The physical and chemical properties of heteronanotubes

Carbon nanotubes undoubtedly take a leading position in nanotechnology research owing to their well-known outstanding structural and electronic properties. Inspired by this, hybrid and functionalized tubular structures have been constructed via several modification paths that involve the presence of molecules, generation of defects, and partial or full replacement of the carbon atoms, always maintaining a nanotube structure. The possibilities are countless. However, this review is mainly dedicated to giving a fundamental insight into the concepts behind wall modification, doping, and formation of a carbon nanotube structure. Theoretical concepts and experimental achievements ranging from carbon nanotubes with low B or N doping to the new physics behind boron nitride nanotubes are covered. Furthermore, special attention is devoted to the bulk and local characterization tools employed with these materials, their suitability and limitations. The theoretical approaches to describing the physical and chemical properties of heteronanotubes are objectively analyzed versus the materials available at this moment.

Figure 1

(Color online) Structure of a single-walled carbon nanotube pictured as a rolled-up graphene sheet (top). In the same manner, a BN nanotube would be the corresponding tubular structure related to a stable planar $h$-BN sheet (bottom).

Figure 2

The figures on the left show a simplified scheme of the hexagonal network in (a) real and (b) reciprocal spaces. This applies for the hexagonal lattices described in this review. The pictures on the right correspond to energy dispersion schemes and LDOS showing the van Hove singularities of a (3,3) C-SWNT and a (3,3) BN-SWNT (two different heteronanotubes with equivalent chirality).

Figure 3

(Color online) Comparative scheme of the electronic structure of a BN (3,3) nanotube and a $h$-BN sheet. The band structure of the nanotube.

Figure 4

(Color online) The molecular models show the different ways to modify the electronic properties of a pristine single-walled carbon nanotube. The diagrams represent (a) a pristine C-SWNT, (b) heteronanotubes with substitutional atoms, (c) intercalation within C-SWNTs in a nanotube bundle, and (d) endohedral doping by filling with different molecular structures such as the example depicted with encaged fullerenes in so-called peapods.

Figure 5

(Color online) N (represented by circles) can be incorporated in the walls of the nanotube in different ways. The most probable configurations for N in the graphitic wall of a C-SWNT are (a) direct substitution or (b) pyridinelike bond. (c) The stability of a nanotube with both type of incorporation is still a matter of discussion.

Figure 6

(Color online) Molecuar models of a $\mathrm{B}{\mathrm{C}}_2\mathrm{N}$ nanotube (left) and a $\mathrm{B}{\mathrm{C}}_3$ nanotube (right).

Figure 7

(Color online) The $\mathrm{B}{\mathrm{C}}_3$-SWNT can be thought as a planar rolled-up sheet of the same stoichiometry. The upper panel shows a single $\mathrm{B}{\mathrm{C}}_3$ planar graphitelike sheet with the corresponding band structure proposed by 153. The lower panels represent the band structure and density of states of a (3,3) $\mathrm{B}{\mathrm{C}}_3$-SWNT. This has been adapted from 153.

Figure 8

(Color online) ${\mathrm{C}}_3{\mathrm{N}}_4$ is one of the promising stable graphitelike structures that contain N. It is expected that these structures form nanotubes of the same stoichiometry. Depicted are the (a) Redemanns model and (b) Alves model for graphitelike ${\mathrm{C}}_3{\mathrm{N}}_4$ as described in Komatsu (2001). (c) Multiplane stacking of ${\mathrm{C}}_3{\mathrm{N}}_4$ layers exhibiting an interplanar distance close to graphite.

Figure 9

(Color online) In the case of BN-NTs no metallic species are observed for the different chiralities. In the left side, contour plots of the charge density corresponding to the bottom of the conduction-band state of an armchair and a zigzag BN-NT are depicted. N atoms are indicated by solid circles and B atoms are represented by empty circles. These structures exhibit a uniform band gap. The DOSs of BN tubes with different chiralities are shown on the right. Image from 217, related to 30.

Figure 10

211 reported a bias-assisted HF-CVD process to synthesize BCN-SWNTs in large scale. Upper panel: TEM bright-field image of bundles of BCN-SWNTs. Electron-energy loss as well as energy-filtered imaging were performed along the SWNT bundles, confirming the BNC structure. Adapted from 211.

Figure 11

One of the successful methods to synthesize $\mathrm{C}{\mathrm{N}}_x$-SWNTs is arc discharge. The images show (a) TEM micrograph of the material produced with this technique. From 69. (b) EELS spectra with inset on the N edge from the same reference.

Figure 12

BN-SWNTs obtained with a catalyst-free arc discharge synthesis method (a) HRTEM of BN-SWNTs emerging from a boron nanoparticle (124). (b)–(d) The transmission TEM images in high resolution are a helicity study of BN-SWNTs produced with this method.

Figure 13

Laser ablation B-doped nanotubes. The TEM images correspond to material produced from targets with nominal boron concentrations of (a) 0 and (b) $2.5\mathrm{at.}\mathrm{\%}$. From 146.

Figure 14

(Color online) One of the recent methods to successfully produce $\mathrm{C}{\mathrm{N}}_x$-SWNTs is the lasere vaporization technique. A TEM image is shown in (a) and the electron-energy-loss spectrum (b) recorded on the $\mathrm{C}{\mathrm{N}}_x$-SWNT bundle is marked by an arrow in the bright-field image. The EELS spectrum displays the C and N $K$ edges (emphasized in the inset). (c) STM images of a nanotube where the tunneling voltages are 900, 800, and $700\mathrm{mV}$ from top to bottom, respectively. (d) The corresponding intensity profiles. From 131.

Figure 15

Loss spectra of the multiwall BN nanotubes between 2 and $15\mathrm{eV}$ as a function of the momentum transfer $q$, as measured by EELS. Inset: Wide energy range between 0 and $30\mathrm{eV}$. From 65.

Figure 16

(Color online) The matrix element weighted conduction-band response of C-SWNTs as a function of doping and the corresponding shift is analyzed by EELS on the example of the C edge of pristine nanotubes and bundles of nanotubes with K for $n$-type doping, as well as $\mathrm{Fe}{\mathrm{Cl}}_3$ for $p$-type doping. The changes in the DOS upon doping for the C $1s$ core level excitation process within the rigid band model are highlighted in the sketches.

Figure 17

(Color online) Comparison of the calculated phonon dispersion of (dashed) graphene and (solid) $h$-BN. From 217.

Figure 18

Shifts and splitting in the $G^{\prime }$ response for different $s{p}^2$ carbon systems. From 138.

Figure 19

Experimental and simulated ED patterns of a BN-SWNT. The intensity profiles of the equatorial line of the experimental and simulated ED patterns of the tube are also represented. Adapted from 7.

Figure 20

(Color online) Spectrum line recorded, after striapping the background, across a bundle of B-doped SWNTs. The inset shows the B:C ratio. Adapted from 32.

Figure 21

(Color online) Spectrum recorded across a BN-SWNT shown in the bright-field image above. The lower panel depicts the fine structures in the B and N edges. Adapted from 10.

Figure 22

(Color online) XPS is a ley method to analyze core level excitations. This method can be used to detect heteroatom percentages as low as 0.2% at., which is the case corresponding to these figures. (a) The N $1s$ core signal corresponding to N-doped C-SWNTs synthesized from benzylamine chemical vapor deposition with a line-shape analysis that shows the presence of pyridinic and $s{p}^2$-like bonding environments in equal proportion at certain synthesis conditions. (b) Nitrogen incorporation profile according to synthesis temperatures. The gray bars correspond to the total N content at different synthesis temperatures. The fractions from the line-shape analysis ($s{p}^2$, pyridine, and N volatile species) are depicted by the symbols specified on the top right labels. Adapted from 19.

Figure 23

(Color online) B-doped SWNTs have been successfully synthesized with CVD. The left side images show low and high magnification TEM micrographs of such kind of material. On the right panel, C $1s$ and B $1s$ XPS signal of the B-SWNT at two different temperatures (1) $890°\mathrm{C}$ and (2) $840°\mathrm{C}$. The inset shows the EFTEM for C, B, and Fe together with the BF image. Adapted from 22.

Figure 24

Spectra corresponding to high-resolution EELS core level excitation spectra of B-containing C-SWNT. This material corresponds to B-doped SWNT in the range of the B $1s$, C $1s$, and N $1s$ edges. Adapted from 33.

Figure 25

Snapshots of ab initio molecular dynamics (MD) calculations revealing the formation of B-doped SWNTs with zigzag chirality (9,0). B acts as a surfactant and inhibits tube closure, thus enhancing the length of the tubules. From 28.

Figure 26

Electronic properties of the unoccupied states highly B-doped SWNT determined by high-resolution core level EELS of the B $1s$ and C $1s$ edges. The inset shows the change of the C $1s$ onset in an expanded scale. From 64.

Figure 27

The electronic configuration of B-doped nanotubes depends on the amount of heteroatomic incorporation in the nanotube structure. The band structure and DOS of a C(16,0) tube with (a) 0%, (b) 6.25%, (c) 12.5%, and (d) 25% boron doping. Zero energy denotes the Fermi edge. Filled states are indicated by gray shading. From 64.

Figure 28

$\mathrm{B}{\mathrm{C}}_2\mathrm{N}$ is one of the stoichiometries, which can lead to the formation of SWNTs. Here are shown one possible configuration of a $\mathrm{B}{\mathrm{C}}_2\mathrm{N}$ sheet (a) and the corresponding $\mathrm{B}{\mathrm{C}}_2\mathrm{N}(4,4)$ semiconducting nanotube obtained rolling the hexagonal stable structure (b). The lower panels correspond to the band structure of $(2,2)\mathrm{B}{\mathrm{C}}_2\mathrm{N}$-SWNTs (c) and the cases where a C atom is substituted as impurity on the N site (left, d) and the B site (right, d). Adapted from 152.

Figure 29

$\mathrm{B}{\mathrm{C}}_3$ nanotubes: (a) Electronic band structure of $\mathrm{B}{\mathrm{C}}_3(3,3)$ nanotube (Fermi energy at $0\mathrm{eV}$). (b) Density of states (filled states shaded with gray). (c) Joint density of states divided by $E^2$ (solid line) and calculated optical absorption (dashed line) of a perfect infinite $\mathrm{B}{\mathrm{C}}_3(3,3)$ tube with light polarization along the tube axis. (d) Sketch of the unit cell. The black dots represent C atoms. From 64.

Figure 30

(Color online) The band structure of a single sheet of $h$-BN (left) and the band structure of a (6,6) BN.

Figure 31

(Color online) Electronic band structure of an (8,0) nitrogen-doped C-NT with one and two N atom substitutions. From 68.

Figure 32

(Color online) Raman spectroscopy study on B-doped C-SWNTs reported by 146. (a) First- and second-order room temperature Raman spectra of products generated from targets with the indicated B concentrations in the feedstock and excited using the $1064\mathrm{nm}$ excitation energy. (b) First-order room temperature Raman spectra of products generated from targets with the indicated B concentrations and excited using the $514\mathrm{nm}$ excitation energy. Adapted from 146.

Figure 33

Optical absorption spectra of boron-doped C-SWNT (solid) and pristine C-SWNT (dashed). From 64.

Figure 34

(Color online) The photoselective resonance Raman response for the $G$ modes in pristine and highly B-substituted $\left(15\mathrm{at.}\mathrm{\%}\right)$ SWNT (left). The relative softening of the $G$-line components as a function of laser energy and B doping is shown on the right panel.

Figure 35

(Color online) Spectrum line of a BN-SWNT with a measurement of the optical gap of a BN-NT from one of the loss spectra extracted from the left panel. Adapted from 10 and 11.

Figure 36

(Color online) Raman spectra excited at $229\mathrm{nm}$ on (a) BN-NT-rich area in a standard TEM carbon grid, (b) a particle of $h$-BN on the same grid (see insets), and (c) highly crystalline powder $h$-BN. Adapted from 6.

Figure 37

(Color online) Random phase approximation spectra of graphene and a $h$-BN sheet, given by the parallel and perpendicular polarizations to the plane, as well as the averaged spectrum. Adapted from 142.

Figure 38

(Color online) RPA absorption spectra of (a) bulk hexagonal BN, (b) a single sheet of hexagonal BN, and (c) six different BN tubes with increasing diameter $d$. Solid lines are calculated with a Lorentzian broadening of $0.025\mathrm{eV}$ and dashed lines with a broadening of $0.1\mathrm{eV}$. From 215.

Figure 39

(Color online) Real $\left({ϵ}_1\right)$ and imaginary $\left({ϵ}_1\right)$ parts of the dielectric function of $h$-BN calculated in the $GW+\mathrm{BSE}$ approach and in the RPA (a). Experimental data from EELS, where ${ϵ}_1$ and ${ϵ}_2$ are calculated from the loss function via a Kramers-Kronig transformation. The calculations include a Lorentzian broadening of $0.2\mathrm{eV}$ (FWHM) in order to represent the estimated experimental broadening. The light polarization is considered as parallel to the BN layers (b). From 214.

Figure 40

(Color online) Band structure of bulk $h$-BN: LDA and $GW$ approximation. From 214.

Figure 41

(Color online) Optical absorption spectrum of $h$-BN. An estimated experimental broadening of $0.1\mathrm{eV}$ calculated with YAMBO (black dashed line) and with VASP (solid line). Inset: Two-dimensional projections of the probability density of degenerate exciton states with $\lambda =\left(\mathrm{a}\right)$ 3 and (b) 4. From 214.

Figure 42

(Color online) Absorption spectra of (0,8) BN-SWNTs are shown on the left side graphic. The spectra are broadened with a Gaussian of $0.0125\mathrm{eV}$. On the right are depicted the wave function of the lowest energy bright exciton and isosurface plots of the electron probability distribution of a (8,0) BN-SWNT (a)–(c) and of a C-SWNT (d)–(f). From 163.

Figure 43

Cathodoluminescence spectrum of $h$-BN crystallite at $T=100\mathrm{K}$. (a) TEM image of the $h$-BN crystallite and (b) polychromatic CL image. From 92.

Figure 44

DOS of (a) a pure (9,9) BN nanotube, (b) a tube with a carbon impurity atom, (c) with a C atom and a perpendicular $E$ field of $0.2\mathrm{V}∕\mathrm{\AA }$, and (d) with a C atom on the opposite side and an $E$ field of $0.2\mathrm{V}∕\mathrm{\AA }$ The light polarization is parallel to the tube axis.

Figure 45

(Color online) N-doped C-SWNTs have exhibited improved conductivity upon optimal incorporation of N in substitutional configuration. Adapted from 209.

Figure 46

(Color online) Use of boron-doped nanotubes as transparent conducting films. The transmittance (at $550\mathrm{nm}$) and sheet resistance of B-doped SWNT and undoped P-SWNT films. From 134.