Abstract
Topological indices are the numerical descriptors that correspond to some certain physicochemical properties of a chemical compound such as the boiling point, acentric factor, enthalpy of vaporisation, heat of fusion, etc. Among these topological indices, the Hyper Zagreb index, is the most effectively used topological index to predict the acentric factor of some octane isomers. In the current work, we investigate the extremal values of the Hyper Zagreb index for some classes of trees.
1. Introduction
Here, we only work with simple, connected, and finite tree graphs. We assume that in a graph , and represent the set of vertices and the set of edges, respectively. The degree of a vertex v is denoted by the symbol or . The symbols or represent a vertex’s highest degree in a graph G. Every acyclic graph is called a tree, which is denoted by T. A leaf or a pendant vertex is a vertex of degree one. A branching vertex contains a degree more than or equal to three. We use the notation to express the set of all neighbouring vertices to v. Further, we assume that . The graph comes into existence when some transformations or operations are applied on G. In such a case, we assume or is the vertex degree of t in G, whereas . For the undefined terminologies and notations related to this work, the reader can consult [1].
Graphs can be used to represent chemical substances (such as hydrocarbons) [2]. A topological descriptor is a number (or combination of numbers) that captures a certain characteristic of the graph. If a certain molecular property resembles the descriptor, it is considered a topological index. This particular type of descriptor, known as a topological index, can be used to analyse the physicochemical characteristics of chemical substances. During the past few years, there has been extensive research into the structural and analytical properties of topological indices in the field of mathematical chemistry [3,4,5]. For instance, topological indices have been used in (QSPR) and (QSAR) studies to model the properties of chemical compounds [6,7,8]. Topological indices have theoretical and practical appeal because they are now essential resources for the investigation of numerous real-world issues in fields such as computer science [9], physics [10], and ecology [11], among others.
Many topological index applications have been documented, the majority of which focus on investigating medical and pharmaceutical problems. A substantial and quickly growing field of study on this subject started midway through the 1990s, yielding a large amount of research, and this marked an important development in the study of topological indices mathematically. The work conducted by Erdös [12,13] should be particularly addressed within this scenario.
As an innovative method of describing heteroatoms in molecules, the idea of generalised molecular descriptors was put forth [14,15], in addition to evaluating structural variations [16]. The Zagreb types of topological indices and Zagreb polynomials for a few nanotubes covered by cycles were discussed in [17]. The regression relation between some topological indices and the sum-based geometric arithmetic index was determined in [18]. In [19], the authors showed the ability to use many topological indices for correlation, given the situation of typical temperatures of production; in addition, the authors assessed the common boiling temperatures of octane isomers. Siddiqui computed the exact formula for Zagreb indices and Zagreb polynomials for certain graphs, , , and , with , and they also determined the corresponding graphs, see [20].
To achieve a forecast of a certain property of molecules, topological indices are primarily used; for more information, see [21,22,23]. Finding the extremal graphs on a collection of graphs that satisfy the limitations imposed by the parameters that minimise (or maximise) the value of a topological index, is hence a straightforward challenge [24,25,26].
Here, we present some topological indices related to our current work. In [27], the variable sum connectivity index was explored as,
with
Note that consequently, and denote the and (H harmonic index), respectively. See [28,29,30,31], etc., for more detail.
In 2004, Zhou et al. [32] developed the Hyper Zagreb index as
which has been widely studied in different areas.
The Hyper Zagreb index and a number of the physicochemical characteristics of alkanes have been well correlated: vapor pressure, boiling points, surface areas, etc. The enthalpy of vaporisation (HVAP), acentric factor, heat of fusion (DHVAP) of octane isomers, and entropy are among the physicochemical parameters that have been assessed using the Hyper Zagreb index HM(G). In particular, the HM(G) highly correlates with the acentric factor [33]. By , we mean the maximum/minimum degree of a graph G; in terms of and , the extreme values on the Hyper Zagreb index of graph G have been examined [34]. In [35], the extreme values for the first and second Hyper Zagreb indices were established. Additionally, using benzenoid hydrocarbons’ boiling points as inputs, linear regression analysis was performed on degree-based indices. By comparison, the models corresponding to the other distance-based indices performed worse than the Hyper Zagreb index-based linear model.
In the current work, we characterize some nonsymmetric special classes of trees, which maximise or minimise the Hyper Zagreb index with n-vertex trees for (i) all odd degree vertices, (ii) fixed even degree vertices, and a (iii) fixed number of vertices of maximum degree.
2. Methodology
The following method was adopted to perform the research:
- Step 1: We categorized the family of tree graphs into three different classes say, , , and .
- Step 2: We constructed the lemmas to support our main results.
- Step 3: Some graph operations were defined in lemmas to search our extremal graphs.
- Step 4: By using the contradiction method and the lemmas, we obtained our main theorems in which we defined the exact formulas for the Hyper Zagreb index.
3. On the Minimum Hyper Zagreb Index of Trees with All Odd Degrees
Let be the collection of all n-vertex trees having an even number of vertices, and each vertex has an odd degree. In this section, we characterize the class of trees from , which contains the minimum value of the Hyper Zagreb index.
Lemma 1.
Let be a tree having a minimum Hyper Zagreb index. Let be a path in , which contains the vertex u, where the vertex u has the maximum degree among all , such that . We also assume that are the vertices adjacent to u in , and is a pendent vertex connected to u via ( may be equal to ). We define , as shown in Figure 1. Clearly, , and we have .
Figure 1.
and in Lemma 1.
Proof of Lemma 1.
We first find the difference − for the Hyper Zagreb index. Note that u and are the only vertices whose degrees differ in and . Therefore,
Since , a contradiction arises with our assumption. □
Theorem 1.
If , then , and the sign of equality holds, if and only if T possesses the degree sequence
Proof of Theorem 1.
Let minimize the Hyper Zagreb index. We claim that there exist only the vertices of degree 1 and degree 3 in T. Suppose that on the contrary, there exists at least one vertex v of a degree greater or equal to , i.e., . According to Lemma 1, we obtain a tree , such that , which contradicts the choice of T, when we apply the transformation defined in Lemma 1 to every vertex v of degree successively. In every step from a tree , we obtain a tree that contains a lower Hyper Zagreb index than its predecessor. We apply this transformation successively to obtain . Clearly, , and , where has the degree sequence
and the proof is complete. □
4. On the Minimum and Maximum Hyper Zagreb Index of Trees with Fixed Even Degree Vertices
Let denote the set of all n-vertex trees in which every tree has a fixed number of even degree vertices. Let the cardinality of the even degree vertices be denoted by r, such that and .
Lemma 2.
Let be a tree having a minimum Hyper Zagreb index. Let be a path in , which contains the vertex u, where u has the maximum degree in , such that , and . We also assume that are the vertices adjacent to u, and are pendent vertices in . We define ( may be equal to ), as shown in Figure 2. Clearly, ; then, .
Figure 2.
and in Lemma 2.
Proof of Lemma 2.
We first find the difference − for the Hyper Zagreb index. Note that u and are the only vertices whose degrees differ in and . Therefore,
and since , is a contradiction. □
Lemma 3.
Let and w be the vertices of a tree such that and , and and . We define from T as ; then, .
Let and t be the vertices of tree such that and , and , and (v may coincide with w). We define as ; then, .
Proof of Lemma 3.
From the structure of T and we have,
and since and , this implies .
From the structure of T and , we have,
and since and , this implies . □
Thus, by Lemma 3, we conclude that in order to minimise the Hyper Zagreb index, we need to place the vertices of degree 2 between the vertices of degree 3, so that there is at least one vertex of degree 2 between any two vertices of degree 3 and the remaining vertices of degree 2 or one vertex of degree 2 and one vertex of degree 3. So, we conclude that the collection of all those trees, which contain n vertices with the degree sequence
must contain a particular arrangement of vertices of degree 2 as described above. Note that if , there are not enough vertices of degree 2 to be placed between any two vertices of degree 3. If , in order to minimise the Hyper Zagreb index, we have to first put at least one vertex of degree 2 between any two vertices of degree 3 (if it is possible), and the remaining vertices of degree 2 are placed arbitrarily between two vertices of degree 2 or between a vertex of degree 3 and a vertex of degree 2.
Theorem 2.
If , where , and , then when , and when The sign of equality holds if and only if T possesses the degree sequence
Proof of Theorem 2.
Let minimise the Hyper Zagreb index. We claim that there does not exist any vertex of degree greater than 3 in T. To prove this claim, we suppose that on the contrary, there exists at least one vertex, say u, of a degree greater than 3 in T. According to Lemma 2, we can find another tree , such that , which contradicts the choice of T. When we apply this transformation successively on every vertex u in T, we will find a sequence of trees with the relation . Having in mind Lemmas 2 and Lemma 3 and the previous discussion, we conclude that the tree which minimises the Hyper Zagreb index contains the degree sequence with the particular arrangement of vertices defined above. Hence, has the degree sequence,
□
Now, we provide some lemmas, which help us to prove Theorem 3.
Lemma 4.
Let , with a maximum Hyper Zagreb index. Let be a path in containing the vertices and , where , such that for , we have . Let be the vertices adjacent to the vertex u. We define , as shown in Figure 3. Clearly, ; then, .
Figure 3.
and in Lemma 4.
Proof of Lemma 4.
We first find the difference − for the Hyper Zagreb index. Note that u and v are the only vertices whose degrees differ in and . Therefore,
and since and , we conclude . □
Lemma 5.
Let be a tree having a maximum Hyper Zagreb index. Let be a path in containing the vertex , where and ; , with . We also assume that are the vertices adjacent to in . We define ( may be equal to ), as shown in Figure 4. Clearly, ; then, .
Figure 4.
and in Lemma 5.
Proof of Lemma 5.
We first find the difference − for the Hyper Zagreb index. Note that u and are the only vertices whose degrees differ in and . Therefore,
and since and , we conclude . □
Lemma 6.
Let and w be the vertices of a tree , such that and , and , with and . We define from T as ; then, .
Proof of Lemma 6.
From the structure of T and , we have,
and since and , this implies . □
Lemma 7.
If there exists a pendent vertex adjacent to a branching vertex in the tree then the tree does not contain a vertex of degree two with both non-pendent neighbours.
Proof of Lemma 7.
Suppose on the contrary that a pendent vertex u is adjacent to a vertex v of a degree greater than two, and there exists a vertex w of degree two, such that , and both and are non-pendent vertices. We define , and clearly, .
Hence, . □
Theorem 3.
If , where and , then
where k denotes the maximum degree of the vertex in T, the sign of equality can be easily obtained, if and only if T possesses the degree sequence
where .
Proof of Theorem 3.
Let maximise the Hyper Zagreb index. We consider the following two cases for completion of the proof of the above theorem.
- Case 1
We claim that there does not exist any vertex of odd degree greater or equal to 3 in T. We suppose, on the contrary, that there exists at least one vertex of odd degree, say v, such that . According to the transformation defined in Lemma 4, we can find another tree , such that , which contradicts the choice of T. If we apply this transformation successively on every vertex v, we find a sequence of trees with the relation . It is easy to understand that in , there does not exist any branching vertex of odd degree.
- Case 2
We claim that there exists only one branching vertex of an even degree in . To prove this claim, we suppose, on the contrary, that there exists more than one branching vertex, say u, of even degree in . Then, by Lemma 5, we can find another tree , such that , and we obtain a contradiction. If we apply this transformation successively on every vertex having an even degree greater or equal to 4 in , we find a sequence of trees with the relation . It is easy to understand that in there exists only one branching vertex of an even degree, and the remaining even degree vertices are of degree 2.
From the above discussion and Lemmas 4–7, it follows that a tree , which maximises the Hyper Zagreb index, is a tree with only one branching vertex with degree k, such that an arbitrary vertex of degree one in is adjacent to a vertex of degree 2 (for ), or there are exactly pendent vertices with neighbours of degree 2 (for ).
Therefore, if , then for , and for . □
5. On the Minimum Hyper Zagreb Index of Trees with Fixed Vertices of Maximum Degree
Suppose that is the collection of all n-vertex trees, where k represents the number of vertices of maximum degree. Here, we discover the graphs of trees that possess the lower bound for the Hyper Zagreb index from . Since is only a member in , we consider for .
Lemma 8.
If is a tree having a minimum Hyper Zagreb index, where , then the maximum degree of a vertex in equals 3.
Proof of Lemma 8.
We suppose that , and the vertex u has the maximum degree in . We assume that is the longest path in containing the vertex u. We also assume that the vertices are the vertices adjacent to u in , and is a pendent vertex connected to u via ( may be equal to ). We define , as shown in Figure 5.
and since and , this implies that .
Figure 5.
and in Lemma 8.
It is easy to observe that contains vertices of degree . In the same way, we apply the transformation defined in Lemma 8 on every vertex of degree . In every step from a tree , we obtain a tree that contains a lesser than its predecessor. We apply this transformation k times, and we reach a tree that contains k vertices of the maximum degree . Clearly, , and , which is a contradiction of the choice of . □
Theorem 4.
If , then when , and when The sign of equality holds if and only if T possesses the degree sequence
Proof of Theorem 4.
Let minimize the Hyper Zagreb index. According to Lemma 8, the degree sequence of the tree is
where . The relation gives ,
Therefore,
- for , we have,
- and for , we have,
□
Corollary 1.
Let be a tree with a minimum Hyper Zagreb index in the class of and be a tree with a minimum Hyper Zagreb index in the class of . If , then .
6. Conclusions
Finding the extremal values (lower/upper bounds) of topological indices of a molecular structure has numerous applications in the field of chemical graph theory. These numerical values give important information regarding the physicochemical properties of the chemical compounds. In particular, investigating a topological index of a chemical tree sometimes gives a very good correlation to a physicochemical property of the chemical compound. The degree-based topological indices, such as the Hyper Zagreb index, have a very good correlation with the boiling point of benzenoid hydrocarbons and the acentric factor of some octane isomers. Therefore, in the current work, we computed the bounds of the Hyper Zagreb index of some specific tree structures, and the corresponding graphs were characterized. These extremal values not only help researchers to predict the properties of the chemical compounds but also significantly reduce the experimental costs.
However, it remains an open problem to study different topological indices for different classes of trees to predict the physicochemical properties of some chemical compounds.
Author Contributions
M.R. contributed to the following: conceptualization, designing the experiments, and analysing the data curation. A.A.B. contributed to the following: supervision, methodology, validation, project administration, and formal analysis. S.S. contributed to the following: performing the experiments, resources, software, some computations, funding, and writing the initial draft of the paper, which was edited and approved by M.J. M.A. wrote the final draft. All authors have read and agreed to the published version of the manuscript.
Funding
This research was supported by the Princess Nourah bint Abdulrahman University Researchers Supporting Project number (PNURSP2023R259), Princess Nourah bint Abdulrahman University, Riyadh, Saudi Arabia.
Data Availability Statement
Data sharing is not applicable to this article, as no data sets were generated or analysed during the current study.
Acknowledgments
The authors would like to thank the referees for their corrections, comments, and useful criticism, which improved the version of this paper.
Conflicts of Interest
The authors are not at cross purposes, as far as their interests in this work are concerned.
References
- Bondy, J.A.; Murty, U.S.R. Graph Theory, 3rd ed.; American Elsevier Publishing Company: Princeton, NJ, USA; Springer: Berlin/Heidelberg, Germany, 2008; pp. 26–264. [Google Scholar]
- Estrada, E.; Bonchev, D. Section 13.1. Chemical Graph Theory. In Handbook of Graph Theory, 2nd ed.; Gross, J.L., Yellen, J., Zhang, P., Eds.; CRC Press: Boca Raton, FL, USA, 2013; pp. 1538–1558. [Google Scholar]
- Gutman, I.; Furtula, B. (Eds.) Recent Results in the Theory of Randi’c Index; Univerzitet Kragujevac: Kragujevac, Serbia, 2008. [Google Scholar]
- Li, X.; Gutman, I. Mathematical Aspects of Randi’c Type Molecular Structure Descriptors; Univerzitet Kragujevac: Kragujevac, Serbia, 2006. [Google Scholar]
- Li, X.; Shi, Y. A survey on the Randi’c index. MATCH Commun. Math. Comput. Chem. 2008, 59, 127–156. [Google Scholar]
- Devillers, J.; Balaban, A.T. Topological Indices and Related Descriptors in QSAR and QSPR; Gordon & Breach: Amsterdam, The Netherlands, 1999; pp. 5055–5056. [Google Scholar]
- Muller, L.A.J.; Kugler, K.G.; Graber, A.; Dehmer, M. A network-base approach to classify the three domains of life. Biol. Direct. 2011, 6, 140–141. [Google Scholar]
- Vukicevic, D. Bond additive modelling 4. QSPR and QSAR studies of the variable Adriatic indices. Croatica Chem. Acta 2011, 41, 87–91. [Google Scholar] [CrossRef]
- Gutman, I.; Furtula, B.; Katani’c, V. Randi’c index and information. AKCE Int. J. Graphs Comb. 2018, 15, 307–312. [Google Scholar] [CrossRef]
- Estrada, E. Quantifying network heterogeneity. Phys. Rev. E 2010, 82, 066102. [Google Scholar]
- Pineda, J.; Martínez, C.; Mendez, A.; Muños, J.; Sigarreta, J.M. Application of Bipartite Networks to the Study of Water Quality. Sustainability 2020, 12, 5143. [Google Scholar] [CrossRef]
- Bollobás, B.; Erdös, P. Graphs of extremal weights. Ars Comb. 1998, 50, 225–233. [Google Scholar] [CrossRef]
- Bollobás, B.; Erdös, P.; Sarkar, A. Extremal graphs for weights. Discr. Math. 1999, 200, 5–19. [Google Scholar] [CrossRef]
- Randi’c, M. Novel graph theoretical approach to heteroatoms in QSAR. Chemom. Intel. Lab. Syst. 1991, 10, 213–227. [Google Scholar] [CrossRef]
- Randi’c, M. On computation of optimal parameters for multivariate analysis of structure-property relationship. J. Chem. Inf. Comput. Sci. 1991, 31, 970–980. [Google Scholar] [CrossRef]
- Randi’c, M.; Plavši´c, D.; Lerš, N. Variable connectivity index for cycle-containing structures. J. Chem. Inf. Comput. Sci. 2001, 41, 657–662. [Google Scholar] [CrossRef] [PubMed]
- Shao, Z.; Siddiqui, M.K.; Muhammad, M.H. Computing Zagreb Indices and Zagreb Polynomials for Symmetrical Nanotubes. Symmetry 2018, 10, 244. [Google Scholar] [CrossRef]
- Sheikh, U.; Tahir, Q.A.; Anjam, Y.N.; Pincak, R. Analysis of a Productive Topological Index Correlated to Fullerenes’ Physical Properties. Symmetry 2023, 15, 58. [Google Scholar] [CrossRef]
- Gutman, I.; Tošovi’c, J. Testing the quality of molecular structure descriptors. Vertex–degreebased topological indices. J. Serb. Chem. Soc. 2013, 78, 805–810. [Google Scholar] [CrossRef]
- Siddiqui, H.M.A. Computation of Zagreb indices and Zagreb polynomials of Sierpiński graphs. Hacettepe J. Math. Stat. 2020, 49, 754–765. [Google Scholar] [CrossRef]
- Das, K.C.; Çevik, A.S.; Cangul, I.N.; Shang, Y. On Sombor Index. Symmetry 2021, 13, 140. [Google Scholar] [CrossRef]
- Das, K.C.; Shang, Y. Some extremal graph with respect to sombor index. Mathematics 2021, 9, 1202. [Google Scholar] [CrossRef]
- Das, K.C.; Akgunes, N.; Togan, M.; Yurttas, A.; Born, I.C.W.; Cevik, A.S. On the first Zagreb index and multiplicative Zagreb coindices of graphs. An. S¸t. Univ. Ovidius Constant¸a 2016, 16, 153–176. [Google Scholar] [CrossRef]
- Rizwan, M.; Bhatti, A.A.; Javaid, M.; Jarad, F. Some bounds on bond incident degree indices with some parameters. Math. Probl. Eng. 2021, 2021, 8417486. [Google Scholar]
- Rizwan, M.; Bhatti, A.A.; Javaid, M.; Bonyah, E. Extremal Values of Variable Sum Exdeg Index for Conjugated Bicyclic Graphs. J. Chem. 2021, 2021, 4272208. [Google Scholar] [CrossRef]
- Rizwan, M.; Bhatti, A.A.; Javaid, M.; Shang, Y. Conjugated tricyclic graphs with maximum variable sum exdeg index. Heliyon 2023, 9, e15706. [Google Scholar] [CrossRef] [PubMed]
- Zhou, B.N.; Trinajsti’c, N. On general sum-connectivity index. J. Math. Chem. 2010, 47, 210–218. [Google Scholar] [CrossRef]
- Fajtlowicz, S. On conjectures of Graffiti—II. Congr. Numer. 1987, 60, 187–197. [Google Scholar]
- Deng, H.; Balachandran, S.; Ayyaswamy, S.K.; Venkatakrishnan, Y.B. On the harmonic index and the chromatic number of a graph. Discrete Appl. Math. 2013, 161, 2740–2744. [Google Scholar] [CrossRef]
- Zhong, L. The harmonic index for graphs. Appl. Math. Lett. 2012, 25, 561–566. [Google Scholar] [CrossRef]
- Zhong, L.; Xu, K. Inequalities between vertex-degree-based topological Indices. Match Commun. Math. Comput. Chem. 2014, 71, 627–642. [Google Scholar]
- Zhou, B.; Gutman, I. Relations between Wiener, hyper-Wiener and Zagreb indices. Chem. Phys. Lett. 2004, 394, 93–95. [Google Scholar] [CrossRef]
- Keerthi, G.; Mirajkar, B.R.; Doddamani, H.H.B. On Corellation of physicochemical properties and the hyper zagreb index for some molecular structures. South East Asian J. Math. Math. Sci. 2021, 3, 331–346. [Google Scholar]
- Elumalai, S.; Mansour, T.; Rostami, M.A.; Xavier, G.B.A. A Short Note On Hyper Zagreb Index. Bol. Soc. Paran. Mat. 2019, 37, 51–58. [Google Scholar] [CrossRef]
- Girish, V.; Rajasekharaiah, U.P. Murthy Hyper-Zagreb indices of graphs and its applications. J. Algebra Comb. Discret. Struct. Appl. 2020, 8, 9–22. [Google Scholar]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).