The healing of N-vacancy in boron nitride nanotube by using NO and NO₂ molecules: a density functional theoretical study

Abstract

Defect healing in boron nitride nanotube (BNNT) would be very helpful for potential applications in various devices. In the present study, density functional theory methods have been used to study the adsorption of NO and NO₂ molecules on the N-vacancy defect in BNNT. Both healing processes mainly undergo three evolution steps: (i) the first-step chemisorption of NO or NO₂ onto the N-vacancy site in BNNT; (ii) the combination of the N atom of the molecule with the N-vacancy in BNNT, accompanied by the formation of B–O–N epoxy structures; and (iii) the reduction of the oxidized BNNT by sequential exposure to NO molecules with the formation of NO₂ and perfect BNNT. In both healing processes, the removal of the last O atom on BNNT is found to be the rate-limiting step, for which the energy barrier decreases with an increase of BNNT diameter, i.e., 0.82 eV in (8, 0) BNNT, 0.54 eV in (12, 0) BNNT and 0.41 eV in hexagonal BN (h-BN), indicating the easy healing of the N-vacancy in BNNT with large diameters and h-BN by NO and NO₂ molecules. Moreover, a molecular dynamics simulation further confirms the healing process of the N-vacancy in h-BN by NO at room temperature. Our study will be helpful not only in the purification of BNNT/h-BN, but also in the removal of NO and NO₂ gases.

---

1. Introduction

Boron nitride nanotube (BNNT), which was first theoretically predicted in 1994 (ref. 1) and synthesized in 1995,² has been attracting much interest because of its unique properties, such as a wide band gap (∼5 eV) that is independent of its helicity and diameter,¹ excellent piezoelectric property,³ high thermal conductivity as well as chemical stability.^4,5 Due to the limitations of most experimental conditions, various kinds of defects such as B or N single vacancies,⁶ Stone–Wales defects,⁷ BN divacancies,⁷ dangling bonds at open ends,⁸ and rehybridization defects are inevitably formed.⁹ Among the various defects, the single vacancy defects, native¹⁰ or created by electron irradiation,¹¹ have been widely studied in many fields, such as native electronic properties,¹² design of metal-free magnetic materials,¹³ hydrogen storage,¹⁴ and their interaction with Fe,¹⁵ Ni¹⁶ and CCl₂.¹⁷ In addition, single vacancies in BNNT could dramatically decrease its tensile strength¹⁸ and their oxidation resistance.^19,20 Accordingly, two aspects should be kept in mind: (i) BNNT with single vacancies could be an active substrate due to its high surface reactivity, and (ii) an effective method is necessary to heal the single vacancies in BNNT in order to preserve its advantageous effects.

As we know, there are two types of single vacancies in BNNT, i.e. N-vacancy and B-vacancy. When compared with B-vacancy, a N-vacancy holds higher chemical stability,¹⁶ and some of the BNNTs produced are slightly yellow due to the existence of N-vacancies.²¹ Moreover, N-vacancies in BNNT could be largely produced under low-energy ion bombardment.²² Thus, in the present paper, only N-vacancy is under consideration. Our previous studies have shown that the single and double vacancies in carbon nanotube (CNT) and graphene could be effectively healed by small molecules, such as CO, CH_y (y = 1, 2, 3), C₂H₂ and C₂H₄.^23–27 Among them, the healing processes are mainly driven by the large energy released after the first-step adsorption of the molecules onto the vacancy site. In view of the similarity between BNNT and CNT (or graphene) with vacancy defects, some small molecules may be potential candidates to heal the N-vacancy in BNNT. In the present study, NO and NO₂ molecules are considered, which are rather common in polluted air and are present in concentrations that are harmful to human health and the environment; therefore, removal of these pollutants has attracted much attention in recent years, and a number of adsorption and catalytic techniques have been employed for this purpose.^28,29 Our results reveal that NO and NO₂ molecules could be potential candidates for healing the N-vacancy in BNNT, and both these molecules could be successfully removed or oxidized to an easily removable form after the healing process.

2. Computational details

2.1. Model of N-vacancy in BNNT

Firstly, we built a perfect (8, 0) BNNT with 64 B atoms, 64 N atoms and 16 H atoms. The H atoms are used to saturate the B and N atoms with dangling bonds at the two ends. Subsequently, one N atom in the middle of BNNT is removed to give an ideal single N-vacancy defect. In this case, local B2–B3 bond reconstruction occurs, changing the original hexagonal network around the vacancy into an atomic arrangement containing a five-membered ring (5MR) and nine-membered ring (9MR), as shown in Fig. 1. We note that such a newly formed B2–B3 bond subsides in the tube and the bond length (1.78 Å) agrees well with the previous theoretical studies.¹⁷ The two-coordinated B atom in the defect protrudes from the circumference of the tube.

Fig. 1 The structure of (8, 0) BNNT with one N-vacancy.

2.2. Computational method

The ONIOM method,³⁰ which has been successfully applied to the reaction studies of nanotubes,^31–34 is adopted to explore the reaction pathways. As shown in Fig. 1, the 9MR of the N-vacancy defect and its connected 5MR and 6MR were chosen to be the high layer. The minima (Lm) and transition state (TSm) calculations are carried out at the ONIOM (B3LYP/6-31G*:B3LYP/3-21G) level, followed by the single-point calculations at the B3LYP/6-31G* level. The calculation is performed with the Gaussian 03 program.³⁵ A first-principles molecular dynamics (MD) simulation was performed by using the VASP package.³⁶ The projector augmented-wave (PAW) method and the generalized gradient approximation (PBE) were adopted to describe the atomic core electrons and electron–electron interactions, respectively.^37–39 A plane wave basis with the cutoff energy of 300 eV was used with a single k point at Γ to sample the Brillouin zone. The time step is 1 fs and the temperature was controlled by velocity scaling at each step.

3. Results and discussion

The characteristic points in the potential energy surfaces (PES), the structures of the transition state (TS) and the intermediate (L) for the healing of the N-vacancy in BNNT by NO and NO₂ molecules are presented in Fig. 2–10. The zero energy corresponds to the non-interacting species, i.e., the NO or NO₂ molecule is at an infinite distance from the defective BNNT.

Fig. 2 (a) The structure of the first-step adsorption of NO on the N-vacancy in BNNT and (b) the corresponding spin density with an isovalue of 0.005 a.u.

3.1 Healing of N-vacancy BNNT by NO

Firstly, we calculate the spin population of the BNNT with a single N-vacancy. The results reveal that BNNT with an N-vacancy defect has a total magnetic moment of 1 μ_B, and the spin density is mainly located on the unsaturated B1 atom according to the Mulliken population analysis. Thus, the B1 atom, which prefers to accept foreign molecules, is the chemically reactive center. For the first-step adsorption of NO onto the N-vacancy BNNT, we initially place a single NO molecule on the B1 atom via N-end or O-end attack after considering both the singlet and triplet spin states. It is found that the triplet N-end attacking structure is the most stable one, as shown in Fig. 2a (³L1a). In ³L1a, the N–O bond length of NO slightly increases from 1.16 Å to 1.22 Å, and the newly formed N1–B1 bond length is 2.38 Å. Mulliken charge population analysis shows that the electron transfer from the N-vacancy BNNT to NO by increasing the charge of the O and N atoms by −0.15e and −0.10e, respectively. The distribution of spin density is shown in Fig. 2b, which shows a large spin density on NO and the B1 atom. This suggests that there is a strong chemical bond between NO and N-vacancy BNNT. It should be noted that a similar N-end attacking structure could also be found in the adsorption of NO on B-doped graphene.⁴⁰ For simplicity, we will only consider the PES based on the triplet N-end attacking structure as shown in Fig. 3. After the first-step adsorption, the ³L1a structure could directly transform into an epoxy structure (³L3a in Fig. 4) by overcoming ³TS1a/3a with intrinsic energy barrier of 0.47 eV. In ³L3a, the O atom is located on the inclined B–N bond of the BNNT, and the newly formed B–O and N–O bond lengths are 1.43 Å and 1.48 Å, respectively. Because of its strong electronegativity, the O atom continues to gain negative electrons up to −0.43e. In addition, another reaction pathway could be found to reach ³L3a: starting from ³L1a, the O atom of NO firstly bonds to the B3 atom in the BNNT, and then the N atom of NO fills the N-vacancy in BNNT by forming the epoxy structure (³L3a). However, such a process needs to overcome an energy barrier of 1.41 eV, which suggests the need for significant chemical activation for this reaction to occur. After the combination of the N atom of NO and the N-vacancy of BNNT, the system will convert from its triplet state (³L3a) to a singlet state (¹L3a). As shown in Fig. 3, the singlet epoxy structure (¹L3a) is energetically (4.64 eV) more stable than the triplet state and a similar epoxy structure could also be found in the adsorption of a single O atom on pure BNNT.¹⁹ To remove the extra O atom, subsequent interaction with another NO molecule is a good choice. A similar strategy has been successfully applied in removing the O atom from an epoxy structure in graphene by NO.²⁶ Starting from ¹L3a, after the adsorption of a second NO molecule, the O atom on BNNT could be removed with an intrinsic energy barrier of 0.82 eV, followed by the formation of a perfect BNNT and NO₂ molecule. It should be noted that thermal relaxation usually occurs more rapidly than the collision of gas molecules on the surface.^41,42 Thus, the large energy released upon the adsorption of the first NO molecule (7.27 eV) might undergo redistribution to the surface to a great extent before the second NO attacks. Thus, the attack of the second NO is the rate-limiting step for the whole healing process. In view of its relatively high energy barrier (0.82 eV), the healing process of N-vacancy in (8, 0) BNNT would hardly take place at room temperature.

Fig. 3 The characteristic points on the potential energy surface (PES) for the adsorption of an NO molecule on the N-vacancy in BNNT. The minima and transition state are represented by L and TS, respectively.

Fig. 4 The structures of the transition states (TS) and intermediates (L) for the healing of the defective BNNT by an NO molecule.

It is noticeable that the relatively high energy barrier for the O-removal process is most likely due to the strong interaction between the O atom and the (8, 0) BNNT. Thus, to explore further possibilities, we consider the curvature effect on the O-removal barrier of NO with various (n, 0) BNNTs (n = 8, 10 and 12) and hexagonal-BN (h-BN) with epoxy structures. As seen in Fig. 5a, both O-removal barriers and O-adsorption energies (E_ads = E_O + E_BNNT – E_O-BNNT) decrease with the increase of BNNT diameters, i.e., the energy barrier and adsorption energy are 0.54 and 1.86 eV in the case of (12, 0) BNNT, which become 0.41 and 1.36 eV in the case of h-BN. Experimentally, the synthesized BNNTs usually have nanoscaled diameters,⁴³ ranging from several to several tens of nanometers (0.95 nanometers in (12, 0) BNNT); thus, the healing barrier of the N-vacancy in BNNT by NO will be within the range from 0.41 to 0.54 eV in the real cases. It should be noted that there are some similarities between BNNT with nanoscaled diameters and h-BN, such as small surface curvatures, low O-removal energy barriers and the same B–N arrangement. Together with the small system of h-BN and as a result the low computational time consumption, we performed the molecule dynamics (MD) simulation to identify the healing process in h-BN with N-vacancy at room temperature, which may give evidence on the case of N-vacancy in BNNT with large diameters. As shown in Fig. 5b and 6, the NO molecule is firstly placed parallel to the N-vacancy in h-BN (Fig. 6a), and subsequently the N atom of NO bonds with one B atom in the N-vacancy site within 80 fs of the MD simulation as shown in Fig. 6b. After 300 fs, the N atom of NO will further attach to another B atom in the vacancy site (Fig. 6c). Finally, the N atom of NO will insert into the h-BN layer after 700 fs and bond with the O atom by forming an epoxy structure as shown in Fig. 6d. The whole process for first NO-vacancy combination releases energy of about 6 eV. Another NO molecule is then introduced into the system, with its position near the O atom in epoxy structure (Fig. 6e). Within 100 fs of the MD simulation at room temperature, the NO molecule will bond to the O atom by forming a NO₂ molecule and desorbs rapidly from the pure h-BN (Fig. 6f) with an energy release of about 4 eV. It should be noted that each snapshot (from Fig. 6a to f) during the healing processes of the N-vacancy in h-BN by NO corresponds to the case of (8, 0) BNNT as shown in Fig. 3 and 4. Based on the results above, we could deduce that NO is a promising candidate to heal the N-vacancy defects in both BNNT with large diameters and h-BN.

Fig. 5 (a) The adsorption energy and removal barrier of a single O atom on various (n, 0) BNNT and h-BN; (b) the energy gain diagram for the healing process of NO molecules on the N-vacancy in h-BN under MD simulation at room temperature.

Fig. 6 The snapshots of the MD simulation of the healing process of NO molecules on N-vacancy in h-BN.

As we know, NO in high concentrations is very reactive and can be oxidized into NO₂ with O₂ by forming HNO₃ in water.⁴⁴ However, NO at levels below 0.1% is very stable and is slowly oxidized in air.⁴⁵ The slow oxidation of NO inhibits the development of its oxidative removal technology. As we have discussed above, our results show that the N-vacancy in BNNT and h-BN could be good oxidation substrates to convert NO into NO₂.

3.1 Healing of the N-vacancy in BNNT by NO₂

For the first-step adsorption of a NO₂ molecule, the singlet O-end attacking configuration is found to be the most stable after considering various possibilities. Similarly, the singlet O-end attacking structure also has been found in the adsorption of NO₂ on B-doped CNT and graphene.^40,46 For simplicity, we will mainly discuss the PES based on the singlet O-end attacking configuration. In detail, for the first-step adsorption, the O1 atom in the NO₂ molecule bonds to the unsaturated B1 atom in the N-vacancy site with a O1–B1 bond length of 1.40 Å as shown in Fig. 7a (L1b). The N1–O1 bond length of NO₂ is increased from 1.20 to 1.50 Å after the adsorption, indicating the single N–O bond property. Mulliken charge population analysis shows that the O1 atom of NO₂ accumulates more electrons by the increase of its charge from −0.24e (in NO₂) to −0.40e and the N1 atom of NO₂ decreases its electron charge from 0.48e (in NO₂) to 0.32e. In Fig. 7b, the corresponding three-dimensional contour plot of the electron charge density difference (i.e., the electron charge density of the molecule adsorbed on the N-vacancy in BNNT minus the electron charge density of the isolated molecule and the N-vacancy BNNT, which is calculated with the same atomic positions of the molecule adsorbed N-vacancy BNNT) clearly shows charge density piling between the O1 atom of NO₂ and the B1 atom of the N-vacancy BNNT, indicating the orbital hybridization between NO₂ and N-vacancy BNNT. The first-step adsorption is a zero-energy barrier with a large energy release of 3.86 eV.

Fig. 7 (a) The structure of the first-step adsorption of NO₂ on the N-vacancy in BNNT and (b) the corresponding electron charge density difference with an isovalue of 0.005 a.u.

As shown in Fig. 8, starting from L1b, the N1 atom of NO₂ could be adsorbed on the B2 atom to form L2b and occurs with a 0.39 eV energy barrier at TS1b/2b. In L2b (Fig. 9), a heptagon-ring-type structure is obtained, the newly formed N1–B2 bond length is 1.66 Å and the B2–B3 bond length in the defect site is slightly increased from 1.82 to 1.88 Å. Passing over the TS2b/3b with the small barrier of 0.17 eV, the N1 atom is completely inserted into the B2–B3 bond of the BNNT, followed by the formation of two symmetrical heptagon rings as shown in L3b (Fig. 9). Starting from L3b, two reaction pathways could occur (hereafter we have named them as (i) and (ii), respectively) as shown in Fig. 8. In process (i), after overcoming the TS3b/4b with a small barrier of 0.42 eV, two O atoms in L3b will insert into two neighbouring B–N bonds as shown in L4b (Fig. 9). Subsequently, the rearrangement of the two O atoms on the surface of the BNNT is considered, among which the most stable structure (L7b) is composed of two O atoms located on the two inclined B–N bonds within the same BN hexagon, which agrees well with Chen et al.'s study.¹⁹ As shown in Fig. 8, the rearrangement process (from L4b to L7b) will release energy of 6.95 eV and overcome an energy barrier of 1.97 eV. For process (ii), after overcoming TS3b/8b with an intrinsic barrier of 1.32 eV, the two O atoms of NO₂ are adsorbed on the axial B–N bond, followed by the formation of a O–B–N–O four-membered ring (L8b). It is noticeable that the perfect BNNT is oxidation resistant;^19,20 thus, two O atoms on BNNT could be desorbed in the form of O₂ via the TS8b with a barrier of 0.91 eV. The completion of process (ii) will release an energy of 7.04 eV and overcome an energy barrier of 1.32 eV.

Fig. 8 The characteristic points on the potential energy surface (PES) for the adsorption of an NO₂ molecule on the N-vacancy in BNNT.

Fig. 9 The structures of transition states (TS) and intermediates (L) for the healing of the defective BNNT by an NO₂ molecule.

Note that both process (i) and (ii) need to overcome high energy barriers of 1.97 and 1.32 eV, respectively, which indicates that both reactions can hardly occur at room temperature. Accordingly, after the adsorption of NO₂ on BNNT with N-vacancy, the most feasible reaction pathway is from L1b to L4b as shown in Fig. 8, during which an energy barrier of only 0.42 eV needs to be overcome. In this case, the NO₂ molecule acts as an oxidant to convert the N-vacancy in BNNT into oxidized BNNT (L4b). As discussed in the NO-BNNT system, NO could be a promising candidate to reduce the oxidized BNNT. Thus, we consider the reduction process of L4b by using NO molecules. In L4b, two types of O atom could be found, which are adsorbed on the inclined and axial B–N bonds of the BNNT. As mentioned above, the epoxy structure located in the inclined B–N bond is more stable than that in the axial bond and as a consequence is difficult to remove. Thus, we take the O1 atom on axial B–N bond as the first one to be removed by NO. As seen in Fig. 10, after the adsorption of NO, the O1 atom could be removed with an intrinsic energy barrier of 0.54 eV, followed by the formation of ¹L3a and a NO₂ molecule. Subsequently, ¹L3a could be reduced by another NO molecule with a low energy barrier (from 0.41 to 0.54 eV) as has been discussed in the NO-BNNT system.

Fig. 10 The structures and characteristic points on the potential energy surface (PES) for the adsorption of an NO molecule on L4b.

According to our results, following two reaction mechanisms can be proposed to heal the single N-vacancy in BNNT: (i) by two NO molecules, releasing one NO₂ molecule as the product and (ii) by sequential exposure to one NO₂ and two NO molecules, releasing two NO₂ molecules as the product. The rate-limiting step in both healing processes is the O-removal by NO, for which the energy barrier decreases with an increase in the BNNT diameter. Since the synthesized BNNT usually have nanoscaled diameters, the estimated energy barrier for the O-removal process is within the range from 0.41 (in case of h-BN) to 0.54 eV (in case of (12, 0) BNNT). In general, a reaction with a barrier of less than 0.5 eV is expected to occur at room temperature. Moreover, our MD simulation confirms that the healing of N-vacancy in h-BN by NO could easily occur at room temperature. Thus, the proposed method not only is theoretically feasible but also has a relatively low reaction energy barrier; therefore, the healing process can be easily achieved on the N-vacancy in BNNT by using NO and NO₂ molecules.

4. Implication

Nanotubes or nanosheet structures, such as CNT, graphene, BNNT and h-BN, always possess very good stability. After removing one X atom (X could be denoted as C, B or N) from the surface of these nanostructures, a single X-vacancy defect could be created. Experimentally, single vacancies could be easily introduced through electron or ion-irradiation,^11,47 and they usually contain unsaturated atoms and thus are highly reactive. Accordingly, the adsorption of certain foreign molecules on the unsaturated atoms is expected to be particularly favorable, followed by release of large amount of energy. Especially for gas molecules containing an X atom or atom with a diameter similar to X, the insertion of the X atom into the X-vacancy nanostructures most likely occurs due to the huge energy released after the first-step adsorption. Note that the insertion processes are always accompanied by bond cleavage of the foreign molecule (containing X atom) and adsorbing leaving atoms (except X atom) on the surface of the host nanostructures. Insertion and bond cleavage processes, such as the CO, C₂H₂ and C₂H₄ adsorbed C-vacancy in CNT and the CO, CH_y (y = 1, 2, 3), C₂H₂ and C₂H₄ adsorbed C-vacancy in graphene, have already been observed in our present and previous theoretical studies.^23–27 Thus, we could say that the single X-vacancy nanostructure (such as CNT, graphene, BNNT or h-BN) could be most likely healed by some small foreign molecules containing an X atom or atom with similar diameter to X; moreover, the nanostructure with a single vacancy could be a potential substrate for the bond cleavage reactions of these foreign molecules.

5. Conclusion

In summary, we have investigated the healing of the N-vacancy in BNNT by using NO or NO₂ molecules. In both cases, low energy barriers (<0.5 eV) are needed for the first NO or NO₂ molecule remerging with the N-vacancy in BNNT. After that, NO molecules could remove the extra O atoms on the BNNT by overcoming an energy barrier of 0.82 eV, accompanied by the formation of pure (8, 0) BNNT and an NO₂ molecule. It is found that the energy barrier is 0.54 eV and 0.41 eV in (12, 0) BNNT in hexagonal BN (h-BN), respectively; therefore, the O-removal energy barrier decreases with an increase in the BNNT diameter. Thus, NO and NO₂ molecules are promising candidates to heal the N-vacancy defects in BNNT with large diameters and h-BN. We believe that our results could be very helpful in both purifying the defective BNNT/h-BN and removing NO and NO₂ gases.

Acknowledgements

This work was funded by the National Basic Research Program of China (973 Program) (2012CB932800), the National Key Basic Research and Development Program of China under Grant no. 2010CB327701, the National Natural Science Foundation of China (no. 20773054, 21073074), the Doctor Foundation by the Ministry of Education (20070183028), the Excellent Young Teacher Foundation of Ministry of Education of China, the Excellent Young People Foundation of Jilin Province (20050103) and the Program for New Century Excellent Talents in University (NCET).

Notes and references

1. X. Blase, A. G. Rubio, S. Louie and M. L. Cohen, Stability and Band Gap Constancy of Boron Nitride Nanotubes, Europhys. Lett., 1994, 28, 335–340.
2. N. G. Chopra, R. J. Luyken, K. Cherrey, V. H. Crespi, M. L. Cohen, S. G. Louie and A. Zettl, Boron Nitride Nanotubes, Science, 1995, 269, 966–967.
3. S. M. Nakhmanson, A. Calzolari, V. Meunier, J. Bernholc and N. M. Buongiorno, Spontaneous polarization and piezoelectricity in boron nitride nanotubes, Phys. Rev. B: Condens. Matter Mater. Phys., 2003, 67, 235406.
4. C. C. Tang and Y. Bando, Effect of BN coatings on oxidation resistance and field emission of SiC nanowires, Appl. Phys. Lett., 2003, 83, 659–661.
5. Y. Xiao, X. H. Yan, J. Xiang, Y. L. Mao, Y. Zhang, J. X. Gao and J. W. Ding, Specific heat of single-walled boron nitride nanotubes, Appl. Phys. Lett., 2004, 84, 4624.
6. Y. Miyamoto, A. Rubio, S. Berber, M. Yoon and D. Tomanek, Spectroscopic characterization of Stone–Wales defects in nanotubes, Phys. Rev. B: Condens. Matter Mater. Phys., 2004, 69, 121413.
7. A. Zobelli, C. P. Ewels, A. Gloter, G. Seifer, O. Stephan, S. Csillag and C. Colliex, Defective Structure of BN Nanotubes: From Single Vacancies to Dislocation Lines, Nano Lett., 2006, 6, 1955–1960.
8. A. Celik-Aktas, J. F. Stubbins and J.-M. Zuo, Electron beam machining of nanometer-sized tips from multiwalled boron nitride nanotubes, J. Appl. Phys., 2007, 102, 024310.
9. X. J. Wu, W. An and X. C. Zeng, Chemical Functionalization of Boron Nitride Nanotubes with NH₃ and Amino Functional Groups, J. Am. Chem. Soc., 2006, 128, 12001–12006.
10. H. S. Kang, Theoretical Study of Boron Nitride Nanotubes with Defects in Nitrogen-Rich Synthesis, J. Phys. Chem. B, 2006, 110, 4621–4628.
11. A. Zobelli, A. Gloter, C. P. Ewels and C. Colliex, Shaping single walled nanotubes with an electron beam, Phys. Rev. B: Condens. Matter Mater. Phys., 2008, 77, 045410.
12. G. Y. Gou and B. C. Pan, Theoretical study of size-dependent properties of BN nanotubes with intrinsic defects, Phys. Rev. B: Condens. Matter Mater. Phys., 2007, 76, 155414.
13. R. Liu, J. Li and G. Zhou, Ab Initio Investigation about the Possibility of Ferromagnetism Induced by Boron Vacancy in BN Nanotubes, J. Phys. Chem. C, 2010, 114, 4357–4361.
14. S. A. Shevlin and Z. X. Guo, Hydrogen sorption in defective hexagonal BN sheets and BN nanotubes, Phys. Rev. B: Condens. Matter Mater. Phys., 2007, 76, 024104.
15. G. Y. Gou, B. C. Pan and L. Shi, Interaction of Iron Atoms with Pristine and Defective (8, 0) Boron Nitride Nanotubes, J. Phys. Chem. C, 2008, 112, 13571–13578.
16. J. X. Zhao and Y. H. Ding, Theoretical Study of Ni Adsorption on Single-Walled Boron Nitride Nanotubes with Intrinsic Defects, J. Phys. Chem. C, 2008, 112, 5778–5783.
17. Y. Li, Z. Zhou and J. Zhao, Functionalization of BN nanotubes with dichlorocarbenes, Nanotechnology, 2008, 19, 015202.
18. M. Griebel, J. Hamaekers and F. Heber, A molecular dynamics study on the impact of defects and functionalization on the Young modulus of boron-nitride nanotubes, Comput. Mater. Sci., 2009, 45, 1097–1103.
19. Y. Chen, C. L. Hu, J. Q. Li, G. X. Jia and Y. F. Zhang, Theoretical study of O₂ adsorption and reactivity on single-walled boron nitride nanotubes, Chem. Phys. Lett., 2007, 449, 149–154.
20. Y. Chen, J. Zou, S. Campbell and G. Caer, Boron nitride nanotubes: Pronounced resistance to oxidation, Appl. Phys. Lett., 2004, 84, 2430.
21. C. Zhi, Y. Bando, C. Tang and D. Golberg, Boron nitride nanotubes, Mater. Sci. Eng., R, 2010, 70, 92–111.
22. M. Petravic, R. Peter, I. Kavre, L. Li, Y. Chen, L. Fan and Y. Yang, Decoration of nitrogen vacancies by oxygen atoms in boron nitride nanotubes, Phys. Chem. Chem. Phys., 2010, 12, 15349–15353.
23. B. Xiao, J. Zhao, Y. Ding and C. Sun, Theoretical investigation of the interaction between carbon monoxide and carbon nanotubes with single-vacancy defects, ChemPhysChem, 2010, 11, 3505–3510.
24. C. Wang, B. Xiao and Y. H. Ding, Role of Hydrocarbon Radicals CH_x (x = 1, 2, 3) in Graphene Growth: A Theoretical Perspective, ChemPhysChem, 2012, 13, 774–779.
25. C. Wang and Y. H. Ding, Theoretical investigation on the healing mechanism of divacancy defect in graphene growth by reaction with ethylene and acetylene, New J. Chem., 2013, 37, 640–645.
26. B. Wang and S. Pantelides, Controllable healing of defects and nitrogen doping of graphene by CO and NO molecules, Phys. Rev. B: Condens. Matter Mater. Phys., 2011, 83, 245403.
27. H. Liu and J. Lee, Electric Field Effects on the Adsorption of CO on a Graphene Nanodot and the Healing Mechanism of a Vacancy in a Graphene Nanodot, J. Phys. Chem. C, 2012, 116, 3034–3041.
28. M. Shelef, Selective Catalytic Reduction of NO_x with N-Free Reductants, Chem. Rev., 1995, 95, 209–225.
29. K. Skalska, J. S. Miller and S. Ledakowicz, Trends in NO_x Abatement: A review, Sci. Total Environ., 2010, 408, 3976–3989.
30. F. Maseras and K. Morokuma, IMOMM – a New Integrated Ab Initio Plus Molecular Mechanics Geometry Optimization Scheme of Equilibrium Structures and Transition states, J. Comput. Chem., 1995, 16, 1170–1179.
31. S. P. Walch, On the Reaction of N and O Atoms with Carbon Nanotubes, Chem. Phys. Lett., 2003, 374, 501–505.
32. A. Ricca, C. W. Bauschlicher and A. Maiti, Comparison of the reactivity of O₂ with a (10, 0) and a (9, 0) carbon nanotube, Phys. Rev. B: Condens. Matter Mater. Phys., 2003, 68, 035433.
33. X. Lu, F. Tian, X. Xu, N. Wang and Q. Zhang, A Theoretical Exploration of the 1,3-Dipolar Cycloadditions onto the Sidewalls of (n,n) Armchair Single-Wall Carbon Nanotubes., J. Am. Chem. Soc., 2003, 125, 10459–10464.
34. L. V. Liu, W. Q. Tian and Y. A. Wang, Ozonization at the vacancy defect site of the single-walled carbon nanotube, J. Phys. Chem. B, 2006, 110, 13037–13044.
35. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgomery, Jr, T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K. Ragha- vachari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong, C. Gonzalez and J. A. Pople, Gaussian 03 (Revision D.02), Gaussian, Inc., Wallingford, CT, 2004.
36. G. Kresse and J. Furthmuller, Phys. Rev. B: Condens. Matter Mater. Phys., 1996, 54, 11169.
37. J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1996, 77, 3865.
38. P. E. Blochl, O. Jepsen and O. K. Andersen, Phys. Rev. B: Condens. Matter Mater. Phys., 1994, 49, 16223.
39. G. Kresse and D. Joubert, Phys. Rev. B: Condens. Matter Mater. Phys., 1999, 59, 1758.
40. W. An and H. Turner, Electronic structure calculations of gas adsorption on boron-doped carbon nanotubes sensitized with tungsten, Chem. Phys. Lett., 2009, 482, 274–280.
41. Z. Y. Ong and E. Pop, Molecular Dynamics Simulation of Thermal Boundary Conductance between Carbon Nanotubes and SiO₂, Phys. Rev. B: Condens. Matter Mater. Phys., 2010, 81, 155408–155414.
42. L. S. Jung and C. T. Campbell, Sticking Probabilities in Adsorption of Alkanethiols from Liquid Ethanol Solution onto Gold, J. Phys. Chem. B, 2000, 104, 11168–11178.
43. D. Golberg, Y. Bando, Y. Huang, T. Terao, M. Mitome, C. C. Tang and C. Y. Zhi, ACS Nano, 2010, 4, 2979–2993.
44. F. A. Cotton and G. Wilkinson, Advanced Inorganic Chemistry, Intersciences, New York, 1972, p. 355.
45. M. Bodenstein, Die Ermittlung des Mechanismus chemischer Reaktionen, Helv. Chim. Acta, 1935, 18, 743–759.
46. J. Dai, J. Yuan and G. Paolo, Gas adsorption on graphene doped with B, N, Al, and S: A theoretical study, Appl. Phys. Lett., 2009, 95, 232105.
47. (a) J. Kotakoski, A. V. Krasheninnikov, U. Kaiser and J. C. Meyer, From Point Defects in Graphene to Two-Dimensional Amorphous Carbon, Phys. Rev. Lett., 2011, 106, 105505; (b) A. V. Krasheninnikov and K. Nordlund, Ion and electron irradiation-induced effects in nanostructured materials, J. Appl. Phys., 2010, 107, 071301.

---