0ni=1rivi implies ri=0 for all i such that 1in. If A is not linearly independent, then A is called linearly dependent.

    (2) Spanning set if each vector of V is contained in a linear combination from A. We say that V is the span of A.

    A basis for a vectorial hyperspace V over a hyperfield F is a set BV which is both spanning set and linearly independent. A vectorial hyperspace is finite-dimensional if it has a finite basis. The number of elements in an arbitrary basis of a vectorial hyperspace is called the dimension of the vectorial hyperspace and we denote it by vdimFV, where "v" stays for "vectorial hyperspace". This is to differentiate it from the dimension of a hyperring R, denoted simply by dimR.

    We conclude this section by recalling one important result in hypermodule theory. It comes from the original Nakayama's lemma in classical algebra, that establishes the interactions between the Jacobson radical ideal of a commutative ring and its finitely generated module. In a commutative hyperring R, the intersection between all maximal hyperideals is called the Jacobson radical of the hyperring, denoted by Jac(R).

    Lemma 2.3. [4] (Nakayama's lemma) Let M be a finitely generated R-hypermodule and I a hyperideal of R contained by the Jacobson radical Jac(R) of R. Then M=IM implies that M={0}.

    The aim of this section is to show that every local regular hyperring is a hyperdomain. For doing this, we use elements from hypermodule theory, where by a hypermodule we mean the structure defined by Stratigopoulos, where the distributivity property is defined using the equality.

    We start with the result about the relationship existing between the dimension of a local hyperring with maximal hyperideal M and the dimension of the associated vectorial hyperspace MM2.

    Theorem 3.1. [5] Let R be a local hyperring with maximal hyperideal M. Then

    dimRvdimRMMM2.

    This naturally leads to a new definition, the one of a regular hyperring, as follows.

    Definition 3.2. [5] Let R be a local Noetherian hyperring with maximal hyperideal M. We say that R is a regular hyperring when dimR=vdimRMMM2. Consider d=dimR. By regular parameter elements of R we mean a set of d elements of R that generate an M-primary hyperideal of R.

    Some properties of a finitely generated R-hypermodule over a local hyperring R are covered in the next results.

    Proposition 3.3. [5] Let R be a local hyperring with maximal hyperideal M and consider the hyperfield F=RM. Let N be a finitely generated R-hypermodule. Then the R-hypermodule NMN has a natural structure as a hypermodule over RM as a F-vectorial hyperspace.

    Moreover, let n1,n2,,ntN. Then the following statements are equivalent.

    (i) N is generated by n1,n2,,nt.

    (ii) The R-hypermodule NMN is generated by the elements n1+MN, n2+MN,,nt+MN.

    (iii) The F-vectorial hyperspace NMN is generated by the elements n1+MN, n2+MN,,nt+MN.

    Corollary 3.3. For a local hyperring R with maximal hyperideal M and dimR=d, the following properties are fulfilled.

    (i) The dimension of the RM-vectorial hyperspace MM2 is the number of elements in each minimal generating set for the hyperideal M. Moreover, at least d elements are needed to generate M, and R is a regular hyperring when the hyperideal M can be generated by exactly d elements.

    (ii) Suppose that R is a regular hyperring and a1,a2,,adM. Using Proposition 3.3, the elements a1,a2,,ad generate M if and only if the elements a1+M2,a2+M2,,ad+M2 in MM2 form a basis for this RM-vectorial hyperspace. Equivalently, the elements a1+M2,a2+M2,,ad+M2 form a linearly independent set.

    In the next theorem, we study the relationship between the dimensions of the two vectorial hyperspaces related to a local hyperring.

    Theorem 3.5. Suppose that R is a local hyperring with maximal hyperideal M. For an arbitrary element aMM2, take ¯R=Ra, ¯M=Ma, and ¯M2=M2a. Moreover, suppose that the application nat:R¯R is the natural surjective hyperring homomorphism defined, for each rR, as follows:

    nat(r)=r+a. (3.1)

    Then, we have

    vdimRMMM2=vdim¯R¯M¯M¯M2+1.

    Proof. Since M is a unique maximal hyperideal of R, it follows that ¯R is a local hyperring with ˉM its maximal hyperideal. Hence, ¯R¯M is a hyperfield and thus ¯M¯M2 is a ¯R¯M-vectorial hyperspace. Suppose that d=vdim¯R¯M¯M¯M2. Then, using Proposition 3.2, there exist the elements m1,m2,,mdM such that their images under the composition of the following two homomorphisms, i.e., the homomorphism gf, is a basis for the ¯R¯M-vectorial hyperspace ¯M¯M2: For any i=1,2,,d,

    Mf¯Mg¯M¯M2,f(mi)=¯mi=mi+a,g(¯mi)=¯mi+¯M2. (3.2)

    Moreover, Proposition 3.3 shows that the hyperideal ¯M of ¯R is generated by d elements ¯m1,¯m2,,¯md. This means that,

    ¯M=di=1¯mi=di=1¯R¯mi=di=1Ra(mi+a)=(di=1mi+a)a.

    Then, based on Proposition 2.1, we can write

    M=di=1mi+a.

    In other words, the set {m1,m2,,md,a} is a generating set for M. Corollary 3.4 part (ii) shows that the RM-vectorial hyperspace MM2 is generated by the elements m1+M2,m2+M2,,md+M2,a+M2. In order to complete the proof, we need to prove that m1+M2,m2+M2,,md+M2,a+M2 are linearly independent elements of MM2 over the hyperfield RM. For doing this, let r1+M,r2+M,,rd+M,rd+1+M be elements in RM such that

    (r1+M)(m1+M2)++(rd+M)(md+M2)+(rd+1+M)(a+M2)=0MM2. (3.3)

    Since M is a hyperideal of R and M2M, we have

    ((r1m1)+M2)++((rdmd)+M2)+((rd+1a)+M2)=0MM2=M2,

    which means that

    r1m1+r2m2++rdmd+rd+1aM2. (3.4)

    Therefore,

    (r1m1+r2m2++rdmd+rd+1a)+a¯M2=M2a.

    Since rd+1aa, we have

    (r1m1+r2m2++rdmd)+a¯M2,

    which means that

    ((r1m1)+a)+((r2m2)+a)++((rdmd)+a)¯M2. (3.5)

    Using (3.1), we know that (rimi)+a=nat(rimi) and clearly nat(rimi)=nat(ri)nat(mi). Thus, we have the following equation in ¯M¯M2:

    (nat(r1)nat(m1)+nat(r2)nat(m2)++nat(rd)nat(md))+¯M2=¯M2,

    equivalently with

    (nat(r1)+¯M)(nat(m1)+¯M2)++(nat(rd)+¯M)(nat(md)+¯M2)=0¯M¯M2. (3.6)

    Since ¯M¯M2 is a vectorial hyperspace over the hyperfield ¯R¯M and using (3.2), it follows that {¯m1+¯M2,¯m2+¯M2,,¯md+¯M2} is a generating set for the vectorial hyperspace ¯M¯M2. Threfore the elements ¯m1+¯M2,¯m2+¯M2,,¯md+¯M2 are linearly independent in ¯R¯M. Based on (3.6) we conclude that ¯ri+¯M=0¯R¯M=¯M for 1id. Hence for each i, 1id, ¯ri¯M and by using Proposition 2.1, we have

    r1,r2,,rdM=0RM. (3.7)

    Therefore in (3.3), for each i, 1id, ri+M=M=0RM. It is enough to show that rd+1M because in this case rd+1+M=M=0RM and therefore in (3.3) the elements m1+M2,m2+M2,,md+M2,a+M2 are all linearly independent.

    For this goal, suppose that rd+1M. In this case, rd+1 should be a unit element of the hyperring R. From (3.4) and (3.7), we can conclude that rd+1aM2. Thus, a=a r1d+1 rd+1M2 which contradicts the hypothesis aMM2. Therefore, rd+1M and the set {m1+M2,m2+M2,,md+M2,a+M2} is a basis for ¯M¯M2, equivalently with

    vdim¯R¯M¯M¯M2=d+1.

    Theorem 3.6. [6] Let R be a commutative Noetherian hyperring, P a prime hyperideal of R and I a proper hyperideal of R generated by n elements, such that IP. Then:

    htRIPIhtRPhtRIPI+n.

    As a consequence of this theorem, we can state the next result.

    Theorem 3.7. Suppose that R is a local regular hyperring with maximal hyperideal M and aMM2 is an arbitrary element. Let ¯R=Ra and ¯M=Ma. Then ¯R=Ra is a regular hyperring.

    Proof. Since R is a regular local hyperring with maximal hyperideal M, using Definition 3.2, we state that

    dimR=vdimRMMM2.

    Moreover, the hyperring ¯R is a local hyperring with maximal hyperideal ¯M. Using Theorem 3.6, we conclude that

    htRM1ht¯R¯M.

    Besides, R is a local hyperring, hence dimR=htRM. Applying Theorem 3.1 to the hyperring ¯R, we get

    dimR1=htRM1ht¯R¯M=dim¯Rvdim¯R¯M¯M¯M2.

    Using now Definition 3.2 and Theorem 3.5, we conclude that

    vdim¯R¯M¯M¯M2=vdimRMMM21=dimR1.

    Thus,

    dim¯R=vdim¯R¯M¯M¯M2.

    Therefore, ¯R is a regular hyperring.

    In the following, we present two results on Noetherian hyperrings, needed in the proof of the Theorem 3.10.

    Proposition 3.8. Let R be a Noetherian hyperring, A be a hyperideal of R and B=n=1An. Then, B=AB.

    Proof. If A=R, then it is clear that B=AB. So, assume that A is a proper hyperideal of R. A routine verification shows that ABB. Since ABBAR, we conclude that AB is a proper hyperideal of R.

    R is a Noetherian hyperring, thus AB has a decomposition. Suppose that AB=P1P2Pn and r(Pi)=Qi for 1in, is a minimal primary decomposition of AB. If there exists i, where 1in, such that BPi, then we can choose an element bBPi. Besides, bAAB=P1P2PnPi. Since Pi is a primary hyperideal and bPi, there exists mN such that APmi. Thus,

    B=n=1AnAPmiPi.

    This is a contradiction. Hence, for every i where 1in, we have BPi. This means that BAB and therefore AB=B.

    Lemma 3.9. Let R be a Noetherian hyperring R and A be a hyperideal such that AJac(R). Then n=1An=0.

    Proof. Suppose that B=n=1An. Using Proposition 3.8, we have B=AB. Since R is a Noetherian hyperring, B has a finitely generated R-hypermodule structure. Thus, using Lemma 2.3, we conclude that B=0.

    Theorem 3.10. Let R be a local hyperring with maximal hyperideal M. Suppose that R is not a hyperdomain and P is a principal prime hyperideal of R. Then htRP=0.

    Proof. Suppose that P is a principal hyperideal of R generated by the element a and htRP0. Then, there exists at least a prime hyperideal Q of R such that QP. If we show that, for all nN, QPn then, using Lemma 3.9, we conclude that Q=0. We will prove this statement by induction on n. If n=1, take an arbitrary qQ. It is clear that qP and QP1. Suppose we have shown that qPn. Since P=<a>, there exists rR such that q=ranQ. Since aQ and Q is a prime hyperideal, it follows that rQ. Hence, rP and q=ranPn+1. Therefore, by induction on n, we conclude that

    Qn=1Pn.

    Lemma 3.9 shows that n=1Pn=0 and therefore Q=0. In this case R should be a hyperdomain and this is a contradiction. So htRP=0.

    Corollary 3.11. Let R be a local hyperring with maximal hyperideal M. Suppose that R is not a hyperdomain and P is a principal prime hyperideal of R. Then P is a minimal prime hyperideal.

    Proposition 3.12. Suppose that P1,P2,,Pn are hyperideals of a hyperring R such that at least n2 of them are prime hyperideals, where n2. Let A be a hyperideal of R with the property

    AP1P2Pn.

    Then, APi, for some i, 1in.

    Proof. In order to prove the statement, we will use the mathematical induction method on n. First, for n=2 suppose that AP1P2. If AP1 and AP2, then there exists xAP1 and yAP2. Since xAP1 and AP1P2, we conclude that xP2 and similarly, yP1. Besides, A is a hyperideal and x,yA, hence, xyAP1P2. For an arbitrary element zxy, we have zP1P2. Thus, zP1 or zP2. Now suppose that zP1. Since zxy, it follows that xz+y. Moreover, z and y are elements of P1 and P1 is a hyperideal. Hence, xz+yP1 which is a contradiction with xAP1.

    Similarly, the other possibility (zP2) leads us to another contradiction (yP2). Therefore, AP1 or AP2.

    Assume that k2 and that for n=k the result has been proved. Now we prove it for n=k+1. Then,

    Ak+1i=1Pi,

    where k1 hyperideals of Pi, for 1ik+1, are prime. Suppose that P1 is one of the prime hyperideals. If for each j, where 1jk+1, Ak+1i=1,ijPi, then there exists ajA such that

    ajAk+1i=1,ijPi.

    Clearly, ajPj, for each 1jk+1. Moreover, since P1 is a prime hyperideal and a2,a3,,ak+1P1 it follows that a2a3ak+1P1. Let define the set B as B=a1+a2a3ak+1. Then B can not be a subset of P1, because if BP1, then

    a2a3ak+1Ba1P1,

    that is a contradiction. Similarly, B can not be a subset of any Pt, where 2tk+1.

    Since A is a hyperideal and a1,a2,,ak+1A, a routine verification shows that

    B=a1+a2a3ak+1A.

    Let bB be an arbitrary element. Then bA and using the hypothesis Ak+1i=1Pi, we conclude that bP1P2Pk+1. Thus, bP1 or bPt for some t, where 2tk+1. In both case this is a contradiction, because B can not be a subset of P1 or Pt for some t, where 2tk+1.

    Therefore, at least for one j, with 1jk+1, we have Ak+1i=1,ijPi and using the induction we conclude that APi for some i, with 1ik+1.

    Now we have all the elements to prove the main result of this article.

    Theorem 3.13. Every regular hyperring is a hyperdomain.

    Proof. Let R be a regular hyperring of dimension d with its maximal hyperideal M. We will use the mathematical induction on d to prove this theorem. If d=0, then using Definition 3.2 and Corollary 3.4, it follows that vdimRMMM2=0 and the number of the elements in each generating set for M is 0. Hence M=0 and R should be a hyperfield. Since every hyperfield is a hyperdomain, the theorem is proved. Now suppose that d>0 and every regular hyperring of dimension less than d is a hyperdomain.

    Suppose by absurd that R is not a hyperdomain. Then M2M because

    0<d=dimR=vdimRMMM2.

    So there exists an element aMM2. Moreover, using Theorem 3.7, we conclude that the hyperring R<a>=ˉR is regular having the dimension d1. Thus, based on the induction assumption, the hyperring ˉR is a hyperdomain. Therefore, the hyperideal <a> is prime. Theorem 3.10 indicates that ht<a>=0 and so <a> is a minimal prime hyperideal.

    Since R is a Noetherian hyperring it follows that every hyperideal of R has a finite set of minimal prime hyperideals. Suppose that the set {P1,P2,,Pn} is the set of minimal prime hyperideals of 0. Then, for an arbitrary element aMM2, there exists i, 1in, such that <a>=Pi. Thus, we conclude that

    MM2P1P2Pn.

    In other words

    MM2P1P2Pn.

    Using Proposition 3.12, we conclude that MM2 or MPi, for some i, 1in. In both cases we get a contradiction since we have shown that M2M and besides if MPi for some i, where 1in, then

    dimR=htRMhtPi=0,

    which contradicts our hypothesis. Therefore, R is a hyperdomin and the proof is complete.

    In the next example, we construct a local hyperring which is also a principle hyperideal hyperdomain, containing non-trivial regular parameter elements.

    Example 3.14. Let R={0,1}{1pnpisaprimenumberandnN}. First we endow the carrier set R with a canonical hypergroup structure, by defining on R the following hyperoperation:

    x+y={max{x,y},if xy{rRrx}=[0,x]R,if x=y. (3.8)

    It is clear that this is a commutative hyperoperation. Besides, also the associativity property holds. Indeed, suppose that x,y and z are arbitrary elements of R. If x=y=z, then obviously,

    x+(y+z)=(x+y)+z=[0,x]R.

    If all three elements are different, then (x+y)+z=max{x,y}+z=max{x,y,z}. Similarly, we have x+(y+z)=x+max{y,z}+z= max{x,y,z}. Thus, in this case, x+(y+z)=(x+y)+z.

    Now suppose that x=yz. Then we distinguish two possibilities.

    Case I: z<x=y. Then

    (x+y)+z=[0,x]R+z=w[0,x]R(w+z). (3.9)

    Since z,w[0,x]R, for the last equation in (3.9), we have

    w+z={[0,w]R,ifw=z,z,ifw<z,w,ifz<w.

    Thus,

    w[0,x]R(w+z)=[0,w]R{wRz<wx}{z}=[0,x]R.

    In a similar way one proves that x+(y+z)=[0,x]R, meaning that the associativity holds.

    Case II: x=y<z. Then,

    (x+y)+z=[0,x]R+z=w[0,x]R(w+z)=z.

    Besides, x+(y+z)=x+z=z, thus again the hyperoperation is associative.

    It is a routine to verify that, for any xR, 0+x=x+0={x}. Moreover, since for each xR, 0x+x=[0,x]R, we conclude that for every element xR, there exist an element yR such that 0x+y.

    Finally, we prove that for all x,y,zR,

    zx+yxz+y.

    Suppose that zx+y and xy. Thus z=max{x,y} and z+y=max{x,y}+y. Hence,

    z+y=max{x,y}+y={x+y,ifx>y,y+y,ifx<y={x,ifx>y,[0,y]R,ifx<y.

    In both cases, it results xz+y.

    Now suppose that zx+y and x=y. Then from zx+y, it follows that z[0,x]R, meaning that zx. Therefore,

    z+y=z+x={max{z,x},ifx>z,[0,x]R,ifx=z={x,ifx>z,[0,x]R,ifx=z.

    And again it results xz+y.

    Therefore, (R,+) is a canonical hypergroup. We endow it now with the classical multiplication of real numbers, which is associative and has the property that, for all xR, 0x=x0=0. In order to show that (R,+,) is a Krasner hyperring, it is enough to prove that the multiplication distributes over the addition, i.e., for all x,y,zR,

    (x+y)z=xz+yz. (3.10)

    If z=0, then Eq (3.10) is clear. So we suppose that z0. Thus z=1pk, for some kN. Now we have two possibilities, the first one is x=y and the second one is xy.

    If xy, then

    (x+y)z=max{x,y}z={xz,ifx>y,yz,ify>x,

    while,

    xz+yz=max{xz,yz}={xz,ifx>y,yz,ify>x.

    This means that equality (3.10) holds.

    Now suppose that x=y0 (if x=y=0, then the Eq (3.10) is clear). Thus, there exists kN such that x=y=1pk and

    (x+y)z=[0,x]Rz=[0,1pk]R1pk=[0,1pk+k]R.

    Besides,

    xz+yz=1pk+k+1pk+k=[0,1pk+k]R.

    Therefore, the equality (3.10) holds and we conclude that (R,+,) is a Krasner hyperring.

    Let search now for the structure of an arbitrary hyperideal of R. Suppose that I is a nonzero hyperideal of R and take an arbitrary element aI. Using the hyperoperation defined by (3.8), we conclude that a+a=[0,a]RI. Thus, aI[0,a]RI, while the reverse inclusion is clear. Therefore,

    I=aI[0,a]R (3.11)

    and each hyperideal I of R has the form aI[0,a]R.

    Moreover, if 1I, then using (3.11), we get I=aI[0,a]R=[0,1]R=1. If 1I, denote A={nN1pnI}. In this case, A and AN. Thus, using the Least Principle Theorem, we know that A has a least element m0. Then we can check that I=1pm0. Therefore, every hyperideal of R is a principle hyperideal.

    Let

    M=1p={0,1p,1p2,1p3,}.

    Then M is a maximal hyperideal because if N is another hyperideal of R such that MN and NR, then there exists n0N such that

    N=1pn0={0,1pn0,1pn0+1,1pn0+2,}.

    It is clear that NM. Thus, we conclude that M=N and M is a maximal hyperideal. Besides, M is the only maximal hyperideal of R. Therefore, R is a local hyperring, which is also a principle hyperideal hyperdomain, with dimR=htRM=1 since 0M is the only chain of prime hyperideals of R.

    Finally, using Definition 3.2, we conclude that the element 1pR is a regular parameter element of R since M=<1p> and M is an M-primary hyperideal of R.

    The main findings of this paper give us the possibility to raise some open questions. Theorem 3.13 states that each regular local hyperring is a hyperdomain. The first question that follows immediately from this is whether a regular local hyperring is a hyperfield or not, or if it could contain a hyperfield or not. It's worth investigating on additional conditions under which the previous statement is true. Moreover, in Theorem 3.7 we use an element aMM2, to show that the quotient ¯R=Ra is a regular local hyperring when R is a regular local hyperring. And now we raise the following question. Can we get the same result if we use an element a in MM3 or generally MMn, when n is a natural number and Mn0? Besides, what can we say about the quotient ¯R for an arbitrary element aM2?

    The third author acknowledges the financial support from the Slovenian Research Agency (research core funding No. P1-0285).

    The authors declare no conflicts of interest.



    [1] R. Ameri, M. Eyvazi, S. Hoskova-Mayerova, Superring of polynomials over a hyperring, Mathematics, 7 (2019), 902. https://doi.org/10.3390/math7100902 doi: 10.3390/math7100902
    [2] H. Bordbar, I. Cristea, Divisible hypermodules, An. Ştiinţ. Univ. Ovidius Constanţa Ser. Mat., 30 (2022), 57–74. https://doi.org/10.2478/auom-2022-0004
    [3] H. Bordbar, I. Cristea, About normal projectivity and injectivity of Krasner hypermodules, Axioms, 10 (2021), 83. https://doi.org/10.3390/axioms10020083 doi: 10.3390/axioms10020083
    [4] H. Bordbar, I. Cristea, Height of prime hyperideals in Krasner hyperrings, Filomat, 31 (2017), 6153–6163. https://doi.org/10.2298/FIL1719153B doi: 10.2298/FIL1719153B
    [5] H. Bordbar, I. Cristea, Regular parameter elements and regular local hyperrings, Mathematics, 9 (2021), 243. https://doi.org/10.3390/math9030243 doi: 10.3390/math9030243
    [6] H. Bordbar, I. Cristea, M. Novak, Height of hyperideals in Noetherian Krasner hyperrings, U.P.B. Sci. Bull., Ser. A, 79 (2017), 31–42.
    [7] H. Bordbar, M. Novak, I. Cristea, A note on the support of a hypermodule, J. Algebra Appl., 19 (2019), 2050019. https://doi.org/10.1142/S021949882050019X doi: 10.1142/S021949882050019X
    [8] B. Davvaz, V. Leoreanu-Fotea, Hyperring theory and applications, Palm Harbor, USA: International Accademic Press, 2008.
    [9] B. Davvaz, T. Musavi, Codes over hyperrings, Matematiski Vesnik, 68 (2016), 26–38.
    [10] S. Jančič-Rašovič, I. Cristea, Hypernear-rings with the defect of distributivity, Filomat, 32 (2018), 1133–1149. https://doi.org/10.2298/FIL1804133J doi: 10.2298/FIL1804133J
    [11] M. Krasner, Approximation des corps values complets de caracteristique p0 par ceux de caracteristique zero, Colloque d'Algèbre Supérieur, 19 (1956), 129–206.
    [12] M. Krasner, A class of hyperrings and hyperfields, Int. J. Math. Math. Sci., 6 (1983), 307–312. https://doi.org/10.1155/S0161171283000265 doi: 10.1155/S0161171283000265
    [13] C. G. Massouros, Free and cyclic hypermodules, Ann. Mat. Pura Appl., 150 (1988), 153–166. https://doi.org/10.1007/BF01761468 doi: 10.1007/BF01761468
    [14] C. G. Massouros, On the theory of hyperrings and hyperfields, Algebra Logic, 24 (1985), 728–742.
    [15] G. Massouros, C. Massouros, Hypercompositional algebra, computer science and geometry, Mathematics, 8 (2020), 1338. https://doi.org/10.3390/math8081338 doi: 10.3390/math8081338
    [16] J. Mittas, Sur certaines classes de structures hypercompositionnelles, Proc. Acad. Athens, 48 (1973), 298–318.
    [17] J. Mittas, Espaces vectoriels sur un hypercorp–Introduction des hyperespaces affines et euclidiens, Math. Balkanica, 5 (1975), 199–211.
    [18] A. Nakassis, Recent results in hyperring and hyperfield theory, Int. J. Math. Math. Sci., 11 (1988), 209–220. https://doi.org/10.1155/S0161171288000250 doi: 10.1155/S0161171288000250
    [19] R. Rota, Strongly distributive multiplicative hyperrings, J. Geom., 39 (1990), 130–138. https://doi.org/10.1007/BF01222145 doi: 10.1007/BF01222145
    [20] D. Stratigopoulos, Hyperanneaux non commutatifs: Hyperanneaux artiniens, centralisateur d'un hypermodule et theoreme de densite, C. R. Acad. Sc. Paris, 269 (1969), 889–891.
    [21] T. Vougiouklis, The fundamental relations in hyperrings. The general hyperfield, In: Algebraic hyperstructures and applications, Teaneck, NJ: World Scientific Publishing, 1991. https://doi.org/10.1142/9789814539555
    [22] T. Vougiouklis, Hv-vector spaces, In: Algebraic hyperstructures and applications, Romania, Hadronic Press, Inc., Florida, 1994,181–190.
  • This article has been cited by:

    1. Christos G. Massouros, Gerasimos G. Massouros, On the Borderline of Fields and Hyperfields, 2023, 11, 2227-7390, 1289, 10.3390/math11061289
    2. Hashem Bordbar, Torsion Elements and Torsionable Hypermodules, 2023, 11, 2227-7390, 4525, 10.3390/math11214525
  • Reader Comments
  • © 2022 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Metrics

Article views(2066) PDF downloads(58) Cited by(2)

/

DownLoad:  Full-Size Img  PowerPoint
Return
Return

Catalog