Abstract
A topological index is a numerical property of a molecular graph that explains structural features of molecules. The potential of topological indices to discriminate between distinct structures is a significant topic to investigate. In this context, the exponential degree-based indices were put forward in the literature. The present work focuses on the exponential augmented Zagreb index (EAZ), which is defined for a graph G as
where \(d_i\) represents the degree of the vertex \(v_i\)and E(G) denotes the edge set of G. This work characterizes the maximal unicyclic graph for EAZ in terms of graph order, which was posed as an open problem in the recent article Cruz et al. (MATCH Commun Math Comput Chem 88:481-503, 2022).
Similar content being viewed by others
Avoid common mistakes on your manuscript.
1 Introduction
Throughout this article, we consider G as a simple connected graph of order n and size m having vertex set \(V(G)=\lbrace v_{1},\,v_{2},\,\ldots ,v_{n}\rbrace \) and edge set E(G). A graph G is said to be unicyclic if \(m=n\). The degree of vertex \(v_{i} \in V(G)\) is defined as the number of vertices adjacent to \(v_i\). Let \(N_G(v_i)\) be the neighbor set of the vertex \(v_i\), that is, \(N_G(v_i)=\{v_k\in V(G):\,v_iv_k\in E(G)\}\). The double star \(DS_{p,q}\) is a graph that is obtained after joining two central vertices of two stars \(S_p\) and \(S_q\).
Topological indices are numerical descriptors used in the field of mathematical chemistry and chem-informatics to characterize the topology of molecular structures. These indices provide a quantitative measure of the structural features of a molecule, focusing on the spatial arrangement of atoms and bonds rather than their chemical nature. The origin of topological indices lies in graph theory, where a molecule is depicted as a graph featuring atoms as vertices and bonds as edges. From a mathematical perspective, it is conceptualized as a function mapping all molecular graphs to real numbers, ensuring its invariance under graph isomorphism. These indices are employed to explain different physical and chemical properties of molecules, which supports researchers working on drug design, material science, and other branches of chemistry (Basak and Vracko 2020). After Wiener’s seminal work in 1947 Wiener (1947), a multitude of indices have been introduced in literature, utilizing various parameters such as degree, distance, eccentricity, and spectrum (Liu 2023; Hayat et al. 2023; Liu 2022; Maitreyi et al. 2023; Liu and Huang 2023; Du and Dimitrov 2020; Ghanbari 2022; Ali et al. 2021; Das et al. 2018; Gutman and Das 2004; Das and Vetrík 2023; Hosseini et al. 2022). While each category has its own set of applications and benefits, the domain of chemical graph theory is notably influenced by degree-based indices. The augmented Zagreb index (AZ) (Furtula et al. 2010) is one of the well-known degree based indices, which is formulated as
Widespread investigation of the AZ index is apparent in Chen et al. (2022); Sun et al. (2018); Cheng et al. (2021); Ali et al. (2021); Li et al. (2019, 2021); Ali (2021). To investigate the discrimination capability of topological indices, Rada (2019) proposed the exponential degree-based indices. A comprehensive identification of extremal trees for such invariants was reported in Cruz and Rada (2019). The path graph was established to be the maximal tree for the exponential Randić index (Cruz et al. 2020). Eliasi provided characterization of the maximal unicyclic and bicyclic graphs with respect to the exponential second Zagreb index in Eliasi (2022). For further insight on this direction, readers are referred to Xu et al. (2023); Cruz et al. (2021); Das et al. (2021); Wang and Wu (2022); Das and Mondal (2023); Carballosa et al. (2023). The present work focuses on the exponential augmented Zagreb index (Rada 2019), which is defined as
Cruz and Rada (2019) determined that the star graph serves as the smallest tree for EAZ. They left the problem of identifying the maximal tree open, and this was resolved in Das et al. (2021). The distinct extremal graphs for EAZ within the set of graphs with n non-isolated vertices were identified in Cruz and Rada (2022). For more works on the EAZ index, readers are referred to Das and Mondal (2024); Das et al. (2024). Let \(C_n\) and \(S_{n}^{\prime }\) be cycle and unicyclic graphs of order n, respectively, where \(S_n^{\prime }\) is generated from the star graph \(S_n\) of order n by attaching two pendent vertices. Most of the degree based indices yield \(C_n\) or \(S_{n}^{\prime }\) as maximal unicyclic graph in terms of graph order n. A unified approach reported by Cruz et al. (2022) to characterize extremal unicyclic graphs for degree-based indices also confirmed this fact. However, in case of EAZ, this approach is inadequate to generate maximal unicyclic graph. As a consequence, the problem of characterizing maximal unicyclic graph for EAZ in terms of graph order n is posed as an open problem in Cruz et al. (2022). The ultimate aim of this work is to solve this problem. We are surprised to state an amazing property of this extremal graph (see, Fig. 1) that the contribution of EAZ corresponding to one edge \(v_1v_2\) of this structure greater than the EAZ value of all other unicyclic graphs. Such scenario is rare to observe in extremal graph theory literature.
2 Main result
In this section, our aim is to explore the maximal unicyclic graph for the EAZ index.Let \(C^{3}_{\lfloor \frac{n-3}{2} \rfloor ,\,\lceil \frac{n-3}{2} \rceil }\) be a unicyclic graph (see, Fig. 1) of order n containing a cycle \(C_{3}:v_{1}v_{2}v_{3}v_{1}\) such that \(\lceil \frac{n-3}{2} \rceil \) pendent edges are incident on \(v_1\) and \(\lfloor \frac{n-3}{2} \rfloor \) pendent edges are incident on \(v_2\). We first obtain the following lemma which will be employed to generate the main outcome.
Lemma 1
Let \(C^{3}_{\lfloor \frac{n-3}{2} \rfloor ,\,\lceil \frac{n-3}{2} \rceil }\) be a unicyclic graph of order n as displayed in Fig. 1. Then there exists an edge \(v_1v_2 \in E\left( C^{3}_{\lfloor \frac{n-3}{2} \rfloor ,\,\lceil \frac{n-3}{2} \rceil }\right) \) such that
Proof
Using Sage (2015), one can easily verify that the assertion holds for \(n \le 9\). We now investigate the case where \(n \ge 10\).
First we assume that \(n=2p+1\). Then \(d_1=p+1=d_2\) and
Now,
Then
Now,
For \(n\ge 11\), we obtain
and hence
Using the above result with (2), from (1), we obtain
Next we assume that \(n=2p\). Then \(d_1=p+1\) and \(d_2=p\). Now,
Now one can write
For \(n\ge 10\), it is easy to check that
which implies
In view of above result, we obtain from (3) that
Hence, the proof is done. \(\square \)
Lemma 2
Let G be a unicyclic graph of even order n. Then there exists \(v_{\alpha },\,v_{\beta } \in V(G)\) such that \(d_{\alpha }=d_{\beta }=\frac{n}{2}\) if and only if \(G \cong H_k\), \(k=1,\,2,\,3,\,4,\,5\), where \(H_k\)’s are reported in Figs. 2, 3, 4 and 5.
Proof
Suppose \(n=2p\). First assume that there exists \(v_{\alpha },\,v_{\beta } \in V(G)\) such that \(d_{\alpha }=d_{\beta }=\frac{n}{2}=p\). We have
Let \(d_G(v_\alpha ,\,v_\beta )\) be the shortest distance between \(v_\alpha \) and \(v_\beta \) in G. Then \(d_G(v_\alpha ,\,v_\beta )\ge 1\). Now we build the proof in the three cases listed below.
\(\mathbf{Case\,1.}\) \(d_G(v_\alpha ,\,v_\beta )=1\). Since G is a unicyclic graph, \(0\le |N_G(v_\alpha )\cap N_G(v_\beta )|\le 1\). First we assume that \(|N_G(v_\alpha )\cap N_G(v_\beta )|=0\). By (4), we have \(|N_G(v_\alpha )\cup N_G(v_\beta )|=d_{\alpha }+d_{\beta }=n\). All the vertices in S are adjacent to either \(v_\alpha \) or \(v_\beta \), where \(S=V(G)\backslash \{v_\alpha ,\,v_\beta \}\). Since \(d_{\alpha }=d_{\beta }=\frac{n}{2}=p\) and \(v_\alpha v_\beta \in E(G)\), \(DS_{p-1,p-1}\) is a subgraph of G. Since G is a unicyclic graph with \(d_{\alpha }=d_{\beta }=\frac{n}{2}=p\), we have \(v_iv_j\in E(G)\), where \(v_i\in N_G(v_\alpha )\), \(v_j\in N_G(v_\beta )\); or \(v_i,\,v_j\in N_G(v_\alpha )\,(\text{ or } ~v_i,\,v_j\in N_G(v_\beta ))\), that is, \(G\cong H_1\) (see, Fig. 2) or \(G\cong H_3\) (see, Fig. 3).
Next we assume that \(|N_G(v_\alpha )\cap N_G(v_\beta )|=1\). Let \(v_\gamma \in N_G(v_\alpha )\cap N_G(v_\beta )\). By (4), we have \(|N_G(v_\alpha )\cup N_G(v_\beta )|=d_{\alpha }+d_{\beta }-1=n-1\). Using this result with \(d_{\alpha }=d_{\beta }=\frac{n}{2}=p\) and \(v_\alpha v_\beta \in E(G)\), we conclude that \(H_0\) (see, Fig. 6) is a subgraph of G with \(|V(H_0)|=n-1\). Let \(v_n\in V(G)\backslash V(H_0)\). Since G is unicyclic of order n, vertex \(v_n\) is adjacent to \(v_\gamma \), or \(v_nv_k\in E(G)\), where \(v_k\in N_G(v_\alpha )\) or \(v_k\in N_G(v_\beta )\), that is, \(G\cong H_4\) (see, Fig. 4) or \(G\cong H_5\) (see, Fig. 5).
\(\mathbf{Case\,2.}\) \(d_G(v_\alpha ,\,v_\beta )=2\). In this case \(|N_G(v_\alpha )\cup N_G(v_\beta )|\le n-2\). Since G is a unicyclic graph, \(1\le |N_G(v_\alpha )\cap N_G(v_\beta )|\le 2\). If \(|N_G(v_\alpha )\cap N_G(v_\beta )|=1\), then by (4), we obtain \(d_\alpha +d_\beta =|N_G(v_\alpha )\cup N_G(v_\beta )|+|N_G(v_\alpha )\cap N_G(v_\beta )|\le n-1<n=d_\alpha +d_\beta \), a contradiction. Otherwise, \(|N_G(v_\alpha )\cap N_G(v_\beta )|=2\). Again by (4), we have \(|N_G(v_\alpha )\cup N_G(v_\beta )|=n-2\). Since \(v_\alpha ,\,v_\beta \notin N_G(v_\alpha )\cup N_G(v_\beta )\), all the vertices in S are adjacent to \(v_\alpha \) or \(v_\beta \) or both, where \(S=V(G)\backslash \{v_\alpha ,\,v_\beta \}\). Since \(d_{\alpha }=d_{\beta }=\frac{n}{2}=p\) with \(|N_G(v_\alpha )\cap N_G(v_\beta )|=2\), we must have \(G\cong H_2\) (see, Fig. 2).
\(\mathbf{Case\,3.}\) \(d_G(v_\alpha ,\,v_\beta )\ge 3\). In this case \(|N_G(v_\alpha )\cap N_G(v_\beta )|=0\) and \(v_\alpha ,\,v_\beta \notin N_G(v_\alpha )\cup N_G(v_\beta )\). This with (4), we obtain \(|N_G(v_\alpha )\cup N_G(v_\beta )|\le n-2<n=d_\alpha +d_\beta =|N_G(v_\alpha )\cup N_G(v_\beta )|\), a contradiction.
Conversely, we assume that \(G \cong H_k,\) \(k=1,\,2,\,3,\,4,\,5\). Then from Figs. 2, 3, 4 and 5, it is evident that there exists \(v_{\alpha },\,v_{\beta } \in V(G)\) such that \(d_{\alpha }=d_{\beta }=\frac{n}{2}\). Hence the proof is done. \(\square \)
Lemma 3
Let G be a unicyclic graph of even order n. If G contains an edge \(v_{\alpha }v_{\beta }\) with \(d_{\alpha }=d_{\beta }=\frac{n}{2},\) then
Proof
Employing Lemma 2, we can write \(G \cong H_k,\) \(k=1,\,2,\,3,\,4,\,5\). From Figs. 2, 3, 4 and 5, it is clear that
One can easily find that \(EAZ(H_k)<EAZ\left( C^{3}_{\lfloor \frac{n-3}{2} \rfloor ,\,\lceil \frac{n-3}{2} \rceil }\right) \) for \(n=4,\,6,\,8,\,10,\) and \(k=1,\,2,\,3,\,4,\,5.\) Now we consider \(n \ge 12.\) Note that \(d_{i} \le p\) for any \(v_i \in V(H_k)\) (\(k=1,\,2,\,3,\,4,\,5.\)). For any \(v_{i}v_{j} \in E(H_k)\) satisfying \(d_i\ge d_j\ge 2\), we obtain
which exerts
For any \(v_{i}v_{j} \in E(H_k)\) satisfying \(d_i\ge d_j=1\), we obtain
that is,
Thus by Lemma 1, we obtain
for \(k=1,\,2,\,3,\,4,\,5\). Hence the proof is completed. \(\square \)
We are now ready to prove our main result of this paper.
Theorem 1
Let G be a unicyclic graph of order n. Then
with equality if and only if \(G\cong C^{3}_{\lfloor \frac{n-3}{2} \rfloor ,\,\lceil \frac{n-3}{2} \rceil }\).
Proof
Using Sage (2015), it is straightforward to examine the result to be true for \(n\le 9\). Therefore, our task is to establish the result for \(n\ge 10\). Suppose \(v_iv_j \in E(G)\) with \(d_i\ge d_j\). As G is a unicyclic graph, we must have \(d_i+d_j\le n+1\). We consider the subsequent two cases:
\(\mathbf{Case\,1.}\) \(n=2p+1\). In this case \(p\ge 5\). Thus we have \(d_j\le p+1\) (Otherwise, \(d_i\ge d_j\ge p+2\), that is, \(d_i+d_j\ge 2p+4=n+3\), a contradiction). If \(d_j=p+1\), then \(d_i=p+1\) (Otherwise, \(d_i\ge p+2\), that is, \(d_i+d_j\ge 2p+3=n+2\), a contradiction) and hence \(G\cong C^{3}_{\lfloor \frac{n-3}{2} \rfloor ,\,\lceil \frac{n-3}{2} \rceil }\) (see, Fig. 1). Thus we have
and hence equality appears in (5). Otherwise, \(d_j\le p\). We take into account the following two cases:
\(\mathbf{Case\,1.1.}\) \(d_i\le p+1\). Now,
that is,
as \(n\ge 11\). Since G is a unicyclic graph, employing the above finding with Lemma 1, it is clear that
Again the result (5) strictly holds.
\(\mathbf{Case\,1.2.}\) \(d_i\ge p+2\). Since \(d_i+d_j\le n+1\), we address the two subcases listed below:
\(\mathbf{Case\,1.2.1.}\) \(d_i+d_j\le n\). First we have to prove that
For \(d_i\ge d_j\ge 2\), we obtain
Consider
Then
Thus f(x) is strictly decreasing on \(2\le x\le p\), and hence
that is,
The result (6) holds. For \(d_i\ge d_j=1\), we obtain
as \(p\ge 5\). Again the result (6) holds. Similarly, as before, we obtain
Again the result (5) strictly holds.
\(\mathbf{Case\,1.2.2.}\) \(d_i+d_j=n+1\). In this case
Similarly, as \(\mathbf{Case\,1.2.1}\), the function \(\displaystyle {\frac{1}{x}+\frac{1}{2p+2-x}\,\left( 1-\frac{2}{x}\right) }\) is a strictly decreasing function on \(x\le p\) and hence
that is,
Now,
Again the result (5) strictly holds.
\(\mathbf{Case\,2.}\) \(n=2p\). In this case \(p\ge 5\). Since \(d_i+d_j\le n+1=2p+1\), we must have \(d_j\le p\). If \(d_j=p\), then either \(d_i=p\) or \(d_i=p+1\). For \(d_i=p+1\), we have \(G\cong C^{3}_{\lfloor \frac{n-3}{2} \rfloor ,\,\lceil \frac{n-3}{2} \rceil }\) (see, Fig. 1) as G is unicyclic. Thus
and hence the equality holds in (5). For \(d_i=p\), by Lemma 3, the result (5) strictly holds. Otherwise, \(d_j\le p-1\). The remaining portion of the proof can be developed by addressing the subsequent two cases:
\(\mathbf{Case\,2.1.}\) \(d_i\le p+2\). Now, we obtain
Combining the above fact with \(d_i\le p+2\) and \(p\ge 5\), we have
which exerts
It is evident from Lemma 1 that
Again the result (5) strictly holds.
\(\mathbf{Case\,2.2.}\) \(d_i\ge p+3\). Since \(d_i+d_j\le n+1\), we consider the following two subcases:
\(\mathbf{Case\,2.2.1.}\) \(d_i+d_j\le n\). First we have to prove that
For \(d_i\ge d_j\ge 2\), we obtain
Let us consider a function
Similarly, as \(\mathbf{Case\,1.2.1}\), the function g(x) is a strictly decreasing function on \(x\le p-1\), and hence
that is,
Thus (7) holds. For \(d_i\ge d_j=1\), we obtain
Again (7) holds. Similarly, as before, we obtain
\(\mathbf{Case\,2.2.2.}\) \(d_i+d_j=n+1\). Let us construct a function
Subsequently, it becomes evident that the function h(x) is increasing for \(x \le \frac{n-2}{2}\), and hence
Employing the aforementioned result, it is apparent that
Similarly, as before, we obtain
Again the result (5) strictly holds. This completes the proof of the theorem. \(\square \)
3 Concluding remarks
In this work, we have solved an open problem that was posed in Cruz et al. (2022). The maximal unicyclic graph has been characterized for the EAZ index in terms of graph order n. We are delighted to report a remarkable characteristic of this extremal graph (see, Fig. 1) that the contribution of EAZ associated with a single edge \(v_1v_2\) of this structure exceeds the EAZ values of all other unicyclic graphs individually. Observing such a scenario is uncommon in the literature of extremal graph theory.
Data availability
No data is associated with this work.
References
Ali A (2021) A note on minimal augmented Zagreb index of tricyclic graphs of fixed order. MATCH Commun Math Comput Chem 85:247–256
Ali A, Furtula B, Gutman I, Vukičević D (2021) Augmented Zagreb index: extremal results and bounds. MATCH Commun Math Comput Chem 85:211–244
Basak SC, Vracko MG (2020) Parsimony principle and its proper use/ application in computer-assisted Drug Design and QSAR. Curr Comput Aided Drug Des 16:1–5
Carballosa W, Quintana Y, Rodríguez JM, Sigarreta JM (2023) Exponential topological indices: optimal inequalities and applications. J Math Chem 61:933–949
Chen C, Liu M, Gu X, Das KC (2022) Extremal augmented Zagreb index of trees with given numbers of vertices and leaves. Discrete Math 345:112753
Cheng K, Liu M, Belardo F (2021) The minimal augmented Zagreb index of k-apex trees for \(k\in \lbrace 1, 2, 3 \rbrace \). Appl Math Comput 402:126139
Cruz R, Monsalve J, Rada J (2020) Trees with maximum exponential Randić index. Discrete Appl Math 283:634–643
Cruz R, Monsalve J, Rada J (2021) The balanced double star has maximum exponential second Zagreb index. J Comb Optim 41:544–552
Cruz R, Rada J (2019) The path and the star as extremal values of vertex-degree-based topological indices among trees. MATCH Commun Math Comput Chem 82:715–732
Cruz R, Rada J (2022) Extremal graphs for exponential VDB indices. Kragujev J Math 46:105–113
Cruz R, Rada J, Sanchez W (2022) Extremal unicyclic graphs with respect to vertex-degree-based topological indices. MATCH Commun Math Comput Chem 88:481–503
Das KC, Elumalai S, Balachandran S (2021) Open problems on the exponential vertex-degree-based topological indices of graphs. Discrete Appl Math 293:38–49
Das KC, Gutman I, Milovanović I, Milovanović E, Furtula B (2018) Degree-based energies of graphs. Linear Algebra Appl 554:185–204
Das KC, Mondal S (2023) On exponential geometric-arithmetic index of graphs. J Math Chem. https://doi.org/10.1007/s10910-023-01542-z
Das KC, Mondal S (2024) On EAZ index of unicyclic and bicyclic graphs, general graphs in terms of the number of cut edges. J Appl Math Comput. https://doi.org/10.1007/s12190-024-02086-4
Das KC, Mondal S, Huh D (2024) On the exponential augmented Zagreb index of graphs. J Appl Math Comput 70:839–865
Das KC, Vetrík T (2023) General Gutman index of a graph. MATCH Commun Math Comput Chem 89:583–603
Du Z, Dimitrov D (2020) The minimal-ABC trees with \(B_2\)-branches. Comput Appl Math 39:85
Eliasi M (2022) Unicyclic and bicyclic graphs with maximum exponential second Zagreb index. Discrete Appl Math 307:172–179
Furtula B, Graovac A, Vukicević D (2010) Augmented Zagreb index. J Math Chem 48:370–380
Ghanbari N (2022) On the Sombor characteristic polynomial and Sombor energy of a graph. Comput Appl Math 41:242
Gutman I, Das KC (2004) The first Zagreb index 30 years after. MATCH Commun Math Comput Chem 50:83–92
Hayat S, Arshad M, Gutman I (2023) Proofs to some open problems on the maximum Sombor index of graphs. Comput Appl Math 42:279
Hosseini SA, Mohar B, Ahmadi MB (2022) The evolution of the structure of ABC-minimal trees. J Comb Theory Ser B 152:415–452
Li F, Ye Q, Broersma H, Ye R (2021) Sharp upper bounds for augmented zagreb index of graphs with fixed parameters. MATCH Commun Math Comput Chem 85:257–274
Li F, Ye Q, Rada J (2019) The augmented Zagreb indices of fluoranthene-type benzenoid systems. Bull Malays Math Sci Soc 42:1119–1141
Liu H (2023) Comparison between Merrifield–Simmons index and some vertex-degree-based topological indices. Comput Appl Math 42:89
Liu H (2022) Extremal problems on Sombor indices of unicyclic graphs with a given diameter. Comput Appl Math 41:138
Liu H, Huang Y (2023) Sharp bounds on the symmetric division deg index of graphs and line graphs. Comput Appl Math 42:285
Maitreyi V, Elumalai S, Balachandran S, Liu H (2023) The minimum Sombor index of trees with given number of pendant vertices. Comp Appl Math 42:331
Rada J (2019) Exponential vertex-degree-based topological indices and discrimination. MATCH Commun Math Comput Chem 82:29–41
Stein WA (2015) Sage mathematics software (Version 6.8). The Sage Development Team, http://www.sagemath.org
Sun X, Gao Y, Du J, Xu L (2018) Augmented Zagreb index of trees and unicyclic graphs with perfect matchings. Appl Math Comput 335:75–81
Wang F, Wu B (2022) The reduced Sombor index and the exponential reduced Sombor index of a molecular tree. J Math Anal Appl 515:126442
Wiener H (1947) Structural determination of paraffin boiling points. J Am Chem Soc 69:17–20
Xu C, Horoldagva B, Buyantogtokh L (2023) The exponential second Zagreb index of \((n,\, m)\)-graphs. Mediterr J Math 20:181–188
Acknowledgements
The authors are grateful to the referees for their valuable comments, which have considerably improved the presentation of this paper. K. C. Das is supported by National Research Foundation funded by the Korean government (Grant No. 2021R1F1A1050646).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare no conflict of interest.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Das, K.C., Mondal, S. & Huh, Dy. Open problem on the maximum exponential augmented Zagreb index of unicyclic graphs. Comp. Appl. Math. 43, 317 (2024). https://doi.org/10.1007/s40314-024-02815-2
Received:
Revised:
Accepted:
Published:
DOI: https://doi.org/10.1007/s40314-024-02815-2