MATCH MATCH Commun. Math. Comput. Chem.

55 (2006) 447-452
Communications in Mathematical
and in Computer Chemistry ISSN 0340 - 6253

PI Index of Zig-Zag Polyhex Nanotubes

Ali Reza Ashrafi• and Amir Loghman

Department of Mathematics, Faculty of Science,

University of Kashan, Kashan, Iran

(Received June 7, 2005)

Abstract
The Padmakar–Ivan (PI) index of a graph G is defined as PI(G) = ∑[neu(e|G)+ nev(e|G)],
where neu(e|G) is the number of edges of G lying closer to u than to v, nev(e|G) is the
number of edges of G lying closer to v than to u and summation goes over all edges of G.
The PI Index is a Szeged-like topological index developed very recently. In this paper an
exact expression for PI index of the zig-zag polyhex nanotubes is given.

1. Introduction

Graph theory was successfully provided the chemist with a variety of very
useful tools, namely, the topological index. A topological index is a numeric quantity
from the structural graph of a molecule.
The Wiener index (W) is the oldest topological indices. Numerous of its
chemical applications were reported and its mathematical properties are well understood
[1-5]. We encourage the reader to consult [6], for a good survey on the topic.
In Refs. [7,8], the authors defined a new topological index and named it
Padmakar-Ivan index. They abbreviated this new topological index as PI. This newly
proposed topological index, PI, does not coincide with the Wiener index (W) for acyclic

•
Author to whom correspondence should be addressed. (E-mail: Ashrafi@kashanu.ac.ir)
 - 448 -

(trees) molecules. The derived PI index is very simple to calculate and has a
discriminating power similar to that of the W index, for details see [9-11].
We now recall some algebraic definitions that will be used in the paper. Let G be
a simple molecular graph without directed and multiple edges and without loops, the
vertex and edge-shapes of which are represented by V(G) and E(G), respectively. If e is
an edge of G, connecting the vertices u and v then we write e=uv. The number of
vertices of G is denoted by n. The distance between a pair of vertices u and w of G is
denoted by dG(u,w). We define for e=uv two quantities neu(e|G) and nev(e|G). neu(e|G) is
the number of edges lying closer to the vertex u than the vertex v, and nev(e|G) is the
number of edges lying closer to the vertex v than the vertex u. Then the Padmakar–Ivan
(PI) index of a graph G is defined as PI(G) = ∑[neu(e|G)+ nev(e|G)]. We notice that the
edges equidistant from both ends of the edge uv are not counted in calculating the PI
index of a graph. In fact, if Gu,e = {x | dG(u,x) < dG(v,x)}, Gv,e = { x | dG(u,x) > dG(v,x)}
and Ge = {x | dG-{e}(u,x) - dG-{e}(v,x) = ±1} then neu(e|G) = |E(Gu,e)|, nev(e|G) = |E(Gv,e)|
and N(e) = |E(Ge)|. Here for any subset U of the vertex set V = V(G), |E(U)| denotes the
number of edges of G between the vertices of U. It is easy to see that |E(G)| = N(e) +
neu(e|G) + nev(e|G).
In a series of papers, Diudea and coauthors [12-18] computed the Wiener index
of some nanotubes. In this paper an exact expression for PI index of zig-zag polyhex
nanotubes is given. Our notation is standard and mainly taken from [12-14] and [19,20].
Throughout this paper T = TUHC6[2p,q] denotes an arbitrary zig-zag polyhex nanotube,
in the terms of their circumference (2p) and their length (q).see Figure 1.

2. PI Index of TUHC6[2p,q]

In this section, the PI index of the graph T = TUHC6[2p,q] were computed. To

do this, we assume that E = E(T) is the set of all edges of T and N(e) = |E| - (neu(e|G) +
nev(e|G)). Then PI(T) = |E|2 - ∑e∈E N(e). But |E(T)| = p(3q-1) and so PI(T) = p2(3q-1)2 -
∑e∈E N(e). Therefore, for computing the PI index of T, it is enough to calculate N(e),
for every e ∈ E. To calculate N(e), we consider two cases that e is horizontal or non-
horizontal.
 - 449 -

Figure 1: Zig-zag TUHC6[20,q] ( The Figure 2: A Zig-Zag Polyhex Lattice

figure is taken from [17]) with p=5 and q=7

Lemma 1. If e is an horizontal edge then N(e) = p.

Proof. Suppose e = UijUi(j+1) denotes an arbitrary horizontal edge of ith row of the zig-
zag polyhex lattice of TUHC6[2p,q], Figure 2. It is obvious that for every k, 0 ≤ k ≤ p-1,
U(i+2k)jU(i+2k)(j+1) is an horizontal edge parallel to e and dT(Uij,U(i+2k)j) =
dT(Ui(j+1),U(i+2k)(j+1)) = 2k. Thus {Uij,U(i+2)j, …, U(i+2p-2)j,Ui(j+1),…,U(i+2p-2)(j+1)} ⊆ Te. We
now prove the equality of two sets. To do this, we assume that UklU(k+1)l is an arbitrary
non-horizontal edge of T. If l ≤ j then dT(Ukl,Uij) < dT(Ukl,U(i+1)j) and dT(U(k+1)l,Uij) <
dT(U(k+1)l,U(i+1)j) and so Ukl ∉ Te. In other case U(k+1)l ∉ Te. A similar argument shows
that every horizontal edge of Te must be parallel to e. Thus {Uij,U(i+2)j, …, U(i+2p-
2)j,Ui(j+1),…,U(i+2p-2)(j+1)} = Te. Therefore, N(e) = p, proving the lemma.

Lemma 2. If e is a non-horizontal edge in the kth column, 1≤ k ≤ p, of the zig-zag

2 p + 2(k − 1) q ≥ p + k − 1
polyhex lattice of T = TUHC6[2p,q], then N(e) =  .
 2q q ≤ p + k −1

Proof. Let Eij denote the non-horizontal edge of T in the ith row and jth column. We first
notice that for every j, 1≤ j ≤ q, N(E1j) = N(E2j) = ⋅⋅⋅ = N(E(2p)j). So it is enough to
calculate N(E11), N(E12), ⋅⋅⋅, N(E1q). Compute the value of N(E11). Suppose q ≥ p. We
 - 450 -

consider the edges E(p+1)1, E(p+1)2, ⋅⋅⋅, E(p+1)p. If 1 ≤ t ≤ p then E(p+1)t = U(p+1)tU(p+2)t and
we have dT(U(p+1)t,U21) = dT(U(p+2)t,U11) = p+t-2. So E(p+1)t ∈ E( TE11 ), 1 ≤ t ≤ p.

Similarly, for 0 ≤ i ≤ p-2, E(2p-i)(i+2) ∈ E( TE11 ) and E( TE11 ) ⊆ {E(p+1)1, E(p+1)2, ⋅⋅⋅, E(p+1)p,

E11, E(2p)2, ⋅⋅⋅, E(p+2)p}. To prove the equality, we assume that UklU(k+1)l is an arbitrary
non-horizontal edge of T. If l ≥ p+1 then dT(Ukl,U21) < dT(Ukl,U11) and dT(U(k+1)l,U21) <
dT(U(k+1)l,U11) and so UklU(k+1)l ∉ E( TE11 ). If l ≤ p then we have exactly two edges in

every column belong to {E(p+1)1, E(p+1)2, ⋅⋅⋅, E(p+1)p, E11, E(2p)2, ⋅⋅⋅, E(p+2)p} and other edges
of this column don't belong to E( TE11 ). Therefore E( TE11 ) = {E(p+1)1, E(p+1)2, ⋅⋅⋅, E(p+1)p,

E11, E(2p)2, ⋅⋅⋅, E(p+2)p}. If q ≤ p by above calculations E( TE11 ) = {E(p+1)1, E(p+1)2, ⋅⋅⋅, E(p+1)q,

E11, E(2p)2, ⋅⋅⋅, E(2p+2-q)q}. We continue our argument by considering the edge E12. To
prove this case, we delete the first column of the zig-zag polyhex lattice and obtain a
TUHC6[2p,q-1]. Since E12 is the (1,1) entry of this lattice, we have
 2p q −1 ≥ p
N(E12) = R +  ,
2 q − 2 q − 1 ≤ p
where R is the number of edges E( TE12 ) in the first column of TUHC6[2p,q]. On the

other hand, E(p+1)1 and E(2p)1 are only edges of TUHC6[2p,q] in the first column.
Therefore,
2 p + 2 q ≥ p + 1
N(E12) =  .
 2q q ≤ p +1

We can continue this method for computing N(E13), ⋅⋅⋅, N(E1p) to complete the proof.

Lemma 3. If q ≤ 2p then N(E11) = N(E1q), N(E12) = N(E1(q-1)), ⋅⋅⋅, N(E1s) = N(E1(s+1+b)),

where s = [q/2], the greatest integer less than or equal to q/2, and b = [(q+1)/2] – [q/2].

Proof. Since the zig-zag polyhex lattice is symmetric, the proof is straightforward.

Lemma 4. If q > 2p then N(E11) = N(E1q), N(E12) = N(E1(q-1)), ⋅⋅⋅, N(E1p) = N(E1(q-p+1)),
and N(E1(p+1)) = N(E1(p+2)) = ⋅⋅⋅ = N(E1(q-p)) = N(E1p).

Proof. The first part of the lemma is a conclusion of this fact that the zig-zag polyhex
lattice is symmetric. To prove the second part, we notice that for a fixed column j there
 - 451 -

are exactly 2p-1 columns with two edges belongs to E( TE1j ). The other columns don't

intersect E( TE1j ). Thus N(E1(p+1)) = N(E1(p+2)) = ⋅⋅⋅ = N(E1(q-p)) = N(E1p).

We now ready to state the main result of the paper. We have:

Theorem. The PI index of the zig-zag polyhex nanotube is as follows:

 p 2 ( 9q 2 − 7q + 2 ) − 4 pq 2 if q≤ p

PI(TUHC6[2p,q]) =  .
 p 2 ( 9q 2 − 15q + 4 p − 2 ) + 4 pq if q≥ p

Proof. Since PI(T) = |E|2 - ∑e∈E N(e), it is enough to compute ∑e∈E N(e). Suppose X

and Y are the set of all horizontal and non-horizontal edges of T. Then

PI(T) = |E|2 – ∑e∈X N(e) – ∑e∈Y N(e)

 4 p ∑ip=1 N ( E1i ) + 2 p ( q − 2 p ) N ( E1 p ) if q ≥ 2p

= p (9q –7q+2)– 4 p ∑iq=−1p +1 N ( E1i ) + 2 p ( 2 p − q − 2) N ( E1( q − p +1) ) if
2 2
p < q < 2p .
 2 pqN ( E11 ) if q≤ p


By Lemma 2, N(E1i) = N(E11) + 2(i-1) and so we have:

2 pqN ( E ) + 4 p( p − 1 )( q − p ) if q ≥ 2p
 11
PI(T) = p2(9q2–7q+2)– 2 pqN ( E11 ) + 4 p( p − 1 )( q − p ) if p < q < 2p
 2 pqN ( E11 ) if q≤ p


 4 pq 2 if q≤ p
2 2 
= p (9q –7q+2)–  ,
4 p( 2 pq − p 2 − q + p ) if q≥ p


which completes the proof.

Acknowledgment. We are greatly indebted to the referees, whose valuable criticisms

and suggestions led us to rearrange the paper.
 - 452 -

REFERENCES

1. H. Wiener, J. Am. Chem. Soc. 69 (1947) 17.

2. R. Todeschini and V. Consonni, Handbook of Molecular Descriptors, Wiley,
Weinheim, 2000.
3. D.E. Needham, I.C. Wei and P.G. Seybold, J. Am. Chem. Soc. 110 (1988) 4186.
4. G. Rucker and C. Rucker, J. Chem. Inf. Comput. Sci. 39 (1999) 788.
5. A.A. Dobrynin, R. Entringer and I. Gutman, Acta Appl. Math. 66 (2001) 211.
6. A.A. Dobrynin, I. Gutman, S. Klavzar and P. Zigert., Acta Appl. Math. 72 (2002)
247.
7. P. V. Khadikar, Nat. Acad. Sci. Lett. 23 (2000) 113.
8. P.V. Khadikar; S. Karmarkar and V.K. Agrawal, Nat. Acad. Sci. Lett. 23 (2000) 124.
9. P. V. Khadikar; P.P. Kale; N.V. Deshpande; S. Karmarkar and V.K. Agrawal, J.
Math. Chem. 29 (2001) 143.
10. P.V. Khadikar; S. Karmarkar, J. Chem. Inf. Comput. Sci. 41 (2001) 934.
11. P.V. Khadikar, S. Karmarkar and R.G. Varma, Acta Chim. Slov. 49 (2002) 755.
12. M.V. Diudea and A. Graovac, MATCH Commun. Math. Comput. Chem. 44 (2001)
93.
13. M.V. Diudea, I. Silaghi-Dumitrescu and B. Parv, MATCH Commun. Math. Comput.
Chem. 44 (2001) 117.
14. M.V. Diudea and P.E. John, MATCH Commun. Math. Comput. Chem. 44 (2001)
103.
15. M. V. Diudea, Bull. Chem. Soc. Jpn. 75 (2002) 487.
16. M. V. Diudea, MATCH Commun. Math. Comput. Chem. 45 (2002) 109.
17. P. E. John and M. V. Diudea, Croat. Chem. Acta, 77 (2004) 127.
18. M.V. Diudea, M. Stefu, B. Parv and P.E. John, Croat. Chem. Acta, 77 (2004) 111.
19. P.J. Cameron, Combinatorics: Topics, Techniques, Algorithms, Cambridge
University Press, Cambridge, 1994.
20. N. Trinajstic, Chemical Graph Theory, CRC Press, Boca Raton, FL. 1992.