Abstract
Let G be a simple graph of order n and \(\mu_{1},\mu_{2},\ldots,\mu_{n}\) the roots of its matching polynomial. The matching energy of G is defined as the sum \(\sum_{i=1}^{n}|\mu_{i}|\). Let \(K_{n-1,1}^{k}\) be the graph obtained from \(K_{1}\cup K_{n-1}\) by adding k edges between \(V(K_{1})\) and \(V(K_{n-1})\). In this paper, we show that \(K_{n-1,1}^{k}\) has the maximum matching energy among the connected graphs with order n and edge connectivity k.
Similar content being viewed by others
1 Introduction
We use Bondy and Murty [1] for terminology and notations not defined in this paper and consider undirected and simple graphs only. Let \(G=(V, E)\) be a graph with order n. Denote by \(m(G,t)\) the number of t-matchings of G. Clearly, \(m(G,1)=e(G)\), the size of G, and \(m(G,t)=0\) for \(t > \lfloor n/2\rfloor\).
Recall that the matching polynomial of a graph G is defined as
and its theory is well elaborated in [2–4].
The energy of G is defined as
where \(\lambda_{1}, \lambda_{2},\ldots, \lambda_{n}\) are the eigenvalues of the adjacency matrix \(A(G)\) of G. The theory of graph energy is well developed nowadays; for details, see reviews [5, 6], the book [7], and the recent related results [8–23].
The Coulson integral formula [24] plays an important role in the study of the graph energy. For an acyclic graph T, its Coulson integral formula is as follows:
Motivated by equation (1), Gutman and Wagner [25] defined the matching energy of a graph G as
Energy and matching energy of graphs are closely related, and they are two quantities of relevance for chemical applications; for details see [26–28].
We now give an equivalent definition of matching energy.
Definition 1.1
[25]
Let G be a graph of order n, and let \(\mu_{1},\mu_{2}, \ldots, \mu _{n}\) be the roots of its matching polynomial. Then
The equation (2) induces a quasi-order relation over the set of all graphs on n vertices: if \(G_{1}\) and \(G_{2}\) are two graphs of order n, then
If \(G_{1}\preceq G_{2}\) and there exists some i such that \(m(G_{1},i) < m(G_{2}, i )\), then we write \(G_{1}\prec G_{2}\). Clearly,
Recall that the Hosoya index of a graph G is defined as \(Z(G)=\sum_{t\geq0} m(G,t)\) [29]. So we also have
In [3, 4], the authors gave two fundamental identities for the number of t-matchings of a graph.
Lemma 1.2
Let G be a graph, \(e=uv\) an edge of G, and \(N(u)=\{v_{1}(=v),v_{2},\ldots,v_{j} \}\) the set of all neighbors of u in G. Then we have
and
From Lemma 1.2, it is easy to get the following result.
Lemma 1.3
[25]
Let G be a graph and e one of its edges. Let \(G-e\) be the subgraph obtained from G by deleting the edge e. Then \(G-e\prec G\) and \(\mathit{ME}(G-e)<\mathit{ME}(G)\).
By Lemma 1.3, among all graphs on n vertices, the empty graph \(E_{n}\) and the complete graph \(K_{n}\) have the minimum and the maximum matching energy [25], respectively. It follows from equations (1) and (2) that \(\mathit{ME}(T)=E(T)\) for any tree T [25]. By using the quasi-order relation, some results have been obtained on extremal graphs with respect to matching energy among some classes of connected graphs with n vertices. For example, the extremal graphs in connected unicyclic, bicyclic graphs were determined in [25] and [30, 31], respectively; the maximal matching energy graphs of tricyclic graphs were obtained in [32], the minimal graphs among connected k-cyclic (\(k\leq n-3\)) graphs and bipartite graphs were characterized in [33]; the maximal connected graph with given connectivity (resp. chromatic number) was determined in [34]. For more as regards the latest results on matching energy, we refer the reader to [30, 35–41].
We now introduce some notations which will be used in the next section.
Let \(\mathcal{G}_{n,k}\) be the set of connected graphs of order n (≥2) with edge connectivity k (\(1\leq k\leq n-1\)). Let \(K_{n-1,1}^{k}\) be the graph obtained from \(K_{1}\cup K_{n-1}\) by adding k edges between \(V(K_{1})\) and \(V(K_{n-1})\); see Figure 1. In this paper, we show that \(K_{n-1,1}^{k}\) is the unique graph with the maximum matching energy (resp. Hosoya index) in \(\mathcal{G}_{n,k}\).
2 Main results
First we recall some notations. We denote by \(\kappa'(G)\) and \(\delta(G)\) the edge connectivity and the minimum degree of a graph G, respectively. Let S be a nonempty proper subset of V. We use \(G[S]\) to denote the subgraph of G induced by S. The edge cut of G, denoted by \(\partial(S)\), is a subset of \(E(G)\) of the form \([S,\bar{S}]\), where \(\bar{S}=V\backslash S\). An edge cut \(\partial(v)\) (\(v\in V\)) is called a trivial edge cut. A k-edge cut is an edge cut of k elements. Let \(G\in\mathcal{G}_{n,k}\). Then G must have a k-edge cut \(\partial(S)\) with \(1\leq|S| \leq\lfloor \frac{n}{2}\rfloor\).
Lemma 2.1
Let \(G\ncong K_{n-1,1}^{k}\) be a graph in \(\mathcal{G}_{n,k}\) with a trivial k-edge cut. Then \(G\prec K_{n-1,1}^{k}\).
Proof
Let \(\partial(S)\) be a trivial k-edge cut of G with \(|S|=1\). Since \(G\ncong K_{n-1,1}^{k}\), \(G[\bar{S}]\) is a proper subgraph of \(K_{n-1}\). Hence G is a proper subgraph of \(K_{n-1,1}^{k}\), and so the result follows from Lemma 1.3. □
Lemma 2.2
Let \(G \in\mathcal{G}_{n,k}\) be a graph without trivial k-edge cuts. Then for any k-edge cut \(\partial(S)\) of G with \(2\leq|S| \leq\lfloor\frac{n}{2}\rfloor\), we have \(|S|\geq k\).
Proof
For \(k\leq2\), the assertion is trivial, so suppose \(k\geq3\). Assume, to the contrary, that G has a k-edge cut \(\partial(S)\) with \(2\leq|S|\leq k-1\). By the facts that \(\delta(G)\geq\kappa'(G)=k\) and G has no trivial k-edge cuts, we have \(\delta(G)\geq k+1\), and thus \(\sum_{v\in S}d_{G}(v)\geq|S|(k+1)\). On the other hand, \(\sum_{v\in S}d_{G}(v)=2e(G[S])+k\leq|S|(|S|-1)+k\). Therefore, we have \(|S|(k+1) \leq|S|(|S|-1)+k\), that is, \((|S|-1)(k-|S|)+|S|\leq0\), which is a contradiction. Therefore the result holds. □
For \(k\leq m\leq\lfloor\frac{n}{2}\rfloor\), let \(K_{n-m,m}^{k}\) be a graph obtained from \(K_{n-m}\cup K_{m}\) by adding k independent edges between \(V(K_{n-m})\) and \(V(K_{m})\), as shown in Figure 1. It is easy to see that \(\kappa'(K_{n-m,m}^{k})=k\) and \(\kappa'(K_{n-1,1}^{k})=k\).
We next show that for a graph \(G \in\mathcal{G}_{n,k}\) without trivial k-edge cuts, \(G\preceq K_{n-m,m}^{k}\) for some m. Before this, we introduce a new graph operation as follows.
Let \(G_{1}\) be a graph in \(\mathcal{G}_{n,k}\) such that \(G_{1}\) has a k-edge cut \(\partial(S)\) with \(G[S]=K_{m}\), \(G[\bar{S}]=K_{n-m}\), and \(k\leq m\leq\lfloor\frac{n}{2}\rfloor\). Suppose that \(u_{1}, u_{2}\in\bar{S}\), \(v_{1}, v_{2}\in S\), \(e_{1}=u_{1}v_{1}\), \(e_{2}=u_{1}v_{2}\) are two edges of \(\partial(S)\), and \(u_{2}\) is not incident with any edge in \(\partial(S)\). If \(G_{2}\) is obtained from \(G_{1}\) by deleting the edge \(e_{2}\) and adding a new edge \(e'_{2}=u_{2}v_{2}\), we say that \(G_{2}\) is obtained from \(G_{1}\) by Operation I, as shown in Figure 2. Clearly, \(G_{2} \in\mathcal{G}_{n,k}\).
Lemma 2.3
If \(G_{2}\) is obtained from \(G_{1}\) by Operation I, then \(G_{1}\prec G_{2}\).
Proof
By equation (4), we have
and
Note that \(G_{1}-e_{2}\cong G_{2}-e'_{2}\), and \(G_{1}-u_{1}-v_{2}\) is isomorphic to a proper subgraph of \(G_{2}-u_{2}-v_{2}\). So, \(m(G_{1}-u_{1}-v_{2},t-1)\leq m(G_{2}-u_{2}-v_{2},t-1)\) for all t and \(m(G_{1}-u_{1}-v_{2},1)< m(G_{2}-u_{2}-v_{2},1)\). The result thus follows. □
Lemma 2.4
Let \(G \in\mathcal{G}_{n,k}\) be a graph without trivial k-edge cuts. Then \(G\preceq K_{n-m,m}^{k}\) for some m with \(\max\{k, 2\} \leq m\leq \lfloor\frac{n}{2}\rfloor\).
Proof
Let \(\partial(S)\) be a k-edge cut of G with \(2\leq|S| \leq\lfloor\frac{n}{2}\rfloor\). Let \(|S|=m\). Then \(m\geq k\) by Lemma 2.2. Let \(G_{1}\) be the graph obtained from G, by adding edges if necessary, such that \(G[S]\) and \(G[\bar{S}]\) are complete graphs. Therefore \(G\preceq G_{1}\) by Lemma 1.3. If \(G_{1} \ncong K_{n-m,m}^{k}\), then by using Operation I repeatedly, we can finally get \(K_{n-m,m}^{k}\) from \(G_{1}\). Hence \(G_{1} \preceq K_{n-m,m}^{k}\) by Lemma 2.3. The proof is thus complete. □
In the following, we show that \(K_{n-m,m}^{k}\prec K_{n-1,1}^{k}\) for \(m\geq2\).
Lemma 2.5
Suppose \(\max\{k, 2\} \leq m\leq\lfloor\frac{n}{2}\rfloor\). Then \(e(K^{k}_{n-m,m})< e(K^{k}_{n-1,1})\).
Proof
Note that
and
Hence we have
The proof is thus complete. □
Lemma 2.6
Let \(m \geq1\) be a positive integer. Then we have
and
Proof
We apply induction on m. For \(m=1\) and \(m=2\), the assertions are trivial since \(K^{1}_{2,2}\) and \(K^{1}_{3,2}\) are proper subgraphs of \(K^{1}_{3,1}\) and \(K^{1}_{4,1}\), respectively. So suppose that \(m\geq3\) and inequalities (6) and (7) hold for smaller values of m. By Lemma 1.2, we obtain
and
By the induction hypothesis, we obtain
Since \(m\geq3\), we have \(m-1\leq2m-3\) and \((m-1)^{2}\leq(2m-3)(2m-5)\) when \(m\geq4\). Notice that for \(m=3\), \(K^{1}_{m-2,m-2}=K_{m-1}\cup K_{m-3}\), and \((m-1)^{2}-1=(2m-3)(2m-5)\). Hence inequality (6) holds.
By Lemma 1.2, we get
and
By the induction hypothesis and inequality (6), we have
Notice that \(m\leq2m-2\) and \(m(m-1)\leq(2m-2)(2m-4)\). Therefore inequality (7) holds.
The proof is thus complete. □
Lemma 2.7
Suppose \(2\leq m\leq\lfloor\frac{n}{2}\rfloor\). Then
Proof
We apply induction on n. As the two cases \(n=2m\) and \(n=2m+1\) were proved by Lemma 2.6, we proceed to the induction step. By Lemma 1.2 and the induction hypothesis, we have
and
Thus the result follows by the fact that \(m\leq n-2\). □
Lemma 2.8
Suppose \(k\leq m\leq\lfloor\frac{n}{2}\rfloor\). Then
Proof
We apply induction on k. As the case \(k=1\) was proved by Lemma 2.7, we suppose that \(k\geq2\) and the assertion holds for smaller values of k. By equation (4), we have
and
By the induction hypothesis and Lemma 1.3, we obtain \(m(K^{k-1}_{n-m,m},t)\leq m(K^{k-1}_{n-1,1},t)\) and \(m(K^{k-1}_{n-m-1,m-1},t-1) \leq m(K_{n-2},t-1)\). Thus the result follows. □
Combining with Lemmas 2.5 and 2.8, we obtain the following result directly.
Corollary 2.9
Suppose \(\max\{k, 2\} \leq m\leq\lfloor\frac{n}{2}\rfloor\). Then \(K^{k}_{n-m,m}\prec K^{k}_{n-1,1}\).
Theorem 2.10
Let G be a graph in \(\mathcal{G}_{n,k}\). Then \(\mathit{ME}(G)\leq \mathit{ME}(K^{k}_{n-1,1})\). The equality holds if and only if \(G\cong K^{k}_{n-1,1}\).
Proof
Notice that \(K^{k}_{n-1,1} \in\mathcal{G}_{n,k}\). Let \(G\ncong K^{k}_{n-1,1}\) be a graph in \(\mathcal{G}_{n,k}\). It suffices to show that \(G\prec K^{k}_{n-1,1}\). If G has a trivial k-edge cut, then we have \(G\prec K_{n-1,1}^{k}\) by Lemma 2.1. Otherwise, by Lemma 2.4 and Corollary 2.9, we obtain \(G\prec K_{n-1,1}^{k}\) again. The proof is thus complete. □
By the proof of Theorem 2.10 and the definition of the Hosoya index, we can get the following result on the Hosoya index.
Theorem 2.11
Let G be a graph in \(\mathcal{G}_{n,k}\). Then \(Z(G)\leq Z(K^{k}_{n-1,1})\). The equality holds if and only if \(G\cong K^{k}_{n-1,1}\).
3 Conclusion
Chen et al. [30] characterized the extremal graphs of the matching energy in unicyclic, bicyclic graphs with a given diameter, respectively. Li et al. [37] obtained the unique graph having extremal matching energy in unicyclic graphs with fixed girth and the graphs with given clique number. Xu et al. [41] got the extremal graph of the matching energy among all t-apex trees. In [38], the authors obtained the extremal graph with matching energy in the complement of graphs. So it is interesting to characterize a graph having an extremal value of the matching energy with some graph invariants. In the paper, we got the maximal graph of the matching energy in graphs with given edge connectivity.
Therefore, our next work is to continue the study of graphs having extremal values of the matching energy with some graph invariants.
References
Bondy, JA, Murty, USR: Graph Theory. Springer, Berlin (2008)
Cvetković, D, Doob, M, Gutman, I, Torgašev, A: Recent Results in the Theory of Graph Spectra. North-Holland, Amsterdam (1988)
Farrell, EJ: An introduction to matching polynomials. J. Comb. Theory, Ser. B 27, 75-86 (1979)
Gutman, I: The matching polynomial. MATCH Commun. Math. Comput. Chem. 6, 75-91 (1979)
Gutman, I: The energy of a graph: old and new results. In: Betten, A, Kohnert, A, Laue, R, Wassermann, A (eds.) Algebraic Combinatorics and Applications, pp. 196-211. Springer, Berlin (2001)
Gutman, I, Li, X, Zhang, J: Graph energy. In: Dehmer, M, Emmert-Streib, F (eds.) Analysis of Complex Networks: From Biology to Linguistics, pp. 145-174. VCH, Weinheim (2009)
Li, X, Shi, Y, Gutman, I: Graph Energy. Springer, New York (2012)
Dehmer, M, Grabner, M: The discrimination power of molecular identification numbers revisited. MATCH Commun. Math. Comput. Chem. 69, 785-794 (2013)
Dehmer, M, Li, X, Shi, Y: Connections between generalized graph entropies and graph energy. Complexity 21(1), 35-41 (2015)
Das, KC, Mojallal, SA: Relation between energy and (signless) Laplacian energy of graphs. MATCH Commun. Math. Comput. Chem. 74(2), 359-366 (2015)
Gong, S, Li, X, Xu, G, Gutman, I, Furtula, B: Borderenergetic graphs. MATCH Commun. Math. Comput. Chem. 74(2), 321-332 (2015)
Huo, B, Ji, S, Li, X, Shi, Y: Complete solution to a problem on the maximal energy of bicyclic bipartite graphs. Linear Algebra Appl. 435, 804-810 (2011)
Huo, B, Ji, S, Li, X, Shi, Y: Complete solution to a conjecture on the fourth maximal energy tree. MATCH Commun. Math. Comput. Chem. 66(3), 903-912 (2011)
Ji, S, Li, J: An approach to the problem of the maximal energy of bicyclic graphs. MATCH Commun. Math. Comput. Chem. 68, 741-762 (2012)
Li, J, Guo, J, Shiu, WC: A note on RandiĆ energy. MATCH Commun. Math. Comput. Chem. 74(2), 389-398 (2015)
Li, X, Qin, Z, Wei, M, Gutman, I, Dehmer, M: Novel inequalities for generalized graph entropies - graph energies and topological indices. Appl. Comput. Math. 259, 470-479 (2015)
Li, X, Shi, Y, Wei, M, Li, J: On a conjecture about tricyclic graphs with maximal energy. MATCH Commun. Math. Comput. Chem. 72(1), 183-214 (2014)
Renqian, S, Ge, Y, Huo, B, Ji, S, Diao, Q: On the tree with diameter 4 and maximal energy. Appl. Comput. Math. 268, 364-374 (2015)
Maden, AD: New bounds on the incidence energy, Randic energy and Randic estrada index. MATCH Commun. Math. Comput. Chem. 74(2), 367-387 (2015)
Ma, H, Bai, L, Ji, S: On the minimal energy of conjugated unicyclic graphs with maximum degree at most 3. Discrete Appl. Math. 186, 186-198 (2015)
Marin, CA, Monsalve, J, Rada, J: Maximum and minimum energy trees with two and three branched vertices. MATCH Commun. Math. Comput. Chem. 74(2), 285-306 (2015)
Rojo, O: Effects on the energy and estrada indices by adding edges among pendent vertices. MATCH Commun. Math. Comput. Chem. 74(2), 343-358 (2015)
Ma, J, Shi, Y, Wang, Z, Yue, J: On Wiener polarity index of bicyclic networks. Sci. Rep. (in press)
Gutman, I, Polansky, OE: Mathematical Concepts in Organic Chemistry. Springer, Berlin (1986)
Gutman, I, Wagner, S: The matching energy of a graph. Discrete Appl. Math. 160(15), 2177-2187 (2012)
Aihara, J: A new definition of Dewar-type resonance energies. J. Am. Chem. Soc. 98, 2750-2758 (1976)
Gutman, I, Milun, M, Trinajstić, N: Topological definition of delocalisation energy. MATCH Commun. Math. Comput. Chem. 1, 171-175 (1975)
Gutman, I, Milun, M, Trinajstić, N: Graph theory and molecular orbitals 19, nonparametric resonance energies of arbitrary conjugated systems. J. Am. Chem. Soc. 99, 1692-1704 (1977)
Hosoya, H: A newly proposed quantity characterizing the topological nature of structural isomers of saturated hydrocarbons. Bull. Chem. Soc. Jpn. 44, 2332-2339 (1971)
Chen, L, Liu, J, Shi, Y: Matching energy of unicyclic and bicyclic graphs with a given diameter. Complexity 21(2), 224-238 (2015)
Ji, S, Li, X, Shi, Y: The extremal matching energy of bicyclic graphs. MATCH Commun. Math. Comput. Chem. 70, 697-706 (2013)
Chen, L, Shi, Y: The maximal matching energy of tricyclic graphs. MATCH Commun. Math. Comput. Chem. 73, 105-119 (2015)
Ji, S, Ma, H: The extremal matching energy of graphs. Ars Comb. 115, 343-355 (2014)
Li, S, Yan, W: The matching energy of graphs with given parameters. Discrete Appl. Math. 162, 415-420 (2014)
Chen, L, Liu, J, Shi, Y: Bounds on the matching energy of unicyclic odd-cycle graphs. MATCH Commun. Math. Comput. Chem. 75(2), 315-330 (2016)
Chen, X, Li, X, Lian, H: The matching energy of random graphs. Discrete Appl. Math. 193, 102-109 (2015)
Li, H, Zhou, Y, Su, L: Graphs with extremal matching energies and prescribed parameters. MATCH Commun. Math. Comput. Chem. 72, 239-248 (2014)
So, W, Wang, W: Finding the least element of the ordering of graphs with respect to their Matching Numbers. MATCH Commun. Math. Comput. Chem. 73, 225-238 (2015)
Wang, W, So, W: On minimum matching energy of graphs. MATCH Commun. Math. Comput. Chem. 74(2), 399-410 (2015)
Wu, T, Yan, Y, Zhang, H: Extremal matching energy of complements of trees. arXiv:1411.7458 (2014)
Xu, K, Zheng, Z, Das, K: Extremal t-apex trees with respect to matching energy. Complexity (2015). doi:10.1002/cplx.21651
Acknowledgements
This work was supported by National Natural Science Foundation of China (Grant Nos. 11101351, 11326216, 11401348, and 11561032), Jiangsu Government Scholarship for Overseas Studies, Natural Science Foundation of the Jiangsu Higher Education Institutions (No. 11KJB110014) and China Postdoctoral Science Foundation.
Author information
Authors and Affiliations
Corresponding author
Additional information
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
All authors contributed equally in the design and progress of the study. All authors modified and approved the final manuscript.
Rights and permissions
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.
About this article
Cite this article
Ji, S., Ma, H. & Ma, G. The matching energy of graphs with given edge connectivity. J Inequal Appl 2015, 415 (2015). https://doi.org/10.1186/s13660-015-0938-3
Received:
Accepted:
Published:
DOI: https://doi.org/10.1186/s13660-015-0938-3