The eccentric connectivity index ECI is a chemical structure descriptor that is currently being used for the modeling of biological activities of a chemical compound. This index has been proved to provide a high degree of predictability as compared to some other well-known indices in case of anticonvulsant, anti-inflammatory, and diuretic activities. The ECI of an infinite class of 1-polyacenic (phenylenic) nanotubes has been recently studied. In this article, we extend this study to generalized polyacenic nanotubes and find ECI of t-polyacenic nanotubes for t>1.
Higher Education Commission, Pakistan5331/Federal/NRPU/R&D/HEC/2016China Postdoctoral Science Foundation2017M621579Postdoctoral Science Foundation of Jiangsu Province1701081BProject of Anhui Jianzhu University2016QD1162017dc031. Introduction
A basic concept of chemistry is that the properties/activities of a molecule depend upon its structural characteristics. Molecular graphs can be used to model the chemical structures of molecules and molecular compounds, by considering atoms as vertices and the chemical bonds between the atoms as edges. In the study of quantitative structure-property and structure-activity relationships (QSPR/QSAR), the topological indices are very helpful in detecting the biological activities of a chemical compound [1–4].
A topological index is a numerical graph invariant that is used to correlate the chemical structure of a molecule with its physicochemical properties and biological activities. Generally, topological indices are classified into five generations: first-generation topological indices are integer numbers obtained by simple operations from local vertex invariants involving only one vertex at a time. Some of the famous topological indices of this class are Wiener index, Hosoya index, and Centric indices of Balaban [5]. Second-generation topological indices are real numbers based on integer graph properties. These indices were obtained via structural operations from integer local vertex invariants, involving more than one vertex at a time. Some examples of this class include molecular connectivity indices, Balaban J index, bond connectivity indices, and kappa shape indices [5]. Third-generation topological indices are real numbers which are based on local properties of the molecular graph. These indices are of recent introduction and have very low degeneracy. These are based on information theory applied to the terms of distance sums or on newly introduced nonsymmetrical matrices. Some examples include information indices [6], the hyper-Wiener index [5], the Kirchhoff index [7], and electrotopological state indices [2]. Recently, fourth- and fifth-generation topological indices are placed as new generations topological indices. Fourth-generation topological indices are of highly discriminating power, i.e., ≥100. The examples of fourth-generation topological indices include eccentric connectivity index [8], superaugmented eccentric connectivity index [9], and superaugmented eccentric connectivity topochemical indices [10]. Detour matrix-based adjacent path eccentric distance sum indices [11] belong to the fifth-generation topological indices.
Let G be a connected molecular graph with vertex set VG and edge EG. Let IG be the set of those edges of G that are incident to a vertex k∈VG, and then the degree of k is denoted by dk and is defined as the cardinality of IG. The distance from a vertex k∈VG to a vertex l∈VG is denoted by dk,l and is defined as the minimum number of edges lying between them. The eccentricity εk of a given vertex k∈VG is defined as the largest distance between k and any vertex l of G.
Sharma et al. in [8] have presented a distance-based chemical structure descriptor, called the eccentric connectivity index (ECI), which is presented as(1)ξcG=∑k∈VGdk⋅εk.
It is recorded in [12–16] that ECI provides good correlations with regard to physicochemical properties and biological activities. This index is reported as a highly discriminating descriptor for QSPR/QSAR studies [8, 9, 17]. The degree of prediction of ECI is better than the Wiener index in case of diuretic activity [18] and anti-inflammatory activity in [19]. Also, this index has been proved to provide a high degree of predictability with regard to anticonvulsant activity [20] in comparison to Zagreb indices. Recently, the eccentric connectivity index has been studied for certain nanotubes [21–26] and for several molecular graphs [27–29].
Polyacenes relate to a family of polycyclic aromatic hydrocarbon (PAH) compounds which are formed by the linearly fused benzene rings. Numerous molecules of this class have interesting optical, thermodynamic, electronic, ferromagnetic, and photoconductive properties [30–33]. In the first organic solid-state injection laser, the lasing was discovered by using the single crystals of tetracene [34, 35]. They have application in rechargeable Li-ion batteries [36] and also have presence in various celestial objects like planetary nebulae [37]. In this sense, the polyacenes have received much attention. The PI index of linear polyacenes has been studied in [38]. The molecular graphs of certain linear polyacene molecules are given in Figure 1.
Recently, the Zagreb indices of 3-polyacenic (anthracenic), 4-polyacenic (tetracenic), and 5-polyacenic (pentacenic) nanotubes have been studied in [39–41], respectively. The ECI of 1-polyacenic (phenylenic) nanotubes has been presented in [25]. In this paper, we generalize these results to t-polyacenic nanotubes for t>1 and present the ECI for these nanotubes.
2. Main Results
The generalized molecular graph of the t-polyacenic nanotube Ptp,q is shown in Figure 2. In this graphical representation, q counts the number of polyacene units in a row and p counts the number of alternative polyacene units in a column of the t-polyacenic nanotube, where a polyacene unit consists of t hexagons. The molecular graph of the t-polyacenic nanotube Ptp,q has 4p rows and q columns. For 1≤t≤6, the t-polyacenic nanotube Ptp,q is known as phenylenic, naphthalenic, anthacenic, tetracenic, pentacenic, and hexacenic nanotubes, respectively. The molecular graphs of these nanotubes are presented in Figure 3. Let G be a molecular graph of the Ptp,q nanotube and then we can observe that 2≤dk≤3 for each k∈VG. So, we have the vertex partitions of G as follows:(2)V2=k∈VG∣dk=2,V3=k∈VG∣dk=3.
The generalized molecular graph of the t-polyacenic nanotube Ptp,q.
The vertex partitions of G along with their cardinalities corresponding to each row are presented in Table 1. In the following theorems, we formulate the eccentric connectivity index for Ptp,q nanotubes for t>1.
The vertex partitions of the Ptp,q nanotube along with their cardinalities corresponding to each row.
Vertex partition
Rows
Cardinality for each row
V1
1,4p
tq
V2
2,3,6,7,…,4p−2,4p−1
t+1q
V3
4,5,8,9,…,4p−4,4p−3
tq
Theorem 1.
Let Ptp,q be the graph of the t-polyacenic nanotube, and then for q even, we have(3)ξcPtp,q=10t2+11t+3q2+25t+3q,ifp=1 andt≥1,92t+1p2q+34t2+4t+1pq2−10t+3pq−t2t+1q2+2tq,if2≤p≤tq2 andt≥1,3t22t+1q3+212t+1p2q−34t2−1pq2−14t+3pq−tq2+2tq,iftq+22≤p≤tq−1 andt≥1,182t+1p2q+32t+1pq2−14t+3pq−tq2+2tq,ifp≥tq,p is even andt≥1,182t+1p2q+32t+1pq2−14t+3pq−tq2+2t−3q,ifp≥tq,p is odd andt≥1.
Proof.
Consider G=Ptp,q. Let vi represents the vertices in the ith row. With respect to εvi or dvi, we have the following cases.
Case 1 (when p=1 and t≥1).
In this case, the eccentricity of each vertex in each row is 2t+1q+2/2. Hence, from Table 1 and (1), we have(4)ξcG=2tq22t+1q+22+2t+1q32t+1q+22=10t2+11t+3q2+25t+3q.
Case 2 (when 2≤p≤tq/2 and t≥1).
In this case,(5)εvi=εvi+1=εv4p+1−i=εv4p−i=4p+2t+1q2−i,where i=1,2,…,p. Hence, from Table 1 and (1), we have(6)ξcG=2tq2+tq+q34p+2t+1q2−1+2tq3+tq+q3∑i=2p4p+2t+1q2−i=92t+1p2q+34t2+4t+1pq2−10t+3pq−t2t+1q2+2tq.
Case 3 (when tq+2/2≤p≤tq−1 and t≥1).
In this case,(7)εvi=εv4p+1−i=8p+q2−i,wherei=1,2,…,4p−tq2.
Also,(8)εv4p−tq/2+i=εv4p−tq/2+1+i=εv2tq+1−i=εv2tq−i=4t+1q2−i,where i=1,2,…,tq−p. Hence, from Table 1 and (1), we have(9)ξcG=2tq28p+q2−1+2tq+q3+2p−tq28p+q2−2+6+⋯+4p−tq2−2+3+7+⋯+4p−tq2−1+2tq32p−tq2−18p+q2−4+8+⋯+4p−tq2+2tq35+9+⋯+4p−tq2−3+2tq3+tq+q3∑i=1tq−p4t+1q2−i,=3t22t+1q3+212t+1p2q−34t2−1pq2−14t+3pq−tq2+2tq.
Case 4 (when p≥tq, p is even and t≥1).
In this case,(10)εvi=εv4p+1−i=8p+q2−i,wherei=1,2,…,2p.
Hence, from Table 1 and (1), we have(11)ξcG=2tq28p+q2−1+2tq+q3p8p+q2−2+6+⋯+2p−2+3+7+⋯+2p−1+2tq3p−18p+q2−4+8+⋯+2p+5+9+⋯+2p−3=182t+1p2q+32t+1pq2−14t+3pq−tq2+2tq.
Case 5 (when p≥tq, p is odd and t≥1).
In this case, we use the eccentricities of vertices as given in case 4. From Table 1 and (1), we have(12)ξcG=2tq28p+q2−1+2tq+q3p8p+q2−2+6+⋯+2p+3+7+⋯+2p−3+2tq3p−18p+q2−4+8+⋯+2p−2+5+9+⋯+2p−1=182t+1p2q+32t+1pq2−14t+3pq−tq2+2t−3q.
Theorem 2.
Let Ptp,q be the graph of the t-polyacenic nanotube, and then for q odd, we have(13)ξcPtp,q=25t2+8t+3,ifp=1,q=1 andt≥2,10t2+11t+3q2+5t+3q,ifp=1,q≥3 andt≥1,92t+1p2q+34t2+4t+1pq2−28t+3pq−t2t+1q2+3tq,if2≤p≤tq2 andtq≥4,3t22t+1q3+212t+1p2q−34t2−1pq2−210t+3pq−tq2+3tq,iftq+22≤p≤tq−1 andt even,3t6t2+5t+1q3−32t+1p2q−34t2+4t+1pq2+62t+1pq+t+18t+3q2+11t+9q,ifp=tq+12 andt odd,3t3q3+310t+3p2q+34t+1pq2+8t+3pq−t9t+5q2−63t−2q,iftq+32≤p≤tq−2 andt odd,67t+3p2q+32t+1pq2−234t+15pq+t48t+23q2−39t+4q,ifp=tq−1 andt odd,182t+1p2q+32t+1pq2−210t+3pq−tq2+3tq,ifp≥tq,p is even andt≥1,182t+1p2q+32t+1pq2−210t+3pq−tq2+3t−1q,ifp≥tq,p is odd andt≥1.
Proof.
Consider G=Ptp,q. Let vi represent the vertices in the ith row of G. With respect to εvi or dvi, we have the following cases.
Case 1 (when p=1, q=1 and t≥2).
In this case, the eccentricity of each vertex in each row is t+1. Hence, from Table 1 and (1), we have(14)ξcG=2t2t+1+2t+13t+1=25t2+8t+3.
Case 2 (when p=1, q≥3 and t≥1).
In this case, the eccentricity of each vertex in each row is 2t+1q+1/2. Hence, from Table 1 and (1), we have(15)ξcG=2tq22t+1q+12+2tq+q32t+1q+12=10t2+11t+3q2+5t+3q.
Case 3 (when 2≤p≤tq/2 and tq≥4).
In this case,(16)εvi=εvi+1=εv4p+1−i=εv4p−i=4p+2t+1q−12−i,where i=1,2,…,p. Hence, from Table 1 and (1), we have(17)ξcG=2tq2+t+1q34p+2t+1q−12−1+2tq3+t+1q3∑i=2p4p+2t+1q−12−i=92t+1p2q+34t2+4t+1pq2−28t+3pq−t2t+1q2+3tq.
Case 4 (when tq+2/2≤p≤tq−1 and t is even).
In this case,(18)εvi=εv4p+1−i=8p+q−12−i,wherei=1,2,…,4p−tq2.
Also,(19)εv4p−tq/2+i=εv4p−tq/2+1+i=εv2tq+1−i=εv2tq−i=4t+1q−12−i,where i=1,2,…,tq−p. Hence, from Table 1 and (1), we have(20)ξcG=2tq28p+q−12−1+2tq+q32p−tq28p+q−12−2+6+⋯+4p−tq2−2+3+7+⋯+4p−tq2−1+2tq32p−tq2−18p+q−12−4+8+⋯+4p−tq2+2tq35+9+⋯+4p−tq2−3+2tq3+t+1q3∑i=1tq−p4t+1q−12−i=3t22t+1q3+212t+1p2q−34t2−1pq2−210t+3pq−tq2+3tq.
Case 5 (when p=tq+1/2 and t is odd).
In this case,(21)εv1=εv2=εv4p=εv4p−1=4tq+q+12,εvi+2=εv4p−1−i=εv4p=εv4p−1=4t+1q+12−i,wherei=1,2.
Also,(22)εvi+4=εvi+5=εv2tq−1−i=εv2tq−2−i=4t+1q−32−i,where i=1,2,…,tq−p−1. Hence, from Table 1 and (1), we have(23)ξcG=2tq2+tq+q34t+1q+12+2tq+q34t+1q+12−1+2tq34t+1q+12−2+2tq3+tq+q3∑i=1tq−p−14t+1q−32−i=3t6t2+5t+1q3−32t+1p2q−34t2+4t+1pq2+62t+1pq+t+18t+3q2+11t+9q.
Case 6 (when tq+3/2≤p≤tq−2 and t is odd).
In this case,(24)εvi=εv4p+1−i=8p+q−12−i,wherei=1,2,…,4p−tq+12,εv4p−tq+1/2+1=εv4p−tq+1/2+2=εv2tq+2=εv2tq+1=4t+1q+12,εv4p−tq+1/2+2+i=εv2tq+1−i=4t+1q+12−i,wherei=1,2.
Also,(25)εv4p−tq+1/2+4+i=εv4p−tq+1/2+5+i=εv2tq−1−i=εv2tq−2−i=4t+1q−32−i,where i=1,2,…,tq−p−1. Hence, from Table 1 and (1), we have(26)ξcG=2tq28p+q−12−1+2tq+q32p−tq+128p+q−12−2+6+⋯+4p−tq+12−2+2tq+q33+7+⋯+4p−tq+12−1+2tq32p−tq+12−18p+q−12−4+8+⋯+4p−tq+12+2tq35+9+⋯+4p−tq+12−3+2tq3+tq+q34t+1q+12+2tq+q34t+1q+12−1+2tq34t+1q+12−2+2tq3+tq+q3∑i=1tq−p−14t+1q−32−i=3t3q3+310t+3p2q+34t+1pq2+8t+3pq−t9t+5q2−63t−2q.
Case 7 (when p=tq−1 and t is odd).
In this case,(27)εvi=εv4p+1−i=8p+q−12−i,wherei=1,2,…,2p−2,εv2p−3=εv2p−2=εv2p+3=εv2p+4=4t+1q+12.
Hence, from Table 1 and (1), we have(29)ξcG=2tq28p+q−12−1+2tq+q3p−28p+q−12−2+6+⋯+2p−2+3+7+⋯+2p−2−1+2tq3p−38p+q−12−4+8+⋯+2p−2+5+9+⋯+2p−2−3+2tq3+tq+q34t+1q+12+2tq+q34t+1q+12−1+2tq34t+1q+12−2=67t+3p2q+32t+1pq2−234t+15pq+t48t+23q2−39t+4q.
Case 8 (when p≥tq, p is even and t≥1).
In this case,(30)εvi=εv4p+1−i=8p+q−12−i,wherei=1,2,…,2p.
Hence, from Table 1 and (1), we have(31)ξcG=2tq28p+q−12−1+2tq+q3p8p+q−12−2+6+⋯+2p−2+3+7+⋯+2p−1+2tq3p−18p+q−12−4+8+⋯+2p+5+9+⋯+2p−3=182t+1p2q+32t+1pq2−210t+3pq−tq2+3tq.
Case 9 (when p≥tq, p is odd and t≥1).
In this case, we use the eccentricities of vertices as given in case 8. From Table 1 and (1), we have(32)ξcG=2tq28p+q−12−1+2tq+q3p8p+q−12−2+6+⋯+2p+3+7+⋯+2p−3+2tq3p−18p+q−12−4+8+⋯+2p−2+5+9+⋯+2p−1=182t+1p2q+32t+1pq2−210t+3pq−tq2+3t−1q.
Remark 1.
The results presented by Rao and Lakshmi in [25] become special cases of the results given in Theorems 1 and 2 for t=1.
3. Conclusion
In this paper, we present generalized formulae of ECI for t-polyacenic nanotubes. The comparability about biological activities of chemical compounds is of immense interest in QSAR/QSPR studies. The eccentric connectivity index ECI provides the best prediction accuracy rate compared to other indices in various biological activities of diverse nature such as anti-inflammatory activity, anticonvulsant activity, and diuretic activity. In this sense, this index can be very helpful in QSAR/QSPR studies, and by using the given results, we can present mathematical models of several biological activities of all chemical compounds, which correspond to t-polyacenic nanotubes such as phenylenic nanotubes, naphthalenic nanotubes, anthracenic nanotubes, tetracenic nanotubes, pentacenic nanotubes, and hexacenic nanotubes.
Data Availability
All data generated or analyzed during this study are included in this article.
Conflicts of Interest
The authors declare that there are no conflicts of interest.
Acknowledgments
The authors would like to express their sincere gratitude to the anonymous referees and the editor for many valuable, friendly, and helpful suggestions, which led to a great deal of improvement of the original manuscript. This work was done under the project supported by the Higher Education Commission, Pakistan, via Grant no. 5331/Federal/NRPU/R&D/HEC/2016. This research was funded by the China Postdoctoral Science Foundation under Grant no. 2017M621579, the Postdoctoral Science Foundation of Jiangsu Province under Grant no. 1701081B, and Project of Anhui Jianzhu University under Grant nos. 2016QD116 and 2017dc03.
de Julián-OrtizJ. V.de Gregorio AlapontC.Rı́os-SantamarinaI.Garcı́a-DoménechR.GálvezJ.Prediction of properties of chiral compounds by molecular topology1998161141810.1016/s1093-3263(98)00013-82-s2.0-0032008459KierL. B.HallL. H.An electrotopological state index for atoms in molecules19907880180710.1023/a:10159526137602-s2.0-0025155575LiuJ.-B.PanX.-F.HuF.-T.HuF.-F.Asymptotic Laplacian-energy-like invariant of lattices201525320521410.1016/j.amc.2014.12.0352-s2.0-84920694057PoglianiL.Modeling enthalpy and hydration properties of inorganic compounds19973803817HuQ.-N.LiangY.-Z.FangK.-T.The matrix expression, topological index and atomic attribute of molecular topological structure20031361389SoofiE. S.RetzerJ. J.Information indices: unification and applications20021071-2174010.1016/s0304-4076(01)00111-72-s2.0-0002603663LiuJ.-B.PanX.-F.Minimizing Kirchhoff index among graphs with a given vertex bipartiteness2016291848810.1016/j.amc.2016.06.0172-s2.0-84977619595SharmaV.GoswamiR.MadanA. K.Eccentric connectivity index: a novel highly discriminating topological descriptor for structure–property and structure–activity studies199737227328210.1021/ci960049h2-s2.0-0002938511DurejaH.MadanA. K.Superaugmented eccentric connectivity indices: new-generation highly discriminating topological descriptors for QSAR/QSPR modeling2007167–933134110.1007/s00044-007-9032-92-s2.0-51749095461DurejaH.GuptaS.MadanA. K.Predicting anti-HIV-1 activity of 6-arylbenzonitriles: computational approach using superaugmented eccentric connectivity topochemical indices20082661020102910.1016/j.jmgm.2007.08.0082-s2.0-37349040711SinghM.JangraH.BharatamP. V.MadanA. K.Detour matrix-based adjacent path eccentric distance sum indices for QSAR/QSPR. Part I: development and evaluation20147429531810.1504/ijcbdd.2014.0665402-s2.0-84920068419AshrafiA. R.GhorbaniM.GutmanI.FurtulaB.Eccentricity connectivity index2010Kragujevac, SerbiaUniversity of Kragujevac169182DošliśT.SaheliM.GutmanI.FurtulaB.Eccentricity connectivity index of fullerenes2010Kragujevac, SerbiaUniversity of Kragujevac183192IlićA.GutmanI.FurtulaB.Eccentricity connectivity index of benzenoid graphs2010Kragujevac, SerbiaUniversity of Kragujevac139168MadanA. K.DurejaH.GutmanI.FurtulaB.Eccentricity based descriptors for QSAR/QSPR2010Kragujevac, SerbiaUniversity of Kragujevac91138MadanA. K.DurejaH.GutmanI.FurtulaB.Applications of eccentricity connectivity index2010Kragujevac, SerbiaUniversity of Kragujevac247268KumarV.SardanaS.MadanA. K.Predicting anti-HIV activity of 2,3-diaryl-1,3-thiazolidin-4-ones: computational approach using reformed eccentric connectivity index2004105-639940710.1007/s00894-004-0215-82-s2.0-10944243937SardanaS.MadanA. K.Application of graph theory: relationship of molecular connectivity index, Wiener’s index and eccentric connectivity index with diuretic activity2011438598GuptaS.SinghM.MadanA. K.Application of graph theory: relationship of eccentric connectivity index and wiener’s index with anti-inflammatory activity2002266225926810.1006/jmaa.2000.72432-s2.0-0037084418SardanaS.MadanA. K.Predicting anticonvulsant activity of benzamides/benzylamines: computational approach using topological descriptors2002168-954555010.1023/a:10219048030572-s2.0-0036706729AshrafiA. R.DošlićT.SaheliM.The eccentric connectivity index of TUC4C8(R) nanotubes2011651221230AshrafiA. R.SaheliM.GhorbaniM.The eccentric connectivity index of nanotubes and nanotori2011235164561456610.1016/j.cam.2010.03.0012-s2.0-77955689506IranmaneshA.AlizadehY.Eccentric connectivity index of HAC5C7[p, q] and nHAC5C6C7[p, q] anotubes201369175182NadeemI.ShakerH.On eccentric connectivity index of TiO2 nanotubes201663236336810.17344/acsi.2016.23372-s2.0-84975288419RaoN. P.LakshmiK. L.Eccentric connectivity index of V-phenylenic nanotubes2010618187SaheliM.AshrafiA. R.The eccentric connectivity index of armchair polyhex nanotubes20102917175IlićA.GutmanI.Eccentric connectivity index of chemical trees201165731744MorganM. J.MukwembiS.SwartH. C.On the eccentric connectivity index of a graph2011311131234129910.1016/j.disc.2009.12.0132-s2.0-79955482851ZhouB.DuZ.On eccentric connectivity index201063181198AnthonyJ. E.Functionalized acenes and heteroacenes for organic electronics2006106125028504810.1021/cr050966z2-s2.0-33846219497BendikovM.WudlF.PerepichkaD. F.Tetrathiafulvalenes, oligoacenenes, and their buckminsterfullerene derivatives: the brick and mortar of organic electronics2004104114891494610.1021/cr030666m2-s2.0-10344259627BredasJ. L.CalbertJ. P.da Silva FilhoD. A.CornilJ.Organic semiconductors: a theoretical characterization of the basic parameters governing charge transport20029995804580910.1073/pnas.0921433992-s2.0-0037197928FirouziR.ZahediM.Polyacenes electronic properties and their dependence on molecular size20088621–371510.1016/j.theochem.2008.04.0392-s2.0-46749100589SchonJ. H.KlocC.DodabalapurA.BatloggB.An organic solid state injection laser2000289547959960110.1126/science.289.5479.5992-s2.0-0034725775SchönJ. H.KlocC.BatloggB.Retraction note to: superconductivity in molecular crystals induced by charge injection2000406679770270410.1038/350210112-s2.0-0034680131YamabeT.YataS.WangS.The structures and properties of conjugated hydrocarbons such as polyacenic materials and polycyclic aromatic hydrocarbons (PAHs) doped with lithium20031371–394995110.1016/s0379-6779(02)00944-x2-s2.0-0037419293BiennierL.Alsayed-AliM.Foutel-RichardA.Laboratory measurements of the recombination of PAH ions with electrons: implications for the PAH charge state in interstellar clouds200613328930110.1039/b516858a2-s2.0-33750591951KhadikarP. V.KarmarkarS.VarmaR. G.On the estimation of PI index of polyacenes200249755771SoleimaniN.NikmehrM. J.TavallaeeH. A.Computation of the different topological indices of nanostructures201543212713310.4038/jnsfsr.v43i2.79402-s2.0-84936763161SoleimaniN.NikmehrM. J.TavallaeeH. A.Theoretical study of nanostructures using topological indices2014594139148VeylakiM.NikmehrM. J.Some degree based topological indices of nanostructures2015473872875