검색
Article Search

JMB Journal of Microbiolog and Biotechnology

OPEN ACCESS eISSN 0454-8124
pISSN 1225-6951
QR Code

Articles

Kyungpook Mathematical Journal 2018; 58(3): 433-451

Published online September 30, 2018

Copyright © Kyungpook Mathematical Journal.

On the Toroidal Comaximal Graph of Lattices

Khadijeh Ahmad Javaheri*, and Atossa Parsapour

Department of Mathematics, Bandar Abbas Branch, Islamic Azad University, Bandar Abbas, Iran
e-mail : javaheri1158kh@yahoo.com and parsa216@gmail.com

Received: August 25, 2015; Revised: April 10, 2016; Accepted: April 12, 2016

In this paper, we study the toroidality of the comaximal graphs of a finite lattice.

Keywords: lattice, comaximal graph, toroidal graph.

The concept of the comaximal graph of a commutative ring R was first defined in [9]. In [9], Sharma and Bhatwadekar defined the comaximal graph of R, denoted by Γ(R), with all elements of R being the vertices, and two distinct vertices a and b are adjacent if and only if Ra + Rb = R. In [5] and [10], the authors considered a subgraph Γ2(R) of Γ(R) consisting of non-unit elements of R, and studied several properties of the comaximal graph. Also the comaximal graph of a non-commutative ring was defined and studied in [11]. Recently, in [1], the comaximal graph of a lattice was defined and studied. The comaximal graph of a lattice L = (L, ∧, ∨), denoted by Γ(L), is an undirected graph with all elements of L being the vertices, and two distinct vertices a and b are adjacent if and only if ab = 1.

First we recall some definitions and notation on lattices and graphs.

Recall that a lattice is an algebra L = (L, ∧, ∨) satisfying the following conditions: for all a, b, cL,

  • aa = a, aa = a,

  • ab = ba, ab = ba,

  • (ab) ∧ c = a ∧ (bc), a ∨ (bc) = (ab) ∨ c, and

  • a ∨ (ab) = a ∧ (ab) = a.

Note that in every lattice the equality ab = a always implies that ab = b. Also, by [7, Theorem 2.1], one can define an order ≤ on L as follows: For any a, bL, we set ab if and only if ab = a. Then (L, ≤) is an ordered set in which every pair of elements has a greatest lower bound (g.l.b.) and a least upper bound (l.u.b.). Conversely, let L be an ordered set such that, for every pair a, bL, g.l.b.(a, b), l.u.b.(a, b) ∈ L. For each a and b in L, we define ab := g.l.b.(a, b) and ab := l.u.b.(a, b). Then (L, ∧, ∨) is a lattice. A lattice L is said to be bounded if there are elements 0 and 1 in L such that 0 ∧ a = 0 and a ∨ 1 = 1, for all aL. Note that every finite lattice is bounded. Recall that for two elements a and b in a partially ordered set (P, ≤), we say that a covers b or b is covered by a, in notation ba, if and only if b < a and there is no element p in P such that b < p < a. An element a in L is called a co-atom if a ≺ 1. We denote the sets of all co-atoms in a lattice L by C(L). Also, for an element aL, we set [a]l = {bL | ba}.

For a positive integer r, an r-partite graph is one whose vertex set can be partitioned into r subsets so that no edge has both ends in any one subset. A complete r-partite graph is one in which each vertex is joined to every vertex that is not in the same subset. The complete bipartite graph (2-partite graph) with part sizes m and n is denoted by Km,n. An elementary contraction consists of the deletion of a vertex or an edge or the identification of two adjacent vertices. A graph G is said to contract to a graph H if there exists a sequence of elementary contractions which transforms G into H. A subdivision of a graph is any graph that can be obtained from the original graph by replacing edges with paths. A graph is said to be planar if it can be drawn in the plane so that its edges intersect only at their ends. A remarkable simple characterization of the planar graphs was given by Kuratowski in 1930. Kuratowski’s Theorem says that a graph is planar if and only if it contains no subdivision of K5 or K3,3 (cf. [2, p.153]).

By a surface, we mean a connected compact 2-dimensional real manifold without boundary, that is a connected topological space such that each point has a neighbourhood homeomorphic to an open disc. The sphere is designated to be the surface S0; the surface formed by adding g handles to the sphere is denoted Sg. It is well-known that every compact surface is homeomorphic to a sphere, or to a connected sum of g tori (Sg), or to a connected sum of k projective planes (Nk) (see [6, Theorem 5.1]). This number g is called the genus of the surface. The torus can be thought of as 1 tori (S1) or as a sphere with 1 handle.

The canonical representation of a torus

A graph G is embeddable in a surface S if the vertices of G are assigned to distinct points in S such that every edge of G is a simple arc in S connecting the two vertices which are joined in G. If G can not be embedded in S, then G has at least two edges intersecting at a point which is not a vertex of G. We say a graph G is irreducible for a surface S if G does not embed in S, but any proper subgraph of G embeds in S. The least non-negative integer g that the graph G can be embedded in Sg is called the genus of G. A toroidal graph is a graph that can be embedded without crossings on the torus. Hence, toroidal graphs have genus 1. Note that a planar graph is a graph that can be embedded on S0. And so a planar graph is not toroidal. Also a complete graph Kn is toroidal if n = 5, 6 or 7, and the only toroidal complete bipartite graphs are K4,4, K3,3, K3,4, K3,5 and K3,6 (see [3] or [8]).

In this paper, we assume that L is a finite lattice. The comaximal graph of a lattice L, denoted by Γ(L), is an undirected graph with all elements of L being the vertices, and two distinct vertices a and b are adjacent if and only if ab = 1. We denote the induced subgraph of Γ(L) with vertex set L (J(L) ∪ {1}), by Γ2(L), where J(L) is the set ∩mC(L)[m]l (see [1]). Consider the l.u.b. of two vertices in J(L) to see that they can not be connected. Therefore, the vertices in J(L) are isolated vertices.

In this section, we explore the toroidality of the graph Γ2(L).

By [1, Lemma 4.1.], if Γ2(L) is planar, then |C(L)| ≤ 4. If |C(L)| = 1, then Γ2(L) is an empty graph. Note that when |C(L)| = 2, we observe that Γ2(L) is a complete bipartite graph (see [1, Corollary 3.5.]). So Γ2(L) is planar if and only if either |[m1]l[m2]l| ≤ 2 or |[m2]l[m1]l| ≤ 2, where C(L) = {m1,m2}. Also one can easily see that Γ2(L) is toroidal if and only if either |[m1]l[m2]l| = |[m2]l[m1]l| = 4 or |[m1]l [m2]l| = 3, |[m2]l [m1]l| ∈ {3, 4, 5, 6}, where C(L) = {m1,m2}. We begin this section with the following lemma.

Lemma 2.1

If Γ2(L) is toroidal, then the size of C(L) is at most 7.

Proof

Assume to the contrary that |C(L)| ≥ 8. Then the induced subgraph of Γ2(L) with vertex set C(L) is isomorphic to K8, which is a contradiction.

By Lemma 2.1., it is sufficient for us to probe the toroidality of the graph Γ2(L) in the cases in which the size of C(L) is 3, 4, 5, 6 or 7. In this paper, we discuss on the case that |C(L)| = 3. First we begin with the following notation.

Notation 2.2

Let |C(L)| = n, where n > 1. To simplify notation, we denote the maximal ideal [m]l, where mC(L), by . We set , where 1 ≤ i, tn. Also , where 1 ≤ t1, t2, …, tkn.

Note that each element in Si is adjacent to each element in Sj, for 1 ≤ ijn, and also it is adjacent to each element in St1t2…tk, where t1, …, tk ∉ {i}.

Remark 2.3

In [1, Theorem 4.3.], Afkhami and Khashyarmanesh completely determined those lattices with 3 co-atoms whose graph Γ2(L) is planar.

Lemma 2.4

Assume thatt=13St10. Then Γ2(L) is not a toroidal graph.

Proof

Set n:=t=13St. Then we have the following cases:

  • Case 1. |Si| ≠ n − 2, for i = 1, 2, 3. Since the contraction of Γ2(L) contains a subgraph isomorphic to K3,7 or K4,5, one can conclude that the graph Γ2(L) is not toroidal.

  • Case 2. There exists 1 ≤ i ≤ 3, such that |Si| = n − 2. If Sjk is an empty set, for j, k ∉ {i} with 1 ≤ i, j, k ≤ 3, then Γ2(L) is planar, which is not a toroidal graph. If Sjk ≠ Ø, for j, k ∉ {i} with 1 ≤ i, j, k ≤ 3, then we can find a copy of K3,8 in the contraction of Γ2(L), and thus the graph Γ2(L) is not toroidal.

Now, by Lemma 2.4., we state necessary and sufficient conditions for toroidality of the graph Γ2(L), when |C(L)| = 3. It should be noted that in the proof of following theorem, according to Remark 2.3., the cases where the graph Γ2(L) is planar is ignored.

Theorem 2.5

Suppose that |C(L)| = 3. Then Γ2(L) is a toroidal graph if and only if one of the following conditions holds:

  • t=13St=5and one of the following conditions is satisfied:

    • There is some Si with |Si| = 3, for 1 ≤ i ≤ 3 and |Si1i2 | ∈ {1, 2, 3, 4}, for i1, i2 ∉ {i}.

    • There is a unique Si with |Si| = 1, for 1 ≤ i ≤ 3, and Sii1 ≠ Ø, for i1 ∉ {i}.

  • t=13St=6and one of the following conditions is satisfied:

    • There exists some i with 1 ≤ i ≤ 3 such that |Si| = 4, and |Si1i2 | ∈ {1, 2}, for i1, i2 ∉ {i}.

    • There exist unique i and j with 1 ≤ i, j ≤ 3 such that |Si| = 3 and |Sj | = 2, also if |Sji1 | = 2, then |Sii1| ≥ 0, and if |Sji1 | = 3, then Sii1 = Ø, for i1 ∉ {i, j}.

    • |Si| = 2, for all i with 1 ≤ i ≤ 3, and Si1i2 ≠ Ø, for 1 ≤ i1, i2 ≤ 3.

  • t=13St=7and one of the following conditions is satisfied:

    • There is some i with 1 ≤ i ≤ 3 such that |Si| = 5, and |Si1i2 | = 1, for i1, i2 ∉ {i}.

    • There is some i with 1 ≤ i ≤ 3 such that |Si| = 4, and |Si1i2 | = 1, for i1, i2 ∉ {i}.

    • |Si| = |Sj | = 3 for some i and j with 1 ≤ i, j ≤ 3, and |Sii1 | = |Sji1 | = 1, for i1 ∉ {i, j}. Also if |Si1i2 | = 2, for some i1 ∈ {i, j}, i2 ∉ {i, j}, then Si2i3 = Ø, for i3 ∈ {i, j} {i1}.

    • There is a unique i with 1 ≤ i ≤ 3 such that |Si| = 3, also |Si1i2 | = 1, and if |Si1i2 | = 2, then Sii1, Sii2are empty sets, for i1, i2 ∉ {i}.

  • t=13St=8and one of the following conditions is satisfied:

    • There exists some i with 1 ≤ i ≤ 3 such that |Si| = 6, and |Si1i2 | = 1, for i1, i2 ∉ {i}.

    • There exists some i with 1 ≤ i ≤ 3 such that |Si| = 5, and Si1i2 = Ø, for i1, i2 ∉ {i}.

    • There exist unique i and j with 1 ≤ i, j ≤ 3 such that |Si| = 4 and |Sj | = 3 and Sii1 = Sji1 = Ø, for i1 ∉ {i, j}.

    • |Si| = |Sj | = 3 for some i and j with 1 ≤ i, j ≤ 3, and if Si1i2 = Ø, for all i1 ∈ {i, j}, i2 ∉ {i, j}, then |Sij| ≥ 0, also if |Si1i2 | = 1, for some i1 ∈ {i, j}, i2 ∉ {i, j}, then Sij = Ø.

    • |Si| = |Sj | = 2 for some i and j with 1 ≤ i, j ≤ 3, and Sij = Ø.

  • t=13St=9and one of the following conditions is satisfied:

    • There is some i with 1 ≤ i ≤ 3 such that |Si| = 6, and Si1i2 = Ø, for i1, i2 ∉ {i}.

    • |Si| = 3 for all i with 1 ≤ i ≤ 3, and Si1i2 = Ø, for i1, i2 ∈ {1, 2, 3}.

Proof

If one of the above statements holds, then one can easily check that Γ2(L) is a toroidal graph.

Conversely, let Γ2(L) be toroidal. By Lemma 2.4., 5t=13St9. Thus we have the following situations:

(i) t=13St=5.

Assume that there is some i, say i = 1, such that |Si| = 3. If S23 is non-empty, then the contraction of Γ2(L) contains a subgraph isomorphic to K3,3, and so it is not planar. Also when |S23| ≥ 5, we have a copy of K3,7 in the contraction of Γ2(L) with vertex set {a1, a2, a3} ∪ {b, c, s1, s2, …, s5}, where a1, a2, a3S1, bS2, cS3 and s1, s2, …, s5S23. It is clear that the graph Γ2(L) is not toroidal. Therefore, we may assume that 1 ≤ |S23| ≤ 4. In this situation, the complement of Γ2(L) contains C603, one of the listed graphs in [4], which is pictured in Figure 1. In Figure 1, we replace x1, x2, …, x9 by a1, a2, a3, b, s1, s2, s3, s4, c, respectively, which a1, a2, a3S1, bS2, cS3 and s1, s2, s3, s4S23. Therefore, Γ2(L) is a toroidal graph (see Figure 2).

Now, if there is a unique i, say i = 1, such that |Si| = 1, and the sets S12 and S13 are non-empty, then we can find a subdivision of K5 in the structure of the contraction of Γ2(L) as it is shown in Figure 3, where aS1, b1, b2S2, c1, c2S3, s12S12 and s13S13. Thus Γ2 (L) is toroidal.

(ii) t=13St=6.

Suppose that there exists only one i, say i = 1, such that |Si| = 4. If the size of S23 is at least 3, then one can easily observe that the contraction of Γ2(L) contains a subgraph isomorphic to K4,5. Hence the graph Γ2(L) is not toroidal. So, for toroidality, the size of S23 is necessarily 1 or 2. In this case, Γ2(L) is contained in K8 (K3K2) (cf. [4, p.55]). Hence Γ2(L) is a toroidal graph (see Figure 4). In Figure 4, we assume that a1, a2, a3, a4S1, bS2, cS3 and s23, s23S23.

Now, assume that there exists a unique i, say i = 1, such that |Si| = 3. If S23 has at least 4 elements, then the contraction of Γ2(L) contains a copy of K3,7, and hence Γ2(L) is not a toroidal graph. Also if |S23| = 3 and S13 is non-empty, then the complement of the contraction of Γ2(L) is contained in U6.6b, one of the listed graphs in [4]. So the graph Γ2(L) is not toroidal (see Figure 5). In Figure 5, we have the vertices a1, a2, a3S1, b1, b2S2, cS3, s13S13 and s23, s23,s23S23.

Therefore, for toroidality of Γ2(L), when |S23| = 3, necessarily, S13 = Ø. In this situation, the complement of Γ2(L) contains C610, one of the listed graphs in [4] (see Figure 6). In Figure 6, we have the vertices a1, a2, a3S1, b1, b2S2, cS3 and s23, s23,s23S23. In this situation, the embedding of the graph Γ2(L) in the torus is pictured in Figure 7.

In addition, if |S23| ≤ 2, then the contraction of Γ2(L) contains a copy of K3,3, and so Γ2(L) is not planar. In this situation, Γ2(L) is contained in K8 (K3K2), (cf. [4, p.55]). Therefore, Γ2(L) is a toroidal graph (see Figure 8). In Figure 8, we have the vertices a1, a2, a3S1, b1, b2S2, cS3, s13S13 and s23, s23S23.

Finally, suppose that |Si| = 2, for all 1 ≤ i ≤ 3 and only one of the sets S12, S13 or S23 is non-empty. Then it is easy to check that the contraction of Γ2(L) contains K3,3 and so it is not planar. Now we assume that at least one of the sets S12, S13 or S23 are non-empty sets. Obviously, we can find a subdivision of K6 in Γ2(L) as it is shown in Figure 9, where a1, a2S1, b1, b2S2, c1, c2S3, s12S12, s13S13 and s23S23. Thus Γ2 (L) is a toroidal graph.

(iii) t=13St=7.

First, suppose that |Si| = 5, for some 1 ≤ i ≤ 3, without loss of generality, we may assume that i = 1. If S23 has exactly 1 element, then Γ2(L) is contained in K8 (K3K2), (cf. [4, p.55]). Hence the graph Γ2(L) is toroidal (see Figure 10). In Figure 10, we have the vertices a1, a2, a3, a4, a5S1, bS2, cS3 and s23S23.

We may assume that S23 has at least 2 elements. In this case, the vertices of the set {a1,a2,,a5}{b,c,s23,s23} form a subgraph isomorphic to K4,5 in the contraction of Γ2(L), where a1, a2, …, a5S1, bS2, cS3 and s23, s23S23. Therefore, Γ2(L) is not a toroidal graph.

Suppose that there is i with 1 ≤ i ≤ 3, say 1, such that |Si| = 4. When S23 is a singleton set, the graph Γ2(L) is contained in K8 (K3K2) (cf. [4, p.55]). So the graph Γ2(L) is toroidal (see Figure 11). In Figure 11, we have the vertices a1, a2, a3, a4S1, b1, b2S2, cS3 and s23S23.

As |S23| ≥ 2, the contraction of Γ2(L) contains a copy of K4,5, which implies that Γ2(L) is not a toroidal graph. If S1 and S2 have exactly 3 elements, and S13 or S23 has at least 3 elements, then we have a subgraph isomorphic to K3,7 in the contraction of Γ2(L), and so Γ2(L) is not a toroidal graph. Therefore, we assume that |S13| ≤ 2, S23 = Ø or |S23| ≤ 2, S13 = Ø. Then the complement of Γ2(L) contains C603, one of the listed graphs in [4] (see Figure 1). In Figure 1, we replace vertices x1, x2, …, x9 by b1, b2, b3, a1, s13, c, a2, a3, s13, respectively, where a1, a2, a3S1, b1, b2, b3S2, cS3 and s13, s13S13. Hence the graph Γ2(L) is toroidal (see Figure 12).

In addition, we may assume S13 and S23 have 1 element, exactly. Then the complement of Γ2(L) contains C402, one of the listed graphs in [4] (see Figure 13). In Figure 13, we have the vertices a1, a2, a3S1, b1, b2, b3S2, cS3, s13S13 and s23S23. So Γ2(L) is a toroidal graph, which is pictured in Figure 14.

If one of the sets S13 or S23 has exactly 2 elements and the other one has only 1 element, then Γ2(L) contains a subgraph isomorphic to G3, one of the listed graphs in [12]. So Γ2(L) is not a toroidal graph. To do this, in Figure 15, we have the vertices a1, a2, a3S1, b1, b2, b3S2, cS3, s13, s13S13 and s23S23.

Now, consider the case that there is a unique i, say 1, such that |Si| = 3. If S23 has at least 3 elements, then the contraction of Γ2(L) contains a copy of K3,7, and so the graph Γ2(L) is not toroidal. When |S23| = 2, and also S12 or S13 is non-empty, the complement of the contraction of Γ2(L) is contained in U6.6b, one of the listed graphs in [4] (see Figure 16). To do this, in Figure 16, we have the vertices a1, a2, a3S1, b1, b2S2, c1, c2S3, s23, s23S23 and s12S12. So the graph Γ2(L) is not toroidal.

Hence we assume that |S23| = 2 and S12 = S13 = Ø. Then the complement of Γ2(L) contains C603, one of the listed graphs in [4] (see Figure 1). In Figure 1, we replace the vertices x1, x2, …, x9 by a1, a2, a3, b1, s23, c1, b2, s23, c2, respectively, where a1, a2, a3S1, b1, b2S2, c1, c2S3 and s23, s23S23. Hence Γ2(L) is a toroidal graph, which is pictured in Figure 17.

When S23 is a singleton set, Γ2(L) is contained in K8 (K3K2), (cf. [4, p.55]). Thus the graph Γ2(L) is toroidal (see Figure 18). In Figure 18, we have the vertices a1, a2, a3S1, b1, b2S2, c1, c2S3, s12S12, s13S13 and s23S23.

(iv) t=13St=8.

Suppose that there exists only one i, say i = 1, such that |Si| = 6. If the size of S23 is at least 2, then one can easily see that the contraction of Γ2(L) contains a subgraph isomorphic to K4,6. Hence Γ2(L) is not a toroidal graph. So the size of S23 is necessarily 1. In this situation, the complement of Γ2(L) contains C603, one of the listed graphs in [4] (see Figure 1). To do this, in Figure 1, we replace the vertices x1, x2, …, x9 by b, s23, c, a1, a3, a5, a2, a4, a6, respectively, where a1, a2, …, a6S1, bS2, cS3 and s23S23. Thus Γ2(L), which is pictured in Figure 19, is a toroidal graph.

Now, suppose that there exists some i, say i = 1, such that |Si| = 5. If S23 is non-empty, then the contraction of Γ2(L) contains a copy of K4,5, and so Γ2(L) is not a toroidal graph. Otherwise, S23 = Ø, and so Γ2(L) is contained in K8 (K3K2), (cf. [4, p.55]). Therefore, Γ2(L) is a toroidal graph (see Figure 20). In Figure 20, we have the vertices a1, a2, …, a5S1, b1, b2S2 and cS3.

Suppose that there exist unique i and j with 1 ≤ i, j ≤ 3, say i = 1 and j = 2, such that |Si| = 4, |Sj| = 3. If S13 is non-empty, then the complement of Γ2(L) is contained in S5.5, one of the listed graphs in [4] (see Figure 21). In fact, in Figure 21, we have the vertices a1, a2, a3, a4S1, b1, b2, b3S2, cS3 and s13S13. And so the graph Γ2(L) is not toroidal.

Also if S23 is non-empty, then one can easily see that the contraction of Γ2(L) contains a copy of K4,5. Hence Γ2(L) is not toroidal. So the size of the sets S13 and S23 is 0. In this case, Γ2(L) is contained in K8 (K3K2) (cf. [4, p.55]), which is a toroidal graph (see Figure 22). In Figure 22, we have the vertices a1, a2, a3, a4S1, b1, b2, b3S2 and cS3.

Suppose that S1 and S2 have exactly 3 elements. If |S13| ≥ 2 or |S23| ≥ 2, then it is easy to see that the contraction of Γ2(L) contains a copy of K3,7. Hence Γ2(L) is not a toroidal graph. Also, if S13 and S23 have only 1 element, then Γ2(L) contains a subgraph isomorphic to G3, one of the listed graphs in [12]. So the graph Γ2(L) not toroidal. To do this, in Figure 23, we have the vertices a1, a2, a3S1, b1, b2, b3S2, c1, c2, ∈ S3, s13S13 and s23S23.

So we may assume that S12 is non-empty and also S13 or S23 has only 1 element. Then the complement of the contraction of Γ2(L) is contained in S5.6, one of the listed graphs in [4] (see Figure 24). Hence the graph Γ2(L) is not toroidal. In Figure 24, we have the vertices a1, a2, a3S1, b1, b2, b3S2, c1, c2S3, s12S12 and s13S13.

Therefore if the size of the set S13 or S23 is 1, then necessarily S12 must be an empty set. Since the complement of Γ2(L) contains C402, one of the listed graphs in [4]. To do this, in Figure 25, we have the vertices a1, a2, a3S1, b1, b2, b3S2, c1, c2S3, s13S13. Hence Γ2(L) is a toroidal graph (see Figure 26).

Consequently, two sets S13 and S23 are both empty. In this situation, Γ2(L) is contained in K8 (K3K2) (cf. [4, p.55]), which is a toroidal graph. To do this, in Figure 27, we have the vertices a1, a2, a3S1, b1, b2, b3S2, c1, c2S3 and s12S12.

Now, suppose that S1 and S2 have 2 elements. If S12 is non-empty, then the contraction of Γ2(L) contains a copy of K4,5, and thus the graph Γ2(L) is not toroidal. So the set S12 is necessarily empty. In this case, Γ2(L) is contained in K8 (K3K2), (cf. [4, p.55]), which is a toroidal graph (see Figure 28). In Figure 28, we have the vertices a1, a2S1, b1, b2S2 and c1, c2, c3, c4S3.

(v) t=13St=9.

First, suppose that |Si| = 7, for some 1 ≤ i ≤ 3. Without loss of generality, we may assume that i = 1. If S23 is non-empty, then the contraction of Γ2(L) contains a copy of K3,7. Hence the graph Γ2(L) is not toroidal.

Also suppose that there is some i with 1 ≤ i ≤ 3, say 1, such that |Si| = 6 and S23 is non-empty. Then we can find a copy of K4,6 in the contraction of Γ2(L). Therefore, Γ2(L) is not a toroidal graph. So, for toroidality of Γ2(L), it is sufficient S23 = Ø, because in this situation the complement of Γ2(L) contains C603, one of the listed graphs in [4]. To do this, in Figure 1, we replace vertices x1, x2, …, x9 by b1, b2, c, a1, a3, a5, a2, a4, a6, respectively, where a1, a2, …, a6S1, b1, b2S2, cS3. The embedding of Γ2(L) in the torus is pictured in Figure 29.

Now, suppose that all of the sets S1, S2 and S3 have 3 elements, exactly. If S12, S13 and S23 are empty, then the complement of Γ2(L) contains C315, one of the listed graphs in [4] (see Figure 30). In Figure 30, we have the vertices a1, a2, a3S1, b1, b2, b3S2, c1, c2, c3S3. Therefore, the graph Γ2(L), which is pictured in Figure 31, is toroidal.

Otherwise, we may assume that at least one of the sets S12, S13 or S23 is nonempty. Then the contraction of Γ2(L) contains a copy of K3,7. Thus Γ2 (L) is not a toroidal graph. Otherwise, there exists some i, with 1 ≤ i ≤ 3, such that |Si| = 4 or |Si| = 5. In these situations, the contraction of Γ2(L) contains a copy of K4,5. Therefore, Γ2(L) is not a toroidal graph.

  1. Afkhami, M, and Khashyarmanesh, K (2014). The comaximal graph of a lattice. Bull Malays Math Sci Soc. 37, 261-269.
  2. Bondy, JA, and Murty, USR (1976). Graph theory with applications. New York: American Elsevier
    CrossRef
  3. Bouchet, A (1978). Orientable and nonorientable genus of the complete bipartite graph. J Combin Theory Ser B. 24, 24-33.
    CrossRef
  4. Hlavacek, AL 1997. 9-vertex irreducible graphs for the torus. PhD thesis. University of Ohio State.
  5. Maimani, HR, Salimi, M, Sattari, A, and Yassemi, S (2008). Comaximal graph of commutative rings. J Algebra. 319, 1801-1808.
    CrossRef
  6. Massey, W (1967). Algebraic topology: an introduction. New York: Harcourt, Brace & World
  7. Nation, JB (1998). Notes on lattice theory. Cambridge studies in advanced mathematics. Cambridge: Cambridge University Press
  8. Ringel, G (1974). Map color theorem. New York/Heidelberg: Springer-Verlag
    CrossRef
  9. Sharma, PK, and Bhatwadekar, SM (1995). A note on graphical representation of rings. J Algebra. 176, 124-127.
    CrossRef
  10. Wang, HJ (2008). Graphs associated to co-maximal ideals of commutative rings. J Algebra. 320, 2917-2933.
    CrossRef
  11. Wang, HJ (2009). Co-maximal graph of non-commutative rings. Linear Algebra Appl. 430, 633-641.
    CrossRef
  12. Zeps, D (2009). Forbidden minors for projective plane are free-toroidal or non-toroidal. IUUK-CE-ITI series.