Three-Dimensional Nitrogen-Doped Multiwall Carbon Nanotube

Oct 17, 2013 - File failed to load: https://cdn.mathjax.org/mathjax/contrib/a11y/accessibility-menu.js .... To the best of our knowledge, this is the ...
0 downloads 0 Views 2MB Size
Download PDF
Letter pubs.acs.org/NanoLett

Three-Dimensional Nitrogen-Doped Multiwall Carbon Nanotube Sponges with Tunable Properties Changsheng Shan,† Wenjie Zhao,†,‡ X. Lucas Lu,‡ Daniel J. O’Brien,§ Yupeng Li,‡ Zeyuan Cao,‡ Ana Laura Elias,∥ Rodolfo Cruz-Silva,# Mauricio Terrones,∥,⊥,# Bingqing Wei,*,‡ and Jonghwan Suhr*,○ †

Center for Composite Materials, University of Delaware, Newark, Delaware 19716, United States Department of Mechanical Engineering, University of Delaware, Newark, Delaware 19716, United States § Weapons and Materials Research Directorate, U.S. Army Research Laboratory, Aberdeen Proving Ground, Maryland 21005, United States ∥ Department of Physics and Center for 2-Dimensional and Layered Materials, The Pennsylvania State University, 104 Davey Lab, University Park, Pennsylvania 16802-6300, United States ⊥ Department of Chemistry and Department of Materials Science and Engineering, The Pennsylvania State University, 104 Davey Lab, University Park, Pennsylvania 16802-6300, United States # Research Center for Exotic Nanocarbons (JST), Shinshu University, Wakasato 4-17-1, Nagano, 380-8553, Japan ○ Department of Polymer Science and Engineering, Sungkyunkwan University, 300 Chunchun-dong, Jangan-gu, Suwon 440-746, South Korea ‡

S Supporting Information *

ABSTRACT: A three-dimensional (3D) nitrogen-doped multiwall carbon nanotube (NMWCNT) sponge possessing junctions induced by both nitrogen and sulfur was synthesized by chemical vapor deposition (CVD). The formation of “elbow” junctions as well as “welded” junctions, which are attributed to the synergistic effect of the nitrogen dopant and the sulfur promoter, plays a critically important role in the formation of 3D nanotube sponges. To the best of our knowledge, this is the first report showing the synthesis of macroscale 3D N-MWCNT sponges. Most importantly, the diameter of N-MWCNT can be simply controlled by varying the concentration of sulfur, which in turn controls both the sponge’s mechanical and its electrical properties. It was experimentally shown that, with increasing diameter of N-MWCNT, the elastic modulus of the sponge increased while the electrical conductivity decreased. The mechanical behaviors of the sponges have also been quantitatively analyzed by employing strain energy function modeling. KEYWORDS: Carbon nanotubes, mechanical and electrical properties, soft materials, CVD

T

macroscopic sponge with randomly interconnected multiwall carbon nanotubes (MWCNTs) using CVD with ferrocene- and 1,2-dichlorobenzene-containing chloride elements.14,15 The MWCNT sponge showed a high structural flexibility and robustness, as well as wettability to organics, but did not show any evidence of junctions established between nanotubes. Hashim et al. fabricated covalently bonded 3D boron-doped MWCNTs (B-MWCNTs) solids via boron induced nanojunctions using a similar CVD approach.2 The boron dopant was believed to be the key factor for creating “elbow-like” junctions and covalent nanojunctions, which give the 3D BMWCNT an excellent mechanical flexibility and oil absorption functionality. However, the diameter of B-MWCNTs and the physical properties of this 3D solid could not be controlled. Sulfur and nitrogen are two common elements that have been extensively but individually used for the synthesis and

hree-dimensional (3D) carbon nanotube (CNT)-based macrostructures have recently received extensive attention because the outstanding properties of one- and two-dimensional nanostructures, such as nanotubes and graphene, have not been successfully translated into key engineering applications.1−6 However, experimental realization of a controlled macroscale 3D CNT architecture represents a huge challenge.2 To create CNT-based 3D frameworks exhibiting nanoscale functional junctions, two strategies have been often used: modification of the straight, tubular morphology of CNTs as well as the formation of junctions established between CNTs.2 Theoretical studies have predicted that the presence of heteroatoms can generate pentagon-heptagon defects, thus causing structural reorganization and the formation of stable bends within CNTs.7−9 Experimental studies have also demonstrated that the presence of heteroatoms, including boron, chlorine, nitrogen, and sulfur, can induce dramatic tubule morphology changes in CNTs, such as nanoscale multijunctions.2,10−14 For example, Gui et al. synthesized a © XXXX American Chemical Society

Received: August 19, 2013 Revised: October 7, 2013

A

dx.doi.org/10.1021/nl403109g | Nano Lett. XXXX, XXX, XXX−XXX

Nano Letters

Letter

Figure 1. (A) Photo of N-MWCNT sponge. (B, C) SEM images of N-MWCNT sponges. (D) TEM image of an “elbow” junction in N-MWCNT sponge. (E) TEM and (F) HRTEM images of “welded” junction between two N-MWCNTs.

property improvement of CNTs. Sulfur is not only a catalyst able to enhance the CNT growth16,17 but has also been used as a branching promoter for the formation of carbon heptagons and for obtaining junctions between CNTs.11,12 However, 3D CNT macrostructures have not been achieved yet using sulfur. Similar to sulfur, nitrogen can also induce structural changes in CNT growth7,18−21 and also improve electrical properties of CNTs, thus expanding their application to fields such as energy, catalysis, and sensing.19,22−27 For example, N-MWCNT showed an extremely high capacity in a lithium ion battery and high electrocatalytic activity for oxygen reduction.25,28 However, only powder- and film-like N-doped CNTs have been reported.18−21 To date, macroscale 3D CNT entities with “welded” junctions induced by nitrogen and sulfur have not been demonstrated. Here we report the facile synthesis of 3D sponge-like Ndoped MWCNT (N-MWCNT) architectures using both nitrogen and sulfur during CVD. “Elbow” and “welded” junctions established between two MWCNTs were clearly

observed in this 3D N-MWCNT sponge. Control experiments revealed that both nitrogen and sulfur significantly affected the formation of these junctions. Most importantly, the tube diameter in the N-MWCNT sponges can be simply controlled by varying the concentration of sulfur, which in turn controls the sponge’s mechanical and electrical properties. Figure 1A shows a typical macroscopic 3D N-MWCNT sponge. A large quantity (0.5−1 g) of N-MWCNT sponges was produced in only 40 min. Scanning electron microscopy (SEM) characterization indicates that the 3D N-MWCNT sponge consists of randomly orientated and entangled CNTs (Figure 1B). Both “elbow” and “welded” junctions established between two CNTs induced by nitrogen and sulfur were observed under high-magnification SEM (Figure 1C). Transmission electron microscopy (TEM) images revealed detailed structures of the “elbow” and “welded” junctions in the 3D N-MWCNT sponge. (see Figure 1D and E, see Supporting Information, Figure S1 for more TEM images). As shown in Figure 1F, the junction between N-MWCNTs was made up of graphitic layers, B

dx.doi.org/10.1021/nl403109g | Nano Lett. XXXX, XXX, XXX−XXX

Nano Letters

Letter

Figure 2. (A) XPS spectrum of N-MWCNT sponge. (B) N1s XPS spectrum of N-MWCNT sponge. (C) S2p XPS spectrum of N-MWCNT sponge. (D) Raman spectra of N-MWCNT sponge (upper spectrum) and pristine MWCNT array (lower spectrum).

Figure 3. (A) SEM image of pristine MWCNTs array synthesized by ferrocene/xylene without nitrogen and sulfur elements. (B) SEM image of NMWCNTs powders synthesized by ferrocene/pyridine in the absence of a sulfur element. (C) SEM image and (D) TEM image of MWCNTs agglomerates synthesized using ferrocene/thiophene/xylene in the absence of a nitrogen element.

indicating that the “welded” junction occurred between NMWCNTs. The chemical composition and structure in the 3D NMWCNT sponge were studied by a series of spectroscopies including X-ray photoelectron (XPS), Raman, and energydispersive X-ray (EDX) as well as element analysis. Figure 2A shows the XPS spectrum of the N-MWCNT sponge. An obvious N1s peak was observed at ca. 400 eV, indicating the

presence of N elements in the sponges. The pyridinic N, graphitic N (coordinated N atoms substituting inner C atoms in the graphene layers), and oxidized N corresponded to binding energy peak positions located at ca. 398.2, 400.9, and 402.6 eV, respectively (see Figure 2B). The intensity of graphitic N at the peak of 400.9 eV was much higher than those of the other two peaks at 398.2 and 402.7 eV, suggesting the dominance of graphitic N doping. The S2p peak in the XPS C

dx.doi.org/10.1021/nl403109g | Nano Lett. XXXX, XXX, XXX−XXX

Nano Letters

Letter

Figure 4. SEM images (A, C, and E) and TEM images (B, D, and F) of diameter-controlled N-MWCNTs sponges synthesized at different concentrations of thiophene with 0.25 (A and B), 0.5 (C and D), and 0.75 vol % (E and F).

spectrum was essentially featureless in Figure 2C. This indicates that there is essentially no presence of sulfur within this NMWCNT sponge. The EDX spectra in the SEM instrument show that the content of nitrogen is 4.26 wt % in this NMWCNT sponge (Figure S2). The content of sulfur (0.03 wt %) was too small to detect, indicating that almost no sulfur was trapped in this N-MWCNT sponge. Furthermore, EDX characterization in the TEM instrument (Figure S3) was used to investigate whether the sulfur was doped into the CNTs lattice. No sulfur was detected within the junction and straight tubular parts of N-MWCNTs contained in this sponge (Figure S3A and B), indicating the absence of sulfur in the NMWCNTs. However, a peak of sulfur was observed in the NMWCNTs with an iron nanoparticle even if the content of

sulfur was as low as 0.09 wt % (Figure S3C). This implies that the sulfur combines first with iron, forming FeSx, as reported previously.29,30 The content of nitrogen was also measured by elemental analysis. The average content of nitrogen corresponded to 4.28 wt % (Figure S4), which is consistent with the EDX spectra. In addition, Raman spectroscopy was used to study the defects of the N-MWCNT sponge. Figure 2D shows the Raman spectrum of the N-MWCNT sponge (upper spectrum) and the reference Raman spectrum of a pristine MWCNT array synthesized in the absence of nitrogen and sulfur (lower spectrum). An intense D-band was observed in the NMWCNT sponge. The ratio of D/G (0.93) of the N-doped MWCNT sponge was significantly larger than that (0.61) of the D

dx.doi.org/10.1021/nl403109g | Nano Lett. XXXX, XXX, XXX−XXX

Nano Letters

Letter

iron catalyst and, as a consequence, results in the formation of an amorphous carbon sphere (Figure S5B). The optimal concentration of thiophene for synthesizing 3D N-MWCNTs sponges appears to be in the range of 0.25−0.75 vol % (Figure 4). Interestingly, the diameters of the N-MWCNTs within the sponges can be controlled by varying the concentration of sulfur (thiophene). N-MWCNT sponges with tube diameters of 40−110 nm, 60−140 nm, and 80−180 nm could be synthesized in the presence of thiophene with concentrations of 0.25, 0.5, and 0.75 vol %, respectively. The diameters of the NMWCNTs within the sponges increased with increasing thiophene concentration, which is consistent with earlier reports.29 Sulfur seems to react with Fe and form a FeSx−Fe eutectic alloy phase. A CNT nucleus would be easier to form on the lower free energy surface of FeSx−Fe eutectic than on the pure Fe because the FeS−Fe eutectic has a lower free energy surface than pure Fe has. The size of the FeS−Fe eutectic increases with increasing sulfur concentration. This results in an increase in the diameter of the CNTs. Figure 5A shows that N-MWCNT sponges exhibit linear stress−strain behavior at low strains (99%)15 and boron doped MWCNT sponge (99%).2 Previous reports showed that 3D CNT macrostructures could not be obtained if the synthesis is performed in the presence of only nitrogen or only sulfur.11,12,18−21 However, as demonstrated in the present work, 3D N-MWCNT sponges could be obtained if synthesis is carried out in the presence of sulfur and nitrogen simultaneously. To investigate the roles of nitrogen and sulfur during the formation of 3D N-MWCNTs macrostructures, three control experiments were investigated. First, in the absence of nitrogen (used xylene instead of pyridine) and sulfur (no thiophene) in the CVD process, no MWCNT sponge was obtained, but pristine MWCNT arrays (Figure 3A) were. Second, a similar CVD process without adding thiophene (sulfur source) was conducted to synthesize N-MWCNTs by using a pyridine/ferrocene mixture. NMWCNT powder with tube diameters of 30−60 nm was produced (Figure 3B). It is noted that there are many “elbow” junctions in N-MWCNT powder, suggesting that the “elbow” junctions can be induced by nitrogen doping. However, the sponge could not be formed only in the presence of nitrogen in the CVD process. Thus it is clear that sulfur plays a critical role in creating the 3D N-MWCNT structure. Third, MWCNTs were synthesized using a ferrocene/thiophene/xylene mixture without nitrogen. Although sulfur was sufficient to promote the formation of carbon heptagons and pentagons, thus obtaining CNT “welded” junctions,2,11,12 only a small amount of MWCNT agglomerates were obtained without nitrogen. The SEM and TEM images showed only a small quantity of “elbow” and “welded” junctions between undoped MWCNTs (Figure 3C and D), suggesting that the formation of these junctions was induced at least in part by the sulfur. However, the MWCNT agglomerates showed fewer “elbow” and “welded” junctions when compared to N-MWCNTs sponge. Therefore, the presence of nitrogen can significantly promote the formation of “elbow” and “welded” junctions. Formation of such junctions in 3D N-MWCNT sponges could possibly be attributed to the synergistic effect of both nitrogen and sulfur. Further study is needed to develop a better understanding of the mechanism controlling junction formation in the presence of nitrogen and sulfur. It is evident that the concentration of sulfur plays an important role in tuning the diameters of the N-MWCNTs. First, when the concentration of thiophene was 0.125 vol %, only a small quantity of N-MWCNT powder could be obtained (Figure S5A). According to our control experiments and previous reports,11,12 sulfur could be used as a “triggering element” for creating “welded” junctions between MWCNTs.13,30 The low concentration of sulfur could not produce sufficient junctions between MWCNTs; that is, the sponge could not be obtained at a low sulfur concentration. Second, when the concentration of thiophene was 1.5 vol %, the sponge could not be produced either. Sulfur could also poison the nanotube growth at high concentrations.31 Therefore, the high concentration of sulfur inhibits the activity of the

Figure 5. Characterization results for N-MWCNT sponges. (A) Representative dots of stress response of N-MWCNT sponges of different diameters under compression. The dots are fitted with strain energy density function (SEDF). (B) Elastic modulus of the sponge increases with the diameter of the N-MWCNT. (C, D) Shear modulus and Lamé constant of N-MWCNT with different diameters modeled by SEDF. (E) The energy dissipation ratio of the sponge decreases with diameter. (F) The conductivity of the sponge decreases with diameter.

behavior at a high strain level, similar to many porous materials.32,33 For N-MWCNT sponges with densities of 35 mg/cm3, the initial modulus (in the curve’s “toe” region) was found to be 20 KPa. The stress−strain curve increases sharply and exhibits a “heel” region with upward concave curvature analogous to many biological connective tissues, for example, E

dx.doi.org/10.1021/nl403109g | Nano Lett. XXXX, XXX, XXX−XXX

Nano Letters

Letter

ligament and cartilage.34 After 40% compressive strain, the modulus of the sponge shows a remarkable increase (∼5 times), reaching 100 KPa. In porous N-MWCNTs sponges, the interstitial space among CNTs gets smaller under larger compressive strain. The heel region could be mainly attributed to densification of the CNTs. The stiffening effect at a high strain level, which is similar to densification stage in open-cell foams,35 can be a result of the packing of bent and buckled CNTs. It should be noted that the stress−strain behavior of the bulk CNT sponges does not reveal the CNT junctions’ properties. Further research is underway to understand the mechanical properties of the junctions themselves as well as their effect on bulk sponge response. Figure 5B correlates the sponge’s elastic modulus with the diameter of the N-MWCNTs. The elastic modulus is calculated as the slope of the stress−strain curve at 20% and 50% strain. Linear regression shows a significant correlation between these two parameters (P = 4 × 10−4). The elastic modulus of the sponge increases with the average diameter of N-MWCNT. In this N-MWCNT sponge, bending of the curved N-MWCNTs dominates the stress response under compression. The bending stiffness (D) can be expressed as D = EI, where E and I represent the Young’s modulus and moment of inertia, respectively. Previous work reported that the Young’s modulus of a single CNT would dramatically decrease with increasing diameter.36 On the other hand, I is proportional to the fourth power of the diameter (I ∝ d4, d is the diameter of NMWCNT), so it is reasonable to see that the elastic modulus of the sponges increases linearly with MWCNT diameter. We employed a strain energy density function (SEDF) for hyper-elastic material to quantitatively characterize the stress− strain behavior of N-MWCNT sponges.37 In this model, the strain energy density (U) is decomposed into two components (see following equations). The Uincomp part is associated with incompressibility of volume-constant distortion, and the other portion, Ucomp, is associated with the compressibility or specific volume change: U = Uincomp + Ucomp n=1

Uincomp =

∑ r

+ n=1

Ucomp =

∑ s

{

shear modulus increases with the N-MWCNT diameter. Results from this constitutive model are consistent with the experimental data in terms of the relationship between the elastic modulus and the N-MWCNT diameter (Figure 5B). The average Lamé constant of sponges with MWCNT diameters ranging from 70 to 160 nm increases from 6 to 10.7 Pa (Figure 5D). Figure 5E shows that the energy dissipation ratio (the ratio between energy loss and elastic energy, see Supporting Figure S6 for detail) decreases with increasing diameter of the NMWCNTs. For N-MWCNT sponges with similar densities, larger diameter CNTs result in fewer CNTs inside the sponge. The energy loss, which originates from the friction between CNTs, might be smaller for sparser CNT sponges. As shown in Figure 5A, the elastic energy may increase with increasing diameter. In addition, the elastic energy is inversely proportional to the energy dissipation ratio, showing that the energy dissipation ratio decreases with increasing diameter. The electrical conductivities of the N-MWCNTs sponges were measured using a four-probe method. These conductivities were varied from 0.15 and 0.21 S/cm. As shown in Figure 5F, the conductivity of sponges with the same density increases with decreasing N-MWCNT diameter. In the case of the same density, the sponges with smaller diameters of N-MWCNTs should have a larger number of N-MWCNTs, which would have more connections in the N-MWCNT sponge. Therefore, the conductivity of sponges with smaller diameters may be better than that of sponges with larger diameters. The two-probe electrical resistivity was monitored in situ during the compression process by attaching thin metal wires to the top and bottom sponge surface between the compression stages (Figure S8). The resistance of sponge showed a reversible change between 20% and 40% strain, decreasing by about 37% (from 650 Ω down to 410 Ω) during compression from 20% to 40% after the fifth cycle. After five cycles, the behavior is consistent over 30 cycles without degradation of sponge conductivity, indicating that the N-MWCNT sponges have potential applications in piezoresistance sensors. In summary, we reported the growth of 3D-networked Ndoped MWCNT sponges directly through an efficient CVD synthetic process. Detailed experiments and characterization revealed that by coupling both nitrogen and sulfur the synergistic effect was mainly responsible for creating “elbowlike” junctions and “welded” junctions. Elemental analysis showed that the nitrogen was doped into this N-MWCNTs sponge with content of ca. 4.28 wt %, while there was no doping of sulfur. The diameters of N-MWCNTs from 40 to 110 nm to 70−180 nm within our sponges could be controlled by varying the concentration of thiophene with developed synthesis processes. Importantly, the resulting 3D N-MWCNTs sponges showed mechanical and electrical properties that could be controlled by varying the diameter of N-MWCNT. If optimized, these 3D N-MWCNT sponges could be used in various applications such as catalysis, energy storage, biological system, 3D porous electrodes, and mechanical/structural applications.

(1)

A n 2n (λ1 + λ 22n + λ32n − 3) 2n

Bn −2n (λ1 + λ 2−2n + λ3−2n − 3) 2n Λn ln(J )2n − 2n

}

(2)

n=1

∑ (A n − Bn)ln(J ) r

where An, Bn, and Λn are material constants and can be derived from SEDF modeling fitting to experimental data; r and s are termination points of the summation; λi are the principal relative stretches in the orthogonal direction; and J (= λ1λ2λ3) is the relative ratio of volume change. A seven-parameter SEDF was adopted for CNT sponge compression in this study (r = 3 and s = 1) according to nonlinearity of stress−strain curves. With SEDF, material parameters were curve-fitted by using a least-squares regression analysis of the experimental data. The shear modulus (G = ∑rn=1n(An + Bn) and Lamé constant (Λ = Λ1) for the NMWCNT sponge were determined. The shear modulus of NMWCNT sponges with diameters of 160, 110, and 70 nm were 0.019, 0.016, and 0.014 MPa, respectively (Figure 5C). The

■

ASSOCIATED CONTENT

S Supporting Information *

Experimental details, additional characterization data, and supporting figures (S1−S8). This material is available free of charge via the Internet at http://pubs.acs.org. F

dx.doi.org/10.1021/nl403109g | Nano Lett. XXXX, XXX, XXX−XXX

Nano Letters

■

Letter

(26) Chen, T.; Cai, Z.; Yang, Z.; Li, L.; Sun, X.; Huang, T.; Yu, A.; Kia, H. G.; Peng, H. Adv. Mater. 2011, 23, 4620−4625. (27) Lee, J. M.; Park, J. S.; Lee, S. H.; Kim, H.; Yoo, S.; Kim, S. O. Adv. Mater. 2011, 23, 629−633. (28) Shin, W. H.; Jeong, H. M.; Kim, B. G.; Kang, J. K.; Choi, J. W. Nano Lett. 2012, 12, 2283−2288. (29) Wei, J. Q.; Zhu, H. W.; Jia, Y.; Shu, Q. K.; Li, C. G.; Wang, K. L.; Wei, B. Q.; Zhu, Y. Q.; Wang, Z. C.; Luo, J. B.; Liu, W. J.; Wu, D. H. Carbon 2007, 45, 2152−2158. (30) Ren, W. C.; Li, F.; Bai, S.; Cheng, H. M. J. Nanosci. Nanotechnol. 2006, 6, 1339−1345. (31) Mohlala, M. S.; Liu, X.-Y.; Witcomb, M. J.; Coville, N. J. Appl. Organomet. Chem. 2007, 21, 275−280. (32) Cha, K. J.; Kim, D. S. Biomed. Microdevices 2011, 13, 877−883. (33) Kai, D.; Prabhakaran, M. P.; Stahl, B.; Eblenkamp, M.; Wintermantel, E.; Ramakrishna, S. Nanotechnology 2012, 23, 095705. (34) Fratzl, P. Collagen: Structure and Mechanics; Springer Science and Business Media: New York, 2008. (35) Gibson, L. J., Ashby, M. F., Harley, B. A., Eds.; Cellular Materials in Nature and Medicine; Cambridge University Press: New York, 2010. (36) Arenal, R.; Wang, M.-S.; Xu, Z.; Loiseau, A.; Golberg, D. Nanotechnology 2011, 22, 265704. (37) Attard, M. M.; Hunt, G. W. Int. J. Solids Struct. 2004, 41, 5327− 5350.

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was financially supported by the U.S. Air Force Office of Scientific Research MURI Grant (FA9550-12-1-0035) entitled “Synthesis and Characterization of 3-D Carbon Nanotube Solid Networks”.

■

REFERENCES

(1) Dag, S.; Senger, R. T.; Ciraci, S. Phys. Rev. B 2004, 70, 205407. (2) Hashim, D. P.; Narayanan, N. T.; Romo-Herrera, J. M.; Cullen, D. A.; Hahm, M. G.; Lezzi, P.; Suttle, J. R.; Kelkhoff, D.; MunozSandoval, E.; Ganguli, S.; Roy, A. K.; Smith, D. J.; Vajtai, R.; Sumpter, B. G.; Meunier, V.; Terrones, H.; Terrones, M.; Ajayan, P. M. Sci. Rep. 2012, 2, 363−370. (3) Wu, M.; Dai, J.; Zeng, X. Prog. Chem. 2012, 24, 1050−1057. (4) Dimitrakakis, G. K.; Tylianakis, E.; Froudakis, G. E. Nano Lett. 2008, 8, 3166−3170. (5) Nardecchia, S.; Carriazo, D.; Luisa Ferrer, M.; Gutierrez, M. C.; del Monte, F. Chem. Soc. Rev. 2013, 42, 794−830. (6) Li, E. Y.; Marzari, N. ACS Nano 2011, 5, 9726−9736. (7) Sumpter, B. G.; Huang, J.; Meunier, V.; Romo-Herrera, J. M.; Cruz-Silva, E.; Terrones, H.; Terrones, M. Int. J. Quantum Chem. 2009, 109, 97−118. (8) Dunlap, B. I. Phys. Rev. B 1992, 46, 1933−1936. (9) Kahaly, M. U. J. Appl. Phys. 2009, 105, 024312. (10) Sharifi, T.; Nitze, F.; Barzegar, H. R.; Tai, C.-W.; Mazurkiewicz, M.; Malolepszy, A.; Stobinski, L.; Wagberg, T. Carbon 2012, 50, 3535−3541. (11) Romo-Herrera, J. M.; Cullen, D. A.; Cruz-Silva, E.; Ramirez, D.; Sumpter, B. G.; Meunier, V.; Terrones, H.; Smith, D. J.; Terrones, M. Adv. Funct. Mater. 2009, 19, 1193−1199. (12) Romo-Herrera, J. M.; Sumpter, B. G.; Cullen, D. A.; Terrones, H.; Cruz-Silva, E.; Smith, D. J.; Meunier, V.; Terrones, M. Angew. Chem., Int. Ed. 2008, 47, 2948−2953. (13) Valles, C.; Perez-Mendoza, M.; Castell, P.; Martinez, M. T.; Maser, W. K.; Benito, A. M. Nanotechnology 2006, 17, 4292−4299. (14) Gui, X.; Wei, J.; Wang, K.; Cao, A.; Zhu, H.; Jia, Y.; Shu, Q.; Wu, D. Adv. Mater. 2010, 22, 617−621. (15) Gui, X.; Cao, A.; Wei, J.; Li, H.; Jia, Y.; Li, Z.; Fan, L.; Wang, K.; Zhu, H.; Wu, D. ACS Nano 2010, 4, 2320−2326. (16) Cheng, H. M.; Li, F.; Su, G.; Pan, H. Y.; He, L. L.; Sun, X.; Dresselhaus, M. S. Appl. Phys. Lett. 1998, 72, 3282−3284. (17) Zhu, H. W.; Xu, C. L.; Wu, D. H.; Wei, B. Q.; Vajtai, R.; Ajayan, P. M. Science 2002, 296, 884−886. (18) Liu, H.; Zhang, Y.; Li, R.; Sun, X.; Desilets, S.; Abou-Rachid, H.; Jaidann, M.; Lussier, L.-S. Carbon 2010, 48, 1498−1507. (19) Jia, N. Q.; Wang, L. J.; Liu, L.; Zhou, Q.; Jiang, Z. Y. Electrochem. Commun. 2005, 7, 349−354. (20) Tang, C. C.; Bando, Y.; Golberg, D.; Xu, F. F. Carbon 2004, 42, 2625−2633. (21) Jang, J. W.; Lee, C. E.; Lyu, S. C.; Lee, T. J.; Lee, C. J. Appl. Phys. Lett. 2004, 84, 2877−2879. (22) Zhang, J.; Lei, J.; Pan, R.; Leng, C.; Hu, Z.; Ju, H. Chem. Commun. 2011, 47, 668−670. (23) Sadek, A. Z.; Zhang, C.; Hu, Z.; Partridge, J. G.; McCulloch, D. G.; Wlodarski, W.; Kalantar-zadeh, K. J. Phys. Chem. C 2010, 114, 238−242. (24) Jiang, S.; Zhu, L.; Ma, Y.; Wang, X.; Liu, J.; Zhu, J.; Fan, Y.; Zou, Z.; Hu, Z. J. Power Sources 2010, 195, 7578−7582. (25) Gong, K.; Du, F.; Xia, Z.; Durstock, M.; Dai, L. Science 2009, 323, 760−764. G

dx.doi.org/10.1021/nl403109g | Nano Lett. XXXX, XXX, XXX−XXX