Tpn

ZC

a

ARRAA

KEEDASD

1

tIgtnnttasipna

h0

Sensors and Actuators B 196 (2014) 555–566

Contents lists available at ScienceDirect

Sensors and Actuators B: Chemical

jo ur nal home page: www.elsev ier .com/ locate /snb

heoretical investigation of ethane and ethene monitoring usingristine and decorated aluminum nitride and silicon carbideanotubes

abiollah Mahdavifar ∗, Maryam Haghbyan, Mehdi Abbasiomputational Chemistry Group, Department of Chemistry, Faculty of Science, Shahid Chamran University, Ahvaz, Iran

r t i c l e i n f o

rticle history:eceived 26 September 2013eceived in revised form 29 January 2014ccepted 14 February 2014vailable online 21 February 2014

eywords:thanetheneetectinglNNT

a b s t r a c t

The adsorption of ethane and ethene molecules on pristine and Ni-doped armchair (4,4) single walledaluminum nitride (AlN) and silicon carbide (SiC) nanotubes are investigated employing density func-tional theory approach. Our results indicate that the ethane/ethene molecule physisorbed onto theouter surface of AlNNT and SiCNT through weak Van der Waals interaction. On the other hand, theencapsulation of ethane/ethene onto the inner surface of considered nanotubes is endothermic and dif-ficult to realize with an appreciable energy barrier. Compared with weak adsorption of ethane/etheneonto the pristine AlNNT and SiCNT, Ni decorated AlN and SiC nanotubes exhibit strong affinity towardthe ethane/ethene molecule with remarkable negative adsorption energies about −61/−179 kJ/mol forAlNNT/Ni and −96/−202 kJ/mol for SiCNT/Ni systems, respectively. Based on our results, it seems thatethene tends to be chemisorbed onto the Ni-doped nanotubes, whereas the ethane-adsorption process

iCNTFT calculations

is through strongly physisorbed process and could serve as a signal of nanosensor due to affect the elec-tronic conductance and structural properties. These observations show that functionalized AlN and SiCnanotubes are highly sensitive toward C2H4/C2H6 molecule. Moreover, these results may be useful for thedesign of new types of nanosensor devices that can detect the presence of small hydrocarbon molecules.

© 2014 Elsevier B.V. All rights reserved.

. Introduction

The field of nanosensors has attracted considerable interest dueo our current concern over environmental and health issues. [1].n this regard, significant attempts have been made to investi-ate new nanomaterial-based sensors from both experimental andheoretical point of views, which some of these One-dimensionalanosensors such as carbon nanotubes (CNTs) [2], semiconductoranowires [3], and graphene nanoribbons [4] can modify many ofhe draw-backs of traditional semiconductor-based gas sensors. Inhe context of sensor development work, CNTs have a set of uniquend outstanding properties such as high sensitivity, fast response,mall size and low operating temperature [5–8] that make themdeal candidates for nanosensors. So far, since the discovery [9],

ure carbon nanotubes have been demonstrated to be promisinganoscale molecular sensors for detecting a few molecule gas suchs O2 [10], NH3 [11], NO2 [12] and H2 [13] with fast response time

∗ Corresponding author. Fax: +98 611 3331042.E-mail addresses: z mahdavifar@scu.ac.ir, zb nojini@scu.ac.ir (Z. Mahdavifar).

ttp://dx.doi.org/10.1016/j.snb.2014.02.048925-4005/© 2014 Elsevier B.V. All rights reserved.

and high sensitivity. This superior sensitivity has been theoreticallyexplained in terms of change of the semiconducting single-wallcarbon nanotube induced by charge transfer from gas moleculesadsorbed on nanotube surfaces, which can dramatically influencethe electrical conductivity of the latter by modifying the electronicstructure of CNTs. To overcome this problem, functionalization ofcarbon nanotubes (CNTs) is a promising approach to increase theirsolubility and reactivity, since the adsorption capability of SWCNTscan be improved through exohedral or substitutional doping andforming active sites in tube walls.

Recently great advances have been made in demonstrating theviability of using inorganic semiconducting nanotubes such as alu-minum nitride (AlN) and silicon carbide (SiC) nanotubes to detectthe presence of chemical gases as well as organic chemical and bio-logical substances and they have led to the design of a new typeof sensor devices. AlNNT and SiCNT could provide very high sensi-tivity due to their large surface to volume ratios and their unique

electronic properties.

The silicon nanotubes (SiNTs), which are analogous to carbonnanotubes in many respects such as electronical and struc-tural characters, have been prepared by various experiments and

5 nd Act

tnlwiStiHoi

scwrneAtdeia[stgau

iettfiasnpmiw

2

ssnudet[etfmadnnAa

56 Z. Mahdavifar et al. / Sensors a

heoretical studies in the recent years [14–20]. Unlike carbonanotubes (CNTs), SiCNTs are semiconducting materials with a

arge band gap, which their stability is diameter-independent andeakly depended on the helicity [21]. Because of unusual mechan-

cal properties, excellent chemical and thermal stabilities, theiCNTs have been promising potential for applications in nanoelec-ronic devices operate at high temperatures, high frequency, andn harsh environment [22]. Theoretical studies show that O2 [23],

2 [24], CO and HCN [25], NO and N2O [26] could be chemisorbedn the exterior surface of SiCNT with large binding energy, whichndicate that SiCNT could act as sensor.

As an important member of semiconductor, 1-D III–V nano-tructures of aluminum nitride nanotubes (AlNTs) have attachedonsiderable attention due to unique properties such as directide band gap (∼6.2 eV), high thermal conductivity [27], supe-

ior mechanical strength, highpiezoelectric response, small or evenegative electron affinity, and so on [28]. Note that before firstxperimentally synthesized by Tondare et al. [29], the synthesis oflNNTs have been a challenging task, but Zhang et al. have been tes-

ified the strain energy and stability of single-walled AlNNT usingensity functional theory and it have been found that AlNNTs arenergetically favorable and arrange in a hexagonal network adopt-ng an sp2 hybridization [30]. Recently, a few theoretical studieslso focus on the tips of AlNNTs with regard to defect properties,31] and functionalization of the AlNNT’s wall [32,33] to modifyensitivity of these nanotubes through different molecules. Fur-hermore, the interaction of AlNNTs with gases, excepting a fewas molecules such as H2O, N2 and O2 [34], CH4 [35], ammonia [36]nd CO2 [37] has seldom been investigated and remains largely annexplored area.

Of particular interest in this paper are those studies related tonteractions of small hydrocarbon molecule including ethane andthene with AlN and SiC nanotubes in different kind including pris-ine and Ni-doped nanotubes. Here we propose to design a newype of nanoscale sensors using AlN and SiC nanotubes with modi-ed electronical and chemical properties through doping impuritytom into them. We demonstrate that this type of nanosen-ors can overcome the weaknesses of existing intrinsic aluminumitride and silicon carbide nanotubes. Hence, the results of thisaper could provide the necessary tools for the molecular levelanipulations that will satisfy some of the fabrication’s needs of

ncreasingly miniaturized devices that the microelectronic industryill demand within the next decades.

. Computational details

The density functional theory are carried out using Gaus-ian03 [38] package to exploring the equilibrium geometries,tabilities, and electronic properties of nanotube/hydrocarbon andanotube/nickel-hydrocarbon systems to search their potentialsage as novel nanosensors. The spin-polarized generalized gra-ient approximation (GGA) with the modified Perdew-Wang91xchange [39] plus the Perdew-Wang91 (MPW1PW91) [40] andhe functional of Perdew-Burke-Ernzerhof (PBEPBE) corrections41] are applied for describing the exchange-correlation term. Anxtra basis set formed by the CEP-121G [42] for Ni atom andhe conventional 6-31G basis set for all other atoms were per-ormed in this paper. We have selected two different nanotube

odels, which include finite single walled armchair (4,4) AlNnd SiC nanotubes with comparable structural properties such asiameter, length and chirality. The length and diameter of both

anotubes is about 16 A and 7.31 A respectively and each of theseanotubes include totally 40 atoms. In addition, the two ends oflNNT and SiCNT are terminated with hydrogen atoms, which thisct can be a barrier for shutting of nanotube edges (see Fig. S1).

uators B 196 (2014) 555–566

The structural-optimization of these pristine nanotubes were per-formed using the PBEPBE/6-31G and MPW1PW91/6-31G level oftheory.

Many attempts have been made during these years to use thequantitative chemical concepts in density-functional-based the-ory [43], namely chemical potential (�) and hardness (�), in theunderstanding of molecular reactivity. For an N-electron systemwith total energy E and external potential (�(r)), chemical poten-tial (�) and hardness(�) are defined as the first- and second-orderpartial derivatives of the total electronic energy (E) with respectto the number of electrons (N) at a fixed external potential (�(r))respectively [44]. Using a Janak’s type of approximation [45], thesedescriptors could be approximated in terms of the obtained ener-gies of highest occupied molecular orbital (HOMO) and the lowestunoccupied molecular orbital (LUMO) from density functional the-ory calculations (εH and εL), respectively.

� =(

∂E

∂N

)�(

�r),T

∼= (∼εL + εH)2

(1)

� = 12

(∂2

E

∂N2

)�(

�r),T

∼= (εL − εH)2

(2)

Parr et al. [46] have proposed electrophilicity index (ω) in termsof the chemical potential and chemical hardness as a measure ofthe electrophilic power of a molecule as:

ω = �2

2�(3)

Using the Janak’s approximations, this relation for electrophilic-ity has the simple forms of Eq. (3).

3. Result and discussion

3.1. Ethane/ethene adsorption on pristine AlN and SiC nanotubes

The structural optimization of pristine armchair (4,4) singlewalled AlNNT and SiCNT is performed in the framework of den-sity functional theory using two MPW1PW91 and PBEPBE methods.Obtained results from relaxed geometries of these nanotubes indi-cate that SiC and AlN nanotubes are semiconductor with energy gapabout 3.41 and 4.88 eV respectively, which these our results are inwell agreement with other previous research works [27,47]. How-ever, because difference in performed methods, there is slightlyquantitative difference between obtained gap energy our researchand other works. In addition, structural parameters of relaxed AlNand SiC were compared with several other research works [48–50],and it is found that the bond length of Al–N and Si–C, which are1.818 and 1.813 A respectively, are well reported. The natural bondorbital (NBO) calculations are also performed to obtain some elec-trical properties of these nanotubes such as partial atomic chargeand bond order using MPW1PW91/3-21G* and PBEPBE/3-21G*level of theory (see Table S1). It is noteworthy that significant dif-ference in electronegativity between Al/Si and N/C atoms in AlNNTand SiCNT causes the ionic character of Al–N and Si–C bonds (TableS1). Du to large ionicity of these bonds, the electronic structures ofAlN and SiC nanotubes are almost independent of tube diameterand chirality, which mean of this phenomenon is that investiga-tions on armchair nanotube could be transferred to zigzag one.Furthermore, the electronic properties of isolated components arecollected in Table S2.

In order to investigate the adsorption of ethane (C2H6) and

ethene (C2H4) molecules onto the pristine AlNNT and SiCNT,different possible adsorption sites including inside (C-position)and outside of nanotubes are selected. The considered hydro-carbon molecules horizontally (H-orientation) and vertically

Z. Mahdavifar et al. / Sensors and Actuators B 196 (2014) 555–566 557

F ding

A

(toneTs

ts

E

waESN

ig. 1. Adsorption of various gas molecule configurations onto the nanotubes inclulN and SiC nanotubes and (c) inside of nanotubes (C-position).

V-orientation) are located on top of the hexagonal ring of enti-led nanotubes with vertical distance about 4 A from outer surfacef AlN and SiC nanotubes (see Fig. 1). The new structures ofanotube/C2H6 and nanotube/C2H4 systems are fully optimizedmploying MPW1PW91 and PEBPEB method with 6-31G basis set.ypically the relaxed geometries of ethane adsorbed onto the outerurface of AlN and SiC nanotubes are presented in Fig. 2.

To examine the nanotube/hydrocarbon interactions, the adsorp-ion energy (Eads) of these molecules adsorbed onto the nanotubeurfaces is defined as:

ads = Etube/gas −(

Etube + Egas)

(4)

here Etube/gas denotes the total energy of the adduct AlNNT

nd SiCNT with the corresponding ethane/ethene molecule andtube and Egas are the total energies of the pristine AlN andiC nanotubes and isolated ethane/ethene molecule respectively.ote that, according to this equation, the negative adsorption

H -and V-orientations of (a) ethane (b) ethene molecules onto the outer surface of

energy imply to formed stable complex and positive adsorptionenergy belongs to the local minimum in which the adsorptionof hydrocarbon molecule onto the nanotubes is prevented by abarrier. The interesting information is obtained if the adsorptionenergies of different situations including C-position, V- and H-orientations of gas molecules are evaluated (see Table 1). First,the most negative adsorption energy is related to H-orientationof ethane/ethene molecule adsorption onto the outer surface ofSiCNT and AlNNT. In other word, when ethane/ethene moleculehorizontally located on top of the nanotube, the most stable struc-ture is obtained. Interestingly, the final optimization structuresof V-orientations indicate that the position of gas molecules isslightly changed from vertical to horizontal orientation. These

results are in well agreement with this fact that H-orientation isfavorable than V-orientation (see Fig. 2). Furthermore, the near-est intermolecular distance of H-orientation of ethane moleculeadsorbed onto the AlNNT and SiCNT surfaces are obtained 2.874

558 Z. Mahdavifar et al. / Sensors and Actuators B 196 (2014) 555–566

ule ad

acseFftt1wmrdamisw

Fig. 2. Fully optimized geometrical structures of H-orientation of ethane molec

nd 3.393 A (using MPW1PW91) respectively. This result is alsoonfirmed that the H-orientation of gas molecule on nanotubeurface is slightly better than others. Details of structural param-ters of nanotube/gas systems are summarized in Table 1 andig. 3. It should be noted that the same trends is obtainedrom two different methods. However, based on our results,he c-position of ethane/ethene molecule has positive adsorp-ion energies (about 72.66/25.27 kJ/mol for AlN–ethane/ethene and10.33/54.42 kJ/mol for SiC–ethane/ethene systems respectively),hich indicate that the adsorption of considered hydrocarbonolecules inside of AlN and SiC nanotubes is prevented by a bar-

ier and the final geometries are not favorable due to the smalliameter of these nanotubes. It is noteworthy that according to thedsorption energies in Table 1, the interaction between C2H6/C2H4

olecule and AlNNT and SiCNT even in the best favorable position

s very weak. The most negative adsorption energy of AlNNT/ethaneystem is only −6.61 kJ/mol (obtained by MPW1PW91 method)hich implies weakly physisorption process.

sorbed onto the pristine (a) AlN (b) SiC nanotubes using MPW1PW91 method.

In continue the natural bond orbital (NBO) calculations suchas partial charge transfers, HOMO and LUMO energies and natu-ral atomic orbital occupancies are investigated. Calculated partialcharges of Al, N, Si, C atoms of nanotubes and gas moleculesadsorbed on AlN and SiC nanotubes are collected in Table 2.It is clear from Table 2 that there is no significant change isoccurred during the adsorption process. In comparison, the elec-tronic properties of nanotube/C2H4 and nanotube/C2H6 systemssuch as HOMO-LUMO gap energy with pristine nanotube demon-strate that there is no significant change in the electronic propertiesof considered nanotubes is observed (see Tables S2 and 3). Theseresults are in well agreement with results of adsorption energygained. In conclusion, the obtained data show that C2H4/C2H6molecule adsorbed onto the AlN and SiC nanotubes through weak

physisorption and also pristine AlN and SiC nanotubes are not goodcandidates for monitoring of C2H4/C2H6 molecule. Furthermore,the Wiberg bond index (WBI) is also studied. The Wiberg bondindex (WBI) which comes from the manipulation of the density

Z. Mahdavifar et al. / Sensors and Actuators B 196 (2014) 555–566 559

Table 1Adsorption energy, Eads (kJ/mol), equilibrium distance re (Å), rNi-gas and rtube-Ni of ethane and ethene adsorption on pristine and Ni decorated nanotubes respectively.

C2H6 C2H4

MPW1PW91 PEBPEB MPW1PW91 PEBPEB

H V C H V C H V C H V C

AlN/gasre 2.874 3.31 3.57 3.35 3.27 3.66 2.70 3.60 3.40 2.65 3.56 3.44Eads −5.43 −4.22 72.66 −11.88 −9.10 35.55 −6.61 10.98 25.27 −28.34 5.30 4.75

SiC/gasre 3.953 3.67 3.59 3.77 3.64 3.63 3.20 4.11 3.55 3.15 4.56 3.54Eads −2.54 −0.26 110.33 −5.27 −2.80 65.46 −2.28 6.68 54.42 −6.94 −2.31 60.26

AlN-Ni/gasrNi-gas 2.189 2.18 – 2.16 2.17 – 1.96 1.98 – 1.86 1.94 –rAlN-Ni 1.837 1.83 – 1.84 1.84 – 1.868 1.87 – 1.85 1.85 –Eads −61.74 −61.58 – −78.26 −76.43 – −179.32 −179.65 −199.87 −115.50 –

SiC-Ni/gasrNi-gas 2.21 2.21 – 2.26 2.26 – 2.00 1.99 – 1.79 1.84 –rSiC-Ni 1.88 1.88 – 1.49 1.46 – 1.93 1.92 – 1.51 1.53 –Eads −96.53 −96.55 – −48.41 −43.38 – −178.12 −202.12 – −183.69 −183.69 –

Fig. 3. Total DOS spectra of (a) AlNNT, (b) AlNNT/C2H6 and (c) AlNNT/C2H4 systems using MPW1PW91 method.

560 Z. Mahdavifar et al. / Sensors and Actuators B 196 (2014) 555–566

Table 2Calculated partial charges of Al, N, Si and C atoms of C2H6 and C2H4 molecule adsorbed on pristine and Ni adsorbed on pristine and Ni functionalized nanotubes (atomnumbering is according to Figs. 1, 4 and 5).

C2H6 C2H4

MPW1PW91 PEBPEB MPW1PW91 PEBPEB

H V H V H V H V

AlN/gasC(1) −0.73 −0.73 −0.72 −0.72 −0.46 −0.44 −0.47 −0.45C(2) – – – – −0.46 −0.45 −0.47 −0.43Al(1) 1.67 1.67 1.58 1.58 1.66 1.67 1.58 1.58N(1) −1.69 −1.69 −1.60 −1.60 −1.69 −1.68 −1.60 −1.60N(2) −1.69 −1.69 −1.60 −1.60 −1.70 −1.68 −1.61 −1.60N(3) −1.69 −1.69 −1.60 −1.60 −1.69 −1.69 −1.60 −1.60

SiC/gasC(5) −0.72 −0.72 −0.72 −0.72 −0.44 −0.44 −0.44C(6) – – – – −0.45 −0.44 −0.45Si(2) 1.84 1.85 1.73 1.74 1.86 1.75 1.72C(1) −1.85 −1.84 −1.73 −1.73 −1.85 −1.74 −1.73C(2) −1.84 −1.84 −1.73 −1.73 −1.85 −1.74 −1.73C(3) −1.84 −1.85 −1.73 −1.73 −1.85 −1.74 −1.73

AlN-Ni/gasC(1) −0.716 −0.716 −0.715 −0.709 −0.520 −0.530 −0.551C(2) – – – – −0.560 −0.530 −0.382H(1) 0.238 0.258 0.266 0.236 0.247 0.249 0.236Ni −0.120 −0.119 −0.153 −0.148 −0.016 −0.030 −0.11N(1) −1.552 −1.552 −1.457 −1.454 −1.580 −1.480 −1.47Al(1) 1.591 1.594 1.506 1.499 1.621 1.520 1.523Al(2) 1.588 1.583 1.506 1.504 1.634 1.520 1.515

SiC-Ni/gasC(1) −1.780 −1.780 −1.710 −1.718 −1.758 −1.796 −1.651 −1.726C(2) −0.737 −0.738 −0.725 −0.724 −0.538 −0.535 −0.526 −0.530

229

243

676

maolw

W

wdfistpmcaevandt

3S

Niaa

H(1) 0.248 0.231 0.229 0.Ni 0.210 0.211 0.238 0.Si(1) 1.560 1.560 1.668 1.

atrix in the orthogonal natural atomic orbital based on the NBOnalysis [51,52], is also considered. WBI demonstrates the strengthf the covalent character (the larger WBI implies to stronger cova-ent character) and closely relates to the bond order. WBI expressed

ith the following mathematical definition:

BI =∑

k

p2jk = 2pjj − p2

jj (5)

here pjk and pjj denote the density matrix elements and chargeensity in the atomic orbital respectively. The WBI calculations con-rmed that the gas molecules adsorbed onto the AlNNT and SiCNTurfaces through weak Van der Waals interaction (see Table 4). Also,he density of state (DOS) spectra of nanotube/ethane systems withristine nanotubes (Fig. 3) demonstrate that the adsorption of C2H6olecule on SiC and AlN nanotubes could not change mechani-

al and electrical properties of these nanotubes as well as ethenedsorption. In short, our results reveal that the interaction betweenthane/ethene molecule and pristine AlN and SiC nanotubes areery weak, so that the adsorption of these molecules onto the AlNNTnd SiCNT is physisorption process. It is seems that pristine AlN/SiCanotube could not be a promising candidate as a sensor devices foretecting ethane/ethene molecule. Therefore, improving the sensi-ivity of these nanotubes for ethane/ethene detecting is necessary.

.2. Ethane/ethene molecule adsorption on Ni-decorated AlN andiC nanotubes

In this section, the electrical and structural influence of doping

i atom onto the pristine AlN and SiC nanotubes are considered to

mprove the weak ability of theses nanotubes for ethane/ethene-dsorption. Five possible adsorption sites on AlN and SiC nanotubesre available for doping Ni atom, which are depicted in Fig. S2:

0.256 0.255 0.259 0.258−0.243 0.108 −0.251 −0.2501.841 1.661 1.727 1.747

T1 (top of the Al or C atoms), T2 (top of the N or Si atom), T3(top of the Al–N or Si–C bond axial zigzag Al–N or Si–C bond), H(top of the center of hexagon) and C (inside the nanotubes). It isfound from previous research works [35,53,54] that the best pos-itions for decorating Ni atom onto the AlN and SiC nanotube areT2 and H adsorption sites respectively. After locating the Ni atomon top of the these most stable sites, the new geometries of nan-otube/metal systems are fully optimized using PBEPBE/6-31G andMPW1PW92/6-31G level of theory with in conjugation CEP-121Gbasis set for Ni atom. The relaxed geometry of these systems isdepicted in Fig. S3. In addition, the binding energy of Ni doped atomonto the AlNNT and SiCNT surfaces is computed as following equa-tion to better understand the nanotube/Ni interaction characters:

Eb = Etube-Ni − (Etube + ENi) (6)

where E(tube-Ni) is the total energy of the Ni-decorated AlN and SiCnanotubes Etube and ENi are the total energies of the isolated AlNand SiC nanotubes and Ni metal atom respectively. The electronicand structural properties of Ni-doped on AlN and SiC nanotubeswith the same level of theories are completely illustrated in ourprevious research work [35]. The binding energies and the near-est distances between surfaces of the nanotubes and Ni atom arealso collected in Table S3 as well as in ref. [35]. It is found thatthe interaction between Ni atom and AlNNT and SiCNT are verystrong and binding Ni atom onto the outer surface of these nano-tubes is chemisorptions process. Furthermore, the detailed analysisof electron population obtained from NBO calculations indicatethat the considerable charge transfers from Ni metal atom to theAlNNT and SiCNT is occurred. This result is in well agreement with

strong interaction between Ni atom and nanotube surfaces. In com-parison, the HOMO-LUMO gaps of pristine AlNNT and SiCNT withnanotube/Ni systems demonstrate that since the adsorption of Niatom on AlN and SiC nanotubes, the HOMO-LUMO gap energy of

Z. Mahdavifar et al. / Sensors and Actuators B 196 (2014) 555–566 561

Table 3Highest occupied molecular orbital (HOMO), lowest unoccupied molecular orbital (LUMO), gap () energy, electronic chemical potential (�), hardness (�), softness (S) andelectrophilicity (ω) of gas molecule adsorbed onto pristine and Ni-doped nanotubes.

C2H6 C2H4

MPW1PW91 PEBPEB MPW1PW91 PEBPEB

H V H V H V H V

AlN/gasHOMO (eV) −6.38 −6.38 −5.23 −5.23 −6.31 −6.39 −5.16 −5.25LUMO (eV) −1.50 −1.49 −2.47 −2.46 −1.42 −1.51 −2.38 −2.48� (eV) 4.88 4.88 2.76 2.77 4.89 4.88 2.77 2.76� (eV) −3.94 −3.94 −3.85 −3.85 −3.87 −3.95 −3.77 −3.86ω (eV) 1.59 1.58 2.68 2.67 1.52 1.60 2.57 2.70S (eV−1) 0.20 0.20 0.36 0.36 0.20 0.20 0.36 0.36 (eV) 4.88 4.88 2.76 2.77 4.89 4.88 2.77 2.76

SiC/gasHOMO (eV) −5.41 −5.41 −4.57 −4.57 −5.37 −5.42 −4.33 −4.58LUMO (eV) −2.00 −1.99 −2.81 −2.81 −1.96 −2.01 −2.77 −2.82� (eV) 3.41 3.41 1.76 1.75 3.41 3.41 1.76 1.76� (eV) −3.71 −3.71 −3.69 −3.69 −3.67 −3.71 −3.65 −3.70ω (eV) 2.01 2.01 3.87 3.86 1.96 2.02 3.78 3.88S (eV−1) 0.29 0.29 0.56 0.56 0.29 0.29 0.56 0.56 (eV) 3.41 3.41 1.76 1.75 3.41 3.41 1.76 1.76

AlN-Ni/gasHOMO (eV) −5.65 −5.65 −4.23 −4.21 −5.92 −5.92 −4.34 −4.48LUMO (eV) −1.50 −1.50 −2.46 −2.45 −1.56 −1.57 −2.56 −2.50� (eV) 4.14 4.14 1.77 1.75 4.35 4.35 1.78 1.98� (eV) −3.57 −3.57 −3.34 −3.33 −3.74 −3.74 −3.45 −3.49ω (eV) 1.54 1.54 3.15 3.16 1.61 1.61 3.34 3.07S (eV−1) 0.24 0.24 0.56 0.56 0.22 0.23 0.56 0.50 (eV) 4.14 4.14 1.77 1.75 4.35 4.35 1.78 1.98

SiC-Ni/gasHOMO (eV) −5.28 −5.28 −4.13 −4.15 −5.26 −5.37 −4.54 −4.54LUMO (eV) −1.89 −1.89 −2.75 −2.74 −1.99 −1.95 −1.80 −2.80� (eV) 3.39 3.39 1.37 1.40 3.43 3.41 1.74 1.74� (eV) −3.59 −3.59 −3.44 −3.45 −3.71 −3.66 −3.67 −3.67

4.0.1.

tt

NiCoSaitsnFbmbaegNtaNwACOtA

ω (eV) 1.90 1.90 4.30

S (eV−1) 0.29 0.29 0.72

(eV) 3.39 3.39 1.37

hese nanotubes is decreased as well as the hardness, thus leadingo increase the reactivity of the systems.

To explore C2H6 and C2H4 gas molecules adsorption onto thei-decorated AlN and SiC nanotubes, ethane/ethene gas molecule

s added to the relaxed geometry of nanotube/Ni systems. The2H6/C2H4 molecule vertically (V-orientation) and horizontally (H-rientation) is located on top of the Ni atom in relaxed AlNNT/Ni andiCNT/Ni systems with vertical distance about 2 A from Ni metaltom. The PBEPBE/6-31G and MPW1PW92/6-31G level of theoryn conjugation with CEP-121G basis set for Ni atom are performedo fully optimization of new structure of nanotube/Ni-hydrocarbonystems without any restriction. The final optimized geometries ofanotube/Ni-C2H6 and nanotube/Ni-C2H4 systems are depicted inigs. 4 and 5. It is obvious from Figs. 4 and 5 that while Ni atomonded to outer surface of considered nanotubes, the C2H4/C2H6olecule also directly bonded to the Ni atom without any bond

reaking, which means that C2H6/C2H4 molecule can be safetydsorbed onto the Ni-decorated nanotubes. The structural param-ters of relaxed nanotube/Ni-C2H6 and nanotube/Ni-C2H4 systemseometries consist of the nearest intermolecular distance betweeni atom and nanotubes as well as the equilibrium distance between

he Ni atom and the C2H4/C2H6 gas molecules are listed in Table 1nd Figs. 4 and 5. It could be found from these figures that thei metal directly bonded to carbon atom of C2H4/C2H6 moleculeith bond lengths about 2.189, 1.960 A for AlNNT/Ni-C2H6 andlNNT/Ni-C2H4; 2.21 and 2.00 A for SiCNT/Ni-C2H6 and SiCNT/Ni-

2H4 (using MPW1PW91 method), respectively (see Figs. 4 and 5).n the other hand, the H atom of C2H6 molecule is directly pointed

o the Ni atom with bond length about 1.729 and 1.730 A forlNNT/Ni-C2H6 and SiCNT/Ni-C2H6 systems respectively. These

23 2.00 1.96 3.86 3.8671 0.29 0.29 0.57 0.5740 3.43 3.41 1.74 1.74

results are related to the fully optimize geometry of H-orientationof C2H4/C2H6 molecule adsorbed onto the nanotube/Ni systems.It is noteworthy that, there is tiny difference between H- and V-orientations of ethane adsorbed onto the relaxed geometries ofNi-doped nanotubes, whereas two stable structures with more dif-ference orientations are obtained for ethene adsorption onto thenanotube/Ni systems.

To better understand the adsorption properties of C2H6/C2H4adsorbed onto the Ni decorated nanotubes, the adsorption energiesare calculated using below equation and obtained data are listed inTable 1.

Eads = Etube/M−gas − (Etube/M + Egas) (7)

where Etube/M-gas denotes the total energy of the adduct Ni-decorated AlNNT and SiCNT with the corresponding molecules andEtube/m, Egas are the total energies of the Ni-decorated nanotubesand isolated ethane and ethene molecules respectively. Accord-ing to the adsorption energies in the Table 1, the interactionbetween C2H6 molecule and Ni decorated AlNNT and SiCNT arevery strong. The adsorption energy of C2H6 molecule onto theouter surface of Ni-doped AlNNT and SiCNT are about −61.74 and−96.53 kJ/mol respectively (using MPW1PW91 method), whichimplies to strongly physisorption process. The above results indi-cate that although ethane molecule adsorbed onto the pristineAlNNT and SiCNT through weak Van der Waals interaction, itcould be strongly physisorbed onto the outer surface of Ni-doped

AlN and SiC nanotubes with considerable negative adsorptionenergies. On the basis of adsorption energy results, the SiCNT/Nisystem is more favorable for C2H6 molecule adsorption com-pared with AlNNT/Ni system. Furthermore, the final optimization

562 Z. Mahdavifar et al. / Sensors and Actuators B 196 (2014) 555–566

ethen

gotostArscNcbCatnt

h

Fig. 4. Fully optimized geometrical structures of (a) ethane and (b)

eometry of H- and V-orientations of C2H6 molecule adsorbednto the Ni-decorated nanotubes is converted to the same struc-ures as shown in Figs. 4 and 5. The obtained adsorption energyf both H- and V-orientations of ethane molecule is nearly theame which confirm the above results. In case of C2H4 adsorp-ion, the adsorption energy of H-orientation of ethene onto thelNNT/Ni and SiCNT/Ni are obtained −179.32 and −178.18 kJ/molespectively (using MPW1PW91 method), which indicate that thetrong adsorption of ethene onto the Ni-decorated nanotubes arehemisorptions process. Hence, obtained result indicates that thei-doped nanotubes are more favorable for the ethene adsorptionompared to the ethane. Not that the interaction can be formedetween Ni atom and ethene molecule due to the �-electron of

C bond in ethene molecule transfer to unoccupied orbital of Nitom as well as the electron of “d” orbital of Ni atom feedback intohe �* of ethene. It seems that the functionalization of AlN and SiC

anotubes with an external impurity such as Ni atom can modifyhe electrical and mechanical characters of AlN and SiC nanotubes.

In short, compared the adsorption energy of nanotube/Ni-ydrocarbon and nanotube/hydrocarbon systems indicate that

e adsorption onto the AlNNT/Ni system using MPW1PW91 method.

functionalized nanotubes with Ni atom are more favorable thanpristine AlN and SiC nanotubes for ethane/ethene-detecting. Fur-thermore, Ni- doped AlN/SiC nanotube is a promising candidate formolecular adsorption of small hydrocarbon molecules due to theirexcellent sensing capabilities.

The sensing mechanism of a nanosensor is related to the changein the electrical conductance, which the electrical conductancechanges also can be directly related to changes in electronic bandstructure and partial electron charge transfer between gas moleculeand nanotube [55]. Therefore, to better understand the ability of Nidecorated nanotubes to sensing small hydrocarbon molecules, NBOanalysis are also performed. The charge transfer from Ni atom tothe AlN and SiC nanotubes in nanotube/Ni systems make the Niatom positively charged. Bonding of Ni atom to the outer surfaceof AlN and SiC nanotubes leads to substantial charge transfer fromNi atom to the nanotubes. The analysis of partial atomic charges

of nanotube/Ni-hydrocarbon systems are collected in Table 2.Upon the C2H4/C2H6 molecule adsorbed onto the nanotube/Ni sys-tems, the negative charge on C atom of C2H4/C2H6 molecule areobserved for H-orientation, which means that there is a charge

Z. Mahdavifar et al. / Sensors and Actuators B 196 (2014) 555–566 563

ethen

tINimtaanoccooma

Fig. 5. Fully optimized geometrical structures of (a) ethane and (b)

ransfer from nanotubes to the C2H4/C2H6 molecule was occurred.n other words, because of inserted C2H4/C2H6 molecule into thei-decorated nanotubes, the direction of the charge transfer was

nverted (from nanotubes to Ni atom and then to the C2H4/C2H6olecule) compared with Ni-doped nanotubes (from Ni to nano-

ubes). However, it is seems that during the adsorption of ethanend ethene hydrocarbons onto the Ni-doped nanotubes, consider-ble atomic charge is transferred between different ingredients ofanotube/Ni/hydrocarbon systems which fullfills the mechanismf the sensing condition. The obtained data of partial charges indi-ate that the electrostatic interaction is one of the major factorsontributed to the overall stabilities of the adsorption C2H4/C2H6

nto the nanotube/Ni systems. In addition, the analysis of bondrder values which are collected in Table 4, show that the C2H4olecule tend to be strongly adsorbed into the Ni-decorated AlN

nd Sic nanotubes with bond order about 0.458 and 0.462 (using

e adsorption onto the SiCNT/Ni system using MPW1PW91 method.

MPW1PW92 method) respectively. As can be seen in Fig. 4b andFig. 5b, the C2H4 molecule bonded to the Ni-decorated AlN and SiCnanotubes with two carbon atoms C(1) and C(2). The bond order ofC2H4 adsorbed onto the AlN/Ni and SiC/Ni systems for one C Ni is0.458 and 0.462 respectively (presented in Table 4). In other words,the total bond orders of C2H4 adsorbed onto the AlN/Ni and SiC/Nisystems are about 0.916 and 0.924 respectively, which suggestthat the C2H4 molecule strongly adsorbed onto the Ni-decoratedAlN and SiC nanotubes. On the other hand, the adsorption energygained for these systems is about −179 kJ/mol, which is in wellagreement with bond order results. Thus, we can conclude that theadsorption of C2H4 molecule onto the AlN/Ni and SiC/Ni systems is

chemisorption. In the case of C2H6, the gas molecule pointed to theNi-decorated AlN nanotube with two C(1) Ni and H(1) Ni inter-actions with the total bond order about 0.292. Compared this resultwith obtained adsorption energy (−61.74 kJ/mol), which imply to

564 Z. Mahdavifar et al. / Sensors and Actuators B 196 (2014) 555–566

thane

pCwa(aNi

atoeiihOh

Fig. 6. Total DOS spectra of H-orientation of (a) AlN/Ni-ethane (b) SiC/Ni-e

hysisorption process (see page 16 in the main text). In the case of2H6, the gas molecule pointed to the Ni decorated AlN nanotubeith two C(1) Ni and H(1) Ni interactions with total bond order

bout 0.292. Compared this result with obtained adsorption energy−61.74 kJ/mol) imply to physisorption process. These results arelso confirmed that the interaction between C2H4 molecule andi-decorated AlN and SiC nanotubes are stronger than that of C2H6

nteractions.To check the sensitivity of the Ni-doped nanotube for the

dsorption of ethane/ethene molecule, the electronic properties ofhe considered configurations are analyzed from their total densityf states (DOS) spectra. The total DOS spectra of nanotube/Ni-thene and nanotube/Ni-ethane systems are shown in Fig. 6. Its obvious from this figure that no significant change is occurred

n the overall feature of nanotube DOS spectra when these smallydrocarbons adsorbed onto the AlN and SiC nanotubes (see Fig. 3).n the other hand, compared the DOS spectra of nanotube/Ni-ydrocarbon systems with pristine nanotubes demonstrate a

(c) AlN/Ni-ethene (d) SiC/Ni-ethene systems using MPW1PW91 method.

dramatically mutation in Fermi level during adsorption processis occurred, which imply to change in HOMO-LUMO gap energy(compared Figs. 3 and 6). The HOMO-LUMO gap energies of con-sidered configurations are presented in Table 3. Compared theHOMO-LUMO gap energy of Ni-decorated nanotubes (see TableS3 in the Supplementary Materials and our previous researchref. [35]) with nanotube/Ni-hydrocarbon systems indicate thatwhen C2H6/C2H4 adsorption onto the Ni-doped nanotubes, the gapenergy of new configuration is increased (from 3.12 eV for SiC/Ni to3.39 eV for SiC/Ni-C2H6 and 3.91 eV for AlN/Ni to 4.14 eV for AlN/Ni-C2H6 respectively). These results indicate that the reactivity ofnanotube/Ni-hydrocarbon systems will be significantly decreased.It should be noted that if the gap energy of a system is increased, thismeans that the reactivity of system is decreased. On the other hand,

compared HOMO-LUMO gap energies of nanotube/hydrocarbonsystems with nanotube/Ni-hydrocarbon systems reveal that thegap energies of nanotube/Ni-hydrocarbon are slightly decreasedwhich indicate that the reactivity of the systems are increased

Z. Mahdavifar et al. / Sensors and Actuators B 196 (2014) 555–566 565

Table 4The WBI bond order characters of nanotube/hydrocarbon and nanotube/Ni–hydrocarbon systems (atom numbering is according to Figs. 1, 4 and 5).

C2H6 C2H4

MPW1PW91 PEBPEB MPW1PW91 PEBPEB

H V H V H V H V

AlN/gasAl(1) N(1) 0.96 0.96 0.98 1.00 0.91 0.95 0.90 0.95Al(1) N(2) 0.98 0.96 0.93 0.98 1.03 0.95 1.03 0.96Al(1) N(3) 1.00 1.03 0.95 1.01 0.91 1.06 0.90 1.06Al(1) C(1) 0.04 0.04 0.05 0.06 0.02 0.02 0.04 0.03Al(1) C(2) – – – – 0.02 – 0.03 –

SiC/gasSi(2) C(1) 1.09 1.09 1.08 1.07 1.07 1.09 1.05 1.17Si(2) C(2) 1.09 1.17 1.08 1.16 1.15 1.19 1.13 1.08Si(2) C(3) 1.17 1.09 1.15 1.07 1.07 1.09 1.05 1.08Si(2) C(5) 0.01 0.01 0.01 0.02 0.04 0.01 0.04 0.00Si(2) C(6) – – – – 0.04 – 0.04 –

AlN-Ni/gasAl(1) N(1) 0.475 0.477 0.483 0.483 0.466 0.508 0.479 0.477Al(2) N(1) 0.474 0.473 0.480 0.482 0.467 0.506 0.481 0.481Al(1) Ni 0.289 0.291 0.311 0.310 0.246 0.211 0.258 0.288Al(2) Ni 0.288 0.287 0.315 0.316 0.218 0.211 0.271 0.295N(1) Ni 0.579 0.579 0.613 0.613 0.521 0.346 0.562 0.581C(1) Ni 0.169 0.169 0.202 0.201 0.458 0.439 0.457 0.378H(1) Ni 0.125 0.125 0.110 0.108 – – – 0.144

SiC-Ni/gasC(1) Si(1) 0.695 0.696 0.834 0.908 0.775 0.702 0.784 0.784C(1) Ni 0.429 0.429 0.256 0.217 0.338 0.428 0.350 0.350

0.10.10.0

(bfcnrennostt

bcsoqpTo

cadei

4

atN

Si(1) Ni 0.526 0.527 0.325

C(2) Ni 0.123 0.123 0.111

H(1) Ni 0.103 0.103 0.05

see Table 3). These results are in well agreement with apprecia-le adsorption energies and considerable charge transfer obtainedrom NBO calculation. The global reactivity indices in the con-eptual of the DFT for C2H4/C2H6 adsorption on Ni-decoratedanotubes are summurized in Table 3. According to the globaleactivity indices (presented in Table 3), it is found that sincethane/ethene adsorbed onto the nanotube/Ni systems, the hard-ess values of these systems are decreased compared to pristineanotubes. Note that based on maximum hardness principle (MHP)f Ralph G. Person [56], the system with more hardness has moretability. Therefore, these hardness value variations suggest thathrough ethane and ethene interaction with Ni-AlN/SiC systemshe reactivity of nanotubes significantly will be increased.

It is noteworthy the influence of adding polarization and diffuseasis functions is also checked using single point calculations for allonsidered structures with the 6-31 + G* basis set. The results areummarized in Tables S4 and S5 in supplementary materials. It isbvious from these tables that only limited impacts on the reporteduantities are observed. The orders of physio or chemisorptionrocesses as well as electronic properties are also unchanged.herefore these alternative results would only verify the originalbtained data.

Based on our obtained results from adsorption energy, NBO cal-ulations as well as global reactivity indices, the Ni-decorated AlNnd SiC single walled nanotubes allow the fabrication of single-chipevice as a C2H4/C2H6 sensor because of appreciable adsorptionnergies, significant charge transfer as well as considerable changen electronic properties of AlN/Ni, and SiC/Ni nanotubes.

. Conclusion

Two different density functional approaches including PBEPBEnd MPW1PW91 methods are employed to systematically inves-igate the adsorption of ethane and ethene onto the pristine andi-doped AlN and SiC nanotubes. The geometrical structures,

42 0.269 0.509 0.299 0.29807 0.462 0.418 0.456 0.45646 0.024 .018 0.024 0.024

electronic properties and natural bond orbital (NBO) analysis areperformed. Based on adsorption energy, structural parametersand electronic properties, C2H6/C2H4 molecule adsorbed onto thepristine AlNNT and SiCNT through weak Van der Waals interac-tions, which means that the adsorption is physisorption process.In addition, the detailed analysis of NBO as well as density of state(DOS) spectra for nanotube/hydrocarbon systems indicate that theadsorption of ethane/ethene onto the considered nanotubes couldnot affect the electronic and structural properties of pristine nano-tubes. To improve the sensitivity of pristine AlNNT and SiCNT withrespect to ethane/ethene detecting, our results release that thefunctionalization of these nanotubes with Ni metal atom is a goodlayout because the ethane/ethene molecule tends to be stronglyadsorbed onto the Ni-decorated nanotubes while no bond breakingin the gas molecule is observed. The NBO analysis reveals that acharge transfer from Ni atom to AlN and SiC has occurred. It is foundevidence for the adsorption of ethane/ethene on Ni-doped nano-tubes is accompanied with considerable charge transfer betweenNi-decorated AlN and SiC and the C2H6/C2H4 molecule, which sig-nificantly changes the gap energy and electronic properties of thesenanotubes. It seems that our results can provide an alternativestrategy to modify the properties of AlN and SiC nanotubes, whichmight be useful for the design of AlN and SiC nanotube-based nan-odevices for adsorbed and storage of small hydrocarbon molecules.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.snb.2014.02.048.

References

[1] N. Yamazoe, Toward innovations of gas sensor technology, Sens. Actuators B108 (2005) 2–14.

[2] J. Kong, N.R. Franklin, C. Zhou, M.G. Chapline, S. Peng, K. Cho, H. Dai, Nanotubemolecular wires as chemical sensors, Science 287 (2000) 622–625.

5 nd Act

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

supervisor Associate Prof. Siamak Noorizadeh, both at the Chemistry Department

66 Z. Mahdavifar et al. / Sensors a

[3] Y. Cui, Q. Wei, H. Park, C.M. Lieber, Nanowire nano sensors for highly sensitiveand selective detection of biological and chemical species, Science 293 (2001)1289–1292.

[4] F. Schedin1, A.K. Geim, S.V. Morozov, E.W. Hill, P. Blake, M.I. Katsnelson, K.S.Novoselov, Nat. Mater 6 (2007) 652–655.

[5] P. Pannopard, P. Khongpracha, M. Probst, J. Limtrakul, Gas sensing propertiesof platinum derivatives of single-walled carbon nanotubes: A DFT analysis, J.Mol. Graphics Modell. 28 (2009) 62–69.

[6] L. Duclaux, Review of the doping of carbon nanotubes (multiwalled and single-walled), Carbon 40 (2002) 1751–1764.

[7] E.S. Snow, F.K. Perkins, J.A. Robinson, Chemical vapor detection using singlewalled carbon nanotubes, Chem. Soc. Rev. 35 (2006) 790–798.

[8] D.R. Kauffman, A. Star, Carbon nanotube gas and vapor sensors, Angew. Chem.Int. Ed. 47 (2008) 6550–6570.

[9] S. Iijima, Helical microtubules of graphitic carbon, Nature 354 (1991) 56–58.10] P.G. Collins, K. Bradley, M. Ishigami, A. Zettl, Extreme oxygen sensitivity of

electronic properties of carbon nanotubes, Science 287 (2000) 1801–1804.11] J. Li, Y.J. Lu, Q. Ye, M. Cinke, J. Han, M. Meyyappan, Carbon nanotube sensors for

gas and organic vapor detection, Nano Lett. 3 (2003) 929–933.12] A. Goldoni, R. Larciprete, L. Petaccia, S.J. Lizzit, Single-wall carbon nanotube

interaction with gases: sample contaminants and environmental monitoring,J. Am. Chem. Soc. 125 (2003) 11329–11333.

13] J. Kong, M.G. Chapline, H. Dai, Functionalized carbon nanotubes for molecularhydrogen sensors, Adv. Mater. 13 (2001) 1384–1386.

14] S.Y. Jeong, J.Y. Kim, H.D. Yang, B.N. Yoon, S.H. Choi, H.K. Kang, C.W. Yang, Y.H.Lee, Synthesis of silicon nanotubes on porous alumina using molecular beamepitaxy, Adv. Mater. 14 (2003) 1172–1176.

15] J. Hu, Y. Bando, Z. Liu, J. Zhan, D. Golberg, T. Sekiguchi, Synthesis of crystallinesilicon tubular nanostructures with ZnS nanowires as removable templates,Angew. Chem. Int. Ed. 43 (2004) 63–66.

16] M.D. Crescenzi, P. Castrucci, M. Scarselli, M. Diociaiuti, P.S. Chaudhari, C. Bal-asubramanian, T.M. Bhave, S.V. Bhoraskar, Experimental imaging of siliconnanotubes, Appl. Phys. Lett. 86 (2005) 231901–231903.

17] P. Castrucci, M. Scarselli, M.D. Crescenzi, M. Diociaiuti, P.S. Chaudhari, C. Bal-asubramanian, T.M. Bhave, S.V. Bhoraskar, Silicon nanotubes: synthesis andcharacterization, Thin Solid Films 508 (2006) 226–230.

18] C. Mu, Q. Zhao, D. Xu, Q. Zhuang, Y.J. Shao, Silicon nanotube array/gold elec-trode for direct electrochemistry of cytochrome c, Phys. Chem B 111 (2007)1491–1495.

19] M.H. Park, M.G. Kim, J. Joo, K. Kim, J. Kim, S. Ahn, Y. Cui, J. Cho, Silicon nanotubebattery anodes, Nano Lett. 9 (2009) 3844–3847.

20] S. Jiuxu, Y. Yintang, L. Hongxia, G. Lixin, Z. Zhiyong, Electronic transport prop-erties of the armchair silicon carbide nanotube, J. Semiconductors 31 (2010)114003–114005.

21] M.W. Zhao, Y. Xia, F. Li, R.Q. Zhao, S.T. Lee, Strain energy and electronic struc-tures of silicon carbide nanotubes: density functional calculations, Phys. Rev.B 71 (2005) 085312–085317.

22] A. Fissel, B. Schroter, W. Richer, Low-temperature growth of SiC thin-films onSi and 6H–SiC by solid-source molecular-beam epitaxy, Appl. Phys. Lett. 66(1995) 3182–3184.

23] F. Cao, X. Xu, W. Ren, C. Zhao, Theoretical study of O2 molecular adsorption anddissociation on silicon carbide nanotubes, J. Phys. Chem. C 114 (2010) 970–976.

24] G. Mpourmpakis, G.E. Froudakis, G.P. Lithoxoos, J. Samios, SiC Nanotubes: anovel material for hydrogen storage, Nano Lett. 6 (2006) 1581–1583.

25] R.Q. Wu, M. Yang, Y.H. Lu, Y.P. Feng, Z.G. Huang, Q.Y. Wu, Silicon carbide nano-tubes as potential gas sensors for CO and HCN Detection, J. Phys. Chem. C 112(2008) 15985–15988.

26] G.H. Gao, H.S. Kang, First principles study of NO and NNO chemisorption onsilicon carbide nanotubes and other nanotubes, J. Chem. Theo. Comput. 4 (2008)1690–1697.

27] K. Rezouali, M.A. Belkhir, A. Houari, J. Bai, Ab initio study of confinement andsurface effects in hexagonal AlN nanotubes, Comput. Mater. Sci. 45 (2009)305–309.

28] D. Zhang, R.Q. Zhang, Theoretical prediction on aluminum nitride nanotubes,Chem. Phys. Lett. 371 (2003) 426–432.

29] V.N. Tondare, C. Balasubramanian, S.V. Shende, D.S. Joag, V.P. Godbole, S.V.Bhoraskar, M. Bhadbhade, Field emission from open ended aluminum nitridenanotubes, Appl. Phys. Lett. 80 (2002) 4813–4815.

30] M. Zhao, Y. Xia, Z. Tan, X. Liu, F. Li, B. Huang, Y. Ji, L. Mei, Strain energy and ther-mal stability of single-walled aluminum nitride nanotubes from first-principlescalculations, Chem. Phys. Lett. 389 (2004) 160–164.

31] M. Simeoni, S. Santucci, S. Picozzi, B. Delley, Defective aluminum nitride nano-tubes: a new way for spintronics? A density functional study, Nanotech. 117(2006) 3166–3174.

32] S.H. Lim, J.Y. Lin, Ab initio study of the hydrogen chemisorption of single-walledaluminum nitride nanotubes, Chem. Phys. Lett. 466 (2008) 197–204.

33] Y. Jiao, A.J. Du, Z.H. Zhu, V. Rudolph, S.C. Smith, A density functional theory

study of CO2 and N2 adsorption on aluminium nitride single walled nanotubes,J. Mater. Chem. 20 (2010) 10426–10430.

34] W.G. Liu, G.H. Chen, X.C. Huang, D. Wu, Y.P. Yu, DFT studies on the interactionof an open-ended single-walled aluminum nitride nanotube (AINNT) with gasmolecules, J. Phys. Chem. C 116 (2012) 4957–4964.

uators B 196 (2014) 555–566

35] Z. Mahdavifar, M. Haghbayan, Theoretical investigation of pristine and func-tionalized AlN and SiC single walled nanotubes as an adsorption candidate formethane, Appl. Surf. Sci. 263 (2012) 553–562.

36] Z. Zhou, J. Zhao, Y. Chen, P.R. Schleyer, Z. Chen, Energetics and electronicstructures of AlN nanotubes/wires and their potential application as ammoniasensors, Nanotech. 18 (2007) 424023–424029.

37] Z. Mahdavifar, N. Abbasi, E. Shakerzadeh, A comparative theoretical study ofCO2 sensing using inorganic AlN, BN and SiC single walled nanotubes, Sens.Actuators B 185 (2013) 512–522.

38] M.J. Frisch, et al. Gaussian03, Revision B. 03, Gaussian, Inc., Pittsburgh, PA, 2003.39] C. Adamo, V. Barone, Toward reliable adiabatic connection models free from

adjustable parameters, Chem. Phys. Lett. 274 (1997) 242–250.40] J.P. Perdew, J.A. Chevary, S.H. Vosko, K.A. Jackson, M.R. Pederson, D.J. Singh, C.

Fiolhais, Atoms, molecules, solids, and surfaces: applications of the generalizedgradient approximation for exchange and correlation, Phys. Rev. B 46 (1992)6671–6677.

41] J.P. Perdew, K. Burke, M. Ernzerhof, Generalized gradient approximation madesimple, Phys. Rev. B 77 (1996) 3865–3868.

42] W. Stevens, H. Basch, J. Krauss, Compact effective potentials and efficientshared-exponent basis sets for the first- and second-row atoms, J. Chem. Phys.81 (1984) 6026–6033.

43] P. Geerlings, F. De Proft, W. Langenaeker, Conceptual density functional theory,Chem. Rev. 103 (2003) 1793–1873.

44] R.G. Parr, R.A. Donelly, M. Levy, W.E. Palke, Electronegativity-density functionalviewpoint, J. Chem. Phys. 68 (1978) 3801–3807.

45] J.F. Janak, Proof that ∂E/∂ni = � in density-functional theory, Phys. Rev. B 18(1978) 7165–7168.

46] R.G. Parr, L. Szentpaly, S. Liu, Electrophilicity index, J. Am. Chem. Soc. 121 (1999)1922–1924.

47] J.X. Zhao, Y.H. Ding, Silicon carbide nanotubes functionalized by transi-tion metal atoms: a density-functional study, J. Phys. Chem. C 112 (2008)2558–2564.

48] M. Menon, E. Richter, A. Mavrandonakis, G. Froudakis, A.N. Andrio-tis, Structure and stability of SiC nanotubes, Phys. Rev. B 69 (2004)115332–115340.

49] M. Zhao, Y. Xia, X. Liu, Z. Tan, B. Huang, C. Song, L. Mei, First-principles cal-culations of AlN nanowires and nanotubes: atomic structures, energetics, andsurface states, J. Phys. Chem. B 110 (2006) 8764–8768.

50] R.L. Zhou, L. Wang, B.C. Pan, Elastic and melting properties of crystalline SiCnanotubes, J. Phys. Chem. C 114 (2010) 8199–8205.

51] A.W. Ehlers, E.J. Baerends, K. Lammertsma, Nucleophilic or electrophilic phos-phinidene complexes MLnPH; What makes the difference? J. Am. Chem. Soc.124 (2002) 2831–2838.

52] K.K. Pandey, G. Frenking, The Nature of the M E Bond: Theoreticalinvestigation of the molecules [(RO)3M E] (M = Mo, W; E = N, P, As, Sb,Bi; R = H, Me) and [(Me3CO)3Mo P], Eur. J. Inorg. Chem. 2004 (2004)4388–4395.

53] J.X. Zhao, Y.H. Ding, Silicon carbide nanotubes functionalized by transi-tion metal atoms: A density-functional study, J. Phys. Chem. C 112 (2008)2558–2564.

54] J.M. Zhang, H.H. Li, Y. Zhang, K.W. Xu, The influence of scalar relativistic effect onthe carbon monoxide adsorption onto small gold clusters, Physica E 43 (2010)2132–2137.

55] M.A. Ryan, A.V. Shevade, C.J. Taylor, M.L. Homer, M. Blanco, ComputationalMethods for Sensor Material Selection, Springer, 2009.

56] R.G. Pearson, The principle of maximum hardness, Acc. Chem. Res. 26 (1993)250–255.

Biographies

Zabiollah Mahdavifar was born in Kazerun, Iran (1978) He received his Ph.D. inComputational & Physical Chemistry under the supervisor of Prof. Amir Abbas Rafatiin 2008 from Bu-Ali Sina University in Iran. His Ph.D. thesis was molecular simula-tion and computational study of gas adsorption on nanotubes. Now, his research isfocused on molecular simulation and computational chemistry with special interestin prediction of novel nanomaterial for rechargeable battery, prediction of nanoma-terial for solar cells and gas sensing, adsorption and separation.

Maryam Haghbyan was born in Kazerun, Iran (1983). She >completed her B.Sc.in Physical Chemistry in 2008 from Chamran University and received her M.Sc. inComputational& Physical Chemistry under the supervision of Assist. Prof. ZabiollahMahdavifar from Chamran University in 2011.

Mahdi Abbasi born in Chadegan, Iran in 1986. He completed his B.Sc. in chemistryfield and received his M.Sc. degree in computational & physical chemistry under the

of Shahid Chamran University in Iran. His major scientific interests are electronicstructure computations including different aspects of molecular modeling and com-putational nano chemistry. At Present, he is studying in physical chemistry Ph.D.degree at the Chemistry Institute of Tehran University.