Next Article in Journal
Stanley Depth of Edge Ideals of Some Wheel-Related Graphs
Next Article in Special Issue
Two-Machine Job-Shop Scheduling with Equal Processing Times on Each Machine
Previous Article in Journal
Stability Analysis of a Fractional-Order Linear System Described by the Caputo–Fabrizio Derivative
Previous Article in Special Issue
Fractional Metric Dimension of Generalized Jahangir Graph
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Further Results on the Resistance-Harary Index of Unicyclic Graphs

1
School of Mathematical Sciences, Anhui University, Hefei 230601, China
2
College of Mathematics, Hunan City University, Yiyang 413000, China
3
School of Mathematics and Physics, Anhui Jianzhu University, Hefei 230601, China
*
Author to whom correspondence should be addressed.
Mathematics 2019, 7(2), 201; https://doi.org/10.3390/math7020201
Submission received: 20 December 2018 / Revised: 14 February 2019 / Accepted: 14 February 2019 / Published: 20 February 2019
(This article belongs to the Special Issue Discrete Optimization: Theory, Algorithms, and Applications)

Abstract

:
The Resistance-Harary index of a connected graph G is defined as R H ( G ) = { u , v } V ( G ) 1 r ( u , v ) , where r ( u , v ) is the resistance distance between vertices u and v in G. A graph G is called a unicyclic graph if it contains exactly one cycle and a fully loaded unicyclic graph is a unicyclic graph that no vertex with degree less than three in its unique cycle. Let U ( n ) and U ( n ) be the set of unicyclic graphs and fully loaded unicyclic graphs of order n, respectively. In this paper, we determine the graphs of U ( n ) with second-largest Resistance-Harary index and determine the graphs of U ( n ) with largest Resistance-Harary index.

1. Introduction

The topological index is the mathematical descriptor of the molecular structure, which can effectively reflect the chemical structure and properties of the material. The famous Wiener index W ( G ) (also Wiener number) introduced by H. Wiener, is a topological index of a molecule, defined as the sum of the lengths of the shortest paths between all pairs of vertices, i.e., W ( G ) = { u , v } V ( G ) d ( u , v ) in the chemical graph representing the non-hydrogen atoms in the molecule.In 1993, Klein and Randić [1] defined a new distance function named resistance distance on the basis of electrical network theory replacing each edge of a simple connected graph G by a unit resistor. Let G be a simple connected graph with vertices set V = { v 1 , v 2 , , v n } . The resistance distance between the vertices v i and v j , denoted by r ( v i , v j ) (if more than one graphs are considered, we write r G ( v i , v j ) to avoid confusion), is defined to be the effective resistance between the vertices v i and v j in G. If the ordinary distance is replaced by resistance distance in the expression for the Wiener index, one arrives at the Kirchhoff index [1,2]
K f ( G ) = { u , v } V ( G ) r ( u , v ) ,
which has been widely studied [3,4,5,6,7,8,9,10,11,12].
Another distance-based graph invariant index named Harary index was introduced independently by Plavšić et al. [13] and by Ivanciuc et al. [14] in 1993 for the characterization of molecular graphs. The Harary index H ( G ) is defined as
H ( G ) = { u , v } V ( G ) 1 d ( u , v ) ,
which is the sum of reciprocals of distances between all pairs of vertices of G. For more related results to Harary index, please refer to [15,16,17,18,19,20,21,22]. In 2017, Chen et al. [23,24] introduced a new graph invariant reciprocal to Kirchhoff index, named Resistance-Harary index, as
R H ( G ) = { u , v } V ( G ) 1 r ( u , v ) .
To understand the results and concepts, we introduce some definitions and notions. All of the graphs considered in this paper are connected and simple. A graph G is called a unicyclic graph if it contains exactly one cycle, simply denoted as G = U ( C l ; T 1 , T 2 , , T l ) , where C l is the unique cycle with vertices v 1 v 2 v l , T i is a tree rooted at v i , 1 i l . A fully loaded unicyclic graph is a unicyclic graph with the property that there is no vertex with degree less than three in its unique cycle. Let S n l denote the graph obtained from cycle C l by adding n l pendant edges to a vertex of C l . Let U ( n ; l ) be the set of unicyclic graphs with n vertices and the unique cycle C l and U ( n ) be the set of unicyclic graphs with n vertices. Let U ( n ; l ) be the set of all fully loaded unicyclic graphs with n vertices and the unique cycle C l , and U ( n ) be the set of unicyclic graphs with n vertices. Let S n and P n be the star and the path on n vertices, respectively.
In this paper, we improve the results of the recent paper (Chen et al. [23]) and we determine the largest Resistance-Harary index among all unicyclic graphs. Additionally, we determine the second-largest Resistance-Harary index among all unicyclic graphs and determine the largest Resistance-Harary index among all fully loaded unicyclic graphs and characterize the corresponding extremal graphs, respectively.

2. Preliminaries

In this section, we introduce some useful lemmas and two transformations. Let R G ( u ) = u V ( G ) { u } 1 r G ( u , v ) , then R H ( G ) = 1 2 u V ( G ) R G ( u ) . Let C g = v 1 v 2 v g v 1 be the cycle on g vertices where g 3 . By Ohm’s law, for any two vertices v i , v j V ( C g ) with i < j , one has
r C g ( v i , v j ) = ( j i ) ( g + i j ) g .
By a simple calculation, we can obtain the Resistance-Harary index of C g , which is
R H ( C g ) = u V ( C g ) 1 2 R C g ( v ) = g i = 1 g 1 1 i .
Lemma 1 
([1]). Let x be a cut vertex of a connected graph G and let a and b be vertices occurring in different components which arise upon deletion of x. Then,
r G ( a , b ) = r G ( a , x ) + r G ( x , b ) .
Definition 1 
([23]). Let v be a vertex of degree p + 1 in a graph G, such that v v 1 , v v 2 , , v v p are pendent edges incident with v, and u is the neighbor of v distinct from v 1 , v 2 , , v p . We form a graph G = σ ( G , v ) by deleting the edges v v 1 , v v 2 , , v v p and adding new edges u v 1 , u v 2 , , u v p . We say that G is a σ-transform of the graph G (see Figure 1).
Lemma 2 
([23]). Let G = σ ( G , v ) be a σ-transform from the graph G, d ( u ) 1 described in Figure 1. Then, R H ( G ) R H ( G ) , with equality holds if and only if G is a star with v as its center.
Definition 2 
([23]). Let u , v be two vertices in a graph G, such that u 1 , u 2 , , u s are pendents incident with u and v 1 , v 2 , , v t are pendents incident with v in G 0 G , respectively. G and G are two graphs β transformed from G, such that G = G { v v 1 , v v 2 , , v v t } + { u v 1 , u v 2 , , u v t } , G = G { v v 1 , v v 2 , , v v s } + { u v 1 , u v 2 , , u v s } , (see Figure 2).
Lemma 3 
([23]). Let G , G be the graphs transformed from the graph G, d ( u ) 1 described in Figure 2. Then, R H ( G ) < R H ( G ) , or R H ( G ) < R H ( G ) .
Corollary 1 
([23]). Let G be a connected graph with u , v V ( G ) . Denote by G ( s ; t ) the graph obtained by attaching s > 1 pendent vertices to vertex u and t > 1 pendent vertices to vertex v. Then, we have R H ( G ( 1 , s + t 1 ) ) > R H ( G ) or R H ( G ( s + t 1 , 1 ) ) > R H ( G ) .
Lemma 4.
The function f ( x ) = 2 ( k 1 ) ( x + 1 ) ( k 1 ) x 2 1 x + 1 1 k x 2 k 2 for k 3 and 1 x k 1 2 is strictly decreasing.
Proof. 
By simple calculation,
f ( x ) = 2 ( 1 k ) ( k 2 x 1 ) ( k ( x + 1 ) x 2 x 1 ) 2 1 ( k x ) 2 + 1 ( x + 1 ) 2 = ( k 2 x 1 ) 2 ( 1 k ) ( k ( x + 1 ) x 2 x 1 ) 2 + ( k + 1 ) ( x + 1 ) 2 ( k x ) 2 ,
Let g ( x ) = 2 ( 1 k ) ( k ( x + 1 ) x 2 x 1 ) 2 + ( k + 1 ) ( x + 1 ) 2 ( k x ) 2 , then we have g ( x ) = 2 ( k + 1 ) ( x + 1 ) 2 ( k x ) 3 2 ( k + 1 ) ( x + 1 ) 3 ( k 2 ) 2 4 ( k 1 ) ( k 2 x 1 ) ( k x + ( 1 k ) + x 2 + x ) 3 < 0 since 1 x k 1 2 and g ( 1 ) = 2 ( 1 k ) 4 k 2 12 k + 9 k + 1 4 k 2 8 k + 4 < 0 since k 3 . Thus, g ( x ) < 0 , since 1 x k 1 2 . It follows that f ( x ) < 0 , since 1 x k 1 2 and k 3 , thus implying the conclusion of the theorem. □

3. Main Results

By Lemmas 2 and 3, we claim that R H ( G ) R H ( S n g ) if G U ( n ; g ) . Next, we will determine the graphs in U ( n ) with the largest Resistance-Harary index and the second-largest Resistance-Harary index.

3.1. The Largest Resistance-Harary Index

Theorem 1.
If G U ( n ) , then
max G U ( n ) { R H ( G ) } = R H ( C n ) i f n 7 , R H ( S 8 5 ) i f n = 8 , R H ( S n 4 ) i f 9 n 15 , R H ( S n 3 ) i f n 16 .
Proof. 
Let H = G C g , by the definition of Resistance-Harary index, one has,
R H ( S n g ) = { u , v } V ( G ) 1 r ( u , v ) = { u , v } V ( C g ) 1 r ( u , v ) + { u , v } V ( H ) 1 r ( u , v ) + u V ( C g ) , v V ( H ) 1 r ( u , v ) = g 1 + 1 2 + 1 3 + + 1 g 1 + 1 4 ( n g ) ( n + 3 g ) + g ( n g ) 1 2 g 1 + 1 3 g 4 + + 1 g 2 ( g 1 ) 2 .
Similarly,
R H ( S n g 1 ) = ( g 1 ) 1 + 1 2 + 1 3 + + 1 g 2 + 1 4 ( n + 1 g ) ( n + 4 g ) + ( g 1 ) ( n + 1 g ) 1 2 g 3 + 1 3 g 7 + + 1 ( g 1 ) 2 ( g 2 ) 2 .
Further, by the symmetry of C g , one has,
= R H ( S n g 1 ) R H ( S n g ) = ( g 1 ) ( n + 1 g ) 1 2 g 3 + 1 3 g 7 + + 1 ( g 1 ) 2 ( g 2 ) 2 + 1 2 ( n g ) g ( n g ) 1 2 g 1 + 1 3 g 4 + + 1 g 2 ( g 1 ) 2 1 + 1 2 + 1 3 + + 1 g 1 = ( n g ) { g 1 2 g 3 + g 1 3 g 7 + + g 1 2 g 3 g 2 + 1 2 g 2 g 1 + g 3 g 4 + + g 2 g 1 g 1 } + ( g 1 ) ( 1 2 g 3 + 1 3 g 7 + + 1 ( g 1 ) 2 ( g 2 ) 2 ) 1 + 1 2 + 1 3 + + 1 g 1 .
To explore the relationship between Δ and parameters g, we first discuss the part of the first brace of Equation (1). Let
Θ 1 = g 1 2 g 3 + g 1 3 g 7 + + g 1 2 g 3 + 1 2 g 2 g 1 + g 3 g 4 + + g 2 g 1 ,
then
Θ 1 = g 1 2 g 3 g 2 g 1 + g 1 3 g 7 g 3 g 4 + + g 1 2 g 3 g 2 g 1 + 1 2 4 g + 4 > 0 , if g 4 and g is even , g 1 2 g 3 g 2 g 1 + g 1 3 g 7 g 3 g 4 + + g 1 2 g 3 g 2 g 1 + 1 2 4 g ( g + 2 ) 2 5 > 0 , if g 5 and g is odd .
Next, we consider the rest of Equation (1). Let
Θ 2 = ( g 1 ) 1 2 g 3 + 1 3 g 7 + + 1 ( g 1 ) 2 ( g 2 ) 2 1 + 1 2 + 1 3 + + 1 g 1 .
(i) If g is even, then
Θ 2 = ( g 1 ) i = 1 g 2 1 ( i + 1 ) ( g 1 ) i 2 i = 1 g 1 1 i = i = 1 g 2 2 2 ( g 1 ) ( i + 1 ) ( g 1 ) i 2 i = 1 g 2 1 i + 1 + 1 = i = 1 g 2 2 2 ( g 1 ) ( i + 1 ) ( g 1 ) i 2 i = 1 g 2 2 1 i + 1 + 1 g i + 2 g 2 + 2 g 2 + + 2 g 2 g 2 2 = i = 1 g 2 2 2 ( g 1 ) ( i + 1 ) ( g 1 ) i 2 1 i + 1 1 g i 2 g 2 .
By Lemma 4, the function
F ( x ) = 2 ( g 1 ) ( x + 1 ) ( g 1 ) x 2 1 x + 1 1 g x 2 g 2
is a monotonically decreasing function on [ 1 , g 1 2 ] . Thus, when x = g 2 2 , F ( x ) get the minimum value
F ( g 2 2 ) = 8 ( g 1 ) g 2 + 2 g 4 2 g 2 2 g 2 g + 2 ,
since
g , F ( g 2 2 ) 2 g > 0 .
Actually, by simple calculation, we have F ( x ) > 0 when g 8 , it follows that Θ 2 > 0 when g 8 .
(ii) Using the same argument as Equation (1), we can check that if g is an odd integer, Θ 2 > 0 for all g 8 .
Comparing Θ 1 and Θ 2 , it is easy to see that
R H ( S n g ) < R H ( S n g 1 ) ,
since g 8 . For g = 3 , 4 , , 7 , we calculate R H ( S n 3 ) , R H ( S n 4 ) , R H ( S n 5 ) , R H ( S n 6 ) , R H ( S n 7 ) and compare the values. We have
R H ( S n 3 ) = R H ( C 3 ) + x , y V ( G C 3 ) 1 r ( x , y ) + x V ( G ) , y V ( C 3 ) 1 r ( x , y ) = 1 20 ( 5 n 2 + 9 n + 18 ) , R H ( S n 4 ) = R H ( C 4 ) + x , y V ( G C 4 ) 1 r ( x , y ) + x V ( G ) , y V ( C 4 ) 1 r ( x , y ) = 1 84 ( 21 n 2 + 33 n + 148 ) .
Then,
R H ( S n 3 ) R H ( S n 4 ) = 2 n 35 181 210 .
Thus, we can get
R H ( S n 3 ) < R H ( S n 4 ) i f n 15 , R H ( S n 3 ) > R H ( S n 4 ) i f n 16 .
In a similar way, by calculating R H ( S n 5 ) , R H ( S n 6 ) , and R H ( S n 7 ) and comparing the values, we can get the following set of inequalities,
R H ( S n 3 ) < R H ( S n 4 ) < < R H ( S n n 1 ) < R H ( S n n ) i f n 7 , R H ( S n 3 ) < R H ( S n 7 ) < R H ( S n 6 ) < R H ( S n 4 ) < R H ( S n 5 ) i f n = 8 , R H ( S n 7 ) < R H ( S n 3 ) < R H ( S n 6 ) < R H ( S n 5 ) < R H ( S n 4 ) i f n = 9 , R H ( S n 7 ) < R H ( S n 6 ) < R H ( S n 3 ) < R H ( S n 5 ) < R H ( S n 4 ) i f n = 10 , R H ( S n 7 ) < R H ( S n 6 ) < R H ( S n 5 ) < R H ( S n 3 ) < R H ( S n 4 ) i f 11 n 15 , R H ( S n 7 ) < R H ( S n 6 ) < R H ( S n 5 ) < R H ( S n 4 ) < R H ( S n 3 ) i f n 16 .
To sum up, we can get R H ( C n ) has the largest value 343 20 if n 7 , R H ( S n 5 ) has the largest value 923 44 if n = 8 , R H ( S n 4 ) has the largest value 1 84 ( 21 n 2 + 33 n + 148 ) if 9 n 15 and R H ( S n 3 ) has the largest value 1 20 ( 5 n 2 + 9 n + 18 ) if n 16 . The proof is completed. □
In [23], the unique element of U ( n ) with the largest Resistance-Harary index is determined. Herewith, we point out some minor errors in [23]. These do not affect the validity of the final result of [23] but deserve to be corrected. We list the error as follows and we give a counterexample.
Theorem 2.
[23] Let G U ( n ) , then we have R H ( G ) 1 20 ( n 2 + 9 n + 18 ) with equality holding if and only if G S n 3 for n 9 and G C n for n 8 .
Counterexample
If n = 9 , according to Theorem 2 in [23], the largest Resistance-Harary index is H R ( S 9 3 ) = 25.2000 . However, H R ( S 9 4 ) = 25.5476 , H R ( S 9 3 ) < H R ( S 9 4 ) , is a contradiction. If n = 8 , according to the Theorem 2 in [23], the result is R H ( C 8 ) = 20.7429 has the largest value. However, R H ( S 8 5 ) = 20.9773 , R H ( C 8 ) < R H ( S 8 5 ) , is a contradiction. Actually, according to our Theorem 1, we have H R ( S 9 4 ) > H R ( S 9 3 ) if n = 9 and R H ( S 8 5 ) > R H ( C 8 ) if n = 8 . Obviously, the result is consistent with our theorem.

3.2. The Second-Maximum Resistance-Harary Index

By Lemmas 2 and 3 and Equation (2) of the proof of Theorem 1, we can conclude that for n 16 G which has the second-largest Resistance-Harary index in U ( n ) and those must be one of the graphs G 1 , G 2 , and G 3 ( S n 4 ) , as shown in Figure 3.
Theorem 3.
If G U ( n ) , let m a x * { R H ( G ) } denote the second-largest Resistance-Harary index of graph G, then
m a x * G U ( n ) { R H ( G ) } = R H ( S n n 1 ) i f n 7 , R H ( S n 4 ) i f n = 8 , R H ( S n 5 ) i f n = 9 , 10 , R H ( S n 3 ) i f 11 n 15 , R H ( S n 4 ) i f n 16 .
Proof. 
(i) For n 16 .
Case 1. Let H 1 be the common subgraph of S n 3 and G 1 . Thus, we can view graphs S n 3 and G 1 as the graphs depicted in Figure 4.
Then, we have
R H ( S n 3 ) = R H ( H 1 ) + 1 2 + 2 x H 1 1 1 + r ( x , v 1 ) = R H ( H 1 ) + 1 2 + 2 1 + n 5 2 + 6 5 , R H ( G 3 ) = R H ( H 1 ) + 1 + x H 2 1 1 + r ( x , v 1 ) + x H 2 1 2 + r ( x , v 1 ) = R H ( H 1 ) + 1 + 1 + 6 5 + n 5 2 + 1 2 + n 5 3 + 3 4 .
Therefore, we can get the difference
R H ( S n 3 ) R H ( G 1 ) = n 6 23 60 .
Case 2. Let H 2 be the common subgraph of S n 3 and G 2 . Thus, we can view graphs S n 3 and G 2 as the graphs depicted in Figure 5.
Then, we have
R H ( S n 3 ) = R H ( S n 3 ) + x H 2 1 1 + r ( x , v 1 ) = R H ( H 2 ) + 1 + n 4 2 + 6 5 , R H ( G 2 ) = R H ( H 2 ) + x H 2 1 1 + r ( x , v 2 ) = R H ( H 2 ) + 1 + 3 ( n 4 ) 8 + 6 5 .
Therefore, we can get the difference
R H ( S n 3 ) R H ( G 2 ) = n 8 1 2 .
Case 3. Let H 3 be the common subgraph of S n 3 and G 3 . Thus, we can view graphs S n 3 and G 3 as the graphs depicted in Figure 6.
Then, we have
R H ( S n 3 ) = R H ( H 3 ) + R H ( S 4 3 ) + x H 3 , y S 4 3 1 r ( x , y ) = R H ( H 3 ) + 67 10 + 27 ( n 4 ) 10 , R H ( G 3 ) = R H ( H 3 ) + R H ( C 4 ) + x H 3 , y C 4 1 r ( x , y ) = R H ( H 3 ) + 22 3 + 37 ( n 4 ) 14 .
Therefore, we can get the difference
R H ( S n 3 ) R H ( G 3 ) = 2 n 35 181 210 .
By the above expressions for the Resistance-Harary index of G 1 , G 2 and G 3 , we immediately have the desired result.
(ii) For 9 n 15 .
By the same arguments as used in (i), we conclude that the possible candidates having the second-largest Resistance-Harary index must be one of the graphs G 4 , G 5 , G 6 , G 7 ( S n 5 ) (as shown in Figure 7) and S n 3 .
Let H 4 , H 5 , H 6 denote the common subgraphs of S n 4 and G 4 G 7 , respectively. Thus, we can view graphs G 4 G 7 as the graphs depicted in Figure 8.
Then, in a similar way, we have
R H ( S n 4 ) R H ( G 4 ) = n 6 193 462 , R H ( S n 4 ) R H ( G 5 ) = 3 n 22 15 22 , R H ( S n 4 ) R H ( G 6 ) = n 6 5 6 , R H ( S n 4 ) R H ( G 7 ) = 85 n 693 2921 2772 .
Therefore, we have H R ( G 4 ) < H R ( G 6 ) < H R ( G 5 ) < H R ( G 7 ) . In connection with Equation (2), we have G 7 ( S n 5 ) < S n 3 if 11 n 15 , so for 11 n 15 , S n 3 is the second largest. For n = 9 , 10 , in connection with Equation (2), we have S n 5 is the second largest.
(iii) For n 7 and n = 8 .
In connection with Equation (2), we have S n n 1 , S n 4 is the second largest, respectively.
The result follows. □

4. Application

Now, we give a specific application of formation mentioned in the Section 3. Fully loaded graphs as a special class of unicyclic graphs also have some special properties about Resistance-Harary index. In this section, we determine the largest Resistance-Harary index among all fully loaded unicyclic graphs.
By a sequence of α and β transformations to a fully loaded graph G, we can obtain a new graph, denoted by Q n l , which is obtained by attaching a pendent edge to each vertex of the unique cycle C l and attaching n 2 l + 1 pendent edges to a vertex of C l . Then, by Lemma 2 and Corollary 1, we arrive at,
Theorem 4.
G U ( n ; g ) , then R H ( G ) R H ( Q n g ) .
Next, we determine the graph in U ( n ) with the largest Resistance-Harary index.
Theorem 5.
If G U ( n ) , then
max G U ( n ) { R H ( G ) } = R H ( Q n 3 ) i f n 7 , R H ( Q n 4 ) i f n = 8 , 9 , R H ( Q n 3 ) i f n 10 .
Proof. 
Using a similar way as in Section 3.2, we can conclude that the unicyclic graphs with n 16 in Figure 9 have the second largest or third largest Resistance-Harary index.
Only one graph Q n 3 is fully loaded (Graph 9 in Figure 9). Thus, Q n 3 has the largest Resistance-Harary index among all fully loaded graphs with n 16 . For n 15 , from Lemmas 2 and 3 we can conclude that the fully loaded graph with largest Resistance-Harary index must be one of the five situations Q n 3 , Q n 4 , Q n 5 , Q n 6 , Q n 7 .
For completeness of the proof, we list all possible values as follows. For n 7 , there is only one situation Q 7 3 with n = 7 and Q 6 3 with n = 6 , so we begin at n = 8 .
Case 1. n = 8 .
R H ( Q 8 3 ) = 19.625 , R H ( Q 8 4 ) = 20.026 .
Then, max G U ( n ) { R H ( G ) } = R H ( Q 8 4 ) .
Case 2. n = 9 .
R H ( Q 9 3 ) = 24.075 , R H ( Q 9 4 ) = 24.229 .
Then, max G U ( n ) { R H ( G ) } = R H ( Q 9 4 ) .
Case 3. n = 10 .
R H ( Q 10 3 ) = 29.025 , R H ( Q 10 4 ) = 28.933 , R H ( Q 10 5 ) = 28.866 .
Then, max G U ( n ) { R H ( G ) } = R H ( Q 10 3 ) .
Case 4. n = 11 .
R H ( Q 11 3 ) = 34.475 , R H ( Q 11 4 ) = 34.136 , R H ( Q 11 5 ) = 33.725 .
Then, max G U ( n ) { R H ( G ) } = R H ( Q 11 3 ) .
Case 5. n = 12 .
R H ( Q 12 3 ) = 40.425 , R H ( Q 12 4 ) = 39.840 , R H ( Q 12 5 ) = 39.085 , R H ( Q 12 6 ) = 38.563 .
Then, max G U ( n ) { R H ( G ) } = R H ( Q 12 3 ) .
Case 6. n = 13 .
R H ( Q 13 3 ) = 46.875 , R H ( Q 13 4 ) = 46.043 , R H ( Q 13 5 ) = 44.944 , R H ( Q 13 6 ) = 44.003 .
Then, max G U ( n ) { R H ( G ) } = R H ( Q 13 3 ) .
Case 7. n = 14 .
R H ( Q 14 3 ) = 53.825 , R H ( Q 14 4 ) = 52.747 , R H ( Q 14 5 ) = 51.303 , R H ( Q 14 6 ) = 49.942 , R H ( Q 14 7 ) = 49.987 .
Then, max G U ( n ) { R H ( G ) } = R H ( Q 14 3 ) .
Case 8. n = 15 .
R H ( Q 15 3 ) = 61.275 , R H ( Q 15 4 ) = 59.950 , R H ( Q 15 5 ) = 58.163 , R H ( Q 15 6 ) = 56.382 , R H ( Q 15 7 ) = 54.946 .
Then, max G U ( n ) { R H ( G ) } = R H ( Q 15 3 ) .
The proof is completed. □

5. Conclusions

This paper focuses on Resistance-Harary index in unicyclic graphs. Let U ( n ) and U ( n ) be the set of unicyclic graphs and fully loaded unicyclic graphs, respectively. Here, we first give a more precise proof about the largest Resistance-Harary index among all unicyclic graphs, then determine the graph of U ( n ) with second-largest Resistance-Harary index and apply this way to fully loaded unicyclic graphs determine the graph of U ( n ) with largest Resistance-Harary index.

Author Contributions

Funding Acquisition, J.L. and J.-B.L.; Methodology, J.-B.L. and J.L.; Supervision, X.-F.P.; and Writing—Original Draft, J.L. and S.-B.C. All authors read and approved the final manuscript.

Funding

This work was supported by the National Natural Science Foundation of China (11601006), the China Postdoctoral Science Foundation (2017M621579), the Postdoctoral Science Foundation of Jiangsu Province (1701081B), the Project of Anhui Jianzhu University (2016QD116 and 2017dc03) and the Anhui Province Key Laboratory of Intelligent Building and Building Energy Saving.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Klein, D.J.; Randić, M. Resistance distance. J. Math. Chem. 1993, 12, 81–95. [Google Scholar] [CrossRef]
  2. Du, J.; Su, G.; Tu, J.; Gutman, I. The degree resistance distance of cacti. Discrete Appl. Math. 2015, 188, 16–24. [Google Scholar] [CrossRef]
  3. Bu, C.; Yan, B.; Zhou, X.; Zhou, J. Resistance distance in subdivision-vertex join and subdivision-edge join of graphs. Linear Algebra Appl. 2014, 458, 454–462. [Google Scholar] [CrossRef]
  4. Feng, L.H.; Yu, G.; Liu, W. Futher results regarding the degree Kirchhoff index of a graph. Miskolc Math. Notes 2014, 15, 97–108. [Google Scholar] [CrossRef]
  5. Feng, L.H.; Yu, G.; Xu, K.; Jiang, Z. A note on the Kirchhoff index of bicyclic graphs. ARS Combin. 2014, 114, 33–40. [Google Scholar]
  6. Gao, X.; Luo, Y.; Liu, W. Resistance distance and the Kirchhoff index in Cayley graphs. Discrete Appl. Math. 2011, 159, 2050–2057. [Google Scholar] [CrossRef]
  7. Gao, X.; Luo, Y.; Liu, W. Kirchhoff index in line, subdivision and total graphs of a regular graph. Discrete Appl. Math. 2012, 160, 560–565. [Google Scholar] [CrossRef] [Green Version]
  8. Gutman, I.; Feng, L.; Yu, G. On the degree resistance distance of unicyclic graphs. Trans. Comb. 2012, 1, 27–40. [Google Scholar]
  9. Liu, J.; Pan, X.; Yu, L.; Li, D. Complete characterization of bicyclic graphs with minimal Kirchhoff index. Discrete Appl. Math. 2016, 200, 95–107. [Google Scholar] [CrossRef]
  10. Liu, J.; Wang, W.; Zhang, Y.; Pan, X. On degree resistance distance of cacti. Discrete Appl. Math. 2016, 203, 217–225. [Google Scholar] [CrossRef]
  11. Liu, J.; Pan, X. Minimizing Kirchhoff index among graphs with a given vertex bipartiteness. Appl. Math. Comput. 2016, 291, 84–88. [Google Scholar] [CrossRef]
  12. Pirzada, S.; Ganie, H.A.; Gutman, I. On Laplacian-Energy-Like Invariant and Kirchhoff Index. MATCH Commun. Math. Comput. Chem. 2015, 73, 41–59. [Google Scholar]
  13. Plavšić, D.; Nikolić, S.; Mihalić, Z. On the Harary index for the characterization of chemical graphs. J. Math. Chem. 1993, 12, 235–250. [Google Scholar]
  14. Ivanciuc, O.; Balaban, T.S.; Balaban, A.T. Reciprocal distance matrix, related local vertex invariants and topological indices. J. Math. Chem. 1993, 12, 309–318. [Google Scholar] [CrossRef]
  15. Furtula, B.; Gutman, I.; Katanić, V. Three-center Harary index and its applications. Iranian J. Math. Chem. 2016, 7, 61–68. [Google Scholar]
  16. Feng, L.H.; Lan, Y.; Liu, W.; Wang, X. Minimal Harary index of graphs with samll parameters. MATCH Commun. Math. Comput. Chem. 2016, 76, 23–42. [Google Scholar]
  17. Li, X.; Fan, Y. The connectivity and the Harary index of a graph. Discrete Appl. Math. 2015, 181, 167–173. [Google Scholar] [CrossRef] [Green Version]
  18. Xu, K.; Das, K.C. On Harary index of graphs. Discrete Appl. Math. 2011, 159, 1631–1640. [Google Scholar] [CrossRef] [Green Version]
  19. Xu, K. Trees with the seven smallest and eight greatest Harary indices. Discrete Appl. Math. 2012, 160, 321–331. [Google Scholar] [CrossRef] [Green Version]
  20. Xu, K.; Das, K.C. Extremal unicyclic and bicyclic graphs with respect to Harary index. Bull. Malays. Math. Sci. Soc. 2013, 36, 373–383. [Google Scholar]
  21. Xu, K.; Wang, J.; Das, K.C.; Klavžar, S. Weighted Harary indices of apex trees and k-apex trees. Discrete Appl. Math. 2015, 189, 30–40. [Google Scholar] [CrossRef]
  22. Yu, G.; Feng, L. On the maximal Harary index of a class of bicyclic graphs. Util. Math. 2010, 82, 285–292. [Google Scholar]
  23. Chen, S.B.; Guo, Z.J.; Zeng, T.; Yang, L. On the Resistance-Harary index of unicyclic graphs. MATCH Commun. Math. Comput. Chem. 2017, 78, 189–198. [Google Scholar]
  24. Wang, H.; Hua, H.; Zhang, L.; Wen, S. On the Resistance-Harary Index of Graphs Given Cut Edges. J. Chem. 2017, 2017, 3531746. [Google Scholar] [CrossRef]
Figure 1. The σ -transform at v.
Figure 1. The σ -transform at v.
Mathematics 07 00201 g001
Figure 2. The β -transform.
Figure 2. The β -transform.
Mathematics 07 00201 g002
Figure 3. The graphs G 1 , G 2 and G 3 .
Figure 3. The graphs G 1 , G 2 and G 3 .
Mathematics 07 00201 g003
Figure 4. The graphs S n 3 and G 1 .
Figure 4. The graphs S n 3 and G 1 .
Mathematics 07 00201 g004
Figure 5. The graphs S n 3 and G 2 .
Figure 5. The graphs S n 3 and G 2 .
Mathematics 07 00201 g005
Figure 6. The graphs S n 3 and G 3 .
Figure 6. The graphs S n 3 and G 3 .
Mathematics 07 00201 g006
Figure 7. The graphs G 4 G 7 .
Figure 7. The graphs G 4 G 7 .
Mathematics 07 00201 g007
Figure 8. The graphs G 4 G 7 .
Figure 8. The graphs G 4 G 7 .
Mathematics 07 00201 g008
Figure 9. The graphs with second maximal or third maximal Resistance-Harary index.
Figure 9. The graphs with second maximal or third maximal Resistance-Harary index.
Mathematics 07 00201 g009

Share and Cite

MDPI and ACS Style

Lu, J.; Chen, S.-B.; Liu, J.-B.; Pan, X.-F.; Ji, Y.-J. Further Results on the Resistance-Harary Index of Unicyclic Graphs. Mathematics 2019, 7, 201. https://doi.org/10.3390/math7020201

AMA Style

Lu J, Chen S-B, Liu J-B, Pan X-F, Ji Y-J. Further Results on the Resistance-Harary Index of Unicyclic Graphs. Mathematics. 2019; 7(2):201. https://doi.org/10.3390/math7020201

Chicago/Turabian Style

Lu, Jian, Shu-Bo Chen, Jia-Bao Liu, Xiang-Feng Pan, and Ying-Jie Ji. 2019. "Further Results on the Resistance-Harary Index of Unicyclic Graphs" Mathematics 7, no. 2: 201. https://doi.org/10.3390/math7020201

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop