675
Views
0
CrossRef citations to date
0
Altmetric
Articles

On the genus of some total graphs

&

Abstract

Let R be a commutative ring with a proper ideal I. A generalization of total graph is introduced and investigated. It is the (undirected) graph with all elements of R as vertices, that two distinct vertices x,yR are adjacent if and only if x+ySH(I) where SH(I)={aR:raI for some rH} and H is a multiplicatively closed subset of R. This version of total graph is denoted by T(ΓHI(R)). We in addition characterize certain lower and upper bounds for the genus of the total graph, and compute genus T(ΓHI(R)) on finite ring R, with respect to some special ideal I.

1 Introduction

Throughout, all rings will be commutative with non-zero identity. Let R be a ring and I a proper ideal of R. The totalgraph of a commutative ring R, denoted by T(Γ(R)), was introduced by Anderson and Badawi in [Citation1] and studied by several authors ([Citation2–4], etc.), where the authors in [Citation3,4] obtained some facts on the genus of total graphs. They considered a total graph with all elements of R as vertices, that two distinct vertices x,yR are adjacent if and only if x+yZ(R) where Z(R) denotes the set of all zero-divisors of R. The total graph is then extended in joint papers [Citation5,6] of the second author in rings and modules, respectively. Furthermore, a generalized total graph was introduced in [Citation7]. For a proper submodule N of M, there is a generalization of the graph of modules relative N under multiplicatively closed subset H denoted by T(ΓHN(M)) which was studied by present authors in [Citation8]. The vertex set of T(ΓHN(M)) is M, that two distinct vertices m and m are adjacent if and only if m+mMH(N) where MH(N)={mM:rmN for some rH} and H is a multiplicatively closed subset of R, i.e. abH for all a,bH. As N is a proper submodule of M and NMH(N), MH(N) is not empty.

We define a generalized total graph over ring R, denoted by T(ΓHI(R)), with all elements of R as vertices, that two distinct vertices x,yR are adjacent if and only if x+ySH(I) where SH(I)={aR:raI for some rH}, I is an ideal of R and H is a multiplicatively closed subset of R.

It follows from the definition that if SH(I)=R, (for example, if I=R, 0H, H(SH(I):R), H(0:R) or H(I:R), by [Citation8]), then T(ΓHI(R)) is complete; so we suppose that SH(I)R. We denote by ΓHI(SH(I)) and ΓHI(SHC(I)) the (induced) subgraphs of T(ΓHI(R)) with vertices in SH(I) and RSH(I), respectively. Based on our assumption, SH(I)R and so ΓHI(SHC(I)) is always nontrivial.

Let G be a simple graph. We say that G is totallydisconnected if none of two vertices of G are adjacent. We use Kn to denote complete graph with n vertices. A bipartitegraph G is a graph whose vertex set V(G) can be partitioned into subsets V1 and V2 such that the edge set consists of precisely those edges which join vertices in V1 to vertices of V2. In particular, if E consists of all possible such edges, then G is called the completebipartitegraph and denoted by Km,n when |V1|=m and |V2|=n. Two subgraphs G1 and G2 of G are disjoint if G1 and G2 have no common vertices and no vertex of G1 (resp., G2) is adjacent (in G) to any vertex not in G1 (resp., G2). The union of two graphs G1=(V1,E1) and G2=(V2,E2) is the graph G1G2 whose vertex set is V1V2 and the edge set is E1E2. The Cartesianproduct of graphs G1 and G2 is defined as the graph G1×G2 which the vertex set is V(G1)×V(G2) and the edge set is the set of all pairs (u1,v1)(u2,v2) such that either u1u2E(G1) and v1=v2 or v1v2E(G2) and u1=u2. Two graphs G and H are said to be isomorphic to each another, written GH, if there exists a bijection f:V(G)V(H) such that for each pair x,y of vertices of G, xyE(G) if and only if f(x)f(y)E(H). For a vertex v of graph G, deg(v) is the degree of vertex v and δ(G)min{deg(v): v is a vertex of G}. For a nonnegative integer k, a graph G is called kregular if every vertex of G has degree k. The genus of a graph G, denoted by g(G), is the minimal integer n such that the graph can be embedded in Sn, where Sn denotes a sphere with n handles. Intuitively, G is embedded in a surface if it can be drawn in the surface so that its edges intersect only at their common vertices. A planargraph is a graph that can be embedded in the plane, i.e., it can be drawn on the plane in such a way that its edges intersect only at their endpoints. For such graphs the genus is zero. A graph with genus one is called a toroidal graph. If G is a subgraph of G, then g(G)g(G). For details on the notion of embedding of a graph in a surface, see White [Citation9, Chapter 6].

In Section 2, we remind some facts and give a lower bound for genus of the graph T(ΓHI(R)). We proceed in Section 3 by determining all isomorphism classes of finite rings R whose T(ΓHI(R)) has genus at most one (i.e. a planar or toroidal graph). Also, we compute genus of the graph over R=Zn under some well-known multiplicatively closed subsets of R.

2 Background problem and some comments

Throughout, x denotes the least integer that is greater than or equal to x. In the following theorem we give some well-known formulas, see, e.g., [Citation9–11]:

Theorem 2.1

The following statements hold:

(1)

For n3 we have g(Kn)=(n3)(n4)12.

(2)

For m,n2 we have g(Km,n)=(m2)(n2)4.

(3)

Let G1 and G2 be two graphs and for each i,pi be the number of vertices of Gi. Then max {p1g(G2)+g(G1),p2g(G1)+g(G2)}g(G1×G2).

(4)

The genus of a graph is the sum of the genuses of its components.

According to Theorem 2.1 we have g(Kn)=0 for 1n4, g(Kn)=1 for 5n7 and g(Kn)2, for other value of n.

Corollary 2.2

If G is a graph with n vertices, then g(G)(n3)(n4)12.

In the following of the section, we characterize a lower bound for the genus of the graph T(ΓHI(R)). Considering the fact that ΓHI(SH(I)) is in the form of K|SH(I)| (see [Citation8, Remark 3.1]), in view of Theorem 2.1, it is enough for us to obtain a lower bound for genus of the graph ΓHI(SHC(I)).

Theorem 2.3

[Citation8, Corollary 3.7] Let SH(I)=α, RSH(I)=β, and 2(SH(I):R).Then ΓHI(SHC(I)) is a disjoint union of β1 copies of Kα.

Theorem 2.4

[Citation8, Theorem 3.10] Let SH(I)=α and RSH(I)=β. If H, a multiplicatively closed subset of R containing some even elements, then ΓHI(SHC(I)) is a disjoint union of β12 copies of Kα,α.

Corollary 2.5

Let |SH(I)|=α and |RSH(I)|=β. Then the following hold:

1.

If 2(SH(I):R), then g(ΓHI(SHC(I)))=(β1)(α3)(α4)12.

2.

If 2rH for some rR, then g(ΓHI(SHC(I)))=β12(α2)24 .

Proof

It is obvious by Theorems 2.3 and 2.4.

Lemma 2.6

[12 , Proposition .2.1] If G is a graph with n vertices and genus g, then δ(G)6+12g12n.

Theorem 2.7

[Citation8 , Theorem 3.13] Suppose that the edge set of ΓHI(SHC(I)) is not empty and x is a vertex of the graph. Then the degree of x is either SH(I) or SH(I)1.

Proposition 2.8

Let ΓHI(SHC(I)) with t vertices have a nonempty edge set, and let |SH(I)|=t. Then (t7)t12+1g(ΓHI(SHC(I))).

Proof

By Lemma 2.6 and Theorem 2.7, t1=|SH(I)|1=δ, so t16+(12g12)t. Then (t7)t12g12. Hence (t7)t12+1g(ΓHI(SHC(I))).

Corollary 2.9

If R is infinite, then g(ΓHI(SHC(I)))is infinite for all ideals I and all closed subsets H of R.

3 The genus of T(ΓHI(R))

Considering Corollary 2.9, the genus will be infinite if R is not a finite ring. In order to compute the genus, we consider a finite ring R.

In view of Theorem 2.1, it is enough for us to study the genus of ΓHI(SHC(I)).

Remark 3.1

If HI, then SH(I)=R and ΓHI(SHC(I)) is trivial so, in the following, we suppose that HI=. It should be noted that if HI=, then HSH(I)=.

Theorem 3.2

Let R be a finite ring such that R=R1×R2××Rt with t4, I=0×R2××Rt and H be a multiplicatively closed subset of R. Then γ(ΓHI(SHC(I)))2.

Proof

It is enough to show that there is a subgraph L of ΓHI(SHC(I)) with γ(L)2; this implies that γ(ΓHI(SHC(I)))2. So, we proceed for t=4.

1.

Let H(Z(R1)×R2××Rt)=. Then I=SH(I). By way of contradiction, let there exists (r1,,rt)RI such that (r1,,rt)(h1,,ht)I for some (h1,,ht)H (note that h1 has inverse in R1). Then r1h1=0 implies that r1=0, a contradiction.

(1) If R1=Z2, then K8ΓHI(SHC(I)).

(2) Let |R1|>2.

(a) If 2Z(R1), then considering the vertices {(a1,a2,a3,a4)|aiRi for i2}, there is a10 belonging to R1 such that 2a1=0. So, K8ΓHI(SHC(I)).

(b). Let 2Z(R1). For a non zero element l1R1, set X1={(l1,l2,l3,l4)R|liRi for 2i4} and Y1={(l1,l2,l3,l4)R|liRi for 2i4}. Then X1,Y1 is a bipartition for Kn,n for n8. Hence, K8ΓHI(SHC(I)) or K8,8ΓHI(SHC(I)); so γ(ΓHI(SHC(I)))2.

2.

Let H(Z(R1)×R2××Rt). Set T=H(Z(R1)×R2××Rt). For (d1,d2,,dt)T, put K(d1,d2,,dt)={(b1,b2,,bt)R|b10,b1d1=0} and define K=(d1,d2,,dt)TK(d1,d2,,dt).

Claim. SH(I)=IK.

Let there exists (r1,,rt)RI such that (r1,,rt)(h1,,ht)I for some (h1,,ht)H. Then r1h1=0 implies that h1Z(R1){0} (by IH=). So (r1,,rt)K. Conversely, let (b1,n2,,nt)K. Then there exists (d,n2...,nt)T such that b1d=0. So (b1,n2,,nt)(d,n2...,nt)I implies that (b1,n2,,nt)SH(I), hence KSH(I). Therefore, SH(I)=IK.

(i) If R1=Z2, then K8ΓHI(SHC(I)).

(ii) Suppose |R1|>2.

(i) Let there exists (2,m2,m3,m4)T4, where T4=H(Z(R1)×R2×R3×R4). Then 2Z(R1){0} (by HI=). So 2a1=0 for non zero element a1R1 and {(a1,n2,n3,n4)|niRi for 2i4}KSH(I), then a12 (otherwise, SH(I)H), also a11 since 20 (by IH=). Considering distinct sets {(a11,a2,a3,a4)|aiRi for i2}, one has K8,8ΓHI(SHC(I)).

(ii) Let (2,m2,m3,m4)T4 for every miRi with i2. By the similar argument of case 1., if 2Z(R1), then K8ΓHI(SHC(I)) and if 2Z(R1), then K8,8ΓHI(SHC(I)).

Hence, γ(ΓHI(SHC(I)))2.

Remark 3.3

It should be noted that, Theorem 3.2 is satisfied for every I=R1×R2××Rn1×0×Rn+1××Rt with n>1 and t4.

Theorem 3.4

Let R=R1×R2×R3 where every Ri is a finite ring for 1i3, I=0×R2×R3 and H be a multiplicatively closed subset of R.

1.

Let H(Z(R1)×R2×R3)=or H(Z(R1)×R2×R3) with (2,m2,m3)H(Z(R1)×R2×R3) for every m2R2 and m3R3.

(i) If 2Z(R1) and |R2||R3|8, then γ(ΓHI(SHC(I)))2.

(ii) If 2Z(R1) and |R2||R3|4, then γ(ΓHI(SHC(I)))1.

2.

Let there exists (2,m2,m3)H(Z(R1)×R2×R3) and |R2||R3|4. Then γ(ΓHI(SHC(I)))1.

Proof

1.

(i) Consider {(a1,a2,a3)|aiRi} which is in the form of Kn for n8, where 2a1=0 for some a10 belonging to R1.

(ii) For a non zero element l1R1, consider {(l1,l2,l3)|liRi,i=2,3}{(l1,m2,m3)|miRi,i=2,3} which is in the form of Kn,n for n4.

2.

If there is (2,m2,m3)H(Z(R1)×R2×R3), then 2Z(R1){0}, so 2a1=0 for non zero element a1R1 and {(a1,n2,n3)|niRi,i=2,3}SH(I), then a12 (otherwise, SH(I)H); furthermore, a11 since 20 ( by IH=). Hence the vertices {(a11,a2,a3)|aiRi,i=2,3}{(1,a2,a3)|aiRi,i=2,3} are in the form of Kn,n for n4.

Corollary 3.5

Let R=R1×R2 where every Ri is a finite ring for i{1,2}, I=0×R2 and H be a multiplicatively closed subset of R.

It is easily proved that the following statements hold.

1.

Let H(Z(R1)×R2)=or H(Z(R1)×R2) with (2,m2)H(Z(R1)×R2) for every m2R2.

(i) If 2Z(R1) and |R2|8, then γ(ΓHI(SHC(I)))2.

(ii) If 2Z(R1) and |R2|4, then γ(ΓHI(SHC(I)))1.

2.

Let there exists (2,m2)H(Z(R1)×R2) and |R2|4. Then γ(ΓHI(SHC(I)))1.

Proof

It is obvious by Theorem 3.4.

Example 3.6

Let R=Z6×Z4, I=0×Z4 and H be a multiplicatively closed subset of R. Then Z(R)=(Z(Z6)×Z4)(Z6×Z(Z4)). If H(Z(Z6)×Z4)=, then SH(I)=I. Let H(Z(Z6)×Z4) and let (2,ns)H (so for every niZ4, (3,ni)H, otherwise HI). Then SH(I)=I{(3,t2)|t2Z4}. Considering the vertices of ΓHI(SHC(I)) in the form of {(1,b2),(4,t2)|b2,t2Z4}{(2,m2),(5,l2)|m2,l2Z4}, one has ΓHI(SHC(I)) is a K8,8.

Theorem 3.7

Let R=Fq×Rn, I=0×Rn where Rn is a ring with |Rn|=n, Fq is a field with q elements and let H be a multiplicatively closed subset of R.

1.

Let q=2m.Then ΓHI(SHC(I)) is planar if and only if 2n4 and it is toroidal if and only if m=1 and 5n7.

2.

If for every mN, q2m, then ΓHI(SHC(I)) is planar if and only if n=2 and it is toroidal if and only if q=3 and n{3,4}.

Proof

Claim; for every nN, I=SH(I). By way of contradiction, let there exists (m,d)RI such that (m,d)(t,l)I for some (t,l)H. So mt=0 which implies that m=0; so m=0, a contradiction.

1.

Let q=2m.

(a) If m=1, for every (1,b)SHC(I), we have (1,b)+(1,b)SH(I) where b,b are disjoint elements of Rn. So ΓHI(SHC(I)) is Kn.

(b) Let m>1. Considering vertices of ΓHI(SHC(I)) in the form of {(l,a)|lF2m,aRn} and by the fact that char(F2m)=2, in the same way of proof of Theorem 3.2, ΓHI(SHC(I)) is a disjoint union of 2m1 copies of Kn. Hence, ΓHI(SHC(I)) is planar if and only if 2n4 and it is toroidal if and only if m=1 and 5n7.

2.

Let for all mN, one has q2m. For non zero element lFq, let Xl={(l,a)|aRn} and Yl={(l,a)|aRn}. Then Xl,Yl is a bipartition of Kn,n. So ΓHI(SHC(I)) is a disjoint union of q12 copies of Kn,n. Hence ΓHI(SHC(I)) is planar if and only if n=2 and it is toroidal if and only if q=3 and n{3,4}.

Theorem 3.8

Let R be a finite ring for which R=R1×R2××Rt with t2, I=0×R2××Rt, H=RZ(R) and let 2Z(R1). If γ(ΓHI(SHC(I)))1, then R is isomorphic to the one of the following rings: Z3×Z2×Z2,R1×Z2,Z3×Z3,Z3×Z4,Z3×F4,Z3×Z2[X](X2).

Proof

Note that t<4, by Theorem 3.2.

1′.

Let t=3.

If γ(ΓHI(SHC(I)))<1, then |R2||R3|<4, by Theorem 3.4. For γ(ΓHI(SHC(I)))=1, consider |R2||R3|=4, by the proof of Theorem 3.4. So, for γ(ΓHI(SHC(I)))1, |R2||R3|4. Hence R2=R3=Z2. For every non zero element b1R1, set Xb1={(b1,n2,n3)|niRi,i=2,3}. Then Xb1,Xb1 is a bipartition of K4,4. If |R1|>3, then γ(ΓHI(SHC(I)))2 (since there exist at least two bipartition of K4,4 in ΓHI(SHC(I)) ). Hence, |R1|3, so R1=Z3 (since 2Z(R1)). Therefore, R=Z3×Z2×Z2.

2′.

Let t=2.

For every non zero element b1R1, set Xb1={(b1,n2)|n2R2}. Then Xb1,Xb1 is a bipartition of Kn,n for n=|R2|. Since 2Z(R1) and γ(ΓHI(SHC(I)))1, then by Corollary 3.5 and by the same way of case 1, |R2|4.

(i) Let |R2|=2. Then Xb1 and Xb1, for nonzero b1R1, are in the form of K2,2 and ΓHI(SHC(I)) is a disjoint union of some copies of K2,2. So for every R=R1×R2 with |R2|=2, ΓHI(SHC(I)) is planar.

(ii) Let |R2|=3,4. For every non zero element b1R1, Xb1,Xb1 is a bipartition of K3,3 or K4,4 and γ(K3,3)=γ(K4,4)=1. So, by 2Z(R1) and the same way of case 1, |R1|=3 and R1=Z3.

Hence, for t=2, R is isomorphic to the one of the following rings: R1×Z2,Z3×Z3,Z3×Z4,Z3×F4,Z3×Z2[X](X2).

Proposition 3.9

Let R be a finite ring for which R=R1×R2××Rt with t2, I=0×R2××Rt, H=RZ(R) and let 2Z(R1). If γ(ΓHI(SHC(I)))1, then R is isomorphic to the one of the following rings: R1×Z2,Z2×Z5,Z2×Z6,Z2×Z7,orR1×Z2×Z2,R1×Z3,R1×R2with|R2|=4.

Proof

Note that t<4, by Theorem 3.2. Put P={a1R1|a10,2a1=0}(|P|1, by 2Z(R1)) and R1=R1{0}.

1.

Let t=3. Put l=|R2||R3|. For every non zero element a1R1, set Xa1={(a1,n2,n3)|niRi,i=2,3}. If a1P, then Xa1 is in the form of Kl. If a1P, then Xa1,Xa1 is a bipartition of Kl,l.

Since 2Z(R1), then by Theorem 3.4, |R2||R3|<8. So |R2||R3|=4 or 6.

(a) Let |R2||R3|=4.

(i) If char(R1)=2 (R1=P), then ΓHI(SHC(I)) is a disjoint union of some copies of K4, which is planar.

(ii) Let char(R1)2. Consider P¯=R1P (note that |P¯|2).

Suppose that |P¯|>2. We claim that |P¯|4. If |P¯|=3 and a1,a1,a2P¯, then a2P¯. So |P¯|=4, a contradiction.

Hence, if |P¯|>2, considering the sets Xa1,Xa1 and Xa2,Xa2 where a1,a1,a2,a2P¯, there are at least two copies of K4,4 in ΓHI(SHC(I)) that implies, γ(ΓHI(SHC(I)))2.

So, |P¯|=2. Then for b1,b1P¯, Xb1,Xb1 is in the form of K4,4. Thus, in this case, ΓHI(SHC(I)) is a disjoint union of some copies of K4 and one copy of K4,4, which is toroidal.

Therefore, if t=3 and |R2||R3|=4, then R=R1×Z2×Z2 with char(R1)=2 (R1=P) or |P¯|=2.

(b) Let |R2||R3|=6. If char(R1)2, then K6,6 is a subgraph of ΓHI(SHC(I)) which implies that γ(ΓHI(SHC(I)))>1. So char(R1)=2. Also, by γ(ΓHI(SHC(I)))1, we should have |P|=1, since for every lP, K6ΓHI(SHC(I)) and γ(K6)=1. Hence, in this case we have R1=Z2 and R is isomorphic to Z2×Z6.

2.

Let t=2. For every non zero element a1R1, set Xa1={(a1,n2)|n2R2}. Since 2Z(R1), then by Corollary 3.5, n=|R2|<8. For every non zero element a1P, Xa1 is in the form of Kn for n7. If a1P, then Xa1,Xa1 is a bipartition of Kn,n.

(a) Let |R2|=2. If char(R1)=2, then ΓHI(SHC(I)) is a disjoint union of some copies of K2, which is planar.

If char(R1)2 (R1P), then for every a1P¯, Xa1,Xa1 is a bipartition of K2,2. So ΓHI(SHC(I)) is a disjoint union of some copies of K2 and K2,2, which is planar. Hence, in this case, R=R1×Z2.

(b) Let |R2|=3. If char(R1)=2, then ΓHI(SHC(I)) is a disjoint union of some copies of K3, which is planar.

Let char(R1)2 (R1P). If |P¯|>2, then |P¯|4. So, there exist at least two copies of K3,3 in ΓHI(SHC(I)) that implies that ΓHI(SHC(I))2.

Hence, |P¯|=2 and ΓHI(SHC(I)) is a disjoint union of some copies of K3 and one copy of K3,3, which is toroidal.

Thus, in this case, R=R1×Z3 with char(R1)=2 or |P¯|=2.

(c) If |R2|=4, then by the same way of case 1(a), R=R1×R2, where |R2|=4 with char(R1)=2 or |P¯|=2.

(d). Let |R2|{5,6,7}. If |P|2, then γ(ΓHI(SHC(I)))2 ( since for every a1P, KnΓHI(SHC(I)) for n{5,6,7}). So |P|=1. If char(R1)2, then Kn,nΓHI(SHC(I)) for n{5,6,7}, that implies that γ(ΓHI(SHC(I)))>1. So char(R1)=2 and R1=Z2, since |P|=1.

Hence, in this case, R is isomorphic to the one of the following rings: Z2×Z5,Z2×Z6,Z2×Z7.

Remark 3.10

Let R be a ring with I=0 and H be a multiplicatively closed subset of R. Then ΓHI(SHC(I)) is a subgraph of T(Γ(R)), since SH(I)={rR:rs=0 for some sH(0H)}Z(R). Recall that T(Γ(R)) is the graph with all elements of R as vertices, and two distinct vertices x,yR are adjacent if and only if x+yZ(R) where Z(R) denotes the set of all zero-divisors of R, so γ(ΓHI(SHC(I)))γ(T(Γ(R))).

For finite ring R with ideal I=0×0××Rn××Rt, n1,t2 and multiplicatively closed subset H of R, ΓHI(SHC(I)) is a subgraph of T(Γ(R)), since SH(I)Z(R). For the recent inclusion, let (r1,r2,,rt)SH(I). Then (r1,r2,,rt)(s1,s2,,st)I for some (s1,s2,,st)H, so risi=0 for every 1in1. If for every 1in1, si=0, then HI, a contradiction, so there exists sl0 for 1ln1. Then rlsl=0 implies that rlZ(Rl), hence SH(I)Z(R).

Theorem 3.11

Let I=kZn with d=(k,n) and H={a|aZn,(a,n)=1}.

1.

Let n be an even integer.

(i) If d is an even integer, then ΓHI(SHC(I)) is a disjoint union of Knd and d22 copies of Knd,nd. Hence g(ΓHI(SHC(I)))=(nd3)(nd4)12+d22(nd2)24.

(ii) If d is an odd integer, then ΓHI(SHC(I)) is a disjoint union of d12 copies of Knd,nd and g(ΓHI(SHC(I)))=d12(nd2)24.

2.

If n is an odd integer, then ΓHI(SHC(I)) is a disjoint union of d12 copies of Knd,nd and g(ΓHI(SHC(I)))=d12(nd2)24.

Proof

We at first show that dZn=I=SH(I). There are t1,t2Z such that t1n+t2k=d, so dI. Hence I=dZn. We claim that I=SH(I). By way of contradiction, if there exists mZnI such that tmI for some tH, then d|tm. But (d,t)=1 ( since d|n), so d|m, a contradiction. Hence I=SH(I).

1.

Let n be an even integer. Then

(i) If d=2, then I=2Zn; so for every mSHC(I), 2mSH(I). Then by Theorem 2.3, ΓHI(SHC(I)) is in the form of Kn2. Therefore, g(ΓHI(SHC(I)))=(n23)(n24)12, by Theorem 2.1.

Let d be an even integer greater than 2, then SHC(I) has the following elements:

1,2,,d2,,d1

d+1,,3d2,,2d1

.

.

.

[(nd1)d]+1,,(2nd1)d2,,ndd1.

Hence, ΓHI(SHC(I)) is a disjoint union of Knd and d22 copies of Knd,nd. So by Theorem 2.1, g(ΓHI(SHC(I)))=(nd3)(nd4)12+d22(nd2)24.

(ii) Let d be an odd integer, then SHC(I) has the following elements: 1,2,,d12,d+12,,d1

d+1,,3d12,3d+12,,2d1

.

.

.

[(nd1)d]+1,,[(2nd1)d]12,[(2nd1)d]+12,,ndd1.

Hence, ΓHI(SHC(I)) is a disjoint union of d12 copies of Knd,nd as in , where 1id12. Then by Theorem 2.1, g(ΓHI(SHC(I)))=d12(nd2)24.

2.

Let n be an odd integer; so, 2H. Then by Theorem 2.4, ΓHI(SHC(I)) is a disjoint union of d12 copies of Knd,nd and by Theorem 2.1, g(ΓHI(SHC(I)))=d12(nd2)24.

Fig. 1 The total graph of Theorem 3.11.

Fig. 1 The total graph of Theorem 3.11.

Example 3.12

Consider R=Z18, I=12Z18 and H={a|aZ18,(a,18)=1}. Then ΓHI(SHC(I)) is a disjoint union of K3 and two copies of K3,3. Then g(ΓHI(SHC(I)))=(33)(34)12+622(32)24=2.

Theorem 3.13

Let R=Zn×Zn, I=0×Zn and H={(a,b)|a,bZn×Zn,(a,n)=1}. Then:

1.

If n is an even integer, then ΓHI(SHC(I)) is a disjoint union of Kn and n22 copies of Kn,n. In this case, g(ΓHI(SHC(I)))=(n3)(n4)12+n22(n2)24.

2.

If n is an odd integer, then ΓHI(SHC(I)) is a disjoint union of n12 copies of Kn,n. In this case g(ΓHI(SHC(I)))=n12(n2)24.

Proof

1.

Let n be an even integer. We claim that I=SH(I). By way of contradiction, if there is (m,d)Zn×Zn(0×Zn) such that (t,l)(m,d)0×Zn for some (t,l)H, then n|tm, but (t,n)=1; so n|m, a contradiction. Hence I=SH(I). The number of elements of (Zn×Zn)SH(I) is n2n. For every element (a,b)SHC(I), we have (a,b)+(a,b)SH(I). Let (a,b)+(c,d)SH(I) for (a,b) and (c,d)SHC(I). Then a+c=0 and this implies that a=c. Hence, each element (a,b) of SHC(I) is just adjacent to (a,d) for every element dZn. Because n is even, so just for one element 0a=n2Zn, (a,b)SHC(I) is adjacent to (a,c) for every b,cZn. Hence, ΓHI(SHC(I)) is a disjoint union of Kn and n22 copies of Kn,n. Now by Theorem 2.1, g(ΓHI(SHC(I)))=(n3)(n4)12+n22(n2)24.

2.

If n is an odd integer, then 2H. By Theorem 2.4, ΓHI(SHC(I)) is a disjoint union of n12 copies of Kn,n and g(ΓHI(SHC(I)))=n12(n2)24, by Theorem 2.1.

Example 3.14

Consider R=Z8×Z8, I=0×Z8 and H={(a,b)|(a,b)Z8×Z8,(a,8)=1}. Then ΓHI(SHC(I)) is a disjoint union of K8 and 3 copies of K8,8. Hence g(ΓHI(SHC(I)))=29.

Theorem 3.15

Let I=kZn, H=ZnpZn where p is a prime number with 1<pn, SH(I)=α and ZnSH(I)=β.

1.

If p2, then ΓHI(SHC(I)) is a disjoint union of β12 copies of Kα,α and g(ΓHI(SHC(I)))=β12(α2)24.

2.

Let p=2. Then ΓHI(SHC(I)) is a disjoint union of Kn2l and 2l22 copies of Kn2l,n2l. Hence g(ΓHI(SHC(I)))=(n2l3)(n2l4)12+2l22(n2l2)24,where (k,n)=2lr with l,rN and (2,r)=1.

Proof

By the proof of Theorem 3.11, I=dZn, where d=(k,n).

1.

If p2, then 2ZnpZn. So, ΓHI(SHC(I)) is a disjoint union of β12 copies of Kα,α by Theorem 2.4 and g(ΓHI(SHC(I)))=(β1)2(α2)24, by Theorem 2.1.

2.

Let p=2. If n is odd, then 0H and this is impossible; so we assume that n is even. It should be noted that d is even; since if d is odd, then dH and IH that this implies that Zn=SH(I), which is not in our assumptions.

Let n=2lr for l,rN where (2,r)=1. We can assume that d=2lr where l,rN and ll, rr. Here, we claim that SH(I)=2lZn. Suppose there is mZn2lZn such that d|tm for some tH, so 2l|tm. Hence 2l|m, a contradiction. Therefore, SH(I)2lZn. Now, let 2lt2lZn for some tZn. It is clear that 2ltrI, so 2ltSH(I). Hence, by an argument similar to the proof of Theorem 3.11, ΓHI(SHC(I)) is a disjoint union of Kn2l and 2l22 copies of Kn2l,n2l. So g(ΓHI(SHC(I)))=(n2l3)(n2l4)12+2l22(n2l2)24 , by Theorem 2.1.

Example 3.16

Consider R=Z64, I=4Z64 and H=Z642Z64. Then ΓHI(SHC(I)) is a disjoint union of K16,16 and K16 such that g(ΓHI(SHC(I)))=62.

Theorem 3.17

Let I=kZn, H={as|s0} such that a|n with (na,a)=1, SH(I)=α and ZnSH(I)=β.

1.

If there is at least one even number in H, then ΓHI(SHC(I)) is a disjoint union of β12 copies of Kα,α. In this case g(ΓHI(SHC(I)))=β12(α2)24.

2.

Let there be no even number in H, (d,na)=l where d=(k,n) and n=rl for some rN.

(1). If l is an even integer, then ΓHI(SHC(I)) is a disjoint union of Kr and l22 copies of Kr,r. In this case g(ΓHI(SHC(I)))=(r3)(r4)12+l22(r2)24.

(2). If l is an odd integer, then ΓHI(SHC(I)) is a disjoint union of l12 copies of Kr,r and g(ΓHI(SHC(I)))=l12(r2)24.

Proof

Note that by the proof of Theorem 3.11, I=dZn.

1.

If there is at least one even element in H, then by Theorem 2.4, ΓHI(SHC(I)) is a disjoint union of β12 copies of Kα,α. Therefore, g(ΓHN(SHC(N)))=(β1)2(α2)24, by Theorem 2.1.

2.

Let there be no even element in H. If (d,na)=1, then d|a, since n=naa where (na,a)=1. Hence IH and this implies that Zn=SH(I), which is not the case. So, we assume that (d,na)=l1. Put I=lZn. We claim that I=SH(I). Suppose there is mZnI such that d|agm, where agH and g0, then l|agm. So, l|m, a contradiction.

Conversely, we show that d|al and this implies that ISH(I). Let d=n1l and na=n2l, for some n1,n2N (so (n1,n2)=1). We show that n1|a; this implies that n1l|al and the proof is complete.

Let (n1,a)=h (so n1=g1h and a=g2h, for some g1,g2N where (g1,g2)=1). By the fact that d=n1l and d|n, one has g1hl|g2hn2l; so g1|g2n2 and (g1,g2)=1 implies that g1|n2. This yields g1=1 and so n1|a.

(1). Let l be an even integer, then by the same way as the proof of Theorem 3.11, ΓHI(SHC(I)) is a disjoint union of Kr and l22 copies of Kr,r . Hence, g(ΓHI(SHC(I)))=(r3)(r4)12+l22(r2)24, by Theorem 2.1.

(2). Let l be an odd integer. By the same way as the proof of Theorem 3.11, ΓHI(SHC(I)) is a disjoint union of l12 copies of Kr,r and g(ΓHI(SHC(I)))=l12(r2)24, by Theorem 2.1.

Example 3.18

Consider R=Z60, I=15Z60 and H={3s|s0}. Then ΓHI(SHC(I)) is a disjoint union of 2 copies of K12,12 and g(ΓHI(SHC(I)))=50.

Theorem 3.19

Let I=kZn, H={1,t} with n|t21, SH(I)=α and ZnSH(I)=β.

1.

If t is an even integer, then ΓHI(SHC(I)) is a disjoint union of β12 copies of Kα,α and g(ΓHI(SHC(I)))=β12(α2)24.

2.

let t be an odd integer and (k,n)=d.

(1). If d is an even integer, then ΓHI(SHC(I)) is a disjoint union of Knd and d22 copies of Knd,nd. In this case g(ΓHI(SHC(I)))=(nd3)(nd4)12+d22(nd2)24.

(2). If d is an odd integer, then ΓHI(SHC(I)) is a disjoint union of d12 copies of Knd,nd and g(ΓHI(SHC(I)))=d12(nd2)24.

Proof

1.

Let t be an even integer. In view of Theorem 2.4, ΓHI(SHC(I)) is a disjoint union of β12 copies of Kα,α and g(ΓHI(SHC(I)))=β12(α2)24.

2.

By the proof of Theorem 3.11, I=dZn. Let t be an odd integer. We claim that I=SH(I). Otherwise, there is aZnI such that d|at. So d|at2. By the assumption d|t21, hence d|a and this implies that aI , a contradiction. Thus, I=SH(I).

(1) If d is an even integer, then we proceed in the proof of Theorem 3.11. So ΓHI(SHC(I)) is a disjoint union of Knd and d22 copies of Knd,nd. Then g(ΓHI(SHC(I)))=(nd3)(nd4)12+d22(nd2)24,by Theorem 2.1.

(2) If d is an odd integer, then we proceed in the proof of Theorem 3.11. Hence, ΓHI(SHC(I)) is a disjoint union of d12 copies of Knd,nd and g(ΓHI(SHC(I)))=d12(nd2)24, by Theorem 2.1.

Example 3.20

Consider R=Z50, I=5Z50 and H={1,49}. Then ΓHI(SHC(I)) is a disjoint union of two copies of K10,10 and g(ΓHI(SHC(I)))=32.

Theorem 3.21

Let R=Zn×Zn, I=0×Zn, H={(1,1),(t,t)} with n|t21, SH(I)=α and ZnSH(I)=β.

1.

If t is an even integer, then ΓHI(SHC(I)) is a disjoint union of β12 copies of Kα,α and g(ΓHI(SHC(I)))=β12(α2)24.

2.

Let t be an odd integer.

(i) If n is an even integer, then ΓHI(SHC(I)) is a disjoint union of Kn and n22 copies of Kn,n. In this case, g(ΓHI(SHC(I)))=(n3)(n4)12+n22(n2)24.

(ii) If n is an odd integer, then ΓHI(SHC(I)) is a disjoint union of n12 copies of Kn,n. In this case g(ΓHI(SHC(I)))=n12(n2)24.

Proof

1.

If t is an even integer, then ΓHI(SHC(I)) is a disjoint union of β12 copies of Kα,α by Theorem 2.4 and g(ΓHI(SHC(I)))=β12(α2)24.

2.

Let t be an odd integer. We claim that I=SH(I). Otherwise, there is (a,b)Zn×Zn such that (a,b)(t,t)I where a0 and b is an arbitrary element of Zn. So n|at; then n|at2. By the assumption n|t21, so n|a and this implies that a=0, a contradiction. Hence I=SH(I). The remaining is similar to proof of Theorem 3.13.

Example 3.22

Consider R=Z21×Z21, I=0×Z21 and H={(1,1),(13,13)}. Then ΓHI(SHC(I)) is a disjoint union of 10 copies of K21,21 and g(ΓHI(SHC(I)))=910.

Theorem 3.23

Let Zp denote the field of p elements where p>2 is a prime number, R=Zp×Z2m where mN and I=0×Z2m. Let H be one of the following sets: {(1,1)},{(1,1),(2,1)},,{(1,1),(2,1),,(p1,1)},{(1,0)},{(1,0),(2,0)}, …, {(1,0),(2,0),,(p1,0)}.

1.

If m=1, then ΓHI(SHC(I)) is planar.

2.

If m>1, then g(ΓHI(SHC(I)))=p12(2m2)24.

Proof

We note that I=SH(I) for every positive integer m and all cases of H. Otherwise, there exists (a,c)Zp×Z2m such that a0 and p|at for some (t,b)H, where bZ2. Because (p,t)=1, so p|a, a contradiction. Hence, in all cases, I=SH(I).

1.

If m=1, for every (a,b)SHC(I), we have (a,b)+(a,b)SH(I), where aZp and bZ2. So, ΓHI(SHC(I)) is a disjoint union of p12 copies of K2,2, where R=Zp×Z2 and I=0×Z2, as , where 1ip12. Hence g(ΓHI(SHC(I)))=0 , by Theorem 2.1.

2.

In the same way of the case 1, for R=Zp×Z2m with mN and m>1, ΓHI(SHC(I)) is a disjoint union of p12 copies of K2m,2m and g(ΓHI(SHC(I)))=p12(2m2)24, where I=0×Z2m.

Fig. 2 The total graph of Theorem 3.23 (1).

Fig. 2 The total graph of Theorem 3.23 (1).

References

  • AndersonD.F.BadawiA., The total graph of a commutative ring J. Algebra 320 7 2008 2706–2719
  • AkbariS.KianiD.MohammadiF.MoradiS., The total graph and regular graph of a commutative ring J. Pure Appl. Algebra 213 12 2009 2224–2228
  • MaimaniH.R.WickhamC.YassemiS., Rings whose total graphs have genus at most one Rocky Mountain J. Math. 42 5 2010 1551–1560
  • Tamizh ChelvamT.AsirT., On the genus of the total graph of a commutative ring Comm. Algebra 412013 142–153
  • AbbasiA.Sh. Habibi., The total graph of a commutative ring with respect to proper ideals J. Korean Math. Soc. 49 1 2012 85–98
  • AbbasiA.Sh. Habibi., The total graph of a module over a commutative ring with respect to proper submodules J. Algebra Appl. 11 3 2012 125004813 pages
  • AndersonD.F.BadawiA., The generalized total graph of a commutative ring J. Algebra Appl. 12 5 2013 125021218 pages
  • AbbasiA.Hamidian JahromiL., A generalization of total graphs of modules over commutative rings under multiplicatively closed subsets J. Math. Ext. 11 3 2017 87–102
  • WhiteA.T., Graphs, Groups and Surfaces1973North-HollandAmsterdam
  • BattleJ.HararyF.KodamaY.YoungsJ.W.T., Additivity of the genus of a graph Bull. Amer. Math. Soc. 68 6 1962 565–568
  • WhiteA.T., The genus of the Cartesian product of two graphs J. Combin. Theory Ser. B 11 2 1971 89–94
  • WickhamC., Classification of rings with genus one zero-divisor graphs Comm. Algebra 36 2 2008 325–345

Further reading

  • GodsilC.RoyleG. Algebraic Graph Theory 2001 SpringerNew York