Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

On the Wiener Polarity Index of Lattice Networks

  • Lin Chen,

    Affiliation Center for Combinatorics and LPMC-TJKLC, Nankai University, Tianjin, China

  • Tao Li ,

    litao@nankai.edu.cn

    Affiliation College of Computer and Control Engineering, Nankai University, Tianjin 300071, China

  • Jinfeng Liu,

    Affiliation Center for Combinatorics and LPMC-TJKLC, Nankai University, Tianjin, China

  • Yongtang Shi,

    Affiliation Center for Combinatorics and LPMC-TJKLC, Nankai University, Tianjin, China

  • Hua Wang

    Affiliation Department of Mathematical Sciences Georgia Southern University, Statesboro, GA 30460-8093, United States of America

On the Wiener Polarity Index of Lattice Networks

  • Lin Chen, 
  • Tao Li, 
  • Jinfeng Liu, 
  • Yongtang Shi, 
  • Hua Wang
PLOS
x

Abstract

Network structures are everywhere, including but not limited to applications in biological, physical and social sciences, information technology, and optimization. Network robustness is of crucial importance in all such applications. Research on this topic relies on finding a suitable measure and use this measure to quantify network robustness. A number of distance-based graph invariants, also known as topological indices, have recently been incorporated as descriptors of complex networks. Among them the Wiener type indices are the most well known and commonly used such descriptors. As one of the fundamental variants of the original Wiener index, the Wiener polarity index has been introduced for a long time and known to be related to the cluster coefficient of networks. In this paper, we consider the value of the Wiener polarity index of lattice networks, a common network structure known for its simplicity and symmetric structure. We first present a simple general formula for computing the Wiener polarity index of any graph. Using this formula, together with the symmetric and recursive topology of lattice networks, we provide explicit formulas of the Wiener polarity index of the square lattices, the hexagonal lattices, the triangular lattices, and the 33 ⋅ 42 lattices. We also comment on potential future research topics.

Introduction

Robustness is the ability of a network to maintain performance when encountering attacks or enduring partial failure. In order to decide whether a given network is robust, a way to quantitatively measure network robustness is needed. Once such a measure has been established, it can serve as a standard for comparing networks or a guidance for improving existing networks and designing new networks. Intuitively robustness is all about back-up possibilities, or alternative paths, but it is rather challenging to capture all these concepts in a simple mathematical formula. During the recent years a lot of robustness measures have been proposed by scientists from different backgrounds, including but not limited to Biology, Chemistry, Computer Science, Engineering, Physics, and Mathematics [17]. Generally a network is considered as a graph consisting of a set of vertices connected by edges, and the study of network robustness relies on the analysis of such underlying graphs.

One way to measure a network structure is through the so called structure descriptors, or topological indices [8]. In theoretical biology and chemistry, for instance, molecular structure descriptors are numerical parameters mathematically derived from the graph structure. They have been found to be useful for modeling physico-chemical, toxicologic, pharmacologic, biological and other properties of molecular compounds. These descriptors are mainly divided into three types: degree-based indices, distance-based indices and spectrum-based indices. Degree-based indices contain the (general) Randić index [911], the (general) zeroth order Randić index [12, 13], the Zagreb index [14, 15], the connective eccentricity index [16] and so on [17]. Distance-based indices [18] include the Balaban index [19, 20], the Wiener index and Wiener polarity index [21], the Kirchhoff index [22, 23] and so forth. Eigenvalues of graphs [24], various of graph energies [2536], the HOMO-LUMO index [37, 38], and the Estrada index [3941] belong to spectrum-based indices. There are also some topological indices defined based on both degrees and distances, such as the degree distance [42] and graph entropies [43, 44]. The study of mathematical properties of such graph indices and the evaluation of them in various graph structures have been of tremendous interest to researchers.

A lattice graph, or simply a lattice, is a graph possessing a drawing whose embedding in a Euclidean space forms a regular tiling. Because of the symmetric nature of its topology, lattice graphs appear to be among the most common network structures. For exactly the same reason, the computation of various physical and chemical indices of various lattice graphs has attracted the attention of many scientists as well as mathematicians. See, for instance, [4551]. We will focus on the square lattices, the hexagonal lattices, the triangular lattices, and the 33 ⋅ 42 lattices, each corresponding to a grid with specific geometric shapes.

In this paper all graphs under consideration are finite, connected, undirected and simple. For standard notations and terminologies we follow [52]. Let G be a graph with vertex set V(G) and edge set E(G). The distance dG(u, v) (or simply d(u, v) when there is no confusion) between two vertices u and v of G is the length of the shortest path that connects u and v. One of the most well-known and well-studied distance-based graph indices is the Wiener number W(G), also termed as Wiener index in chemical or mathematical chemistry literatures. It is defined as the sum of distances over all unordered vertex pairs in G [21]. I.e., As a representative of successful structure-descriptors, the Wiener index has received much attention. For further details we refer the readers to some recent papers [5356] and the comprehensive survey of Dobrynin, Entringer and Gutman [57].

Another important molecular descriptor was also introduced by Wiener [21], called the Wiener polarity index. Denoted by Wp(G), it is defined as the number of unordered pairs of vertices that are at distance 3 in G. That is, (1) In organic compounds, say paraffin, the Wiener polarity index is the number of pairs of carbon atoms which are separated by three carbon-carbon bonds. Based on the Wiener index and the Wiener polarity index, the formula was used to calculate the boiling points tB of the paraffins, where a, b and c are constants for a given isomeric group.

In an acyclic structure, the Wiener polarity index can be expressed in terms of vertex degrees (see, for instance, Lemma 2). This unique characteristic of the Wiener polarity index makes it interesting for studies from both distance-based and degree-based points of view. However, compared with the Wiener index, surprisingly little attention has been paid to the Wiener polarity index until very recently. Nevertheless, the study of the Wiener polarity index has indeed caught the attention of many researchers. By using the Wiener polarity index, Lukovits and Linert demonstrated quantitative structure-property relationships in a series of acyclic and cycle-containing hydrocarbons in [58]. Hosoya in [59] found a physical-chemical interpretation of Wp(G). Du et al. [60] described a linear time algorithm for computing the Wiener polarity index of trees and characterized the trees maximizing the index among all the trees of the given order. Later, Deng, Xiao and Tang characterized the extremal trees with respect to this index among all trees of order n and diameter k [61]. While for cycle-containing graphs, the maximum Wiener polarity index of unicyclic graphs and the corresponding extremal graphs were determined in [62]. In [63] Ma et al. determined the sharp upper bound of the Wiener polarity index among all bicyclic networks based on some graph transformations. Moreover, the extremal values of catacondensed hexagonal systems, hexagonal cacti and polyphenylene chains with respect to the Wiener polarity index were computed in [64]. It was proved that the Wiener polarity index of fullerenes with n carbon atoms is (9n − 60)/2 in the same paper. Recently, Hua and Das [65] established an upper bound on the Wiener polarity index in terms of Hosoya index and characterized the corresponding extremal graphs. They also obtaind Nordhaus-Gaddum-type results for Wp(G). Other recent work on Wiener polarity index can be found in [6670].

In this paper we study the Wiener polarity index of several common classes of lattices. In the section of Preliminaries and the square lattices, we first introduce some notations and previously established fundamental results on the Wiener polarity index of graphs. Through the concept of 3rd neighborhoods, we provide a simple but extremely useful formula for calculating the Wiener polarity index. We also present the computation of the Wiener polarity index of square lattices as an example of applications of such results. In the sections that follow, we discuss in details the computation of the Wiener polarity index of the hexagonal lattices, the triangular lattices, and the 33 ⋅ 42 lattices. We summarize our findings and propose some future directions of research in the section of Concluding remarks.

Preliminaries and the square lattices

In this section, we first recall some notations. Then a general formula is presented for the Wiener polarity index of any graph. In mathematics, a Cartesian product is a mathematical operation which returns a product set (or simply product) from multiple sets. Given sets A and B, the Cartesian product of A and B, generally denoted by AB, is the set of all ordered pairs (a, b), where aA and bB. That is, Given graphs G and H with vertex sets U and V, the Cartesian product GH of graphs G and H is a graph such that the vertex set of GH is UV, and any two vertices (u, u′) and (v, v′) are adjacent in GH if and only if either u = v and u′ is adjacent to v′ in H, or u′ = v′ and u is adjacent to v in G.

For a graph G and vertex vV(G), let NG(v) denote the neighborhood of v and dG(v) = |NG(v)| denote the degree of v. The greatest distance between any two vertices in G is the diameter of G, denoted by diam(G). The girth g(G) of G, is the length of a shortest cycle in G. For any integer i, we call the ith neighborhood of v, and the vertices in are called the ith neighbors of v. In particular, is precisely the neighborhood NG(v) of v, , while for i > diam(G). In addition, let Pn and Cn denote the path and the cycle with n vertices, respectively.

The following lemmas are useful in the study of the Wiener polarity index of the lattice networks.

Lemma 1 ([69]) Let G and H be two non-trivial connected graphs, then where m(G) and m(H) are the number of edges of G and H, respectively, and W2(G) = |{{u, v}|d(u, v) = 2, u, vV(G)}| is the number of unordered pairs of vertices {u, v} of G such that dG(u, v) = 2.

Lemma 2 ([60]) Let T = (V, E) be a tree. Then

Lemma 3 ([62, 67]) Let U = (V, E) be a unicyclic graph. Let C denote the unique cycle of U:

  • If g(U) = 3 with V(C) = {v1, v2, v3}, then
  • If g(U) = 4 with V(C) = {v1, v2, v3, v4}, then
  • If g(U)≥5, then

Lastly, we provide a simple general formula for the Wiener polarity index of a graph, which plays an important role in the proofs of our main results.

Lemma 4 For any graph G, the Wiener polarity index Wp(G) of G can be expressed as (2)

Proof. By the definitions of the Wiener polarity index and the ith neighborhood, we have and Formula (2) then follows immediately.

To study the Wiener polarity index of lattice networks, we first consider the simple case concerning the square lattices.

Let PmPn(m ≥ 2, n ≥ 2), PmCn(m ≥ 2, n ≥ 3), and CmCn(m ≥ 3, n ≥ 3) denote the square lattices with free, cylindrical and toroidal boundary conditions respectively, it is easy to see that PmPn is a sequence of spanning subgraphs of the sequence PmCn of finite graphs, and PmCn is a sequence of spanning subgraphs of the sequence CmCn of finite graphs. Following the aforementioned lemmas, the Wiener polarity index of the square lattices can be easily calculated. In the rest of this section, we assume without loss of generality, that nm in PmPn and CmCn.

Theorem 1 Let PmPn, PmCn and CmCn denote the square lattices with free, cylindrical and toroidal boundary conditions, respectively. Then (3) (4) (5)

Proof. It follows from Lemma 2 that Wp(P2) = 0 and Wp(Pm) = m − 3 for m ≥ 3. On the other hand, Lemma 3 yields that Wp(C3) = Wp(C4) = Wp(C5) = 0, Wp(C6) = 3, and Wp(Cn) = n for n ≥ 7. By the definition of W2(G), it is not difficult to see that W2(Pm) = m − 2 for m ≥ 2, W2(C3) = 0, W2(C4) = 2, and W2(Cn) = n for n ≥ 5. In addition, by Lemma 1 we have proving Eq (3).

Similarly, Eqs (4) and (5) are also direct consequences of Lemma 1 and the above discussion.

Remark 1 For small values of m and n, still by Lemma 1, we have the followings:

Wp(P2P2) = 0; Wp(P2C3) = 0, Wp(P2C4) = 4, Wp(P2C5) = 10, Wp(P2C6) = 18; Wp(C3C3) = 0, Wp(C3C4) = 12, Wp(C3C5) = 30, Wp(C3C6) = 45, Wp(C4C4) = 32, Wp(C4C5) = 60, Wp(C4C6) = 84, Wp(C5C5) = 100, Wp(C5C6) = 135, Wp(C6C6) = 180.

The Wiener polarity index of the hexagonal lattices

Next we consider the Wiener polarity index of the hexagonal lattices. We follow the notations in [50]. The hexagonal lattices with toroidal, cylindrical and free boundary conditions, are denoted by Ht(n, m), Hc(n, m) and Hf(n, m), respectively, where (a1, b1), (a2, b2), …, (am+1, bm+1); , , , …, , (cn, bm+1) are edges in Ht(n, m) (as illustrated in Fig 1). The hexagonal lattice Hc(n, m) is obtained from Ht(n, m) by deleting edges , , , …, , (cn, bm+1). If the edges (a1, b1), (a2, b2), …, (am+1, bm+1) are also removed, then the hexagonal lattice Hf(n, m) with free boundary condition is obtained. It is easy to see that |V(Ht(n, m))| = |V(Hc(n, m))| = |V(Hf(n, m))| = 2(n + 1)(m + 1). Furthermore, from the definitions it is obvious that Hf(n, m) and Hc(n, m) are spanning subgraphs of Ht(n, m). In the following result we assume, without loss of generality, that mn in Ht(n, m) and Hf(n, m).

Theorem 2 For the hexagonal lattices Ht(n, m), Hc(n, m) and Hf(n, m) with toroidal, cylindrical and free boundary conditions, we have

Remark 2 As illustrated in Fig 2, we generally have with the common asymptotic value 9mn as both m an n approaches infinity.

Proof. We consider each of these three lattices through different cases.

(1) First consider Ht(n, m). Our argument relies on the simple fact that every vertex in Ht(n, m) has the same number of 3rd neighbors. Take, for instance, vertex a1 from Fig 1:

  • If n = 1 and m ≥ 3, then Hence the number of 3rd neighbors of any vertex in Ht(1, m) is 5 and Lemma 4 implies that
  • If n = 2 and m ≥ 3, then Thus the number of 3rd neighbors of any vertex in Ht(2, m) is 8 and Lemma 4 implies that
  • If n ≥ 3 and m ≥ 3, then Similarly we have

(2) In the case of Hc(n, m) (Fig 1), we can make similar observations on vertices whose 3rd neighborhood share the same cardinality. For this purpose we partition the vertex set Hc(n, m) into m + 1 disjoint classes:

  • when m is odd, or when m is even.

Since vertices in the same class have the same number of 3rd neighbors, we only need to consider one vertex from each class.

  • If n = 1, it is easy to see that Wp(Hc(1, 1)) = 6 and Wp(Hc(1, 2)) = 16. When m ≥ 3, one can verify the followings: That is, for vV0, for vV1 or vV2, and for any vertex v in V′ = V(Hc(1, m)) − V0V1V2. By Lemma 4, It is not hard to check that Wp(Hc(1, 1)) (with m = 1) and Wp(Hc(1, 2)) (with m = 2) also satisfy this expression.
  • If n = 2, then Wp(Hc(2, 1)) = 21 and Wp(Hc(2, 2)) = 45. When m ≥ 3, the 3rd neighborhoods of the representative vertices are for , and for .
    Hence for vV0, for vV1, for vV2, and for any vertex v in V′ = V(Hc(2, m)) − V0V1V2. Consequently which also holds for m = 1 and m = 2.
  • If n ≥ 3, we have, for m = 1 or m = 2, and When m ≥ 3, similar to before we have and Hence also satisfied by Wp(Hc(n, 1)) and Wp(Hc(n, 2)).

(3) Lastly, we will evaluate the Wiener polarity index of Hf(n, m) recursively.

First consider the case n = 1. Through direct computation we have, in Hf(1, m), the followings when m ≥ 3:

,

,

,

for 2 ≤ im − 1,

for 3 ≤ im,

,

, and

.

Thus also satisfied by Wp(Hf(1, 1)) = 7 and Wp(Hf(1, 2)) = 16.

Now let n ≥ 2. The structure of Hf(n, m) yields Here and appear because of the double counted terms. By applying this recursion we can explicitly compute Wp(Hf(n, m)) for n ≥ 2.

For m ≥ 3, it is not difficult to check the 3rd neighborhoods individual vertices, we skip the details. As a result we have and for m ≥ 3. It is easy to check Wp(Hf(n, 1)) = 9n − 2 and Wp(Hf(n, 2)) = 18n − 2, and hence Wp(Hf(n, m)) = 9nm − 2 for n ≥ 2 and m ≥ 1. Furthermore, we have Wp(Hf(1, m)) = 9m − 2. Thus we conclude that Wp(Hf(n, m)) = 9nm − 2 for n ≥ 1 and m ≥ 1.

Remark 3 For small values of n and m, direct computation yields

Wp(Ht(1, 1)) = 4, Wp(Ht(1, 2)) = Wp(Ht(2, 1)) = 18, Wp(Ht(2, 2)) = 54.

The Wiener polarity index of the triangular lattices

We now turn our attention to the Wiener polarity index of the triangular lattices. Again our notations follow [50]. The triangular lattices with toroidal, cylindrical and free boundary conditions are respectively denoted by Tt(n, m), Tc(n, m) and Tf(n, m). It is not hard to see, that the triangular lattice with toroidal boundary condition Tt(n, m) can be considered as an n × m square lattice CnCm with toroidal boundary condition with an additional diagonal edge added to every square. As in Fig 3, , , …, ; , , …, ; , , …, , ; , , …, are edges. Note that a1 = b1, , and . The triangular lattice with cylindrical boundary condition Tc(n, m) is obtained by deleting the edges , , …, ; , , …, , from Tt(n, m). If we further remove the edges , , …, ; , , …, from Tc(n, m), the triangular lattice with free boundary condition Tf(n, m) is then obtained. Since it has been established that CnCm exists for n ≥ 3 and m ≥ 3, in what follows we assume n ≥ 3 and m ≥ 3 for all the triangular lattices considered.

Theorem 3 Let Tf(n, m), Tc(n, m) and Tt(n, m) be the triangular lattices with free, cylindrical and toroidal boundary conditions respectively. Then we have

(i) For n ≥ 3 and m ≥ 3, Wp(Tf(n, m)) = 9nm − 18n − 18m + 31;

(ii)

(iii) Moreover, for small values of n and m we have the followings:

Wp(Tt(n, m)) = 0 if n = 3, 3 ≤ m ≤ 5 or n = m = 4;

Wp(Tt(3, 6)) = 27; Wp(Tt(4, 5)) = 20; Wp(Tt(4, 6)) = 72;

Wp(Tt(5, 5)) = 75; Wp(Tt(5, 6)) = 165; Wp(Tt(6, 6)) = 270.

Remark 4 As illustrated in Fig 4, we generally have with the common asymptotic value 9mn as both m an n approaches infinity. It is interesting to note that the Wiener polarity index of the triangular lattices and that of the hexagonal lattices are approximately the same.

Proof. We consider each of the three statements.

(i) First we consider the triangular lattice with free boundary condition, Tf(n, m), with the assumption nm.

Case 1 When m = 3:

  • If n = 3, Wp(Tf(3, 3)) = 4.
  • If n ≥ 4, we have Thus also satisfied by Wp(Tf(3, 3)) = 4.

Case 2 When m = 4, for nm we have Hence

Case 3 When m = 5, similar to Case 2 we have Wp(Tf(n, 5)) = 27n − 59 for nm = 5.

Case 4 When m ≥ 6 and nm. Simple computation yields Noting that Wp(Pm) = m − 3 and Wp(Tf(3, m)) = Wp(Tf(m, 3)) = 9m − 23, we have Since this formula coincides with our findings for smaller values of m, we conclude that Wp(Tf(n, m)) = 9nm − 18n − 18m + 31 for n ≥ 3 and m ≥ 3.

(ii) Next we consider the triangular lattice with cylindrical boundary condition Tc(n, m). The symmetric structure of Tc(n, m) indicates that the vertices from the same row have the same number of 3rd neighbors. Thus it suffices to compute for i = 1, 2, …, m to obtain Wp(Tc(n, m)).

Case 1 If n = 3; first suppose m ≥ 7. Direct computation yields and Thus Lemma 4 implies that On the other hand, for 3 ≤ m ≤ 6, one can verify Wp(Tc(3, m)) = 9m − 27 through direct computation.

Case 2 If n = 4, 5, or 6; Following essentially the same arguments as that of Case 1, we have

Case 3 If n ≥ 7:

  • When m = 3, it is easy to see that for i = 1, 2, 3, and hence Lemma 4 implies
  • When m = 4, we have and . Thus
  • Similarly, when m = 5 or 6 we have
  • When m ≥ 7, examining the 3rd neighborhoods yields Consequently Again this formula can be verified with m = 3, 4, 5, 6, and hence we may conclude that Wp(Tc(n, m)) = 9nm − 18n for n ≥ 7 and m ≥ 3.

(iii) Lastly, we consider Tt(n, m) with the assumption that mn. It is easy to see that all vertices share the same number of 3rd neighbors.

For small values of n and m, we have Wp(Tt(n, m)) = 0 for n = 3 and 3 ≤ m ≤ 5 or n = m = 4; Wp(Tt(3, 6)) = 27; Wp(Tt(4, 5)) = 20; Wp(Tt(4, 6)) = 72; Wp(Tt(5, 5)) = 75; Wp(Tt(5, 6)) = 165; Wp(Tt(6, 6)) = 270.

For m ≥ 7 we consider different cases depending on the value of n.

  • If n = 3 and m ≥ 7, then for any vertex vV(Tt(3, m)) and
  • If n = 4 and m ≥ 7, for any vertex vV(Tt(4, m)) and Wp(Tt(4, m)) = 20m.
  • If n = 5 and m ≥ 7, for any vertex vV(Tt(5, m)) and Wp(Tt(5, m)) = 35m.
  • If n = 6 and m ≥ 7, for any vertex vV(Tt(6, m)) and Wp(Tt(6, m)) = 51m.
  • If n ≥ 7 and m ≥ 7, direct computation shows for any vertex vV(Tt(n, m)). Hence

The Wiener polarity index of the 33 ⋅ 42 lattices

We conclude our study by considering the Wiener polarity index of the 33 ⋅ 42 lattices, following the notations of [50]. The 33 ⋅ 42 lattice with toroidal boundary condition, denoted by St(n, 2m), can be constructed from the square lattice C2mCn by adding a diagonal edge in each square of every other row, as shown in Fig 5. Here a1 = b1, , , , and , , …, ; , , …, ; , , …, are edges. If we remove the edges , , …, of St(n, 2m), then the 33 ⋅ 42 lattice with cylindrical boundary condition, denoted by Sc(n, 2m), is obtained. The 33 ⋅ 42 lattice Sf(n, 2m) with free boundary condition is obtained by further removing edges , , …, ; , , …, from Sc(n, 2m). Similar to before, we will assume n ≥ 3 and m ≥ 2 when the 33 ⋅ 42 lattices are discussed.

Theorem 4 Let Sf(n, 2m), Sc(n, 2m) and St(n, 2m) be the 33 ⋅ 42 lattices with free, cylindrical and toroidal boundary conditions, respectively. Then

(i) For n ≥ 3 and m ≥ 2, Wp(Sf(n, 2m)) = 15nm − 13n − 25m + 15;

(ii)

(iii)

In addition, for small n and m we have Wp(St(3, 4)) = 0; Wp(St(3, 6)) = 45; Wp(St(4, 4)) = 16; Wp(St(4, 6)) = 108; Wp(St(5, 4)) = 50; Wp(St(5, 6)) = 165; Wp(St(6, 4)) = 84; Wp(St(6, 6)) = 234.

Remark 5 As illurstrated in Fig 6, we generally have with the common asymptotic value 15mn as both m an n approaches infinity.

Proof. We consider each of the three cases as follows:

(i) We start with the 33 ⋅ 42 lattice Sf(n, 2m) with free boundary condition. First we consider Wp(Sf(n, 2m)), noting that for 1 ≤ i ≤ 2m, thus for m ≥ 4 we have Plugging into Eq (2), we have Wp(Sf(3, 2m)) = 20m − 24 for m ≥ 4. Since Wp(Sf(3, 4)) = 16 = 20 × 2 − 24 when m = 2 and Wp(Sf(3, 6)) = 36 = 20 × 3 − 24 when m = 3, we conclude that Wp(Sf(3, 2m)) = 20m − 24 for m ≥ 2.

Now suppose n ≥ 4. Observe that if the vertices in the first column are removed from Sf(n, 2m), then Sf(n − 1, 2m) is obtained. We now consider the vertices in the first column of Sf(n, 2m) and their 3rd neighbors: Applying Lemma 4 together with the fact that Wp(Pn) = n − 3, have, for m ≥ 4, Again this formula can be verified for small values of n and m. Hence Wp(Sf(n, 2m)) = 15nm − 13n − 25m + 15 for m ≥ 2.

(ii) Next we consider the 33 ⋅ 42 lattice Sc(n, 2m) with cylindrical boundary condition. In this case the vertices of Sc(n, 2m) in the same row have the same number of 3rd neighbors. Hence it is sufficient to compute for i = 1, 2, …, 2m.

Case 1 If n = 3 and m ≥ 2; When m ≥ 4 we have Then Lemma 4 implies This formula can be easily verified for Wp(Sc(3, 4)) = 15 (m = 2) and Wp(Sc(3, 6)) = 36 (m = 3).

Case 2 If n = 4 and m ≥ 2; When m ≥ 4, we have Thus Again this can be verified for m = 2 or 3.

Case 3 Similarly, we have

Case 4 If n ≥ 7 and m ≥ 2;

  • When m = 2 and n ≥ 7, we have Hence
  • When m = 3, we have and hence Wp(Sc(n, 6)) = 32n.
  • When m ≥ 4, we have Hence

Note that the formula Wp(Sc(n, 2m)) = 15nm − 13n also hold for both m = 2 and m = 3.

(iii) Lastly, we calculate the Wiener polarity index of St(n, 2m), the 33 ⋅ 42 lattice with toroidal boundary condition. As in Ht(n, m) and Tt(n, m), all vertices of St(n, 2m) have the same number of 3rd neighbors. Also note that |V(St(n, 2m))| = 2nm.

Case 1 If n = 3 and m ≥ 2, it is easy to see that Wp(St(3, 4)) = 0 and Wp(St(3, 6)) = 45. When m ≥ 4, we have for any vV(St(3, 2m)). Hence

Case 2 If n = 4 and m ≥ 2;

  • When m = 2, we have Wp(St(4, 4)) = 16.
  • When m = 3, we have Wp(St(4, 6)) = 108.
  • When m ≥ 4, we have |V(St(4, 2m))| = 8m and for any vV(St(4, 2m)). Hence Wp(St(4, 2m)) = 40m by Lemma 4.

Case 3 If n = 5 and m ≥ 2, similarly we have and

Case 4 If n = 6 and m ≥ 2, similarly we have and

Case 5 If n ≥ 7 and m ≥ 2:

  • When m = 2, we have for any vV(St(n, 4)) and Wp(St(n, 4)) = 16n.
  • When m = 3, we have for any vV(St(n, 6)) and Wp(St(n, 6)) = 42n.
  • When m ≥ 4, we have for any vV(St(n, 2m)) and hence

Concluding remarks

Evaluation of topological indices of network structures is an important problem in the study of network robustness [7173]. In particular, the computation of distance-based graph indices of various lattices has attracted the attention of researchers from many different backgrounds. By using a fundamental general formula of the Wiener polarity index of graphs, we determined the explicit formulas for the Wiener polarity index of the square lattices, the hexagonal lattices, the triangular lattices, and the 33 ⋅ 42 lattices with free, cylindrical and toroidal boundary conditions. The results of Theorems 2, 3, and 4 are plotted in Figs 46 respectively.

There exist other interesting graph structures of practical interests, such as the polyomino chains and the triangular Kagomé lattices. It would be worthwhile to explore their structure through computation of similar graph indices [7477]. Given our findings, it may also be interesting to study the asymptotic behavior of a given topological index of these lattices structures. Furthermore, it could be a challenge to develop theoretical bounds on such indices, when certain restrictions (such as the sizes and number of “holes” in a hexagonal system) accommodated by the graph structure.

Author Contributions

  1. Conceptualization: LC TL JL YS HW.
  2. Data curation: LC TL JL YS HW.
  3. Formal analysis: LC TL JL YS HW.
  4. Funding acquisition: LC TL JL YS HW.
  5. Investigation: LC TL JL YS HW.
  6. Methodology: LC TL JL YS HW.
  7. Project administration: LC TL JL YS HW.
  8. Resources: LC TL JL YS HW.
  9. Software: TL.
  10. Supervision: LC TL JL YS HW.
  11. Validation: LC TL JL YS HW.
  12. Visualization: LC TL JL YS HW.
  13. Writing – original draft: LC TL JL YS HW.
  14. Writing – review & editing: LC TL JL YS HW.

References

  1. 1. Boccaletti S, Latora V, Moreno Y, Chavez M, Hwanga D (2006) Complex networks: structure and dynamics. Physics Reports 424: 175–308.
  2. 2. Costa L. da F., Rodrigues F, Travieso G, Villas Boas PR, Characterization of complex networks: A survey of measurements, Adv. Phys.56 (2007) 167–242.
  3. 3. Dorogovtsev S, Mendes J (2002) Evolution of networks. Adv Phys 51: 1079–1187.
  4. 4. Ellens W, Kooij R (2013) Graph measures and network robustness. arXiv:1311.5064v1 [cs. DM].
  5. 5. Gutman I, Li X, Zhang J, (2009) Graph energy, in: Dehmer M, Emmert-Streib F(Eds.). Analysis of Complex Networks—From Biology to Linguistics. Wiley-VCH, Weinheim, pp. 145–174.
  6. 6. Kraus V, Dehmer M, Emmert-Streib F (2014) Probabilistic inequalities for evaluating structural network measures. Inform Sciences 288: 220–245.
  7. 7. Sydney A, Scoglio C, Schumm P, Kooij R (2008) Elasticity: topological characterization of robustness in complex networks. IEEE/ACM Bionetics.
  8. 8. Hosoya H (1971) Topological index, a newly proposed quantity characterizing the topological nature of structural isomers of saturated hydrocarbons. Bull Chem Soc Jpn 44: 2332–2339.
  9. 9. Bollobás B, Erdös P (1998) Graphs of extremal weights. Ars Combin 50: 225–233.
  10. 10. Li X, Shi Y (2008) A survey on the Randić index. MATCH Commun Math Comput Chem 59: 127–156.
  11. 11. Randić M (1975) On characterization of molecular branching. J Amer Chem Soc 97: 6609–6615.
  12. 12. Kier LB, Hall LH (1977) The nature of structure-activity relationships and their relation to molecular connectivity. Europ J Med Chem 12: 307–312.
  13. 13. Li X, Zheng J (2005) A unified approach to the extremal trees for different indices. MATCH Commun Math Comput Chem 54: 195–208.
  14. 14. Gutman I, Ruščić B, Trinajstić N, Wilcox CF (1975) Graph theory and molecular orbitals. XII. Acyclic polyenes. J Chem Phys 62: 3399–3405.
  15. 15. Gutman I, Trinajstić N (1972) Graph theory and molecular orbitals. Total π-electron energy of alternant hydrocarbons. Chem Phys Lett 17: 535–538.
  16. 16. Gupta S, Singh M, Madan AK (2000) Connective eccentricity index: a novel topological descriptor for predicting biological activity. J Mol Graph Model 18: 18–25. pmid:10935202
  17. 17. Rada J, Cruz R, Gutman I (2014) Benzenoid systems with extremal vertex-degree-based topological indices. MATCH Commun Math Comput Chem 72: 125–136.
  18. 18. Xu K, Liu M, Das KC, Gutman I, Furtula B (2014) A survey on graphs extremal with respect to distance-based topological indices. MATCH Commun Math Comput Chem 71(3): 461–508.
  19. 19. Balaban AT (1982) Highly discriminating distance based numerical descriptor. Chem Phys Lett 89: 399–404.
  20. 20. Balaban AT (1983) Topological indices based on topological distances in molecular graphs. Pure Appl Chem 55: 199–206.
  21. 21. Wiener H (1947) Structural determination of paraffin boiling points. J Am Chem Soc 69: 17–20. pmid:20291038
  22. 22. Klein DJ, Randić M (1993) Resistance distance. J Math Chem 12: 81–95.
  23. 23. Pirzada S, Ganie HA, Gutman I (2015) On Laplacian-energy-like invariant and Kirchhoff index. MATCH Commun Math Comput Chem 73: 41–59.
  24. 24. Nosal E (1970) Eigenvalues of Graphs. Master’s thesis, University of Calgary.
  25. 25. Li X, Shi Y, Gutman I (2012) Graph Energy. Springer.
  26. 26. Bozkurt SB, Güngör AD, Gutman I, Cevik AS (2010) Randić matrix and Randić energy. MATCH Commun Math Comput Chem 64: 239–250.
  27. 27. Gutman I (1978) The energy of a graph. Ber Math Statist Sekt Forsch Graz 103: 1–22.
  28. 28. Gutman I, Wagner S (2012) The matching energy of a graph. Discrete Appl Math 160(15): 2177–2187.
  29. 29. Jooyandeh M, Kiani D, Mirzakhah M (2009) Incidence energy of a graph. MATCH Commun Math Comput Chem 62: 561–572.
  30. 30. Chen L, Liu J (2016) Extremal values of matching energies of one class of graphs. Appl Math Comput 273: 976–992.
  31. 31. Chen L, Liu J, Shi Y (2015) Matching energy of unicyclic and bicyclic graphs with a given diameter. Complexity 21: 224–238.
  32. 32. Chen L, Liu J, Shi Y (2016) Bounds on the Matching Energy of Unicyclic Odd-cycle Graphs. MATCH Commun Math Comput Chem 75: 315–330.
  33. 33. Chen L, Shi Y (2015) The maximal matching energy of tricyclic graphs. MATCH Commun Math Comput Chem 73: 105–119.
  34. 34. Das KC, Sorgun S (2014) On Randić energy of graphs. MATCH Commun Math Comput Chem 72: 227–238.
  35. 35. Das KC, Sorgun S, Gutman I (2015) On Randić energy. MATCH Commun Math Comput Chem 73: 81–92.
  36. 36. Gu R, Li X, Liu J (2015) Note on three results on Randić energy and incidence energy. MATCH Commun Math Comput Chem 73: 61–71.
  37. 37. Jaklič G, Fowler PW, Pisanski T (2012) HL-index of a graph. Ars Math Contemp 5: 99–105.
  38. 38. Li X, Li Y, Shi Y, Gutman I (2013) Note on the HOMO-LUMO index of graphs. MATCH Commun Math Comput Chem 70(1): 85–96.
  39. 39. Chen X, Hou Y (2015) Some results on Laplacian Estrada index of graphs. MATCH Commun Math Comput Chem 73: 149–162.
  40. 40. Chen X, Qian J (2015) On resolvent Estrada index. MATCH Commun Math Comput Chem 73: 163–174.
  41. 41. Gao N, Qiao L, Ning B, Zhang S (2015) Coulson-type integral formulas for the Estrada index of graphs and the skew Estrada index of oriented graphs. MATCH Commun Math Comput Chem 73: 133–148.
  42. 42. Feng L, Liu W, Ilić A, Yu G (2013) The degree distance of unicyclic graphs with given matching number. Graphs Combin 29(3): 449–462.
  43. 43. Cao S, Dehmer M, Shi Y (2014) Extremality of degree-based graph entropies. Inform Sci 278: 22–33.
  44. 44. Chen Z, Dehmer M, Shi Y (2014) A note on distance-based graph entropies. Entropy 16(10): 5416–5427.
  45. 45. Li S, Yan W, Tian T (2016) The spectrum and Laplacian spectrum of the dice lattice. J Stat Phys 164: 449–462.
  46. 46. Liu J, Pan X (2015) A unified approach to the asymptotic topological indices of various lattices. Appl Math Comput 270: 62–73.
  47. 47. Liu J, Pan X (2015) Asymptotic incidence energy of lattices. Physica A 422: 193–202.
  48. 48. Liu J, Pan X, Hu F, Hu F (2015) Asymptotic Laplacian-energy-like invariant of lattices. Appl Math Comput 253: 205–214.
  49. 49. Shrock R, Wu F (2000) Spanning trees on graphs and lattices in d dimensions. J Phys A: Math Gen 33: 3881–3902.
  50. 50. Yan W, Zhang Z (2009) Asymptotic energy of lattices. Physica A 388: 1463–1471.
  51. 51. Ye L (2011) On the Kirchhoff index of some toroidal lattices. Linear Multilinear A 59: 645–650.
  52. 52. Bondy JA, Murty USR (2008) Graph Theory, Springer.
  53. 53. Dobrynin AA, Gutman I, Klavžar S, Žigert P (2002) Wiener index of hexagonal systems. Acta Appl Math 72: 247–294.
  54. 54. Ghebleh M, Kanso A, Stevanovic D (2016) On trees having the same Wiener index as their quadratic line graph. MATCH Commun Math Comput Chem 76: 731–744.
  55. 55. Khalifeh MH, Yousefi-Azari H, Ashrafi AR, Wagner SG (2009) Some new results on distance-based graph invariants. European J Comb 30: 1149–1163.
  56. 56. Wu X, Liu H (2010) On the Wiener index of graphs. Acta Appl Math 110: 535–544.
  57. 57. Dobrynin AA, Entringer RC, Gutman I (2001) Wiener index of trees: theory and applications. Acta Appl Math 66: 211–249.
  58. 58. Lukovits I, Linert W (1998) Polarity-numbers of cycle-containing structures. J Chem Inf Comput Sci 38: 715–719.
  59. 59. Hosoya H (2002) Mathematical and chemical analysis of Wiener’s polarity number, in: Rouvray DH, King RB(Eds.). Topology in Chemistry-Discrete Mathematics of Molecules. Horwood, Chichester.
  60. 60. Du W, Li X, Shi Y (2009) Algorithms and extremal problem on Wiener polarity index. MATCH Commun Math Comput Chem 62: 235–244.
  61. 61. Deng H, Xiao H, Tang F (2010) On the extremal Wiener polarity index of trees with a given diameter. MATCH Commun Math Comput Chem 63: 257–264.
  62. 62. Hou H, Liu B, Huang Y (2012) The maximum Wiener polarity index of unicyclic graphs. Appl Math Comput 218: 10149–10157.
  63. 63. Ma J, Shi Y, Wang Z, Yue J (2016) On Wiener polarity index of bicyclic networks. Sci Rep 6: 19066. pmid:26750820
  64. 64. Behmaram A, Yousefi-Azari H, Ashrafi AR (2012) Wiener polarity index of fullerenes and hexagonal systems. Appl Math Lett 25: 1510–1513.
  65. 65. Hua H, Das KC (2016) On the Wiener polarity index of graphs. Appl Math Comput 280: 162–167.
  66. 66. Deng H (2010) The wiener polarity index of molecular graphs of alkanes with a given number of methyl groups. J Serb Chem Soc 75: 1405–1412.
  67. 67. Liu M, Liu B (2011) On the Wiener polarity index. MATCH Commun Math Comput Chem 66: 293–304.
  68. 68. Liu M, Liu B (2013) A Survey on recent results of variable Wiener index. MATCH Commun Math Comput Chem 69: 491–520.
  69. 69. Ma J, Shi Y, Yue J (2014) The Wiener polarity index of graph products. Ars Combin 116: 235–244.
  70. 70. Zhang Y, Hu Y (2016) The Nordhaus-Gaddum-type inequality for the Wiener polarity index. Appl Math Comput 273: 880–884.
  71. 71. Chen M, Wang L, Sun S, Wang J, Xia CY (2016) Evolution of cooperation in the spatial public goods game with adaptive reputation assortment. Physics Letters A 380(1): 40–47.
  72. 72. Xia CY, Meloni S, Moreno Y (2012) Effects of environment knowledge on agglomeration and cooperation in spatial public goods games. Advances in Complex Systems 15 (supp01): 1250056.
  73. 73. Xia CY, Miao Q, Wang J, Ding S (2014) Evolution of cooperation in the traveler’s dilemma game on two coupled lattices. Appl Math Comput 246: 389–398.
  74. 74. Feng L, Yu G, Xu k, Jiang Z (2014) A note on the Kirchhoff index of bicyclic graphs. Ars Comb 114: 33–40.
  75. 75. Yu G, Feng L, Ilić A (2011) On the eccentric distance sum of trees and unicyclic graphs. J Math Anal Appl 375: 934–944.
  76. 76. Yu G, Feng L, Wang Q (2013) Bicyclic graphs with small positive index of inertia. Linear Algebra Appl 438: 2036–2045.
  77. 77. Yu G, Liu X, Qu H (2017) Singularity of Hermitian (quasi-)Laplacian matrix of mixed graphs. Appl Math Comput 293: 287–292.