We will now state a few results regarding the genus properties of a zero-divisor graph of a commutative ring which will be further used to prove the main result.
First, we provide some results needed to classify the genus three zero-divisor graphs.
Proof. Assume that . Since R is local, there is a prime power q and integers such that , and . Let t be a largest positive integer such that . Therefore, .
Suppose that . Then, . Since , we have for every . Therefore, the subgraph induced by in is complete and so . Thus, by Proposition 2, is a contradiction. Hence, .
Case 1: Suppose . The possibilities for q are or 9. If , then and . Since , every element of is adjacent to all the elements of in So and by Proposition 2, we obtain . Therefore, . That is, . If , then . Since the collection of non-zero elements of form a complete graph in and , we have , a contradiction. Therefore, is a principal ideal and . In this case, . The assumption implies that , a contradiction.
Case 2: Suppose . If , then . This statement implies that and . Since , so that , a contradiction. Thus, . That is, .
Moreover, suppose that for some . Here, . If , then the subgraph induced by forms a complete graph in , which leads to , a contradiction. If , then and . This implies that , a contradiction. Therefore, means that . But by Theorems 1 and 2, the zero-divisor graph of any local ring of order less than or equal to 16 has a genus less than 2. Thus, .
Case 3: Suppose . If , then and . Since , a contradiction. Therefore, . Hence, .
Suppose that for some . This implies that .
Case 3.1: If , then the subgraph induced by is complete in , leading to , a contradiction.
Case 3.2: If
, then
. Let
. Since
, we have
, a contradiction. Therefore,
and so
. Now, by Proposition 2.8 [
17],
for some
. Without loss of generality, we may assume that either
or
.
Assume that . Then by Proposition 7, for . Since and , we have the sets and which form in . Therefore .
Assume that . Since any element of generates , we may assume that . Now, the cosets and are mutually disjoint. In particular, and . Let , and so that , and . Therefore, in , the subgraph induced by the set A is isomorphic to , and the subgraph induced by the sets B and C is isomorphic to . Thus, so that .
Case 3.3: If
, then
and
. Let
. Then,
contains
as a subgraph, a contradiction. Therefore,
and
. Consequently,
. Then, by Proposition 2.8 [
17],
for some
. So we may assume that
or
.
Assume that . Then, by Proposition 7, for every i. Notice that the cosets and are mutually disjoint and every element in is adjacent to all the elements of in . So contains as a subgraph and thus .
Assume that . Since any element of generates , we may assume that . Clearly, . If and , then for some . Since and , we have . Therefore, the subgraph induced by contains in and so .
Thus, with . But by Theorems 1 and 3, for all local rings R of order less than or equal to 27. Hence, .
Case 4: Suppose . If , then and . Since , we have so that , a contradiction. Therefore, , which leads to .
Suppose that for some . Notice that .
Case 4.1: If , then , a contradiction.
Case 4.2: If , then . Let . Then, contains as a subgraph, a contradiction. Therefore, and so This implies that for and without loss of generality, we may assume or .
Assume that . Then, by Proposition 7, we have . Let and . Then, for every , so that , a contradiction.
Assume . Then, by Proposition 8, for . Also, any element of generates and so we may assume that . Now, let and let . Then, for every and therefore , a contradiction.
Case 4.3: If
, then
and
. Let
. Then,
contains
as a subgraph, a contradiction. Therefore,
and
. Consequently,
Now, by Proposition 2.8 [
17],
for some
, and without loss of generality, let
or
. In either case, by Propositions 7 and 8, we have
for
. Since
and
, we can choose two non-zero elements
such that
. Let
and
Clearly
for every
so that
. By Proposition 2,
, a contradiction.
A similar proof technique is valid for the cases or .
Thus, . But by Theorems 1 and 2, for all local rings R of order less than or equal to 16. Hence, or 64. □
Proof. Assume to the contrary that
. Then, by Theorem 4,
and
. Note that
R is isomorphic to one of the following 22 commutative local rings [
18].
(1) If , then the zero-divisor graph is empty, a contradiction.
(2) Let
R be isomorphic to any one of the following seven rings:
Then, by Theorem 3, , a contradiction.
(3) If , then every vertex of the set is adjacent to all the vertices of the set . Therefore, contains as a subgraph and so, by Proposition 2, .
(4) If , then the subgraph induced by the set forms in and so, by Proposition 1, .
(5) If , then the partite sets and form in , which implies that .
(6) If , then the partite sets and form as a subgraph in . Therefore, .
(7) Let
R be isomorphic to any one of the following five rings:
Then, the partite sets
and
form a subgraph
in
. Hence,
.
(8) Let or or . Then, contains a subgraph , where the partite sets are and . Thus, .
(9) Let or . Then, contains as a subgraph with the bipartite sets and . Thus, .
Hence, for all the 22 local rings of order 81 with , which is in contradiction to our assumption. □
Proof. Let be a local ring. Suppose Then, by Theorem 4, the number of elements in the residue field is 2 or 3 or 4.
If , then, by Lemma 3, we have and also has a nilpotency index of 3. This implies that , and
If , then, by Lemma 4, we obtain a contradiction.
Let . By Theorem 4, we have or 32 with .
Case 1: Suppose with . Then, .
We claim that the nilpotency index of is 6. In order to prove this claim, assume to the contrary that for some
Case 1.1: If , then the 31 non-zero elements in together form a complete graph in . So, by Proposition 1, we have , a contradiction.
Case 1.2: Suppose . If , then contains as a subgraph and so, by Proposition 2, . Therefore, . This implies that . So we assume for some . Without loss of generality, we may assume that or .
Case 1.2.1: Assume that . Then, by Proposition 7, we have .
(i). If for all , then , where the bipartite sets are and . Therefore, .
(ii). Suppose exactly one element of is non-zero, say . That is, . Let and . Clearly, for all and and so . Thus .
(iii). Suppose exactly two elements of are non-zero. If and , then . This implies that . Let us take . Clearly, and so Therefore, . Similarly, if and , then , so that . Let , and we have and . So .
Case 1.2.2: Assume that . Then, none of and are in . Thus, . By Proposition 8, we obtain . If necessary, replacing by and by , we have . Now, let and . Then, for every , so that . Therefore, .
Case 1.3: Suppose . Then, and . If , then contains as a subgraph and so . Therefore, and . Consequently, . So, let for some , and without loss of generality, assume or .
First, assume that . Then, . Since , we let . This implies that . Let and Since for every , we have so that .
Second, assume that . Then, by Lemma 8, we have . Moreover, we may assume that each of and are in . However, , a contradiction.
Case 1.4: Suppose . Then, , and . If , then Since , we have contains as a subgraph and so , a contradiction. Therefore, . Consequently, . Thus, just two cases can occur: either or .
First, suppose that . Then, , and we have for some . Without loss of generality, or . Let . Then, and so . Clearly, and are mutually disjoint sets. Note that, in , every elements in the set is adjacent to all the elements of so that . Therefore, . Let . Since the cosets and are mutually disjoint and since , we have the sets , and which are mutually disjoint. Then, in , every element in is adjacent to every element of so that . Therefore, .
Second, suppose that . Then, and so is principal. Therefore, is principal for every i. This implies that for all , which contradicts the fact that .
Case 1.5: Suppose
. Then,
. Consequently,
for all
and
’s are the principal ideal for all
. Note that, in
, the subgraph induced by
is complete, every element in
is adjacent to all the vertices of
and every element in
is adjacent to the single non-zero element of
By labeling the vertices of
by
,
and
, it follows that
is isomorphic to the graph given in
Figure 1. Thus, by Lemma 2, we obtain
.
Hence, in Case 1, if and only if such that , and .
Case 2: Suppose with . This implies that . Then, by Lemma 1, there exist 53 (non-field) local rings.
(1). Assume that
R is isomorphic to any one of the following eight rings:
Then, by Theorem 2, we have
as toroidal.
(2). Assume that
R is isomorphic to one of the following five rings:
Then, by Theorem 3, we have that
is double toroidal.
(3). Assume that
R is isomorphic to one of the following 19 rings:
Then, the corresponding zero-divisor graphs of R are provided in a tabular format.
Thus, all the zero-divisor graphs displayed in
Table 1 are either the graph
H (refer
Figure 1) or a subgraph of
H. Hence, by Lemma 2, in this case,
for all the 17 local rings.
(4). Let and . Then, and . It is clear that and are isomorphic to and so .
(5). Assume that
R is isomorphic to one of the following three rings:
If , then the partite sets and forms in . Therefore, .
If , then the partite sets and forms in . Therefore, .
If , then every vertex in the set is adjacent to all the vertices of in . So , a contradiction.
If , then contains , as a subgraph with bipartite sets and , a contradiction.
(6). Assume that
R is isomorphic to one of the following three rings:
Note that the number of vertices and number of edges in
is
and
, respectively. If
, then, by Proposition 3, there are 43 faces in any
-embedding of
. So the average length of the faces is
, which is less than the girth value(=3) of
, a contradiction.
(7). Assume that
R is isomorphic to one of the following five rings:
Let . Then, in , every vertex of the set is adjacent to all the vertices of . Therefore, is a subgraph of . Clearly, . By Euler’s formula, there are 17 faces when drawing on . Fix a representation of -embedding of and denote it as L. Let be the number of i-gons of corresponding to the representation L. Since , we have . Therefore, and . This implies that there are fifteen rectangular and two hexagonal faces in L. Note that, in each vertex in is adjacent to all the four vertices in . If , then we have to embed all these 12 edges into L. Choose an arbitrary . Then, and so v is a part of exactly three faces of L. Notice that any rectangular face can adopt at most one new edge into it and any hexagonal face can adopt at most three edges between the vertices of . So to embed the four edges between v and a vertex of B, two rectangular and one hexagonal face are required. Therefore, to embed all the edges between A and B, six rectangular and three hexagonal faces are required, a contradiction to L which has two hexagonal faces.
If , , or , then it is not hard to verify that the graph is isomorphic to a subgraph of so that
(8). Assume that
R is isomorphic to one of the following five rings:
Let
. Then, consider the subgraph
. The diagrammatic representation of
is given in
Figure 3. Note that
is a triangle-free graph and the number vertices and edges of
is
and
, respectively. Therefore, by Remark 2, we obtain
. Suppose that
. Then,
and also, by Euler’s formula, there are 19 faces in any
-embedding of
. Let
be the number of
i-gons in a
-embedding of
. Then,
and
This implies that all the faces in any
-embedding of
are rectangular. Now, to recover a
-embedding of
from a
-embedding of
, we have to embed all the 15 omitted edges of
into a
-embedding of
. Notice that each rectangular face can adapt at most one edge into it. So to embed the eight edges incident with the vertex
, namely
and
, eight rectangular faces which contain
are required. But
. This implies that
is a part of exactly six rectangular faces of any
-embedding of
, a contradiction. Therefore,
.
If or or or , then and so .
(9). Assume that
R is isomorphic to one of the following four rings:
Let
. Take
and
. Then,
is isomorphic to
Figure 4 and
, since a copy of
is contained in
. Suppose that
. Consider the subgraph
. Then,
is isomorphic to
and clearly
.
By the Euler formula, the faces of any -embedding of have fifteen rectangular and two hexagonal faces. Now, we try to embed the 21 omitted edges of into an arbitrary -embedding of .
First, we try to insert the edges
into a
-embedding of
. To embed edges, we require one hexagonal and three rectangular faces, which we name as
(see
Figure 5a). Similarly, to embed the next six edges
, we require another one hexagonal and three rectangular faces, which we name as
. Here, we notice that the vertices
, for
and
are placed in two faces of
. Since
, one more face should have
’s and those which form three rectangular faces, we name as
,
and
. The edges
and
can be embedded in the
, respectively.
Next, to embed the three edges
, we requires three rectangular faces, which we name as
and
, all of which contain
(
Figure 5b). Furthermore, to embed two edges
, two rectangular faces are required, which we name as
, all of which contain
(
Figure 5b). Since
and 17 faces are in
-embedding of
, exactly one more face should have
, and we shall denote it as
. Notice that, in any
-embedding, every edge of a graph is in exactly two faces. Here, the edges
and
are used twice in the faces
. Now, the choices for another one diagonal of
are in
. Obviously, we have to select distinct vertices for
in which one is from
or
. This is a contradiction to the fact that the edges
and
are used twice in the faces
. Therefore,
.
Furthermore, if or , then and so . □