Metric dimension and Zagreb indices of essential ideal graph of a finite commutative ring

Abstract

Let $R$ be a commutative ring with unity. The essential ideal graph $\mathcal{E}_{R}$ of $R$ is a graph whose vertex set consists of all nonzero proper ideals of R. Two vertices $\hat{I}$ and $\hat{J}$ are adjacent if and only if $\hat{I}+\hat{J}$ is an essential ideal. In this paper, we characterize the graph $\mathcal{E}_{R}$ as having a finite metric dimension. Additionally, we identify that the essential ideal graph and annihilating ideal graph of the ring $\mathbb{Z}_{n}$ are isomorphic whenever $n$ is a product of distinct primes. We also estimate the metric dimension of the essential ideal graph of the ring $\mathbb{Z}_{n}$. Furthermore, we determine the topological indices, namely the first and the second Zagreb indices, of $\mathcal{E}_{\mathbb{Z}_{n}}$.

Keywords— Essential ideal graph, metric dimension, first and second Zagreb indices
AMS $2010$ Subject Classification: $05C07,05C12,05C25$

1 Introduction

Let $\Gamma$ be a simple graph with vertex set $V(\Gamma)=\{v_{1},v_{2},\cdots,v_{n}\}$ and edge set $E(\Gamma)$. If a vertex $u$ is adjacent to a vertex $v$ in $\Gamma$, we write $u\sim v$ in $\Gamma$. The set $N(u)=\{v\in V(\Gamma):v\sim u\ \text{in}\ \Gamma\}$, is called the set of neighbors of $u$ and $deg(u)=|N(u)|$ is called the degree of a vertex $u$. Also, $N[u]=N(u)\cup\{u\}$. The distance $d(u,v)$ between two vertices $u$ and $v$ of a connected graph $\Gamma$ is the number of edges in the shortest path between $u$ and $v$. The complete graph $K_{n}$, is a graph in which any two vertices are adjacent. A graph $\Gamma$ is a $k-\textit{partite graph}$ if $V(\Gamma)$ can be partitioned into $k$ subsets $V_{1},\ V_{2},\ \cdots,\ V_{k}$ (named partite sets) such that the vertices $u$ and $v$ form an edge in $\Gamma$ if they belong to different partite sets. If, in addition, there exists an edge between every two vertices belonging to different partite sets, then graph $\Gamma$ can be classified as complete k-partite graph. The graph denoted as $K_{m,n}$ represents a complete bipartite graph consisting of two sets with sizes $m$ and $n$ respectively. The induced subgraph, $\Gamma[S]$, is formed by taking the subset $S$ of vertices from $\Gamma$, along with all the edges that connect vertices solely within $S$. The complement of a graph $\Gamma$ is denoted by $\overline{\Gamma}$. The join of two graphs, $\Gamma_{1}$ and $\Gamma_{2}$, represented as $\Gamma_{1}\vee\Gamma_{2}$, is formed by adding edges between any two vertices $v_{1}$ and $v_{2}$, where $v_{1}\in\Gamma_{1}$ and $v_{2}\in\Gamma_{2}$.

The concept of metric dimension of a graph was introduced by Slater in [slater1975leaves], and was called locating sets and locating numbers. An equivalent terminology was also introduced by Harary and Melter independently in [harary1976metric], and used the term resolving set. Slater described the usefulness of these ideas in long-range aids to navigation. Also, these concepts have some applications in chemistry for representing chemical compounds [johnson1993structure, johnson1998browsable], or in problems of pattern recognition and image processing, some of which involve the use of hierarchical data structures [melter1984metric]. Other applications of this concept to the navigation of robots in networks and other areas appear in [chartrand2000resolvability, hulme1984boolean, khuller1996landmarks]. Hence, according to its applicability resolving sets has become an interesting and popular topic of investigation in graph theory.

Topological Indices play a vital role in mathematical chemistry. They give ideas about structural characteristics with easy identification for a molecule. Hence there are a lot of molecular descriptors called graph invariants. A graph invariant is a number that is invariant under graph isomorphisms in graph theory. The graphical invariant is considered as a structural invariant related to a graph. Since the topological index is constructed as a graphical invariant in molecular graph theory, the computing of topological indices of many graph structures has been an attractive research area for scientists especially chemists and mathematicians for a long time [consonni2009molecular, gutman2012mathematical]. The first and second Zagreb indices of a graph $\Gamma$ introduced in [gutman1972graph], and elaborated in [gutman1975graph] are degree-based topological indices defined respectively as follows:
$M_{1}=\displaystyle\sum_{v\in V(\Gamma)}deg(v)^{2}$ and $M_{2}=\displaystyle\sum_{\begin{subarray}{c}u\sim v\\ u,v\in V(\Gamma)\end{subarray}}deg(u)deg(v).$

Let $R$ be a commutative ring with nonzero unity. An element $z\in R$ is said to be a zero divisor of $R$ whenever there exists a nonzero element $w\in R$ such that $zw=0$. An ideal $I$ of a ring $R$ is said to be an annihilating ideal of $R$ if there exists a nonzero ideal $J$ of $R$ such as $IJ=0$. An ideal $I$ of a ring $R$ which has a nonzero intersection with every other nonzero ideal of $R$ is called an essential ideal.

The study of metric dimension and topological indices of graphs related to various algebraic structures has emerged as a compelling area of research in recent times. In [pirzada2014locating], S. Pirzada and R. Raja introduced and investigated the metric dimension of the zero divisor graph of a commutative ring $R$. The results on topological indices of this graph can be seen in [selvakumar2022wiener]. In [banerjee2022spectra, banerjee2023adjacency], S. Banerjee determined the metric dimension and topological indices like the Wiener index, the first and the second Zagreb index of comaximal graph of the ring $\mathbb{Z}_{n}$. In [aijaz2020annihilating], M. Aijaz and S. Pirzada computed the metric dimension of annihilating ideal graphs of commutative rings. The annihilating ideal graph $\mathbb{AIG}(R)$, of a commutative ring $R$, introduced and studied by M. Behboodi and Z. Rakeei in [behboodi2011annihilating], is a graph in which the vertex set consists of the set of all nonzero annihilating ideals of $R$ and two distinct vertices $\hat{I}$ and $\hat{J}$ are joined by an edge if and only if $\hat{I}\hat{J}=0$.

Being motivated by these works, in this paper, we study the metric dimension and topological indices of the essential ideal graph of the ring $\mathbb{Z}_{n}$. The essential ideal graph $\mathcal{E}_{R}$ of a commutative ring $R$, introduced and studied by J. Amjadi in [amjadi2018essential], is a graph in which the vertex set is the set of all nonzero proper ideals of $R$ and two vertices $\hat{I}$ and $\hat{J}$ are joined by an edge whenever $\hat{I}+\hat{J}$ is an essential ideal. To date, there is no information about the metric dimension and topological indices of the essential ideal graph of $\mathbb{Z}_{n}$ in literature.

This paper has been organized as follows: In Section $2$, we list the results and definitions that are needed for the present study. In Section $3$, we determine the metric dimension of the essential ideal graph of $\mathbb{Z}_{n}$. Also, we prove that the essential ideal graph and annihilating ideal graph of the ring $\mathbb{Z}_{n}$ are equal (up to isomorphism) whenever $n$ is a product of $k\geq 2$ distinct primes. Moreover, we provide an alternate proof to show that the metric dimension of $\mathcal{E}_{\mathbb{Z}_{n}}$ is $\leq k$ when $n=\prod_{i=1}^{k}p_{i}$. In section $4$, we calculate the first and the second Zagreb index of the graph $\mathcal{E}_{\mathbb{Z}_{n}}$ for any $n\geq 4$.

Throughout this paper, $\mathbb{Z}_{n}=\mathbb{Z}/n\mathbb{Z}$, where $n\geq 4$ and $n$ is not a prime.

2 Preliminaries

In this section, we list some definitions and results that are needed for the present study.

Definition 2.1.

A subset $W$ of vertices of a connected graph $\Gamma$ is said to resolve $\Gamma$, if each vertex of $\Gamma$ is uniquely determined by its vector of distances to the vertices of $W$. In general, for an ordered subset $W=\{w_{1},w_{2},\cdots,w_{k}\}$ of vertices of a connected graph $\Gamma$ and a vertex $v\in V(\Gamma)\backslash W$ of $\Gamma$, the metric representation of $v$ with respect to $W$ is the $k-$vector $r(v|W)=(d(v,w_{1}),d(v,w_{2}),\cdots d(v,w_{k}))$. The set $W$ is a resolving set for $\Gamma$ if $r(v|W)\neq r(u|W)$, for any pair of distinct vertices $u,v\in V(\Gamma)\backslash W$.

The resolving set, the metric representation of a vertex, and the metric dimension of a graph are also called the locating set, locating code of a vertex, and locating number of a graph respectively.

Definition 2.2.

Let $\Gamma$ be a connected graph with order $n\geq 2$. The metric dimension $dim(\Gamma)$ of $\Gamma$, is defined as $dim(\Gamma)=min\{|W|:W\textit{is a resolving set of}\ \Gamma\}$ and such a set $W$ is the metric basis for $\Gamma$.
For every connected graph $\Gamma$ of order $n\geq 2$, $1\leq dim(\Gamma)\leq n-1$.

Definition 2.3.

Let $\Gamma$ be a connected graph with order $n\geq 2$. Two distinct vertices $u$ and $v$ are said to be distance similar if $d(u,x)=d(v,x)$, for all $x\in V(\Gamma)\backslash\{u,v\}$. It can be verified that the distance relation is an equivalence relation on $V(\Gamma)$ and two vertices are distance similar if either $uv\notin E(\Gamma)$ and $N(u)=N(v)$ or $uv\in E(\Gamma)$ and $N[u]=N[v]$.

Theorem 2.4.

[chartrand2000resolvability] Let $\Gamma$ be a connected graph with order $n\geq 2$ and $W$ be a metric basis for $\Gamma$. Then $dim(\Gamma)=n-1$ if and only if $\Gamma\cong K_{n}$.

Theorem 2.5.

[pirzada2014locating] Let $\Gamma$ be a connected graph and $V(\Gamma)$ is partitioned into $k$ distinct distance similar classes $X_{1},X_{2},\cdots X_{k}$. Then

1. Any resolving set $W$ contains all but at most one vertex from each $X_{i}$.

2. If $t$ is the number of distance similar classes that consist of a single vertex, then $|V(\Gamma)|-k\leq dim(\Gamma)\leq V(\Gamma)|-k+t$.

Theorem 2.6.

[amjadi2018essential] Let $R$ be a commutative ring with unity. Then, $\mathcal{E}_{R}$ is a finite graph if and only if every vertex of $\mathcal{E}_{R}$ has finite degree.

In [p2023structure], the authors determined the structure of essential ideal graph of the ring $\mathbb{Z}_{n}$ by defining an equivalence relation on the set $\mathscr{U}$ of nonessential ideals of $\mathbb{Z}_{n}$ as follows:

Definition 2.7.

Let $\Xi=\{1,2,\cdots,k\}$ be an index set. For an ideal $\hat{I}$ of $\mathscr{U}$, define a subset $\Xi_{\hat{I}}$ of $\Xi$ by, $\Xi_{\hat{I}}=\{i:r_{i}=m_{i}\ \text{in}\ \hat{I}\}$.

Definition 2.8.

Let $\hat{I}$ and $\hat{J}$ be any two ideals of $\mathscr{U}$. We define a relation $\preccurlyeq$ on $\mathscr{U}$ by $\hat{I}\preccurlyeq\hat{J}$ if and only if $\Xi_{\hat{I}}=\Xi_{\hat{J}}$.

Thus, $\mathscr{U}$ is partitioned into $2^{k}-2$ equivalent classes, and each equivalent class is denoted by $[\hat{I}]$. For example, if $n=p_{1}^{2}p_{2}^{3}p_{3}p_{4}p_{5}^{4}$ and $\hat{I}=\langle p_{2}^{3}p_{4}\rangle$ is the representative ideal then, the corresponding equivalent class $[\hat{I}]$ is the set $X_{\hat{I}}=\{\langle p_{1}^{r_{1}}p_{2}^{3}p_{4}p_{5}^{r_{5}}\rangle:0\leq r_{1}\leq 1,\ \text{and}\ 0\leq r_{5}\leq 3\}$.

Lemma 2.9.

Let $\hat{K}$ and $\hat{L}$ be two vertices of any two of the $2^{k}-2$ equivalent classes, say $[\hat{I}]$ and $[\hat{M}]$ respectively. Then $\hat{K}$ and $\hat{L}$ are adjacent in $\mathcal{E}_{\mathbb{Z}_{n}}$ if and only if $\Xi_{\hat{I}}\cap\Xi_{\hat{M}}=\phi$.

The next theorem can be found in [p2023structure], which determines the structure of the induced subgraph $\mathcal{E}_{\mathbb{Z}_{n}}(\mathscr{U})$. The next theorem gives the structure of the induced subgraph $\mathcal{E}_{\mathbb{Z}_{n}}(\mathscr{U})$.

Theorem 2.10.

[p2023structure] Let $n=p_{1}^{m_{1}}p_{2}^{m_{2}}\cdots p_{k}^{m_{k}}$, where $p_{1}<p_{2}<\cdots<p_{k}$ are primes, $k\geq 2$, and $m_{i}>1$ for at least one $i$.Then, the induced subgraph $\mathcal{E}_{\mathbb{Z}_{n}}(\mathscr{U})$ is the generalized join of certain null graphs given by,

$$\begin{split}\mathcal{E}_{\mathbb{Z}_{n}}(\mathscr{U})=\mathscr{G}&[\mathcal{E}_{\mathbb{Z}_{n}}([\langle p_{1}^{m_{1}}\rangle]),\cdots,\mathcal{E}_{\mathbb{Z}_{n}}([\langle p_{k}^{m_{k}}\rangle]),\mathcal{E}_{\mathbb{Z}_{n}}([\langle p_{1}^{m_{1}}p_{2}^{m_{2}}\rangle]),\cdots,\\ &\mathcal{E}_{\mathbb{Z}_{n}}([\langle p_{k-1}^{m_{k}-1}p_{k}^{m_{k}}\rangle]),\cdots,\mathcal{E}_{\mathbb{Z}_{n}}([\langle p_{2}^{m_{2}}\cdots p_{k-1}^{m_{k-1}}p_{k}^{m_{k}}\rangle])],\end{split}$$

where $\mathcal{E}_{\mathbb{Z}_{n}}([\hat{I}])=\overline{K}_{\displaystyle\prod_{i\notin\Xi_{\hat{I}}}m_{i}}$ for the representative ideal $\hat{I}(\text{vertex of}\ \mathscr{G})$ of the equivalent class $[\hat{I}]$.

The following theorem determines the structure of $\mathcal{E}_{\mathbb{Z}_{n}}$ as the join of a complete graph induced by the essential ideals of $\mathbb{Z}_{n}$ and the induced subgraph $\mathcal{E}_{\mathbb{Z}_{n}}(\mathscr{U})$.

Theorem 2.11.

[p2023structure] Let $n=p_{1}^{m_{1}}p_{2}^{m_{2}}\cdots p_{k}^{m_{k}}$, where $p_{1}<p_{2}<\cdots<p_{k}$ are primes, and $m_{i}>1$ for at least one $i$. Then, the essential ideal graph $\mathcal{E}_{\mathbb{Z}_{n}}\cong K_{m}\vee H$, where $K_{m}$ is the complete graph on $m=\prod_{i=1}^{k}m_{i}-1$ vertices and

$$\begin{split}H=\mathscr{G}&[\mathcal{E}_{\mathbb{Z}_{n}}([\langle p_{1}^{m_{1}}\rangle]),\cdots,\mathcal{E}_{\mathbb{Z}_{n}}([\langle p_{k}^{m_{k}}\rangle]),\mathcal{E}_{\mathbb{Z}_{n}}([\langle p_{1}^{m_{1}}p_{2}^{m_{2}}\rangle]),\cdots,\\ &\mathcal{E}_{\mathbb{Z}_{n}}([\langle p_{k-1}^{m_{k}-1}p_{k}^{m_{k}}\rangle]),\cdots,\mathcal{E}_{\mathbb{Z}_{n}}([\langle p_{2}^{m_{2}}\cdots p_{k-1}^{m_{k-1}}p_{k}^{m_{k}}\rangle])].\end{split}$$

3 Metric Dimension of $\mathcal{E}_{\mathbb{Z}_{n}}$

In this section, we compute the metric dimension of the essential ideal graph of $\mathbb{Z}_{n}$.

Theorem 3.1.

Let $R$ be a commutative ring with unity. Then, $dim(\mathcal{E}_{R})$ is finite if and only if $R$ is finite.

Proof.

If $R$ is finite, obviously, $dim(\mathcal{E}_{R})$ is finite. Conversely, suppose that $dim(\mathcal{E}_{R})=k<\infty$. This ensures that each vertex of $\mathcal{E}_{R})$ has a unique $k$-vector metric representation with respect to a minimum resolving set $W$ of cardinality $k$. Since $diam(\mathcal{E}_{R})=3<\infty$, for every vertex $v\in V(\mathcal{E}_{R})\backslash W$, there are only $4^{k}$ choices for $r(v|W)$. Hence, $|V(\mathcal{E}_{R})|\leq 4^{k}+k$.
∎

The next result follows directly from Theorems 2.6 and 3.1.

Corollary 3.2.

Let $R$ be a commutative ring with unity. Then, $dim(\mathcal{E}_{R})$ is finite if and only if every vertex of $\mathcal{E}_{R}$ has finite degree.

Lemma 3.3.

Let $n=p_{1}p_{2}\cdots p_{k}$, where $p_{1}<p_{2}<\cdots<p_{k}$ are primes and let $d_{1}$ and $d_{2}$ be two distinct nontrivial proper divisors of $n$. Then, $gcd(d_{1},d_{2})=1$ if and only if $n\mid(\frac{n}{d_{1}})(\frac{n}{d_{2}})$

Proof.

Assume that $gcd(d_{1},d_{2})=1$. Then, there exist integers $x$ and $y$ such that $1=d_{1}x+d_{2}y$. Now,

$$\begin{split}(\frac{n}{d_{1}})=&nx+(\frac{n}{d_{1}})d_{2}y\\ (\frac{n}{d_{2}})=&(\frac{n}{d_{2}})d_{1}x+d_{2}y\\ \text{and hence}\ (\frac{n}{d_{1}})(\frac{n}{d_{2}})=&n(x_{1}+2nxy+y_{1}),\\ \end{split}$$

where $x_{1}=(\frac{n}{d_{2}})d_{1}x^{2},\ y_{1}=(\frac{n}{d_{1}})d_{2}y^{2}$. Thus, $n|(\frac{n}{d_{1}})(\frac{n}{d_{2}})$. For the converse, suppose that $gcd(d_{1},d_{2})=d>1$. Then $d=p_{i_{1}}p_{i_{2}}\cdots p_{i_{t}}$, where $p_{i_{1}},p_{i_{2}},\cdots,p_{i_{t}}$ are primes such that $i_{1}<i_{2}<\cdots<i_{t}$ and $1\leq i_{t}\leq k-1$ so that $d_{1}=r_{1}d$, and $d_{2}=r_{2}d$. Consequently, both divisors $(\frac{n}{d_{1}})$ and $(\frac{n}{d_{2}})$ of $n$ do not have $d$ as a factor and hence $n\nmid(\frac{n}{d_{1}})(\frac{n}{d_{2}})$. ∎

Theorem 3.4.

Let $R_{1}=\displaystyle\prod_{i=1}^{k}F_{i}$, where each $F_{i}$ is a field and let $R_{2}=\mathbb{Z}_{n}$ for $n=p_{1}p_{2}\cdots p_{k}$, where $p_{i}$’s are distinct primes for $1\leq i\leq k$. Then, $\mathcal{E}_{R_{1}}\cong\mathcal{E}_{R_{2}}\cong\mathbb{AIG}(R_{2})$.

Proof.

We first note that the vertices of $\mathcal{E}_{R_{1}}$ are the nonzero proper ideals of the ring $\prod_{i=1}^{k}F_{i}$, given by $\hat{I}=\prod_{i=1}^{k}\hat{I_{i}}$, where $\hat{I_{i}}=\langle 0\rangle$ for at least one $i$ and $\hat{I_{i}}=F_{i}$ for at least one $i$. Thus, $|V(\mathcal{E}_{R_{1}})|=2^{k}-2=|V(\mathcal{E}_{R_{2}})|=|V(\mathbb{AIG}(R_{2}))|$. Also,
$V(\mathcal{E}_{R_{2}})=V(\mathbb{AIG}(R_{2}))=\{\langle d\rangle:d\ \text{is a positive proper divisor of}\ n\}$. For the divisor $d$ of $n$, define a map $\varphi:V(\mathcal{E}_{R_{2}})\rightarrow V(\mathbb{AIG}(R_{2}))$ by $d\longmapsto\frac{n}{d}$(divisor conjugate of $d$). Since each divisor $d$ of $n$ has a unique divisor conjugate $\frac{n}{d}$, and $1<\frac{n}{d}<n$ for $1<d<n$, it follows immediately that $\varphi$ is both one-one and onto. Now, Lemma 3.3 assures that two vertices $\langle d_{1}\rangle$ and $\langle d_{2}\rangle$ are adjacent in $\mathcal{E}_{R_{2}}$ if and only if $\varphi(\langle d_{1}\rangle)$ and $\varphi(\langle d_{2}\rangle)$ are adjacent in $\mathbb{AIG}(R_{2})$. Thus, $\varphi$ is an isomorphism and hence $\mathcal{E}_{R_{2}}\cong\mathbb{AIG}(R_{2})$.

Now, for each vertex $\hat{I}=\prod_{i=1}^{k}\hat{I_{i}}$ of $\mathcal{E}_{R_{1}}$, we define a subset $\Theta_{\hat{I}}$ of the index set $\{1,2,\cdots,k\}$ such that $\hat{I_{i}}=\begin{cases}\langle 0\rangle,&\text{if}\ i\in\Theta_{\hat{I}}\\ F_{i},&\text{otherwise}.\end{cases}$ Obviously, two distinct vertices $\hat{I}$ and $\hat{J}$ are adjacent in $\mathcal{E}_{R_{1}}$ if and only if $\Theta_{\hat{I}}\cap\Theta_{\hat{J}}=\phi$. Define a map $\psi:V(\mathcal{E}_{R_{1}})\rightarrow V(\mathbb{AIG}(R_{2}))$ by $\psi(\hat{I})=\langle\displaystyle\prod_{i\notin\Theta_{\hat{I}}}p_{i}\rangle$. Clearly, $\psi$ is a well defined bijection preserving adjacencies and nonadjacencies in $\mathcal{E}_{R_{1}}$ and $\mathbb{AIG}(R_{2})$, and hence $\mathcal{E}_{R_{1}}\cong\mathbb{AIG}(R_{2})$. ∎

In [aijaz2020annihilating], the authors computed the metric dimension of the annihilating ideal graph of the rings $\displaystyle\prod_{i=1}^{k}F_{i}$ and $\mathbb{Z}_{n}$, $n=p_{1}p_{2}\cdots p_{k}$.

Theorem 3.5.

[aijaz2020annihilating] For $R=\displaystyle\prod_{i=1}^{k}F_{i}\ \text{or}\ \mathbb{Z}_{n}$, $n=p_{1}p_{2}\cdots p_{k}$, the following hold:

1. $dim(\mathbb{AIG}(R))=k-1$ for $1\leq k\leq 4$.

2. $dim(\mathbb{AIG}(R))=5$ for $k=5$.

3. $dim(\mathbb{AIG}(R))\leq k$ for $k\geq 6$.

The proof is developed by showing that the annihilating ideal graph $\mathbb{AIG}(R)$ for $R=\displaystyle\prod_{i=1}^{k}F_{i}\ \text{or}\ \mathbb{Z}_{n}$, $n=p_{1}p_{2}\cdots p_{k}$ is isomorphic to the zero divisor graph ($\mathbb{ZDG}$) of the boolean ring $\displaystyle\prod_{i=1}^{k}\mathbb{Z}_{2}$ and applying the result on metric dimension of zero divisor graph of $\displaystyle\prod_{i=1}^{k}\mathbb{Z}_{2}$ [Proposition $6.2$ and Theorem $6.3$ of [raja2016locating]]. Hence by Theorems 3.4 and 3.5, we can have the following result.

Proposition 3.6.

Let $R=\displaystyle\prod_{i=1}^{k}F_{i}\ \text{or}\ \mathbb{Z}_{n}$, $n=p_{1}p_{2}\cdots p_{k}$. Then,

1. $dim(\mathcal{E}_{R})=k-1$ for $1\leq k\leq 4$.

2. $dim(\mathcal{E}_{R})=5$ for $k=5$.

3. $dim(\mathcal{E}_{R})\leq k$ for $k\geq 6$.

In the following theorem, we give another proof for computing the metric dimension of $\mathcal{E}_{\mathbb{Z}_{n}}$ for $n=p_{1}p_{2}\cdots p_{k}$, $k\geq 6$, by finding a minimal resolving set of $\mathcal{E}_{\mathbb{Z}_{n}}$. For this, we make use of the following Lemma.

Lemma 3.7.

Let $R=\mathbb{Z}_{n}$, $n=p_{1}p_{2}\cdots p_{k}$. Then, for any two vertices $\hat{I}$ and $\hat{J}$ of $\mathcal{E}_{R}$

1. $d(\hat{I},\hat{J})=2$ if and only if $\hat{I}+\hat{J}\neq R$ and $\hat{I}\cap\hat{J}\neq 0$.

2. $d(\hat{I},\hat{J})=3$ if and only if $\hat{I}+\hat{J}\neq R$ and $\hat{I}\cap\hat{J}=0$.

Proof.

$(i)$ First, suppose that $d(\hat{I},\hat{J})=2$. Obviously, $\hat{I}+\hat{J}\neq R$. Thus, it remains to prove that $\hat{I}\cap\hat{J}\neq 0$. If possible, let $\hat{I}\cap\hat{J}=0$. Then, any prime not in the generator of the ideal $\hat{I}$ must be in the generator of the ideal $\hat{J}$ and vice versa. Hence, if $\hat{K}$ is a vertex adjacent to the vertex $\hat{I}$, then it cannot be adjacent to the vertex $\hat{J}$ as the generators of both $\hat{K}$ and $\hat{J}$ have at least one common prime factor. This leads to the conclusion that $d(\hat{I},\hat{J})>2$, is a contradiction. Thus, $\hat{I}\cap\hat{J}\neq 0$. For the converse, assume that $\hat{I}+\hat{J}\neq R$ and $\hat{I}\cap\hat{J}\neq 0$. Then $d(\hat{I},\hat{J})>1$. Since $\hat{I}\cap\hat{J}\neq 0$, there must exist at least one prime number $p_{s}$ such that $p_{s}$ is not a prime factor of generators of both ideals $\hat{I}$ and $\hat{J}$. Thus, if $\hat{S}=\langle p_{s}\rangle$, we have $\hat{I}\sim\hat{S}\sim\hat{J}$. Consequently, $d(\hat{I},\hat{J})=2$.
$(ii)$ Result follows as a direct consequence of the proof of Case $1$. ∎

In the following theorem, we give another proof for computing the metric dimension of $\mathcal{E}_{\mathbb{Z}_{n}}$ for $n=p_{1}p_{2}\cdots p_{k}$.

Theorem 3.8.

Let $n=p_{1}p_{2}\cdots p_{k}$, where $p_{1}<p_{2}<\cdots<p_{k}$ are primes and $k\geq 6$. Then $dim(\mathcal{E}_{\mathbb{Z}_{n}})\leq k$.

Proof.

Consider the set $W$ consisting of all minimal ideals of $\mathcal{E}_{\mathbb{Z}_{n}}$ as in the following order:

$$W=\{\langle p_{1}p_{2}\cdots p_{k-1}\rangle,\langle p_{1}p_{2}\cdots p_{k-2}p_{k}\rangle,\cdots,\langle p_{2}p_{3}\cdots p_{k}\rangle\}.$$

claim: $W$ is a resolving set of $\mathcal{E}_{\mathbb{Z}_{n}}$
We need to show that each vertex $v\in V(\mathcal{E}_{\mathbb{Z}_{n}})\backslash W$ has a unique representation of distances with respect to $W$. For this, take any two vertices of $V(\mathcal{E}_{\mathbb{Z}_{n}})\backslash W$ of the form $\hat{I}=\langle p_{i_{1}}p_{i_{2}}\cdots p_{i_{t}}\rangle$ and $\hat{J}=\langle p_{j_{1}}p_{j_{2}}\cdots p_{j_{s}}\rangle$, where $p_{i_{1}},p_{i_{2}},\cdots,p_{i_{t}},p_{j_{1}},p_{j_{2}},\cdots,p_{j_{s}}$ are primes such that $i_{1}<i_{2}<\cdots<i_{t}$ and $j_{1}<j_{2}<\cdots<j_{s}$ not necessarily distinct and $1\leq i_{t},j_{s}\leq k-2$. Then three cases may occur- either $t<s$, or $t=s$, or $t>s$.
Case $1:t<s$
Then, there exists at least one prime $p_{j_{l}}$ which is in the generator of $\hat{J}$ but not in that of $\hat{I}$. Now, consider a vertex $\hat{P}$ in $W$ such that $p_{j_{l}}$ is not in the generator of $\hat{P}$. Then $d(\hat{I},\hat{P})=2$, by Lemma 3.7$(1)$. That is, $\hat{I}\sim\langle p_{j_{l}}\rangle\sim\hat{P}$. However, since $\hat{J}$ is not adjacent to the vertex $\langle p_{j_{l}}\rangle$ to which $\hat{P}$ is only adjacent, $d(\hat{J},\hat{P})=3$. Then, the coordinate corresponding to the vertex $\hat{P}$ of $W$ in the $k$-vector of both $\hat{I}$ and $\hat{J}$ are distinct. Hence, $r(\hat{I}|W)\neq r(\hat{J}|W)$.
Case $2:t=s$
In this case, at least one prime is not common in the generators of both $\hat{I}$ and $\hat{J}$. Without loss of generality, assume that $p_{i_{h}}$ is in the generator of $\hat{I}$ but not in that of $\hat{J}$ and $p_{j_{l}}$ is in the generator of $\hat{J}$ but not in that of $\hat{I}$. Consider the vertex $\hat{Q}\in W$ such that the generator of $\hat{Q}$ contains $p_{j_{l}}$ as a factor but not $p_{i_{h}}$. Then, $\hat{Q}$ is adjacent only to the vertex $\langle p_{i_{h}}\rangle$ and the latter is not adjacent to $\hat{I}$. Hence, by Lemma 3.7, $d(\hat{I},\hat{Q})=3$ and $d(\hat{J},\hat{Q})=2$.
Case $3:t>s$
Here, there is at least one prime $p_{i_{h}}$ in the generator of $\hat{I}$ but not in that of $\hat{J}$. Then, by Lemma 3.7, $d(\hat{I},\hat{K})=3$, and $d(\hat{J},\hat{K})=2$, for the vertex $\hat{K}\in W$ having $p_{i_{h}}$ not in the generator of $\hat{K}$. This proves that $r(\hat{I}|W)\neq r(\hat{J}|W)$, for any two distinct vertices $\hat{I}$ and $\hat{J}$ in $V(\mathcal{E}_{\mathbb{Z}_{n}})\backslash W$. Hence $W$ is a resolving set of cardinality $k$ and $dim(\mathcal{E}_{\mathbb{Z}_{n}})\leq k$. ∎

Proposition 3.9.

Let $T=|V(\mathcal{E}_{\mathbb{Z}_{n}})|$. Then, $dim(\mathcal{E}_{\mathbb{Z}_{n}})=T-1$ if and only if either $n=p^{m},\ m>1$ or $n=p_{1}p_{2}$.

Proof.

It is obvious that $dim(\mathcal{E}_{\mathbb{Z}_{n}})=T-1$ when $n=p^{m},\ m>1$ or $n=p_{1}p_{2}$. For the converse, assume that $dim(\mathcal{E}_{\mathbb{Z}_{n}})=T-1$. Then, $\mathcal{E}_{\mathbb{Z}_{n}}$ is complete by Theorem 2.4. Suppose $n\neq p_{1}p_{2}$. To prove $n=p^{m},\ m>1$, assume to the contrary that $n=p_{1}^{\alpha_{1}}p_{2}^{\alpha_{2}}\cdots p_{k}^{\alpha_{k}}$, $k\geq 2$ and $\alpha_{i}>1$ for at least two $i$ (say, $\alpha_{1},\alpha_{2}$). Now, consider the two vertices $\hat{I}=\langle p_{1}^{\alpha_{1}}\rangle$ and $\hat{J}=\langle p_{1}^{\alpha_{1}}p_{2}^{\alpha_{2}}\rangle$. Obviously, $\hat{I}$ and $\hat{J}$ are nonadjacent in $\mathcal{E}_{\mathbb{Z}_{n}}$ contradicting the fact that $\mathcal{E}_{\mathbb{Z}_{n}}$ is complete. ∎

By Theorem 2.11, $\mathcal{E}_{\mathbb{Z}_{n}}\cong K_{m}\vee\mathscr{G}[\Gamma_{1},\Gamma_{2},\cdots,\Gamma_{2^{k}-2}]$, where $\Gamma_{i}=\mathcal{E}_{\mathbb{Z}_{n}}([\hat{I}])$ for each of the equivalence class $[\hat{I}]$ of the partition on the set of nonessential ideals of $\mathcal{E}_{\mathbb{Z}_{n}}$. This can be further viewed as $\mathcal{E}_{\mathbb{Z}_{n}}\cong\mathscr{G}[K_{m},\Gamma_{1},\Gamma_{2},\cdots,\Gamma_{2^{k}-2}]$, since the vertices of the subgraph $K_{m}$ are adjacent to all the vertices of the subgraphs $\Gamma_{i}$ for $1\leq i\leq 2^{k}-2$. Also, note that the vertices in each of the induced subgraphs $K_{m}$ and $\Gamma_{i}$ for $1\leq i\leq 2^{k}-2$ are distance similar so that $V(\mathcal{E}_{\mathbb{Z}_{n}})$ is partitioned into $2^{k}-1$ distance similar classes $X,X_{1},X_{2},\cdots,X_{2^{k}-2}$ as follows.

$$X= \{\langle p_{1}^{r_{1}}p_{2}^{r_{2}}\cdots p_{k}^{r_{k}}\rangle:0\leq r_{i}\leq m_{i}-1\ \text{for}\ 1\leq i\leq k\}\backslash\mathbb{Z}_{n},$$

$$X_{1}= X_{\langle p_{1}^{m_{1}}\rangle}=\{\langle p_{1}^{m_{1}}p_{2}^{r_{2}}\cdots p_{k}^{r_{k}}\rangle:0\leq r_{i}\leq m_{i}-1\ \text{for}\ 2\leq i\leq k\},$$

$$\vdots$$

$$X_{k}= X_{\langle p_{1}^{m_{k}}\rangle}=\{\langle p_{1}^{r_{1}}p_{2}^{r_{2}}\cdots p_{k}^{m_{k}}\rangle:0\leq r_{i}\leq m_{i}-1\ \text{for}\ 1\leq i\leq k-1\},$$

$$X_{k+1}= X_{\langle p_{1}^{m_{1}}p_{2}^{m_{2}}\rangle}=\{\langle p_{1}^{m_{1}}p_{2}^{m_{2}}\cdots p_{k}^{r_{k}}\rangle:0\leq r_{i}\leq m_{i}-1\ \text{for}\ 3\leq i\leq k\},$$

$$\vdots$$

$$X_{2^{k}-2}= X_{\langle p_{2}^{m_{2}}\cdots p_{k}^{m_{k}}\rangle}=\{\langle p_{1}^{r_{1}}p_{2}^{m_{2}}\cdots p_{k}^{m_{k}}\rangle:0\leq r_{1}\leq m_{1}-1\}.$$

Here, $X=V(K_{m})$ and $X_{i}=V(\Gamma_{i})=[\hat{I}]$ for each of $2^{k}-2$ equivalent class $[\hat{I}]$. By Theorem 2.5, any resolving set $W$ of $\mathcal{E}_{\mathbb{Z}_{n}}$ must contain all but at most one vertex from each of the partitioned sets $X,\ X_{i}$ for $i=1,2,\cdots,2^{k}-2$. Hence, for any resolving set $W$,

$$\begin{split}|W|\geq&|X|-1+|X_{1}|-1+|X_{2}|-1+\cdots+|X_{2^{k}-2}|-1\\ \geq&m-1+T-m\underbrace{-1-1\cdots-1}_{2^{k}-2\ \text{times}}\\ \geq&T-(2^{k}-1).\\ \end{split}$$ (1)

Now, we identify the values of $n$ for which these bounds are attained by computing the metric dimension of $\mathcal{E}_{\mathbb{Z}_{n}}$.

Theorem 3.10.

Let $n=p_{1}^{m_{1}}p_{2}^{m_{2}}\cdots p_{k}^{m_{k}}$, where $p_{1}<p_{2}<\cdots<p_{k}$ are primes, $k\geq 2$, and $m_{i}>1$ for at least one $i$. Then

$$dim(\mathcal{E}_{\mathbb{Z}_{n}})=\begin{cases}T-(2^{k}-1),&\text{if}\ m_{i}>1\ \text{for at least two}\ i,\\ T-(2^{k}-2),&\text{if}\ m_{i}>1\ \text{for exactly one}\ i.\\ \end{cases}$$

Proof.

By Equation $(1)$, we see that any resolving set of $\mathcal{E}_{\mathbb{Z}_{n}}$ must contain at least $T-(2^{k}-1)$ vertices consisting of all but at most one vertex of each of the distance similar partitioned sets. Case $1$: $m_{i}>1$ for at least two $i$
Here, it remains to show that there exists a resolving set of cardinality $T-(2^{k}-1)$. Take $W$ as an ordered set consisting of $m-1$ vertices of $X$, followed by $|X_{i}|-1$ vertices of the sets $X_{i}$, for $1\leq i\leq 2^{k}-2$. Without loss of generality, let

$$\begin{split}W=&X\backslash\{\langle p_{1}^{m_{1}-1}p_{2}^{m_{2}-1}\cdots p_{k}^{m_{k}-1}\rangle\}\bigcup X_{1}\backslash\{\langle p_{1}^{m_{1}}p_{2}^{m_{2}-1}\cdots p_{k}^{m_{k}-1}\rangle\}\bigcup\cdots\\ &\bigcup X_{2^{k}-2}\backslash\{\langle p_{1}^{m_{1}-1}p_{2}^{m_{2}}\cdots p_{k}^{m_{k}}\rangle\}.\\ \end{split}$$

Since $\langle p_{1}^{m_{1}-1}p_{2}^{m_{2}-1}\cdots p_{k}^{m_{k}-1}\rangle$ is the only essential ideal of the set $V\backslash W$, we see that

$$r(\langle p_{1}^{m_{1}-1}p_{2}^{m_{2}-1}\cdots p_{k}^{m_{k}-1}\rangle|W)=(1,1,\cdots,1)\neq r(v|W)\ \text{for any}\ v\in V\backslash W.$$

Now, take any two vertices $\hat{I}$ and $\hat{J}$ of $V\backslash W$ with respective index sets $\Xi_{\hat{I}}$ and $\Xi_{\hat{J}}$. we claim that $r(\hat{I}|W)\neq r(\hat{J}|W)$. For $\hat{I}$ and $\hat{J}$, either $\Xi_{\hat{I}}\cap\Xi_{\hat{J}}=\phi$ or $\Xi_{\hat{I}}\cap\Xi_{\hat{J}}\neq\phi$. If $\Xi_{\hat{I}}\cap\Xi_{\hat{J}}=\phi$, then there exist at least two distinct primes $p_{i}$ and $p_{j}$ such that $p_{i}^{m_{i}}\in\hat{I}\ \text{but}\notin\hat{J}$ and $p_{j}^{m_{j}}\in\hat{J}\ \text{but}\ \notin\hat{I}$. Now, $d(\hat{I},v)=2$ and $d(\hat{J},v)=1$ for any $v\in X_{\langle p_{i}^{m_{i}}\rangle}$. Consequently, $r(\hat{I}|W)$ will have $2$ in all the co-ordinates corresponding to the elements from the set $X_{\langle p_{i}^{m_{i}}\rangle}$ whereas $r(\hat{J}|W)$ will have $1$ in the respective coordinates. So $r(\hat{I}|W)\neq r(\hat{J}|W)$. If $\Xi_{\hat{I}}\cap\Xi_{\hat{J}}\neq\phi$, then it can be either $\Xi_{\hat{I}}$ or $\Xi_{\hat{J}}$ or none of these. Let $\Xi_{\hat{I}}\cap\Xi_{\hat{J}}=\Xi_{\hat{I}}$. Since $\Xi_{\hat{J}}\neq\Xi_{\hat{I}}$, there exists at least one prime $p_{j}$ such that $p_{j}^{mj}$ is in $\hat{J}$ but not in $\hat{I}$. Hence, $r(\hat{I}|W)$ will have $1$ in all the co-ordinates corresponding to the elements from the set $X_{\langle p_{j}^{m_{j}}\rangle}$ and $r(\hat{J}|W)$ will have $2$ in all the co-ordinates corresponding to the elements from the set $X_{\langle p_{j}^{m_{j}}\rangle}$. Thus, $r(\hat{I}|W)\neq r(\hat{J}|W)$. Through a similar argument, we see that $r(\hat{I}|W)\neq r(\hat{J}|W)$ whenever $\Xi_{\hat{I}}\cap\Xi_{\hat{J}}=\Xi_{\hat{J}}$. Now, in the last case, there must exist at least two primes $p_{i}$ and $p_{j}$ such that $p_{i}^{m_{i}}$ is in $\hat{I}$ but not in $\hat{J}$ and $p_{j}^{m_{j}}$ is in $\hat{J}$ but not in $\hat{I}$ leading to $r(\hat{I}|W)\neq r(\hat{J}|W)$.
Case $2$: If $m_{i}>1$ for exactly one $i$
without loss of generality take $n=p_{1}^{m_{1}}p_{2}\cdots p_{k}$, $m_{1}>1$. We know that any resolving set contains all but at most one vertex of each of the distance similar partitioned sets and by Equation $(1)$, $|W|\geq T-(2^{k}-1)$, for any resolving set $W$. At first, we show that there is no resolving set of cardinality $T-(2^{k}-1)$. For this, take $W$ as an ordered set consisting of $m-1$ vertices of $X$ followed by $|X_{i}|-1$ vertices of $X_{i}$ for each $i$. That is,

$$\begin{split}W=&\{\langle p_{1}\rangle,\langle p_{1}^{2}\rangle,\langle p_{1}^{m_{1}-2}\rangle,\langle p_{2}\rangle,\langle p_{1}p_{2}\rangle,\cdots,\langle p_{1}^{m_{1}-2}p_{2}\rangle,\cdots,\langle p_{k}\rangle,\langle p_{1}p_{k}\rangle,\cdots,\langle p_{1}^{m_{1}-2}p_{k}\rangle,\\ &\langle p_{2}p_{3}\rangle,\langle p_{1}p_{2}p_{3}\rangle,\cdots,\langle p_{1}^{m_{1}-2}p_{2}p_{3}\rangle,\cdots,\langle p_{2}p_{3}\cdots p_{k}\rangle,\cdots,\langle p_{1}^{m_{1}-2}p_{2}p_{3}\cdots p_{k}\rangle\}.\\ \end{split}$$

Consider the vertices $\hat{I}=\langle p_{1}^{m_{1}-1}\rangle$ and $\hat{J}=\langle p_{1}^{m_{1}}\rangle$ of $V\backslash W$. Since $\hat{I}$ is essential, $r(\hat{I}|W)=(1,1,\cdots,1)$. For $\hat{J},\ \Xi_{\hat{J}}=\{1\}$ and for any vertex $w\in W$, $1\notin\Xi_{w}$. Consequently, $d(\hat{J},w)=1$ and $r(\hat{J}|W)=(1,1,\cdots,1)=r(\hat{I}|W)$. Thus, there is no resolving set of cardinality $T-(2^{k}-1)$. Now, take $W^{\prime}$ as an ordered set obtained by adjoining one more vertex (say $\langle p_{1}^{m_{1}}\rangle$) to $W$. Let $\hat{I}$ and $\hat{J}$ be two distinct vertices of $V\backslash W$ with index sets $\Xi_{\hat{I}}$ and $\Xi_{\hat{J}}$ respectively. Then there may occur two cases- either $\Xi_{\hat{I}}\cap\Xi_{\hat{J}}=\phi$ or $\Xi_{\hat{I}}\cap\Xi_{\hat{J}}\neq\phi$. Proceeding in the same manner as in the proof of case $1$, we see that $W^{\prime}$ is a resolving set of $\mathcal{E}_{\mathbb{Z}_{n}}$ of minimum cardinality. ∎

Corollary 3.11.

Let $n=p_{1}^{m_{1}}p_{2}^{m_{2}}$, where $p_{1}<p_{2}$ be primes. Then

$$dim(\mathcal{E}_{\mathbb{Z}_{n}})=\begin{cases}2m-2,&\textit{if}\ m_{1}=m\geq 2,m_{2}=1\textit{or vice versa},\\ m_{1}m_{2}+m_{1}+m_{2}-4,&\textit{if}\ m_{1},m_{2}>1.\end{cases}.$$

Example 3.12.

• Consider the graph $\mathcal{E}_{\mathbb{Z}_{n}}$ for $n=p_{1}^{m_{1}}p_{2}^{m_{2}}p_{3}$, $m_{1},m_{2}>1$. Then the distance similar partition of vertices is given by,

  $$\begin{split}X=&\{\langle p_{1}^{r_{1}}p_{2}^{r_{2}}\rangle:0\leq i\leq m_{i}-1\ \text{for}\ i=1,2\}\backslash\mathbb{Z}_{n},|X|=m_{1}m_{2}-1\\ X_{1}=&\{\langle p_{1}^{m_{1}}p_{2}^{r_{2}}\rangle:0\leq r_{2}\leq m_{2}-1\},|X_{1}|=m_{2},\\ X_{2}=&\{\langle p_{1}^{r_{1}}p_{2}^{m_{2}}\rangle:0\leq r_{1}\leq m_{1}-1\},|X_{2}|=m_{1},\\ X_{3}=&\{\langle p_{1}^{r_{1}}p_{2}^{r_{2}}p_{3}\rangle:0\leq r_{i}\leq m_{i}-1\ \text{for}\ i=1,2\},|X_{3}|=m_{1}m_{2},\\ X_{4}=&\{\langle p_{1}^{m_{1}}p_{2}^{m_{2}}\rangle\},\\ X_{5}=&\{\langle p_{1}^{m_{1}}p_{2}^{r_{2}}p_{3}\rangle:0\leq r_{2}\leq m_{2}-1\},|X_{5}|=m_{2},\\ X_{6}=&\{\langle p_{1}^{r_{1}}p_{2}^{m_{2}}p_{3}\rangle:0\leq r_{1}\leq m_{1}-1\},|X_{6}|=m_{1}.\\ \end{split}$$

  Since any resolving set $W$ contains all but at most one vertex of each of the distance similar vertex partitioned sets, take $W$ as follows:

  $$\begin{split}W=&X\backslash\{\langle p_{1}^{m_{1}-1}p_{2}^{m_{2}-1}\rangle\}\bigcup X_{1}\backslash\{\langle p_{1}^{m_{1}}p_{2}^{m_{2}-1}\rangle\}\bigcup X_{2}\backslash\{\langle p_{1}^{m_{1}-1}p_{2}^{m_{2}}\rangle\}\bigcup X_{3}\backslash\{\langle p_{1}^{m_{1}-1}p_{2}^{m_{2}-1}p_{3}\rangle\}\\ &\bigcup X_{5}\backslash\{\langle p_{1}^{m_{1}}p_{2}^{m_{2}-1}p_{3}\rangle\}\bigcup X_{6}\backslash\{\langle p_{1}^{m_{1}-1}p_{2}^{m_{2}}p_{3}\rangle\},|W|=2(m_{1}m_{2}+m_{1}+m_{2})-7=T-7\\ \end{split}$$

  To prove $W$ is a minimum resolving set, it is enough to show that each vertex of $V\backslash W$ has a unique metric representation. The representations of the seven vertices of $V\backslash W$ are given as follows:

  $$\begin{split}r(\langle p_{1}^{m_{1}-1}p_{2}^{m_{2}-1}\rangle|W)=&(\underbrace{1,1,\cdots,1}_{T-7\text{times}}),\\ r(\langle p_{1}^{m_{1}}p_{2}^{m_{2}-1}\rangle|W)=&(\underbrace{1,1,\cdots,1}_{m_{1}m_{2}-2},\underbrace{2,2,\cdots,2}_{m_{2}-1},\underbrace{1,1,\cdots,1}_{m_{1}-1},\underbrace{1,1,\cdots,1}_{m_{1}m_{2}-1},\underbrace{2,2,\cdots,2}_{m_{2}-1},\underbrace{1,1,\cdots,1}_{m_{1}-1}),\\ r(\langle p_{1}^{m_{1}-1}p_{2}^{m_{2}}\rangle|W)=&(\underbrace{1,1,\cdots,1}_{m_{1}m_{2}-2},\underbrace{1,1,\cdots,1}_{m_{2}-1},\underbrace{2,2,\cdots,2}_{m_{1}-1},\underbrace{1,1,\cdots,1}_{m_{1}m_{2}-1},\underbrace{1,1,\cdots,1}_{m_{2}-1},\underbrace{2,2,\cdots,2}_{m_{1}-1}),\\ r(\langle p_{1}^{m_{1}-1}p_{2}^{m_{2}-1}p_{3}\rangle|W)=&(\underbrace{1,1,\cdots,1}_{m_{1}m_{2}-2},\underbrace{1,1,\cdots,1}_{m_{2}-1},\underbrace{1,1,\cdots,1}_{m_{1}-1},\underbrace{2,2,\cdots,2}_{m_{1}m_{2}-1},\underbrace{2,2,\cdots,2}_{m_{2}-1},\underbrace{2,2,\cdots,2}_{m_{1}-1}),\\ \end{split}$$

  $$\begin{split}r(\langle p_{1}^{m_{1}}p_{2}^{m_{2}}\rangle|W)=&(\underbrace{1,1,\cdots,1}_{m_{1}m_{2}-2},\underbrace{2,2,\cdots,2}_{m_{2}-1},\underbrace{2,2,\cdots,2}_{m_{1}-1},\underbrace{1,1,\cdots,1}_{m_{1}m_{2}-1},\underbrace{2,2,\cdots,2}_{m_{2}-1},\underbrace{2,2,\cdots,2}_{m_{1}-1}),\\ r(\langle p_{1}^{m_{1}}p_{2}^{m_{2}-1}p_{3}\rangle|W)=&(\underbrace{1,1,\cdots,1}_{m_{1}m_{2}-2},\underbrace{2,2,\cdots,2}_{m_{2}-1},\underbrace{1,1,\cdots,1}_{m_{1}-1},\underbrace{2,2,\cdots,1}_{m_{1}m_{2}-1},\underbrace{2,2,\cdots,2}_{m_{2}-1},\underbrace{2,2,\cdots,2}_{m_{1}-1}),\\ r(\langle p_{1}^{m_{1}-1}p_{2}^{m_{2}}p_{3}\rangle|W)=&(\underbrace{1,1,\cdots,1}_{m_{1}m_{2}-2},\underbrace{1,1,\cdots,1}_{m_{2}-1},\underbrace{2,2,\cdots,2}_{m_{1}-1},\underbrace{2,2,\cdots,1}_{m_{1}m_{2}-1},\underbrace{2,2,\cdots,2}_{m_{2}-1},\underbrace{2,2,\cdots,2}_{m_{1}-1}).\\ \end{split}$$

  It can be seen that any two distinct vertices of $V(\mathcal{E}_{\mathbb{Z}_{n}})\backslash W$ have different metric representations with respect to $W$. Thus $W$ is a resolving set having $T-7=2(m_{1}m_{2}+m_{1}+m_{2})-7$ vertices. Also, any resolving set must contain more than $T-7$ elements, we conclude that $dim(\mathcal{E}_{\mathbb{Z}_{n}})=T-7$.

• Let $n=p_{1}^{m_{1}}p_{2}p_{3}$, $m_{1}>1$. Then the distance similar partition of vertices of $\mathcal{E}_{\mathbb{Z}_{n}}$ is given by:

  $$\begin{split}X=&\{\langle p_{1}^{r_{1}}\rangle:1\leq r_{1}\leq m_{1}-1\},|X|=m_{1}-1,\\ X_{1}=&\{\langle p_{1}^{m_{1}}\rangle\},\\ X_{2}=&\{\langle p_{1}^{r_{1}}p_{2}\rangle:0\leq r_{1}\leq m_{1}-1\},|X_{2}|=m_{1},\\ X_{3}=&\{\langle p_{1}^{r_{1}}p_{3}\rangle:0\leq r_{1}\leq m_{1}-1\},|X_{3}|=m_{1},\\ X_{4}=&\{\langle p_{1}^{m_{1}}p_{2}\rangle\},\ X_{5}=\{\langle p_{1}^{m_{1}}p_{3}\rangle\},\\ X_{6}=&\{\langle p_{1}^{r_{1}}p_{2}p_{3}\rangle:0\leq r_{1}\leq m_{1}-1\},|X_{6}|=m_{1}.\\ \end{split}$$

  Now, take $W$as an ordered set consisting of first $m_{1}-2$ vertices of $X$, first $m_{1}-1$ vertices of $X_{2}$, and so on. Then, for the vertices $\langle p_{1}^{m_{1}-1}\rangle$ and $\langle p_{1}^{m_{1}}\rangle$ of $V(\mathcal{E}_{\mathbb{Z}_{n}})\backslash W$,

  $$r(\langle p_{1}^{m_{1}-1}\rangle|W)=(\underbrace{1,1,\cdots,1}_{m_{1}-2},\underbrace{1,1,\cdots,1}_{m_{1}-1},\underbrace{1,1,\cdots,1}_{m_{1}-1},\underbrace{1,1,\cdots,1}_{m_{1}-1})=r(\langle p_{1}^{m_{1}}\rangle|W)$$

  Hence, $W$ cannot be a resolving set of $\mathcal{E}_{\mathbb{Z}_{n}}$. Without loss of generality, take $W^{\prime}=W\cup\{\langle p_{1}^{m_{1}}\rangle\}$. That is,

  $$W^{\prime}=X\backslash\{\langle p_{1}^{m_{1}-1}\rangle\}\bigcup X_{1}\bigcup X_{2}\backslash\{\langle p_{1}^{m_{1}-1}p_{2}\rangle\}\bigcup X_{3}\backslash\{\langle p_{1}^{m_{1}-1}p_{3}\rangle\}\bigcup X_{6}\backslash\{\langle p_{1}^{m_{1}-1}p_{2}p_{3}\rangle\}.$$

  The representations of vertices of $V(\mathcal{E}_{\mathbb{Z}_{n}})\backslash W^{\prime}$ are given by,

  $$\begin{split}r(\langle p_{1}^{m_{1}-1}\rangle|W^{\prime})=&(\underbrace{1,1,\cdots,1}_{m_{1}-2},1,\underbrace{1,1,\cdots,1}_{m_{1}-1},\underbrace{1,1,\cdots,1}_{m_{1}-1},\underbrace{1,1,\cdots,1}_{m_{1}-1}),\\ r(\langle p_{1}^{m_{1}-1}p_{2}\rangle|W^{\prime})=&(\underbrace{1,1,\cdots,1}_{m_{1}-2},1,\underbrace{2,2,\cdots,2}_{m_{1}-1},\underbrace{1,1,\cdots,1}_{m_{1}-1},\underbrace{2,2,\cdots,2}_{m_{1}-1}),\\ r(\langle p_{1}^{m_{1}-1}p_{3}\rangle|W^{\prime})=&(\underbrace{1,1,\cdots,1}_{m_{1}-2},1,\underbrace{1,1,\cdots,1}_{m_{1}-1},\underbrace{2,2,\cdots,2}_{m_{1}-1},\underbrace{2,2,\cdots,2}_{m_{1}-1}),\\ r(\langle p_{1}^{m_{1}}p_{2}\rangle|W^{\prime})=&(\underbrace{1,1,\cdots,1}_{m_{1}-2},2,\underbrace{2,2,\cdots,2}_{m_{1}-1},\underbrace{1,1,\cdots,1}_{m_{1}-1},\underbrace{2,2,\cdots,2}_{m_{1}-1}),\\ r(\langle p_{1}^{m_{1}}p_{3}\rangle|W^{\prime})=&(\underbrace{1,1,\cdots,1}_{m_{1}-2},2,\underbrace{1,1,\cdots,1}_{m_{1}-1},\underbrace{2,2,\cdots,2}_{m_{1}-1},\underbrace{2,2,\cdots,2}_{m_{1}-1}),\\ r(\langle p_{1}^{m_{1}-1}p_{2}p_{3}\rangle|W^{\prime})=&(\underbrace{1,1,\cdots,1}_{m_{1}-2},1,\underbrace{2,2,\cdots,2}_{m_{1}-1},\underbrace{2,2,\cdots,2}_{m_{1}-1},\underbrace{2,2,\cdots,2}_{m_{1}-1}).\\ \end{split}$$

This unique representation of vertices of $V(\mathcal{E}_{\mathbb{Z}_{n}})\backslash W^{\prime}$ ensures that $W^{\prime}$ is a minimum resolving set of $\mathcal{E}_{\mathbb{Z}_{n}}$. Hence, the metric dimension of $\mathcal{E}_{\mathbb{Z}_{n}}$ is $4(m_{1}-1)=T-6$.

4 Zagreb Indices of $\mathcal{E}_{\mathbb{Z}_{n}}$

In this section, we calculate the $1$st and $2$nd Zagreb indices of $\mathcal{E}_{\mathbb{Z}_{n}}$.

Proposition 4.1.

Let $n=p^{m}$, $m>2$. Then

1. The first Zagreb index of $\mathcal{E}_{\mathbb{Z}_{n}}$ is, $M_{1}(\mathcal{E}_{\mathbb{Z}_{n}})=(m-1)(T-1)^{2}$.

2. The second Zagreb index of $\mathcal{E}_{\mathbb{Z}_{n}}$ is, $M_{2}(\mathcal{E}_{\mathbb{Z}_{n}})=\binom{m-1}{2}(T-1)^{2}$.

Lemma 4.2.

[jamsheena2023adjacency] Let $n=p_{1}p_{2}\cdots p_{k}$, where $p_{1},p_{2},\cdots,p_{k}$ are distinct primes. Then any two vertices $\langle x\rangle$ and $\langle y\rangle$ of the essential ideal graph of $\mathbb{Z}_{n}$ are adjacent if and only if $gcd(x,y)=1$, provided $x$ is the product of $i$ distinct primes and $y$ is the product of $j$ distinct primes for $1\leq i,j\leq k-1$.

Theorem 4.3.

Let $n=p_{1}p_{2}\cdots p_{k}$. Then, the fist and second Zagreb indices of $\mathcal{E}_{\mathbb{Z}_{n}}$ are,

1. $M_{1}(\mathcal{E}_{\mathbb{Z}_{n}})=\displaystyle\sum_{i=1}^{k-1}\binom{k}{i}(2^{k-i}-1)^{2}.$

2. $M_{2}(\mathcal{E}_{\mathbb{Z}_{n}})=\displaystyle\sum_{t=1}^{\lfloor\frac{k}{2}\rfloor}\binom{k}{t}(2^{k-t}-1)[\frac{1}{2}\binom{k-t}{t}(2^{k-t}-1)+\displaystyle\sum_{s=t}^{k-t}\binom{k-t}{s}(2^{k-s}-1)].$

Proof.

For $n=p_{1}p_{2}\cdots p_{k}$, the vertex set of $\mathcal{E}_{\mathbb{Z}_{n}}$ can be partitioned as follows:

$$\begin{split}&V_{1}=\{\langle p_{i}\rangle:1\leq i\leq k\}\\ &V_{2}=\{\langle p_{i}p_{j}\rangle:1\leq i\leq k-1\textit{and}\ i+1\leq j\leq k\}\\ &V_{3}=\{\langle p_{i}p_{j}p_{l}\rangle:1\leq i\leq k-2,\ i+1\leq j\leq k-1\textit{and}\ j+1\leq l\leq k\}\\ &\vdots\\ &V_{k-1}=\{\langle p_{1}p_{2}p_{3}\cdots p_{k-1}\rangle,\langle p_{1}p_{2}p_{3}\cdots p_{k-2}p_{k}\rangle,\cdots,\langle p_{2}p_{3}\cdots p_{k-1}p_{k}\rangle\}\end{split}$$

Clearly $|V_{1}|=\binom{k}{1},\ |V_{2}|=\binom{k}{2},\cdots,\ \text{and}\ |V_{k-1}|=\binom{k}{k-1}$. Also, by Lemma 4.2, two vertices $\langle x\rangle$ and $\langle y\rangle$ of $\mathcal{E}_{\mathbb{Z}_{n}}$ are adjacent if and only if the generators $x$ and $y$ have no prime factors in common.

$$\text{For a vertex}\ v\in V(\mathcal{E}_{\mathbb{Z}_{n}}),\ deg(v)=\begin{cases}2^{k-1}-1,&\textit{if}\ v\in V_{1},\\ 2^{k-2}-1,&\textit{if}\ v\in V_{2},\\ \vdots&\vdots\\ 3,&\textit{if}\ v\in V_{k-2},\\ 1,&\textit{if}\ v\in V_{k-1}.\\ \end{cases}$$

Also, for a fixed $i$,

$$\sum_{v\in V_{i}}deg(v)^{2}=\binom{k}{i}(2^{k-i}-1)^{2}.$$

Hence,

1. $$\begin{split}M_{1}(\mathcal{E}_{\mathbb{Z}_{n}})=&\sum_{v\in V_{1}}deg(v)^{2}+\sum_{v\in V_{2}}deg(v)^{2}+\cdots+\sum_{v\in V_{k-1}}deg(v)^{2}\\ =&\sum_{i=1}^{k-1}\sum_{v\in V_{i}}deg(v)^{2}\\ =&\sum_{i=1}^{k-1}\binom{k}{i}(2^{k-i}-1)^{2}.\\ \end{split}$$

2. For a fixed $t$, $1\leq t\leq\lfloor\frac{k}{2}\rfloor$, Lemma 4.2 assures that any vertex $u\in V_{t}$ are adjacent only to $\binom{k-t}{s}$ vertices of $V_{s}$ for $t\leq s\leq k-t$. Then, for $u,v\in V_{t},\ 1\leq t\leq\lfloor\frac{k}{2}\rfloor$,

   $$\sum_{u\sim v}deg(u)deg(v)=\frac{1}{2}\binom{k}{t}\binom{k-t}{t}(2^{k-t}-1)^{2}.$$ (2)

   Now, consider $u\in V_{t}$ for a fixed $t$ such that $1\leq t\leq\lfloor\frac{k}{2}\rfloor$. Then we have,

   $$\begin{split}\displaystyle\sum_{u\sim v\atop v\in V_{s}}deg(u)deg(v)=&\displaystyle\sum_{u\sim v\atop v\in V_{t}}deg(u)deg(v)+\displaystyle\sum_{u\sim v\atop v\in V_{t+1}}deg(u)deg(v)+\cdots+\displaystyle\sum_{u\sim v\atop v\in V_{k-t}}deg(u)deg(v)\\ =&\frac{1}{2}\binom{k}{t}\binom{k-t}{t}(2^{k-t}-1)^{2}+\binom{k}{t}\binom{k-t}{t+1}(2^{k-t}-1)(2^{k-(t+1)}-1)+\cdots\\ +&\binom{k}{t}\binom{k-t}{k-t}(2^{k-t}-1)(2^{t}-1)\\ =&\frac{1}{2}\binom{k}{t}\binom{k-t}{t}(2^{k-t}-1)^{2}+\binom{k}{t}(2^{k-t}-1)\displaystyle\sum_{s=t+1}^{k-t}(2^{k-s}-1).\end{split}$$ (3)

   Now, by Equation (3),

   $$\begin{split}M_{2}(\mathcal{E}_{\mathbb{Z}_{n}})=&\displaystyle\sum_{u\sim v\atop u,v\in V(\mathcal{E}_{\mathbb{Z}_{n}})}deg(u)deg(v)\\ =&\displaystyle\sum_{u\sim v\atop u\in V_{1},\ v\in V_{s}}deg(u)deg(v)+\displaystyle\sum_{u\sim v\atop u\in V_{2},\ v\in V_{s}}deg(u)deg(v)+\cdots+\displaystyle\sum_{u\sim v\atop u\in V_{\lfloor\frac{k}{2}\rfloor},\ v\in V_{s}}deg(u)deg(v)\\ =&\frac{1}{2}\displaystyle\sum_{t=1}^{\lfloor\frac{k}{2}\rfloor}\binom{k}{t}\binom{k-t}{t}(2^{k-t}-1)^{2}+\displaystyle\sum_{t=1}^{\lfloor\frac{k}{2}\rfloor}\binom{k}{t}(2^{k-t}-1)\displaystyle\sum_{s=t}^{k-t}\binom{k-t}{s}(2^{k-s}-1)\\ =&\displaystyle\sum_{t=1}^{\lfloor\frac{k}{2}\rfloor}\binom{k}{t}(2^{k-t}-1)[\frac{1}{2}\binom{k-t}{t}(2^{k-t}-1)+\displaystyle\sum_{s=t}^{k-t}\binom{k-t}{s}(2^{k-s}-1)].\end{split}$$

∎

Example 4.4.

• Let $n=p_{1}p_{2}p_{3}$. Then by Theorem 4.3, the first and second Zagreb indices are given by,

  $$\begin{split}M_{1}(\mathcal{E}_{\mathbb{Z}_{n}})=&\displaystyle\sum_{i=1}^{2}\binom{3}{i}(2^{3-i}-1)^{2}\\ =&\binom{3}{1}3^{2}+\binom{3}{2}=30,\end{split}$$

  and

  $$\begin{split}M_{2}(\mathcal{E}_{\mathbb{Z}_{n}})=&\displaystyle\sum_{t=1}^{\lfloor\frac{3}{2}\rfloor}\binom{3}{t}(2^{3-t}-1)\displaystyle\sum_{s=t}^{3-t}\binom{3-t}{s}(2^{3-s}-1)\\ =&\binom{3}{1}3[3\binom{2}{1}+1]=63.\end{split}$$

• Let $n=p_{1}p_{2}p_{3}p_{4}$. Then

  $$\begin{split}M_{1}(\mathcal{E}_{\mathbb{Z}_{n}})=&\displaystyle\sum_{i=1}^{3}\binom{4}{i}(2^{4-i}-1)^{2}\\ =&\binom{4}{1}7^{2}+\binom{4}{2}3^{2}+\binom{4}{3}=254,\end{split}$$

  and

  $$\begin{split}M_{2}(\mathcal{E}_{\mathbb{Z}_{n}})=&\displaystyle\sum_{t=1}^{2}\binom{4}{t}(2^{4-t}-1)\displaystyle\sum_{s=t}^{4-t}\binom{4-t}{s}(2^{4-s}-1)\\ =&\binom{4}{1}7[\binom{3}{1}7+3\binom{3}{2}+1]+\binom{4}{2}9=922.\end{split}$$

This can be easily verified from Figure 1.

Figure 1: $\mathcal{E}_{\mathbb{Z}_{n}}$ for $n=\prod_{i=1}^{3}p_{i}$ and $n=\prod_{i=1}^{4}p_{i}$

Theorem 4.5.

Let $n=p_{1}^{m_{1}}p_{2}^{m_{2}}\cdots p_{k}^{m_{k}}$, where $p_{1}<p_{2}<\cdots<p_{k}$ are primes, $k\geq 2$ and $m_{i}>1$ for at least one $i$. Then,

1. $M_{1}(\mathcal{E}_{\mathbb{Z}_{n}})=m(T-1)^{2}+\displaystyle\sum_{\hat{I}}|X_{\hat{I}}|(m+\displaystyle\sum_{\hat{J}:\ \Xi_{\hat{I}}\cap\Xi_{\hat{J}}=\phi}|X_{\hat{J}}|)^{2}$.

2. $$\begin{split}M_{2}(\mathcal{E}_{\mathbb{Z}_{n}})=&\binom{m}{2}(T-1)^{2}+m(T-1)\displaystyle\sum_{\hat{I}}|X_{\hat{I}}|(m+\displaystyle\sum_{\hat{J}:\ \Xi_{\hat{I}}\cap\Xi_{\hat{J}}=\phi}|X_{\hat{J}}|)\\ +&\frac{\delta_{i}^{j}}{2}|X_{\hat{I}}||X_{\hat{J}}|\displaystyle\sum_{u,v\in\mathscr{U}}deg(u)deg(v),\text{where}\quad\delta_{i}^{j}=\begin{cases}1,&\text{if}\ \Xi_{\hat{I}}\cap\Xi_{\hat{J}}=\phi,\\ 0,&\text{otherwise},\end{cases}\end{split}$$

   and $deg(u)=|X|+\displaystyle\sum_{\hat{J}:\ \Xi_{\hat{I}}\cap\Xi_{\hat{J}}=\phi}|X_{\hat{J}}|$ for any vertex $u\in X_{\hat{I}}$ of $\mathscr{U}$.

Proof.

By Theorem 2.11, $\mathcal{E}_{\mathbb{Z}_{n}}\cong K_{m}\vee\mathscr{G}[\Gamma_{1},\Gamma_{2},\cdots,\Gamma_{2^{k}-2}]$, where $K_{m}$ is the subgraph induced by the set $X$ of essential ideals of $\mathbb{Z}_{n}$, and $\Gamma_{i}=\mathcal{E}_{\mathbb{Z}_{n}}(X_{\hat{I}})$ for each of the $2^{k}-2$ equivalent class $X_{\hat{I}}$ of the set $\mathscr{U}$. Thus, $V(\mathcal{E}_{\mathbb{Z}_{n}})=X\cup_{\hat{I}}X_{\hat{I}}$, where the union is taken over all the equivalent classes. Then, for any vertex $v\in X$, $deg(v)=T-1$ and by Lemma 2.9,

$$\text{for any vertex}\ v\in X_{\hat{I}},\hskip 42.67912ptdeg(v)=|X|+\displaystyle\sum_{\hat{J}:\ \Xi_{\hat{I}}\cap\Xi_{\hat{J}}=\phi}|X_{\hat{J}}|,$$ (4)

1. $$\begin{split}M_{1}(\mathcal{E}_{\mathbb{Z}_{n}})=&\displaystyle\sum_{v\in X}deg(v)^{2}+\displaystyle\sum_{v\in\cup_{\hat{I}}X_{\hat{I}}}deg(v)^{2},\\ =&m(T-1)^{2}+\displaystyle\sum_{\hat{I}}|X_{\hat{I}}|(m+\displaystyle\sum_{\hat{J}:\ \Xi_{\hat{I}}\cap\Xi_{\hat{J}}=\phi}|X_{\hat{J}}|)^{2},\end{split}$$

   where the summation runs over all the equivalent classes.

2. First consider $v\in X$. Since, $|X|=m$ and there are $\binom{m}{2}$ pairs of elements in $X$ , we have

   $$\sum_{u,v\in X}deg(u)deg(v)=\binom{m}{2}(T-1)^{2}.$$ (5)

   Now, each vertex $u\in X$ is adjacent to every vertex $v\in X_{\hat{I}}$, for each of the equivalent class $X_{\hat{I}}$. Then,

   $$deg(u)deg(v)=(T-1)(m+\displaystyle\sum_{\hat{J}:\ \Xi_{\hat{I}}\cap\Xi_{\hat{J}}=\phi}|X_{\hat{J}}|),$$

   and hence,

   $$\begin{split}\displaystyle\sum_{u\in X\atop v\in\mathscr{U}}deg(u)deg(v)=&\displaystyle\sum_{\hat{I}}m(T-1)|X_{\hat{I}}|(m+\displaystyle\sum_{\hat{J}:\ \Xi_{\hat{I}}\cap\Xi_{\hat{J}}=\phi}|X_{\hat{J}}|)\\ =&m(T-1)\displaystyle\sum_{\hat{I}}|X_{\hat{I}}|(m+\displaystyle\sum_{\hat{J}:\ \Xi_{\hat{I}}\cap\Xi_{\hat{J}}=\phi}|X_{\hat{J}}|).\\ \end{split}$$ (6)

   Now, if we take $u,v$ from the vertex subset $\mathscr{U}=\bigcup\limits_{\hat{I}}X_{\hat{I}}$, then $u\in X_{\hat{I}}$ and $v\in X_{\hat{J}}$ for some equivalent classes $X_{\hat{I}}$ and $X_{\hat{J}}$. By Lemma 2.9, $u$ and $v$ are adjacent if and only if $\Xi_{\hat{I}}\cap\Xi_{\hat{J}}=\phi$ and hence

   $$\sum_{u\sim v\atop u,v\in\mathscr{U}}deg(u)deg(v)=\delta_{i}^{j}|X_{\hat{I}}||X_{\hat{J}}|\displaystyle\sum_{u,v\in\mathscr{U}}deg(u)deg(v),\text{where}\quad\delta_{i}^{j}=\begin{cases}1,&\text{if}\ \Xi_{\hat{I}}\cap\Xi_{\hat{J}}=\phi,\\ 0,&\text{otherwise}.\end{cases}$$ (7)

   Using Equations (5), (6) and (7), we have

   $$\begin{split}M_{2}(\mathcal{E}_{\mathbb{Z}_{n}})=&\displaystyle\sum_{u,v\in X}deg(u)deg(v)+\displaystyle\sum_{u\in X\atop v\in\mathscr{U}}deg(u)deg(v)+\displaystyle\sum_{u\sim v\atop u,v\in\mathscr{U}}deg(u)deg(v)\\ =&\binom{m}{2}(T-1)^{2}+m(T-1)\displaystyle\sum_{\hat{I}}|X_{\hat{I}}|(m+\displaystyle\sum_{\hat{J}:\Xi_{\hat{I}}\cap\Xi_{\hat{J}}=\phi}|X_{\hat{J}}|)\\ +&\delta_{i}^{j}|X_{\hat{I}}||X_{\hat{J}}|\displaystyle\sum_{u,v\in\mathscr{U}}deg(u)deg(v),\text{where}\quad\delta_{i}^{j}=\begin{cases}1,&\text{if}\ \Xi_{\hat{I}}\cap\Xi_{\hat{J}}=\phi,\\ 0,&\text{otherwise},\end{cases}\end{split}$$

   $deg(u)$ and $deg(v)$ are given by Equation (4).

∎

Corollary 4.6.

Let $n=p_{1}^{m_{1}}p_{2}^{m_{2}}$, where $p_{1}<p_{2}$ are primes and $m_{i}>1$ for at least one $i$. Then

1. $M_{1}(\mathcal{E}_{\mathbb{Z}_{n}})=m(T-1)^{2}+m_{1}(m+m_{2})^{2}+m_{2}(m+m_{1})^{2}$

2. $M_{2}(\mathcal{E}_{\mathbb{Z}_{n}})=\binom{m}{2}(T-1)^{2}+m(T-1)[m(m_{1}+m_{2})+2m_{1}m_{2}]+m_{1}m_{2}(m+m_{1})(m+m_{2}).$

Proof.

By Theorem 2.11, $\mathcal{E}_{\mathbb{Z}_{n}}\cong K_{m}\vee K_{2}[\overline{K_{m_{2}}},\ \overline{K_{m_{1}}}]$, where $K_{m}$ is the subgraph induced by the set $X$ of essential ideals of $\mathbb{Z}_{n}$. Here,

$$\begin{split}X_{1}=&X_{\langle p_{1}^{m_{1}}\rangle}=\{\langle p_{1}^{m_{1}}p_{2}^{r_{2}}\rangle:0\leq r_{2}<m_{2}\};|X_{1}|=m_{2},\\ X_{2}=&X_{\langle p_{2}^{m_{2}}\rangle}=\{\langle p_{1}^{r_{1}}p_{2}^{m_{2}}\rangle:0\leq r_{1}<m_{1}\};|X_{2}|=m_{1}.\end{split}$$

Then, by Theorem 4.5, $M_{1}(\mathcal{E}_{\mathbb{Z}_{n}})=m(T-1)^{2}+m_{1}(m+m_{2})^{2}+m_{2}(m+m_{1})^{2}$, and

$$\begin{split}M_{2}(\mathcal{E}_{\mathbb{Z}_{n}})=&\binom{m}{2}(T-1)^{2}+m(T-1)[m_{2}(m+m_{1})+m_{1}(m+m_{2})]+m_{1}m_{2}(m+m_{1})(m+m_{2})\\ =&\binom{m}{2}(T-1)^{2}+m(T-1)[m(m_{1}+m_{2})+2m_{1}m_{2}]+m_{1}m_{2}(m+m_{1})(m+m_{2}).\end{split}$$

∎

Example 4.7.

Let $n=p_{1}^{2}p_{2}^{3}p_{3}^{2}$. Then, $T=|V(\mathcal{E}_{\mathbb{Z}_{n}})|=34$, and $\mathcal{E}_{\mathbb{Z}_{n}}\cong K_{m}\vee\mathscr{G}[\overline{K_{6}},\overline{K_{4}},\overline{K_{6}},\overline{K_{2}},\overline{K_{3}},\overline{K_{2}}]$, where $m=11$. The partitioned sets of nonessential ideals of $\mathbb{Z}_{n}$ are

$$\begin{split}X_{1}=&X_{\langle p_{1}^{2}\rangle}=\{\langle p_{1}^{2}p_{2}^{r_{2}}p_{3}^{r_{3}}\rangle:0\leq r_{i}\leq<m_{i}\ \text{for}\ i=2,3\};\ |X_{1}|=6,\\ X_{2}=&X_{\langle p_{2}^{3}\rangle}=\{\langle p_{1}^{r_{1}}p_{2}^{3}p_{3}^{r_{3}}\rangle:0\leq r_{i}<m_{i}\ \text{for}\ i=1,3\};\ |X_{2}|=4,\\ X_{3}=&X_{\langle p_{3}^{2}\rangle}=\{\langle p_{1}^{r_{1}}p_{2}^{r_{2}}p_{3}^{3}\rangle:0\leq r_{i}<m_{i}\ \text{for}\ i=1,2\};\ |X_{3}|=6,\\ X_{4}=&X_{\langle p_{1}^{2}p_{2}^{3}\rangle}=\{\langle p_{1}^{2}p_{2}^{3}p_{3}^{r_{3}}\rangle:0\leq r_{3}<m_{3}\};\ |X_{4}|=2,\\ X_{5}=&X_{\langle p_{1}^{2}p_{3}^{2}\rangle}=\{\langle p_{1}^{2}p_{2}^{r_{2}}p_{3}^{2}\rangle:0\leq r_{2}<m_{2}\};\ |X_{5}|=3,\\ X_{6}=&X_{\langle p_{2}^{3}p_{3}^{2}\rangle}=\{\langle p_{1}^{r_{1}}p_{2}^{3}p_{3}^{2}\rangle:0\leq r_{1}<m_{1}\};\ |X_{6}|=2.\end{split}$$

$deg(u)=\begin{cases}11+|X_{2}|+|X_{3}|+|X_{6}|=23,&\ \text{for}\ u\in X_{1},\\ 11+|X_{1}|+|X_{3}|+|X_{5}|=26,&\ \text{for}\ u\in X_{2},\\ 11+|X_{1}|+|X_{2}|+|X_{4}|=23,&\ \text{for}\ u\in X_{3},\\ 11+|X_{3}|=17,&\ \text{for}\ u\in X_{4},\\ 11+|X_{2}|=15,&\ \text{for}\ u\in X_{5},\\ 11+|X_{1}|=17,&\ \text{for}\ u\in X_{6}.\\ \end{cases}$
Then, by Theorem,

$$\begin{split}M_{1}(\mathcal{E}_{\mathbb{Z}_{n}})=&11\times 33^{2}+6\times 23^{2}+4\times 26^{2}+6\times 23^{2}+2\times 17^{2}+3\times 15^{2}+2\times 17^{2}\\ =&22,862.\end{split}$$

$$\begin{split}M_{2}(\mathcal{E}_{\mathbb{Z}_{n}})=&\binom{11}{2}33^{2}+11\times 33[6\times 23+4\times 26+6\times 23+2\times 17+3\times 15+2\times 17]+\frac{1}{2}[6\times 4\times 23\times 26\\ +&6\times 6\times 23\times 23+6\times 2\times 23\times 17+4\times 6\times 26\times 23+4\times 6\times 26\times 23+4\times 3\times 26\times 15\\ +&6\times 6\times 23\times 23+6\times 4\times 23\times 26+6\times 2\times 23\times 17+2\times 6\times 17\times 23+3\times 4\times 15\times 26\\ +&2\times 6\times 17\times 23]\\ .=&3,00,666.\end{split}$$

5 Conclusion

In this article, we have proved that the metric dimension of the essential ideal graph $\mathcal{E}_{R}$ of a commutative ring $R$ is finite whenever each vertex of $\mathcal{E}_{R}$ is of finite degree. Also, for the ring $\mathbb{Z}_{n}$, it is identified that the graphs $\mathcal{E}_{\mathbb{Z}_{n}}$ and $\mathbb{AIG}(\mathbb{Z}_{n})$ coincide (up to isomorphism) when $n$ is a product of distinct primes. Furthermore, we have calculated the metric dimension of $\mathcal{E}_{\mathbb{Z}_{n}}$. Additionally, an alternative method has been provided to establish an upper limit for $dim(\mathcal{E}_{\mathbb{Z}_{n}})$ when $n=p_{1}p_{2}\cdots p_{k};\ k\geq 6$. Finally, the first and second Zagreb indices of $\mathcal{E}_{\mathbb{Z}{n}}$ are computed for arbitrary values of $n$.

6 Declarations

Conflict of interest On behalf of all authors, the corresponding author states that there is no conflict of interest.