Studies on the Encapsulation of Fâ in Single ... - ACS Publications

Article pubs.acs.org/JPCA

Studies on the Encapsulation of F− in Single Walled Nanotubes of Different Chiralities Using Density Functional Theory Calculations and Car−Parrinello Molecular Dynamics Simulations P. Ravinder, R. Mahesh Kumar, and V. Subramanian* Chemical Laboratory, Council of Scientific and Industrial Research-Central Leather Research Institute, Adyar, Chennai 600 020, India S Supporting Information *

ABSTRACT: In this study, the encapsulation of F− in different nanotubes (NTs) has been investigated using electronic structure calculations and Car−Parrinello molecular dynamics simulations. The carbon atoms in the single walled carbon nanotube (CNT) are systematically doped with B and N atoms. The effect of the encapsulation of F− in the boron nitride nanotube (BNNT) has also been investigated. Electronic structure calculations show that the (7,0) chirality nanotube forms a more stable endohedral complex (with F−) than the other nanotubes. Evidence obtained from the band structure of CNT calculations reveals that the band gap of the CNT is marginally affected by the encapsulation. However, the same encapsulation significantly changes the band gap of the BNNT. The density of states (DOS) derived from the calculations shows significant changes near the Fermi level. The snapshots obtained from the CPMD simulation highlight the fluctuation of the anion inside the tube and there is more fluctuation in BNNT than in CNT.

■

INTRODUCTION It is found from previous studies that the encapsulated guest molecules influence properties of single walled carbon nanotubes (CNTs).1−3 To the best of our knowledge, the first publication on the encapsulation of foreign molecule inside CNTs appeared in 1998 by Green and co-workers.4 In this study, authors have described the chemical technique that can selectively open the tube and deposit the material within it.4 In the same year, Smith et al. reported that tubules prepared via pulsed laser vaporization (PLV) contained close-packed 1D chains consisting of C60 and other fullerene molecules (i.e., Cn).5 These encapsulated CNTs are referred to as “peapods” and are denoted as [email protected] In a subsequent study, Smith and co-workers have explained the strategies to fill the nanotubes.6 Quantitative filling of CNTs with specific fullerenes has been achieved by incorporation of endohedral metalofullerenes such as [La2@C80]@CNT and [Gd@C82]@CNT.7,8 Scanning tunneling microscopy (STM) studies on C60@CNT and [Gd@C82]@CNT have revealed significant band gap modulations and provided credibility to the argument that similar molecular materials could form components in solid state electronic devices.2,9 In similar lines, several types of molecules with increasing complexity have been encapsulated, from chains of atoms, to molecules, and to inorganic crystals.9,10 Ilie et al. have studied the encapsulation of KI in CNT and its effect on bond-length, charge transfer, and electronic density of states.11 They have observed that encapsulation of KI leads to strain that shrinks both the C−C bonds and the overall nanotube along the axial direction.11 Zettl and co-workers have also investigated the encapsulation of potassium halides in BN nanotubes (BNNTs).12 Their findings reveal that the BN nanotubes not only confine the halide crystals but also act © 2012 American Chemical Society

effectively as chemically and electrically inert protecting layers and could enhance the stability of potassium crystals in harsh chemical, thermal, and electrical/optical environments.12 Recently, Jindal and co-workers have studied the encapsulation of Al clusters in CNTs.13 It is found that lower metal clusters prefer linear geometry whereas higher metal clusters form 3D structure due to confinement.13 Furthermore, the metal−metal interactions are vital in the case of lower clusters whereas C− metal interactions are important in the case of larger clusters.13 The size-dependence of the encapsulation of C60 into CNTs and the applications of such intercalated molecules in nonvolatile memory devices have been predicted by electronic-structure calculations.14,15 The electronic properties of the different nanotubes and their functionalized forms have been investigated extensively.16−20 It is found from the band structure calculations that CNTs are metallic and semiconductor depending on their chirality. Recently, a number of studies have focused on BNNT due to its morphology being similar to that of CNT; however, they exhibit distinctly different properties.21 BNNTs are also found to be nontoxic to health and environment due to their chemical inertness and structural stability.22,23 It is well-known that electronic properties of CNT vary with diameter, chirality, concentric layers, etc. On the other hand, the same properties of BNNTs do not vary with their diameter and chirality.24 BNNT is semiconductor regardless of the diameter and chirality.25−27 In addition, the electrostatic potential inside the BNNT is different from that of CNT due to the electronegativity difference between Received: November 7, 2011 Revised: May 10, 2012 Published: May 14, 2012 5519

dx.doi.org/10.1021/jp210668b | J. Phys. Chem. A 2012, 116, 5519−5528

The Journal of Physical Chemistry A

Article

the B−N bond.28 Thus, it is of immense interest to explore the differences between the endohedral complex formed by the anion inside CNT and BNNT. Although significantly large quantity of literature exists in this field, the encapsulation of anions in CNTs and other NTs are scarce. Therefore, a systematic attempt has been made to understand the endohedral complex formation of F− inside CNT, BN-doped CNTs, and BNNTs using electronic structure calculations. Further, the effect of encapsulation of F−on the band structures of these tubes has also been investigated. In this study, the following points have also been addressed: (i) comparing the geometrical parameters and energies of endohedral complexes formed by F− and neutral F atom inside (5,0)CNT, BN-doped (5,0)CNT and (5,0)BNNTs; (ii) probing the effect of BN doping on the geometrical parameters and energies of endoohedral complex formation; (iii) comparing the geometrical parameters and energies of endohedral complexes formed by F− with (5,0)CNT, BNdoped (5,0)CNT and BNNTs with corresponding exhohedral counterparts; (iv) assessing the role of different chirality and types of NT on the geometrical parameters and energies of endohedral complex formation.

■

COMPUTATIONAL METHODS The geometries of all the encapsulated tubes were optimized without any geometrical constrain using density functional theory (DFT) based M05-2X method employing 6-31+G* basis set.29,30 The interaction energy (IE) was calculated using a supermolecule approach, applying the following equation: IE = Etotal − (E NT + E F−)

where Etotal is the total energy of the complex, ENT is the total energy of the corresponding nanotube, and EF− is the total energy of F− ion. The monomer energies were calculated from the respective monomer geometries in the complexes. The calculated IEs were corrected for the basis set superposition error (BSSE) using the counterpoise method suggested by Boys and Bernardi.31 All the calculations were performed using Gaussian 03 suit of programs.32 To understand the nature of interaction between NTs and the F−, the interaction energy decomposition analysis was carried out employing Ziegler−Morokuma energy decomposition analysis scheme.33−36 These calculations were performed using empirical dispersion corrected B3LYP-D employing Slater type basis DZP. The Amsterdam density functional (ADF) theory package was used for the decomposition analysis.37 AIM (Atoms in Molecule) Calculation. The atoms in molecule (AIM) approach was used for the characterization of the nature of the noncovalent interaction.38,39 The wave functions for all the systems were generated using HF/6-31+G* level of theory. The AIM calculations were carried out using AIM-2000 program.40 Band Structure calculations. The density of states and band structures of various systems were calculated using the Vienna ab initio simulation package (VASP).41,42 DFT with the generalized gradient approximation (GGA) of Perdew, Burke, and Ernzerhof (PBE) and the projector-augmented wave (PAW) method were used for the calculations.43−45 The PAW technique is essentially an all-electron approach and provides the full wave functions that are not directly accessible with the

Figure 1. Optimized geometries of various endohedral and exohedral complexes. Color key: gray = C, white = H, blue = N, pink = B, and cyan = F.

pseudopotential approach. One-dimensional (1-D) periodic boundary conditions (PBCs) were applied along the tube axis (z axis) to simulate infinitely long nanotube systems. Four unit cells of (5,0)CNT, and (5,0)BNNT along the z-axis were included to build the hexagonal supercell. A plane wave basis set, truncated with an energy cutoff of 400 eV, was adopted for all of the calculations. The positions of all of the atoms in the supercell were fully relaxed during the geometry optimizations. The convergence threshold was set to be 10−7 eV for the total energy in the electronic self-consistent-loop and 10−2 eV/Å of force on each atom. On the basis of the equilibrium structures, 1 × 1 × 9 k-points were then used to compute the electronic structures. CPMD Simulation. The M05-2X/6-31+G* optimized geometries were taken as the initial geometries for the Car− Parrinello molecular dynamics (CPMD) simulations. The complex was placed in the center of a cubic box with size of 40 Å. The Kohn−Sham orbitals were expanded in a plane wave basis set with a cutoff of 70 Ry, and Goedecker−Teter−Hutter dual space Gaussian Pseudopotentials were used.46 The equations of motion were integrated with a time step of 0.1 fs. Furthermore, all these calculations were performed in gas phase using 300 K temperature and 1 atm pressure. The trajectories were collected 5520



Article

Figure 2. Optimized geometries of F− encapsulated nanotubes of different chirality (nanotube endohedral complexes). Color key: gray = C, white = H, blue = N, pink = B, and cyan = F.

of the CNT and F/F−. It can be seen from the table that in all other NTs there is only negligible effect on the radius of the tubes upon encapsulation. For all NTs considered in this study, F− ion was kept at the center of gravity in initial guess for the geometry. However, during geometry optimization, in some cases F− shifts away from the center of the tube. The calculated distance of the F− ion from the center of gravity (DCg−F−) is given in the Table 1. Results show that the F− ion lies almost closer to the center of gravity in CNTs and BN substituted CNTs. The close analysis of the optimized geometries of BNNTs indicates that in these complexes, the F− ion is relocated closer to electron deficient B atom. This can be attributed to significant changes in the electron density distribution of BNNTs in comparison to that of CNTs. Due to the electronegativity difference between the boron and nitrogen atoms, charge localization is more on the nitrogen atom. As a result, the π-electron distribution in the B−N substituted unit is “lumpy” in comparison to the distribution for the carbon-based ring.28 It is also interesting to observe from the table that the F− remain at the center as diameter of tube increases. Therefore, it is interesting to mention that F− ion does not experience the impact of confinement as the diameter of NTs increases. Interaction Energies. Calculated interaction energies (IEs) at M05-2X level using 6-31+G* basis set for various exohedral and endohedral NT complexes are listed in Table 2. It can be seen from the table that the IE values of exohedral anionic complexes are significantly larger than those for anionic and neutral endohedral complexes. It is interesting to mention that F/F− encapsulation in (5,0)CNT is energetically not favorable. The trend in the stability of complexes varies as F−...(5,0)CNT > F@(5,0)CNT > F−@(5,0)CNT. Further, doping of CNTs by B and N atoms significantly alters the IEs. Doping of one BN pair in (5,0)CNT results a more stable complex than the pristine CNT.

for every ≈0.1 fs. The CPMD program package was utilized for the computation.47 Models and Notation. In this investigation, different zigzag CNTs with different chiralities such as (5,0), (6,0), (7,0), (8,0), (9,0), and (10,0) were selected. The molecular formulas of these CNTs are C50H10, C60H12, C70H14, C80H16, C90H18, and C100H20, respectively. The BNNTs with same chiralities were also selected for the study. Corresponding molecular formulas are B25N25H10, B30N30H12, B35N35H14, B40N40H16, B55N45H18, and B50N50H20. Armchair type of CNT and BNNT of (5,5) chirality were selected as models systems. Molecular formulas of these tubes are C100H20 and B50N50H100, respectively. The endohedral complexes formed upon encapsulation of F− and F were represented as F−@(m,n)NT and F@(m,n)NT, respectively. The exohedral complex formed by the F− ion is designated as F−...(m,n)NT.

■

RESULTS AND DISCUSSION Geometry. Optimized geometries of various endohedral and exohedral complexes are given in Figures 1 and 2. Because the sizes of all the systems considered in this investigation are significantly large to handle computationally, only a few model systems have been considered to probe the effect of encapsulation of F in a NT. They are (i) F@(5,0)CNT, (ii) F@1 BN-unit-doped (5,0)CNT, F@3-unit-doped (5,0)CNT, and F@(5,0)BNNT. Similarly, all these tubes were used for the formation of exohedral complexes. The calculated geometrical parameters of various NTs and their complexes using M05-2X/6-31+G* method are depicted in Table 1. It can be noted that the average radius of the (5,0)CNT is slightly influenced by the encapsulation of both F and F−. Typically, the average radius of the tube before encapsulation is 1.96 Å. The after encapsulation of F and F−, the radius becomes 2.04 and 2.03 Å, respectively. The increase in the radius is may be attributed to the repulsion between the π−electronic cloud 5521



Article

It can be noticed that the BN-doped CNTs and BNNTs show significantly different MESP when compared with those of the CNTs. Therefore, the role of electrostatic interaction involved in the stabilization of these complexes is different. It is evident from the IEs as obtained from M05-2X/6-31+G* level of theory that F− is not stable inside (6,0)CNT. Further, results illustrates that F− forms a stable endohedral complex inside (7,0)CNT with IE of −40.88 kcal/mol. It is interesting to observe from the results that the stability of the F− ion encapsulated BNNTs complex increases with the increase in the chirality of tube from (5,0) to (7,0). It can be see that the encapsulation of F− depends on the radius of the CNT. As the radius of the CNT increases, repulsion between the π-cloud of CNT and F− decreases. Consequently, F− forms a stable endohedral complex with (7,0)CNT. Moreover, calculated IE value decreases as chirality of tube increases from (7,0) to (10,0). Therefore, it is noteworthy to mention that the diameter of (7,0)CNT and (7,0)BNNT is more suitable for F− encapsulation, which facilitates the formation of highly stable endohedral complex when compared with other NTs with different chiralities. The IE values of F−@(5,5)CNT and F−@(5,5)BNNT are −30.07 and −52.9 kcal/mol, respectively. The IE values of these tubes are between (8,0) and (9,0) NTs ,which are in agreement with the trend in their respective diameter ((8,0)NT < (5,5)NT < (9,0)NT). Therefore, it is noteworthy to mention that the stability of F− encapsulated complexes is governed mainly by sterical factors and the stability is independent of type of NTs. The dipole moment of CNT is zero Debye. Both F−@(5,0)CNT and F−@(6,0)CNT have zero dipole moments due to the presence of F− ion at the center of gravity. The displacement of F− ion from the center of gravity alters the charge distribution and as a result F−@(7,0)CNT exhibits dipole moment of 1.07 D. Similarly, the substitution of BN units in CNT changes its dipole moment. The dipole moments of (5,0)CNT substituted with one and three BN units are 3.43 and 2.29 D, respectively, and the corresponding values after encapsulation of F−are 1.71 and 3.43 D. The dipole moments of (5,0), (6,0) and (7,0)BNNTs are 5.80, 7.95, and 9.89 D, respectively. The encapsulation of F− modulates these dipole moments to 6.24, 4.28, and 10.06 D, respectively. The location of the F− inside the NTs significantly affects the charge distribution and the resulting dipole−dipole interaction plays an important role in the stabilization of anion inside the NTs. Calculated NBO charge of encaged F−/F in various complexes is given in Supporting Information (Table S1). Charge analysis clearly demonstrates that significant electronic distribution in smaller nanotube endohedral complexes. It is noteworthy to mention that F− significantly donates the electron to the (5,0)

Table 1. Calculated Geometrical Parameters (Using M05-2X/ 6-31+G*) of Various Nanotubes and Their Encapsulated (Exohedral) Complexesa average radius of tube

length of tube NTs

pristine

F−@

pristine

F−@

Rcg‑F−

(5,0) CNT (6,0) CNT (7,0) CNT (8,0) CNT (9,0) CNT (10,0) CNT (5,5) CNT (5,0) BN NT (6,0) BN NT (7,0) BN NT (8,0) BN NT (9,0) BN NT (10,0) BN NT (5,5) BN NT 1BN-doped (5,0) CNT 3BN-doped (5,0) CNT

11.34 11.42 11.40 11.42 11.42 11.42 12.92 11.45 11.50 11.52 11.54 11.54 11.55 13.16 11.33 11.41

11.17 11.39 11.37 11.39 11.39 11.39 12.88 11.44 11.48 11.51 11.53 11.53 11.54 13.15 11.23 11.36

1.96 2.03 2.34 2.35 2.72 2.71 3.11 3.09 3.50 3.48 3.90 3.87 3.42 3.40 1.97 1.99 2.37 2.37 2.77 2.76 3.17 3.16 3.56 3.54 3.95 3.94 3.46 3.45 1.96 1.99 1.95 1.98 average radius of tube

0.00 0.00 0.36 0.00 0.00 0.00 0.00 0.63 0.61 0.55 0.01 0.00 0.00 0.00 0.68 0.76

NTs

pristine

F− on

pristine

F− on

Rcg‑F−

(5,0) CNT 1BN-doped (5,0) CNT 3BN-doped (5,0) CNT (5,0) BN NT

11.34 11.33 11.41 11.45

11.23 11.24 11.31 11.44

1.96 1.96 1.96 1.98 1.95 1.95 1.97 2.00 average radius of tube

3.63 3.76 3.89 3.82

length of tube

length of tube

a

NTs

pristine

F@

pristine

F@

Rcg‑F

(5,0) CNT 1BN-doped (5,0) CNT 3BN-doped (5,0) CNT (5,0) BN NT

11.34 11.33 11.41 11.45

11.29 11.18 11.28 11.45

1.96 1.96 1.95 1.97

2.04 2.01 2.00 1.99

0.00 0.52 0.62 0.58

All the distances are in Å.

Substitution of one BN pair enhances the IE by ∼75 kcal/mol when compared to the IE of the unsubstituted F−@(5,0)CNT complex. Similarly, substitution of three BN units in (5,0)CNT gives the IE of −33.81 kcal/mol. Therefore, doping of BN units significantly influences IE of F−@CNT due to the changes in the inner electron density distribution and electrostatic potential of CNT. Calculated molecular electrostatic potential (MESP) of various (5,0)NTs and corresponding endohedral and exohedral complexes are presented in Supporting Information (Figure S1).

Table 2. Interaction Energies of the F− Endohedral (Exohedral) Complexation with CNTs, BN-Doped NTs, and BNNTs with Different Chiralities Calculated Using M05-2X/6-31+G*a (5,0)NTs

a

−

other NTs

−

complex

F ...

F @

F@

complex

CNT

BNNT

(5,0) CNT 1BN-doped (5,0) CNT 1BN-doped (5,0) CNT (5,0) BNNT

−68.25 −120.29 −138.29 −129.79

47.94 −26.3 −33.81 −46.7

32.37 −66.92 −56.63 −9.00

F−@(6,0) F−@ (7,0) F−@ (8,0) F−@ (9,0) F−@ (10,0) F−@ (5,5)

0.57 −40.8 −37.82 −29.11 −21.96 −30.07

−76.29 −75.43 −61.15 −47.34 −40.73 −52.9

All the energies are in kcal/mol. 5522



Article

Table 3. Interaction Energy Components of F− Endohedral (Exohedral) Complexation with CNTs, BN-Doped NTs, and BNNTs of Different Chiralities (Calculated Using B3LYP-D/DZP)a VPauli

complexes

a

F−@(5,0) CNT F−@(6,0) CNT F−@(7,0) CNT F−@(5,5) CNT F−@(5,0) BNNT F−@(6,0) BNNT F−@(7,0) BNNT F−@(5,5) BNNT F−@1BN-doped (5,0) CNT F−@3BN-doped (5,0) CNT

309.95 129.14 41.44 4.15 286.59 103.22 33.21 3.97 415.88 415.45

F−...(5,0) CNT F−...1BN-doped (5,0) CNT F−...3BN-doped (5,0) CNT F−...(5,0) BNNT

366.72 216.93 230.25 189.89

EElec

ESteric

Endohedral Complexes −170.78 139.17 −76.45 52.69 −44.83 −3.40 −11.13 −6.98 −224.74 61.85 −116.63 −13.41 −69.92 −36.71 −38.19 −34.22 −268.36 147.52 −275.50 139.96 Exohedral Complexes −182.71 184.01 −169.02 47.91 −182.15 48.10 −170.07 19.82

EOrbital

EDispersion

IE

−109.67 −62.17 −46.40 −30.44 −118.57 −68.85 −47.90 −33.70 −189.56 −188.86

−6.98 −8.56 −10.42 −5.86 −6.60 −9.90 −10.35 −5.93 −5.86 −5.63

22.52 −18.04 −60.22 −43.29 −63.33 −92.16 −94.95 −73.86 −47.90 −54.54

−270.24 −186.24 −198.77 −188.02

−2.52 −2.50 −2.85 −2.90

−88.75 −140.83 −153.51 −146.21

All the energies are in kcal/mol.

Table 4. Calculated HOMO−LUMO Energies (Using M05-2X/6-31+G* Level of Calculation)a F−...

pristine CNTs (5,0)CNT 1 BN-doped (5,0)CNT 3 BN-doped (5,0)CNT BNNT

a

EHOMO

ELUMO

EHOMO

ELUMO

EHOMO

−5.20 −5.92 −6.84 −8.36 pristine

−2.79 −3.05 −3.27 −1.89

−2.77 −2.85 −2.58 −5.10

0.13 0.05 0.37 1.04

−2.70 −2.58 −2.21 −5.45

F−@t

F−@

F@ ELUMO 0.16 0.03 0.18 1.15 pristine

EHOMO

ELUMO

−5.09 −6.38 −6.02 −8.57

−2.43 −2.58 −2.48 −1.73

F−@t

CNTs

EHOMO

ELUMO

EHOMO

ELUMO

BNNTs

EHOMO

ELUMO

EHOMO

ELUMO

(6,0)CNT (7,0)CNT (8,0)CNT (9,0)CNT (10,0)CNT (5,5)CNT

−4.28 −4.96 −4.50 −4.54 −4.71 −5.57

−2.60 −3.12 −3.71 −3.73 −3.63 −2.55

−1.19 −1.88 −2.14 −2.19 −2.50 −1.40

0.19 −0.99 −1.19 −1.39 −1.40 −0.04

(6,0)BNNT (7,0)BNNT (8,0)BNNT (9,0)BNNT (10,0)BNNT (5,5)BNNT

−8.47 −8.44 −8.36 −8.43 −6.50

−1.18 −0.68 −0.43 −0.54 −0.46

−5.64 −5.81 −5.84 −5.86 −4.70 −5.35

1.67 1.66 1.65 1.69 1.72 2.26

All the energies are in electronvolts.

repulsion between the filled orbitals of the two systems is very high. The calculated ESteric value is larger than EOrbital + EDispersion. Therefore, it is evident that F− encapsulation in the (5,0)SWCNT is not feasible due to steric interaction. Close analysis of the F− encapsulation in (6,0)CNT provides that the ESteric value is 52.69 kcal/mol and the sum of EOrbital and EDispersion values is −70.73 kcal/mol; therefore, the encapsulation F− is energetically favorable and the respective IE is −18.04 kcal/ mol. Same value obtained from M05-2X/6-31+G* level of calculation is 0.57 kcal/mol. To understand the inconsistency in results, we have calculated IE values using M05-2X employing different basis sets (6-31++G*, 6-311+G*, 6-311++G*, and 6311++G**) and results are given in Supporting Information (Figure S2). It can be noted that M05-2X predicts only positive IE values. Therefore, the inconsistency between M05-2X and B3LYP-D method may due to the differences in the functionals. The validation of the prediction of these functionals calls for generation of high quality experimental results on these systems and further theoretical investigations. The IE components clearly show that the repulsion arises due to the encapsulation of F− in (7,0)CNT is significantly less (−3.40 kcal/mol). The calculated ESteric values for the (5,0),

and (6,0) NTs as compared with the case of larger nanotubes. It can be seen that donation of electron from F− to (7,0)NT is marginally less than that of both (5,0) and (6,0)NTs. Overall the charge transfer between the two systems plays a significant role in the stabilization of these complexes. To understand the nature of the interaction between the two systems, energy decomposition analysis was carried out using B3LYP-D/DZP method including dispersion correction. Energy Decomposition Analysis. Various energy components such as Pauli repulsion (VPauli), electrostatic interaction (EElec), orbital interaction (EOrbital), and dispersion interaction (EDispersion) are listed in Table 3. Further, the ESteric component is combination of Pauli repulsion (VPauli) and electrostatic interaction (EElec). The IE is given as the algebraic sum of ESteric, EOrbital, and EDispersion. It can be seen from Tables 2 and 3 that the IE obtained using the M05-2X/6-31+G* level of theory is lower than that from the B3LYP-D/DZP level of theory due to the differenct functionals used in the calculation. The B3LYP−D/ DZP calculation reveals that F− can be encaspsulated inside the (6,0)-SWCNT with the IE of −18.04 kcal/mol. The IE components of F−@(5,0)CNT reveals that the EPauli contribution is 309.95 kcal/mol, which clearly indicates that the 5523



Article

Figure 3. Electron density contour maps of various (5,0)NTs and their complexes.

NTs are lower than those of the F− encapsulated (and exohedral interaction) NTs. The encapsulation of F shows only marginal effect on the frontier orbital energies. Typically, the isolated HOMO and LUMO energies of isolated (5,0)CNT are −5.20 and −2.79 eV, respectively. The same values for the F−@(5,0)CNT, F@(5,0)CNT, and F−...(5,0)CNT are −2.70, −0.16, −5.09, −2.43, −2.77, and +0.13 eV, respectively. It is interesting to note that the difference of HOMO and LUMO energies decreases for the BN-doped systems when compared to those of the isolated systems. But in the cases of BNNTs, the difference between HOMO and LUMO energies increases as diameter of CNT increases (except (10,0)BNNT). AIM Analysis. Calculated AIM molecular graphs for the endohedral (exohedral) complexes of various (5,0)NTs are given in Supporting Information along with electron density (ρ(rc)) and its Laplacian (∇2ρ(rc)) at bond critical points (BCPs) (Figure S3 and Table S2). It can be noticed that the ρ(rc) and ∇2ρ(rc) values are higher for the exohedral complexes than for

(6,0), and (7,0)BNNTs are +61.85, −13.41, and −36.71 kcal/ mol, respectively. Overall results elicit that steric interaction plays predominant role in the stabilization of endohedral complexes. The stepwise doping of BN units in (5,0)CNT improves the endohedral anion binding affinity. It can be seen from Table 3 that the ESteric contributions of F−@1BN-substituted (5,0)CNT and F−@3BN-substituted (5,0)CNT complexes are higher than those of the F−@(5,0)CNT. It is evident from the EOrbital contributions that the maximum orbital overlapping between the two systems stabilizes the endohedral complexes. Furthermore, the involvement of dispersion interaction is significant in the formation of anionic endohedral complexes involving CNTs in comparison to BN-doped CNTs. Frontier Orbitals. The energies of HOMO and LUMO of the isolated systems were calculated using the geometries of endohedral complexes at M05-2X/6-31+G* level. The HOMO and LUMO energies of various NTs and their complexes are depicted in Table 4. The HOMO and LUMO energies of isolated 5524



Article

Figure 4. Calculated band structure and density of states (DOS) for pristine and F−encapsulated (5,0)CNTs and (5,0)BNNT.

Figure 5. Potential energy surfaces of various systems.

(5,0)CNT clearly unveil the interaction between B and F− in complexes. Electron Density of States (DOS) and Band Structure. Calculated band structure and density of states (DOS) for pristine and F− encapsulated (5,0)CNTs and (5,0)BNNT are depicted in Figure 4. The pristine (5,0)CNT shows the smaller energy gap between valence and conduction band in comparison to that of (5,0)BNNT. Typically, calculated band gaps for the (5,0)CNT and (5,0)BNNT are 0.22 and 3.35 eV, respectively. With a view to understand the influence of ion encapsulation on

the endohedral complexes. Furthermore, the high interaction tendency of F− with doped B atoms can be seen from the molecular graphs. The calculated ∇2ρ(rc) values clearly elicit the noncovalent nature between F−/F with NTs. The electron density contour maps of all (5,0)NTs and their complexes are given in Figure 3. Electron density contour maps clearly differentiate the electron density distribution on CNT and BNNTs. The contour maps of complexes reveal that the electron density distribution between F−/F and NTs. The contour maps of F−@1 BN-unit-doped (5,0)CNT and F−@3 BN-unit-doped 5525



Article

Figure 6. Snapshots of F−@ various (5,0) NTs during CPMD run at different time.

Hence, CPMD simulations were performed on the anion inside the various (5,0)-NTs. The variation of total energy of the complexes with time is plotted in Figure 5. It can be noted that the anion fluctuates inside the CNT around the center of the tube. With the doping of CNT by BN, the periodicity in the fluctuation increases. Various snapshots obtained from the CPMD simulation are presented in Figure 6. Evidence shows that anion does not lie at the center of the CNT. However, energetics information obtained from the CPMD simulations elicits that the endohedral phase of the tube can also be involved in the formation of anion···π complexes. From the trajectories collected from the CPMD simulations, the radial distribution function (g(r)) was calculated. The calculated g(r) for the various systems are depicted in Figure 7. In the case of F−@(5,0)CNT the gC−F−(r) shows broad peak between ≈2−5 Å, which clearly emphasizes the contact between the C and F− . All the NTs containing B atom shows maxima between ≈1−2 Å corresponding to gB−F− (r). The maxima corresponding to the gN−F−(r) is greater than 2 Å. These findings clearly reveal that the contact between the B−F− is stronger than the N−F−. Thus the B−F− contacts govern the stability of the BN-doped NTs and BNNTs.

the electronic structure of nanotubes, band structures for same nanotubes have been calculated. The calculated band gap for (5,0)CNT with F− is 0.27 eV. Thus, the encapsulation of F− in (5,0)CNT does not significantly influence the band gap of CNT. However, it can be seen from Figure 4, due to encapsulation of F− additional states near the Fermi level are observed in the F−@(5,0)CNT (red lines in Figure 4). The calculated band gap of F− encapsulated BNNT is 2.91 eV. Comparison of band gaps of (5,0)BNNT and F−@(5,0)BNNT shows that the encapsulation of F− inside BNNT significantly alters its band gap and conductivity. Similar to CNT, additional states have also been found near the Fermi level of encapsulated BNNT. The calculated DOS is given Figure 4. It shows that density of states near Fermi level undergoes changes upon encapsulation. Therefore, the conductivity of the tube enhances upon the F− encapsulation. Overall, it is clear from Figure 4 that the band structures of ion encapsulated nanotubes are superposition of pristine nanotube band structures. The weak interaction between these two systems (from IE calculation) reinforces this observation. CPMD of F−@ various (5,0) NTs. The investigation of anion dynamics in the confinement is another interesting issue. 5526



Article

Figure 7. Radial distribution functions of all possible dissimilar combination for the various complexes of different NTs.

■

F−@(6,0)CNT complex using M05-2X employing various basis sets. This material is available free of charge via the Internet at http://pubs.acs.org.

CONCLUSIONS It can be noted that the diameter of the NT plays vital role in the energetics of the formation of nano peapods with anionic seeds. Evidence from the decomposition analysis highlights the nature of interactions. It can be observed that the stability of F− inside CNTs with small diameters is exclusively destabilized by the steric interaction, whereas the same in the case of BNNTs is governed by the electrostatic and orbital interactions. The exohedral complexes formed by F− with CNT is more stable than the corresponding endohedral counterpart. The encapsulation of atomic F does not form stable complex with (5,0)CNT. However, it forms stable complexes with BN-doped CNTs and BNNT. The type of CNT does not significantly influences formation of endohedral complex. The scrutiny of band gap and electron density of state elucidates the significant modulation in the electronic properties of the BNNT upon encapsulation. The density of the states near the Fermi level of the complexes undergoes appreciable changes upon encapsulation in comparison with that cases of isolated tubes. An increase in the conductivity of the NTs is evident from the band gap and DOS. Results from the CPMD simulation reveal that the anion fluctuates inside the CNT around the center of the tube.

■

■

AUTHOR INFORMATION

Corresponding Author

*Tel.: +91 44 24411630. Fax: +91 44 24911589. E-mail: [email protected], [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank the Center of Excellence for Computational Chemistry (Project No. NWP-53), Council of Scientific and Industrial Research (CSIR), New Delhi, for financial support. P.R. and R.M.K. thank CSIR, New Delhi, India, for the award of Senior Research Fellowship.

■

REFERENCES

(1) Zhou, C.; Kong, J.; Yenilmez, E.; Dai, H. Science 2000, 290, 1552. (2) Lee, J.; Kim, H.; Kahng, S. J.; Kim, G.; Son, Y. W.; Ihm, J.; Kato, H.; Wang, Z. W.; Okazaki, T.; Shinohara, H.; Kuk, Y. Nature 2002, 415, 1005. (3) Yang, S. H.; Shin, W. H.; Kang, J. K. J. Chem. Phys. 2006, 125, 084705. (4) Sloan, J.; Hammer, J.; Zweifka-Sibley, M.; Green, M. L. H. J. Chem. Soc., Chem. Commun. 1998, 347. (5) Smith, B. W.; Monthioux, M.; Luzzi, D. E. Nature 1998, 396, 323.

ASSOCIATED CONTENT

S Supporting Information *

Calculated chargers of various complexes, AIM data, AIM molecular graphs and MESP of (5,0)CNT, BN-doped (5,0)CNTs, and (5,0)BNNT and interaction energies of 5527



Article

(6) Smith, B. W.; Luzzi, D. E. Chem. Phys. Lett. 2000, 321, 169. (7) Smith, B. W.; Luzzi, D. E.; Achiba, Y. Chem. Phys. Lett. 2000, 331, 137. (8) Suenaga, K.; Tence, M.; Mory, C.; Colliex, C.; Kato, H.; Okazaki, T.; Shinohara, H.; Hirahara, K.; Bandow, S.; Iijima, S. Science 2000, 290, 2280. (9) Faist, J.; Capasso, F.; Sirtori, C.; Sivco, D. L.; Hutchinson, A. L.; Cho, A. Y. Appl. Phys. Lett. 1995, 67, 3057. (10) Toth, G.; Lent, C. S. Phys. Rev. A 2001, 63, 052315. (11) Ilie, A.; Bendall, J. S.; Roy, D.; Philp, E.; Green, M. L. H. J. Phys. Chem. B 2006, 110, 13848. (12) Han, W. Q.; Chang, C. W.; Zettl, A. Nano Lett. 2004, 4, 1355. (13) Garg, I.; Sharma, H.; Dharamvir, K.; Jindal, V. K.; Kanhere, D. G. J. Phys. Chem. C 2010, 114, 18762. (14) Kwon, Y.-K.; Tománek, D.; Iijima, S. Phys. Rev. Lett. 1999, 82, 1470. (15) Service, R. F. Science 2001, 292, 45. (16) White, C. T.; Mintmire, J. W. Nature 1998, 394, 29. (17) Carbon nanotubes Topics in Applied Physics; Dresselhaus, M. S.; Dresselhaus, G.; Avouris, P., Eds.; Springer: New York, 2001. (18) Hamada, N.; Sawada, S.; Oshiyama, A. Phys. Rev. Lett. 1992, 68, 1579. (19) Akagi, K.; Tamura, R.; Tsukada, M. Phys. Rev. B 1996, 53, 2114. (20) Chico, L.; Crespi, V. H.; Benedict, L. X.; Louie, S. G.; Cohen, M. L. Phys. Rev. Lett. 1996, 76, 971. (21) Yu, J.; Chen, Y.; Cheng, B. M. Solid State Commun. 2009, 149, 763. (22) Chen, X.; Wu, P.; Rousseas, M.; Okawa, D.; Gartner, Z.; Zettl, A.; Bertozzi, C. R. J. Am. Chem. Soc. 2009, 131, 890. (23) Cohen, M. L.; Zettl, A. Phys. Today 2010, 63, 34. (24) Wang, J.; Lee, C. H.; Bando, Y.; Golberg, D.; Yap, Y. K. In B-C-N Nanotubes and Related Nanostructures, Yap, Y. K., Ed.; Springer: New York, 2009; pp 23−44. (25) Ni, S.; He, W.; Li, Z.; Yang, J. J. Phys. Chem. C 2011, 115, 12760. (26) Rubio, A.; Corkill, J. L.; Cohen, M. L. Phys. Rev. B 1994, 49, 5081. (27) Khoo, K. H.; Mazzoni, M. S. C.; Louie, S. G. Phys. Rev. B 2004, 69, 201401R. (28) Ravinder, P.; Subramanian, V. J. Phys. Chem. A 2010, 114, 5565. (29) Zhao, Y.; Schultz, N. E.; Truhlar, D. G. J. Chem. Theory Comput. 2006, 2, 364. (30) Zhao, Y.; Truhlar, D. G. Acc. Chem. Res. 2008, 41, 157. (31) Boys, S. F.; Bernardi, F. Mol. Phys. 1970, 19, 553. (32) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; et al. Gaussian 03, revision E.01; Gaussian, Inc.: Wallingford, CT, 2004. (33) Morokuma, K.; Kitaura, K. Energy decomposition analysis of molecular interactions. In Chemical Applications of Atomic and Molecular Electrostatic Potentials; Politzer, P.; Truhlar, D. G., Ed.; Plenum: New York, 1981. (34) Ziegler, T.; Rauk, A. Theor. Chim. Acta 1977, 46, 1. (35) Ziegler, T.; Rauk, A. Inorg. Chem. 1979, 18, 1558. (36) Bickelhaupt, F. M.; Baerends, E. J. Reviews in Computational Chemistry; Wiley: New York, 2000; Vol. 15. (37) Amsterdam Density Functional (ADF) version 2009.01. http:// www.scm.com. (38) Bader, R. F.W. Atoms in Molecules: A QuantumTheory; Clarendon Press: Oxford, United Kingdom, 1990. (39) Popelier, P. L. A. Atoms in Molecules: An Introduction; Prentice Hall: New York, 2000. (40) Biegler-Konig, F.; Schonbohm, J.; Derdav, R.; Bayles, D.; Bader, R. F. W. AIM2000, version 1; Bielefeld, Germany, 2000. (41) Kresse, G.; Furthmuller. J. Phys. Rev. B 1996, 54, 11169. (42) Kresse, G.; Furthmuller. J. Comput. Mater. Sci. 1996, 6, 15. (43) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. Rev. Lett. 1996, 77, 3865. (44) Kresse, G.; Joubert, D. Phys. Rev. B 1999, 59, 1758. (45) Blochl, P. E. Phys. Rev. B 1994, 50, 17953. (46) Goedecker, S.; Teter, M.; Hutter, J. Phys. Rev. B 1996, 54, 1703. (47) CPMD Version 3.7.0; IBM Corp. and Max-Planck-Institut für Festkörperforschung: Stuttgart, Germany. 5528


Studies on the Encapsulation of Fâ in Single ... - ACS Publications

Recommend Documents

Studies on the Encapsulation of Fâ in Single ... - ACS Publications