4 9 ( 2 0 1 1 ) 1 8 2 1 –1 8 2 8
available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/carbon
Observations of the electrical behaviour of catalytically grown scrolled graphene Andreas K. Schaper Dmitri Golberg c a b c
Houqing Hou b, Mingsheng Wang c, Yoshio Bando c,
Center for Materials Science, Philipps University of Marburg, 35032 Marburg, Germany Chemistry College of Jiangxi University, Nanchang, JX 330027, PR China WPI Center for Materials Nanoarchitectonics (MANA), National Institute for Materials Science (NIMS), Tsukuba, Ibaraki 3050044, Japan
A R T I C L E I N F O
A B S T R A C T
Carbon nanoscrolls, made of rolled up graphene, are expected to show unique electronic
Received 8 June 2010
properties different from those of nested multi-wall tubes. Here we report in situ TEM
Accepted 30 December 2010
observations and electrical measurements of the transport and breakdown behaviour of
Available online 8 January 2011
catalytically grown, multi-turn monochiral graphene scrolls, 30–65 nm in outer diameter. Generally, the low-bias IV region proved strictly linear Ohmic behaviour, non-linear increase in conductance occurred beyond an applied voltage of 0.4 V. Excellent maximum conductance values up to G 63 G0 and sustainable current-carrying capacities up to J = 8.5 · 108 A/cm2 were found in the most successful samples right before electric breakdown. Inferior values are ascribed to defect-rich or semiconducting scrolls. This study emphasizes the promising nature of carbon nanoscrolls for a number of electronic device applications. 2011 Elsevier Ltd. All rights reserved.
Increasing the functional density and power of electronic, magnetic, and optical devices requires utilization of nanostructures, or nanostructured elements, with well defined properties and high efficiency. Carbon nanotubes (CNTs) have appeared to be promising materials for the construction of photoconductors, light-emitting diodes, field-effect transistors, sensors, spintronic devices or nano-electromechanical systems [1–3]. The electronic states of an isolated graphene sheet [4,5] which is truly two-dimensional carbon and considered a zero-gap semiconductor, vary with changing boundary conditions under rolling-up into a closed seamless nanotube. Such a single-walled CNT (SWCNT) shows either metallic or semiconducting properties with an energy gap inversely propor-
tional to its diameter . Multi-walled CNTs (MWCNTs) are of a more complex electronic behaviour due to interwall interferences between the adjacent graphene layers [7–9]. First measurements of the electrical conductance by de Heer and coworkers [10,11] yielded conductance values nearly equal to the quantum conductance G G0 = 2 e2/h (e = electron charge; h = Planck constant). This suggested that the electrical current flow had mainly occurred through the outermost shells of the nested cylinder structure, and that the transport had almost been ballistic. Such a behaviour has been confirmed by the observations of fractional conductance steps in the order of G0, and of a sequential destruction of the concentric-shell geometry in electrical breakdown experiments [12–15]. Measurements also suggested the possibility of several layers to contribute to the current conduction, and of electron tunneling between the layers [7,8,16].
* Corresponding author: Address: Philipps University of Marburg, Center for Materials Science, EM&Mlab, Hans-Meerwein-Strasse, D-35032 Marburg, Germany. Fax: +49 6421 28 23383. E-mail address: [email protected]
(A.K. Schaper). 0008-6223/$ - see front matter 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.carbon.2010.12.066
4 9 ( 2 0 1 1 ) 1 8 2 1 –1 8 2 8
Recent reports have presented exceptionally high room temperature conductance of MWCNTs under vacuum, 20 G0  and up to 490 G0 , and current densities >109 A/ cm2 at 250 C [18–20]. A crucial point in regard of most of the transport measurements of MWCNTs is the quality and design of the electrode/nanotube junctions. Consequently, different variants of ‘‘end contact’’ geometries were realized [12,15,18,21–23] to overcome the complications peculiar to ‘‘side contacts’’ to MWCNTs. While the electronic properties of MWCNTs and, recently, of flat graphene have been to a significant extent elucidated, our understanding of the behaviour of scrolled graphene structures is still rather fragmentary. Carbon nanoscrolls (CNSs) represent a special type of CNTs in that they do not consist of nested graphene cylinders but are formed by rolling up of a single (or a few) graphene sheet(s) [24–29] (Sua´rez-Martinez et al.  have proposed the more convenient term ‘‘scrolled graphene’’ instead of ‘‘carbon nanoscrolls’’). Performing ab initio calculations based on a density-functional theory and local density approximations, Pan et al.  and Chen et al.  found distinct differences in the electronic and optical properties between CNSs and MWCNTs, the electroactuation behaviour was calculated by Rurali et al. . The first electrical transport measurements, performed using CNSs which were prepared by rolling up graphene layers on a SiO2/Si substrate, have recently been reported by Xie et al. . The authors obtained high current densities up to 5 · 107 A/cm2 facilitating the application of CNSs as microcircuit interconnects. Other possible electronic applications include electronic paper, terahertz oscillators, varactors, or water and ion channels [35,36], further potential interests in the peculiar properties of CNSs are summarized in . Here we report the first in situ transmission electron microscope (TEM) measurements of the electrical properties of catalytically grown CNSs using a scanning tunneling microscope (STM) probe.
Graphene nanoscrolls with outer diameters in the range 30– 65 nm, and inner diameters between 15 and 40 nm, were synthesized by thermal decomposition of an iron phthalocyanine (FePc) precursor with subsequent catalyst-assisted growth by chemical vapor deposition (CVD) at 800–1000 C on a silicon substrate using an apparatus described in a previous paper . The FePc precursor was introduced by Huang et al.  and Li et al.  to successfully promote aligned CNT growth. The preferred formation of scrolls using this precursor, and clear evidence by electron diffraction of their dominating monochiral nature was reported by Ruland et al.  (see also Lucas and Lambin ). Fig. 1a shows a diffraction pattern of such a monochiral scroll (Fig. 1b) with chirality angle h = 10.9 that has metallic properties. The chirality was determined according to the procedure of Gao et al.  and Qin . Additional proof for the monochirality comes from high-resolution TEM micrographs such as shown in Fig. 1c–e. Imaging of opposite portions of the side walls of a tube exhibits 22 and 23 graphene layers, respectively, which is clear indication of the scroll character of the tube. The texture of this scroll is of slight conical shape which produces periodically terminating graphene layers [28,43]. Identification of the scrolls during the in situ measurements occurred via their characteristic helical morphology, instead of using electron diffraction, in order to limit the total electron dose. The electrical in situ measurements were performed in a JEM-3100FEF (Omega Filter) TEM using an STM–TEM specimen holder  (for experimental details see [45–47]). The tubes under study were assembled on a copper wire as unbiased electrode, and were biased by side-contacting using a tungsten tip. This is a suitable set-up of electrically contacting of a scroll for two-terminal current–voltage (IV) measurements. In each experiment, the IV characteristics at low bias could naturally be recorded, with increasing bias voltage in some cases either the electric
Fig. 1 – (a) SAED pattern of a monochiral scrolled graphene tube (h = 10.9); (b) scroll model; (c) TEM overview of a scroll with a dislocation defect arrowed; (d and e) close-ups of the areas marked in (c) of the left and right tube wall. The number of graphene layers on the two sides counts to 22 and 23 layers, respectively, clearly designating the tube as a scroll (the image in Fig. 1e has been mirrored with respect to the tube axis for clarity).
4 9 ( 20 1 1 ) 1 8 2 1–18 2 8
contacting got worse, or early breaking of the tube occurred at dangerous structure defects. Altogether, eleven carbon tubes have been selected and analysed, seven scroll-type tubes and four nested tubes (the latter observations will be reported elsewhere).
Results and discussion
Fig. 2a–c show low-magnification micrographs and Fig. 2d the corresponding low-bias IV plots recorded in a measuring cycle with probing a single scroll of 60 nm outer diameter at successive surface points along the tube axis direction. That we have indeed a scroll structure of the type described above is confirmed by the detailed views of the CNS tip in Fig. 2e. They reveal a continuous helical twist of the flattened tube end due to the basically scroll-like nature of the tube (see similar feature in Fig. 3a). The symmetric IV relation in Fig. 2d proves perfect Ohmic behaviour in the low-bias region, the resistance changes in dependence on the probed effective length of the scroll from 14.1 kX (corresponding to a conductance of 0.92 G0) at 1 lm (a) to 10 kX (1.3 G0) at 0.75 lm (b), and 8.8 kX (1.5 G0) at 0.2 lm (c). Steadily increasing the bias, electric breakdown of this scroll was witnessed at a 0.45 mA threshold current and a bias of 1.76 V measured at 1 lm effective length, which corresponds to a maximum current density of 2.9 · 107 A/cm2. The failure of the scroll occurred abruptly as clearly seen in the accompanying video sequence . The breakdown behaviour of the scrolls was then studied in more detail. For the scroll shown in Fig. 3a and b before
and after the break, an originally linear behaviour was observed with a nonlinear rising IV curve beyond 0.4 V, as indicated in Fig. 3c. Some fluctuations in the high-bias region (before the current drops to zero after complete breakdown at 1.45 V) are apparent. No tendency toward current saturation at high biases was detected, in striking contrast to the behaviour of many MWCNTs. The scroll in Fig. 3 with an effective length of 250 nm and an outer diameter of 65 nm showed an initial (below 0.4 V) resistance of 910 X (G 14.2 G0) that decreased up to a minimum value of 660 X (G 19.5 G0) before the break. The scroll sustained a maximum current of 2.2 mA relating to a current density of 0.9 · 108 A/cm2. Final break of this and the other scrolls happened at some point between the two contacts which, in case of MWCNTs, is taken as an indication of resistive heating and of diffusive rather than ballistic transport . However, the existence of morphological disturbances along with the electron irradiation may dominate that behaviour as discussed below. The more than linear increase in conductance, which we normally detected with increasing bias voltage, and which is monitored in Fig. 3c, is also known from some reports on MWCNTs [15,17,49–52]. It was noted that this behaviour depended on geometrical conditions in such a way that large tube diameters strongly favour the conductance increase, as do shorter tube lengths [17,51–55]. With SWCNTs and smalldiameter MWCNTs of sufficient length, usually current saturation was observed [13,14,56] due to the onset of inelastic phonon scattering [57,58]. Coupling of phonons across several
Fig. 2 – (a–c) Low-magnification images indicating three different contact points during in situ electrical measurements along the tube axis; (d) IV curves at low bias voltage obtained by contacting the scroll according to the situations displayed in Fig. 2a–c. A decrease of the resistance is noted with decreasing effective tube length; (e) magnified views of the helically shaped tip of the tube revealing its scroll-type nature.
4 9 ( 2 0 1 1 ) 1 8 2 1 –1 8 2 8
Fig. 3 – Images of a short scroll before (a) and after (b) its break during the in situ breakdown experiment along with the corresponding IV characteristics (c). The dashed line indicates the linear part of the curve that extends up to 0.4 V bias, beyond a superlinear behaviour is revealed. Shortly before break some fluctuations of the current are detected.
atomic layers in few-layer graphene has been proved recently the reason for a moderate decrease also in thermal conductivity . On the other hand, electric conductance is determined by the competing processes of Bragg reflection and its influence on the electron transport in non-crossing subbands, and of Zener tunneling [51–54,60]. Hence, the observed superlinear increase in conductance and the improvement in the current carrying capability can be attributed to decreased scattering rates and increasing interlayer tunneling. Other possible reasons considered for MWCNTs are poor contact or contact modifications, which facilitate additional charge injection directly from the electrodes into non-crossing subbands of particular shells [7,16,17,52]. Whether those arguments are applicable to CNSs with their continuous graphene structure remains open. Use of the in situ method as in the present work raises further questions regarding the superlinear conductance behaviour. Under room temperature conditions, lattice defects, which may be either pre-existing defects going back to the catalytic fabrication of the samples, or interstitials and vacancies induced by the knock-on collisions between electrons and carbon atoms, accumulate to high-density agglomerates during electron irradiation leading to partial or even complete disorder of the graphene lattice . However, it was shown that holding the nanotube in situ at temperatures above
300–400 C , or between 600 and 700 C [61,63], thermal diffusion of interstitials and vacancies and their recombination, or divacancy formation, prevents clustering of the defects . Therefore, in the in situ electric experiments, moderate resistive heating may lend the carbon nanotubes, and also the scrolls, behave like self-healing structures. A TEM image sequence that illustrates the final 40 s in the electrical breakdown of another scroll is presented in Fig. 4a– d along with the related part of the IV curve in Fig. 4e. According to the diagram in Fig. 4e, this quite short scroll (length 150 nm) of 40 nm diameter reached 8 mA current at break at a bias of 1.7 V, yielding a remarkably high current density of 8.5 · 108 A/cm2. The resistance at low bias of 840 X has fallen down to 205 X, corresponding to an increase in the differential conductance from 15.4 G0 to 62.9 G0. The opposite broken ends of the scroll after failure in Fig. 4d are visualized at higher magnification in Fig. 4f and g. The images reveal graphene layers wound over each other in a typically scrolllike fashion, partially polygonized traces of the wall are discernable. Other scrolls did not show such polygonization. The wall terminates in a rather sharp breaking edge, a stepwise breakdown or fractional ablation of individual shells as it is expected to happen in MWCNTs through steps of 12 lA hight or multiples of it [12–15], and as it has convincingly been demonstrated in the excellent in situ study by
4 9 ( 20 1 1 ) 1 8 2 1–18 2 8
Fig. 4 – Snapshots of the final steps in the in situ breakdown process of a scroll at time intervals of 14 s (a and b), 17 s (b and c), and 7 s (c and d), together with the corresponding part of the IV curve (e). The close-ups in (f and g) of the broken ends reveal the turning of the graphene layer in the scroll and the non-fractional breaking process. Huang et al.  on six-walled CNTs, has never been observed with the scrolls.
The occasional polygonization of the scrolls deserves some further discussion. Polygonal cross-sections were
4 9 ( 2 0 1 1 ) 1 8 2 1 –1 8 2 8
suggested for MWCNTs by X-ray diffraction and TEM studies [65–67] but have been observed just as well in scroll-type carbon nanotubes [27–29,67]. Polygonization is concomitant with the graphene sheets loosing some of their sp2 character through hybridization towards sp3 along the flat sections of the tube or scroll with more or less coherent interfaces. Recently, Tibbetts et al.  have shown how the shape of the cross-section is determined by the chirality of the tube, providing a reasonable tube diameter. According to Charlier et al. , a polygonized cross-section induces more or less pronounced changes in the electronic band structure, the effects proved strongly depending on the actual tube configuration. The electron microscope images in Figs. 3 and 4 show, at high bias, the occurrence of other types of defects such as rippling and buckling of the scrolls. The formation of these defects is reflected by the irregularities in the upper part of the IV curves. The defects are rather the result of anisotropic knock-on damage at irradiation with high-energy electrons than induced by dissipative heating [61,62,70]. Nevertheless, it is suggested that those defects locally disturb the integrity of the graphene layers to an extent which leads to electric weakening of the scroll structure and, finally, to early failure where beam damage has become most advanced. In this scenario, the local point of breakdown could not be used as an indicator of either a basically ballistic, or an Ohmic behaviour of the originally intact scroll. An effect, particularly related to the scroll structure, eventually contributes to the fluctuations at high bias voltage: the thermally activated dislocation-mediated scroll-to-tube transformation as proposed by Berber and Toma´nek  and Sua´rez-Martinez et al. . A pre-existing dislocation in a scroll structure is seen in Fig. 1c, such dislocations generally act as scattering centres which reduce conductivity. Under strong applied electric field and in conjunction with local heating, however, the mobility of the dislocation will become enhanced forcing the above zipping-like transformation process. In comparison with the peak values of CNSs, the conductance and current density values determined for MWCNTs using the same method were significantly smaller (G = 0.3– 1.1 G0, J = 0.9–2.5 · 107 A/cm2) as was to be expected, and as we will show in an accompanying paper. The discrepancies between the results for different scroll species obtained in the present experiments may be suspected to be caused by unevenly provided electric contacting, different effective scroll lengths or diameters, or by substantial differences in the intrinsic scroll structures. The quality of contacting is always an issue in those experiments . A rough analysis of the length dependence of the resistance in Fig. 2 in fact shows a contact resistance nearly half the resistance of the scroll itself which indicates a not perfect contact in this case. Length and diameter variations may provide another explanation for the differences in properties of the scrolls in Figs. 2–4: the sample of Fig. 4 with the smallest diameter (40 nm) and shortest effective length (150 nm) shows the highest current-bearing capacity at all, the values for the samples in Fig. 3 (65/250 nm) and Fig. 2 (60 nm/1 lm) are 1–1.5 orders of magnitude smaller. On the other hand, since it is known that structural imperfections
may strongly affect the resistivity of CNTs , also deviations from the perfect scroll structure have to be considered, such as multiple scrolls in which several graphene sheets are wrapped one over each other thereby introducing a number of unsaturated graphene edges [28,74], mixtures of scroll and tube structures [75,76], or the polygonization of the cross-section discussed above. Finally, there are reasons to believe that large scatters in the electric properties of the scrolls may be caused by differences in their chirality-related structure: inferior data as, e.g., in Fig. 2 are eventually related to a semiconducting instead of a metallic scroll. While semiconducting scrolls should principially be able to carry current owing to a very small band gap, particularly at elevated temperature, they are expected to show an averaged shorter mean free path due to back-scattering by long-range disorder and thus a reduced conductance [77–79]. In the low-bias region a semiconducting scroll exhibits a linear IV behaviour as in Fig. 2d, which proves very similar to the one in MWCNTs [7,54,56,77,78,80–82].
Summary and outlook
In summary, we have shown in this paper that the particular structure of scrolled graphene tubes implicates interesting electronic behaviour, and that the scroll structure indeed shares properties with both SWCNTs and MWCNTs . High current densities of 108–109 A/cm2 were determined, they represent the ultimate electric power capacity of metallic CNSs studied in this work under in situ conditions, and suggest a near-ballistic transport behaviour. Values considerably smaller than these are supposed to belong to scrolls with numbers of non-healing structure defects, or scrolls which are semiconducting by nature. The statement in the paper by Chen et al.  according to which ‘‘chirality plays a great role in the electronic structure of scrolls’’ holds equally true for scrolls as well as for SWCNTs and MWCNTs in general, however, only the continuous rolling up of graphene into a scroll allows to get a multi-layered carbon nanotube of uniform chirality with its unprecedented properties. The high conductivity and current density values of metallic scrolls as well as the switching capabilities predicted for the semiconducting state, make the CNS-type of nanotubes strong candidates as interconnects or conductive channel material in nanoelectronics and integrated circuit applications. Their high mechanical stability will be favourable to those applications. Concerning the issue of electric contacting, the continuity of the scrolled graphene sheet guarantees immediate involvement of all wall levels at once in electrical transport, a clear advantage in side-contacting . Future work will account for our ability to synthesize intermediate diameter CNSs of high structural quality, precisely controlled number of graphene turns, and defined chirality.
Author contributions A.K.S., Y.B. and D.G. designed research; H.H. synthesized nanoscrolls; M.W. and A.K.S. performed research, A.K.S. and D.G. analyzed data; A.K.S. wrote the paper.
4 9 ( 20 1 1 ) 1 8 2 1–18 2 8
Acknowledgement A.K.S. is grateful to JSPS/DAAD for support of this research work.
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.carbon.2010.12.066.
R E F E R E N C E S
 Dekker C. Carbon nanotubes as molecular quantum wires. Phys Today 1999;52:22–8.  Avouris Ph. Molecular electronics with carbon nanotubes. Acc Chem Res 2002;35:1026–34.  Anantram MP, Le´onard F. Physics of carbon nanotube electronic devices. Rep Prog Phys 2006;69:507–62.  Novoselov KS, Geim AK, Morozov SV, Jiang D, Zhang Y, Dubonos SV, et al. Electric field effect in atomically thin carbon films. Science 2004;306:666–9.  Novoselov KS, Jiang D, Schedin F, Booth TJ, Khotkevich VV, Morozov SV, et al. Two-dimensional atomic crystals. PNAS 2005;102:10451–3.  Saito R, Dresselhaus G, Dresselhaus MS. Physical properties of carbon nanotubes. 5th ed. London: Imperial College Press; 2003.  Collins PG, Avouris P. Multishell conduction in multiwalled carbon nanotubes. Appl Phys A 2002;74:329–32.  Bourlon B, Miko C, Forro´ L, Glattli DC, Bachtold A. Determination of the intershell conductance in multiwalled carbon nanotubes. Phys Rev Lett 2004;93:176806-1–176806-10.  Scho¨nenberger C, Bachtold A, Strunk C, Salvetat J-P, Forro´ L. Interference and interaction in multi-wall carbon nanotubes. Appl Phys A 1999;69:283–95.  Frank S, Poncharal Ph, Wang ZL, deHeer WA. Carbon nanotube quantum resistors. Science 1998;280:1744–6.  Poncharal P, Berger C, Yi Y, Wang ZL, de Heer WA. Room temperature ballistic conduction in carbon nanotubes. J Phys Chem B 2002;106:12104–18.  Sanvito S, Kwon Y-K, Toma´nek D, Lambert CJ. Fractional quantum conductance in carbon nanotubes. Phys Rev Lett 2000;84:1974–7.  Collins PG, Hersam M, Arnold M, Martel R, Avouris P. Current saturation and electrical breakdown in multiwalled carbon nanotubes. Phys Rev Lett 2001;86:3128–31.  Collins PG, Arnold MS, Avouris P. Engineering carbon nanotubes and nanotube circuits using electrical breakdown. Science 2001;292:706–9.  Huang JY, Chen S, Jo SH, Wang Z, Han DX, Chen G, et al. Atomic-scale imaging of wall-by-wall breakdown and concurrent transport measurements in multiwall carbon nanotubes. Phys Rev Lett 2005;94:236802-1–4.  Kociak M, Suenaga K, Hirahara K, Saito Y, Nakahira T, Iijima S. Linking chiral indices and transport properties of doublewalled carbon nanotubes. Phys Rev Lett 2002;89:155501-1–4.  Tsutsui M, Taninouchi Y, Kurokawa S, Sakai A. Electrical breakdown of short multiwalled carbon nanotubes. J Appl Phys 2006;100:094302-1–5.  Li HJ, Lu WG, Li JJ, Bai XD, Gu CZ. Multichannel ballistic transport in multiwall carbon nanotubes. Phys Rev Lett 2005;95:086601-1–3.  Wei BQ, Vajtai R, Ajayan PM. Reliability and current carrying capacity of carbon nanotubes. Appl Phys Lett 2001;79:1172-1–3.
 Wei BQ, Spolenak R, Kohler-Redlich Ph, Ru¨hle M, Arzt E. Electrical transport in pure and boron-doped carbon nanotubes. Appl Phys Lett 1999;74:3149-1–3.  Takesue I, Haruyama J, Kobayashi N, Chiashi S, Maruyama S, Sugai T, et al. Superconductivity in entirely end-bonded multiwalled carbon nanotubes. Phys Rev Lett 2006;96:0570011–4.  Haruyama J, Tokita A, Kobayashi N, Nomura M, Miyadai S, Takazawa K, et al. End-bonding multiwalled carbon nanotubes in alumina templates: superconducting proximity effect. Appl Phys Lett 2004;84:4714-1–3.  Wang MS, Golberg D, Bando Y. Interface dynamic behavior between a carbon nanotube and metal electrode. Adv Mater 2010;22:93–8.  Bacon R. Growth, structure, and properties of graphite whiskers. J Appl Phys 1960;31:283-1–8.  Spreadborough J. The frictional behaviour of graphite. Wear 1962;5:18–30.  Bandow S. Radial thermal expansion of purified multiwall carbon nanotubes measured by X-ray diffraction. Jpn J Appl Phys 1997;36:L1403–5.  Ruland W, Schaper AK, Hou H, Greiner A. Multi-wall carbon nanotubes with uniform chirality: evidence for scroll structures. Carbon 2003;41:423–7.  Schaper AK. Carbon nanoscrolls. In: Blank VD, Kulnitskiy BA, editors. Carbon nanotubes and related structures. Trivandrum (India): Research Signpost; 2008. p. 1–49 [chapter 1].  Chuvilin AL, Kuznetsov VL, Obraztsov AN. Chiral carbon nanoscrolls with a polygonal cross-section. Carbon 2009;47:3099–105.  Sua´rez-Martinez I, Grobert N, Ewels CP. Nomenclature of carbon nanoforms. Advances in carbon nanomaterials: science and applications. Pan Stanford Publishing, submitted for publication.  Pan H, Feng Y, Lin J. Ab initio study of electronic and optical properties of multiwall carbon nanotube structures made up of a single rolled-up graphite sheet. Phys Rev B 2005;72:085415-1–5.  Chen Y, Lu J, Gao Z. Structural and electronic study of nanoscrolls rolled up by a single graphene sheet. J Phys Chem C 2007;111:1625–30.  Rurali R, Coluci VR, Galva˜o DS. Prediction of giant electroactuation for papyruslike carbon nanoscroll structures: first-principles calculations. Phys Rev B 2006;74:085414-1–4.  Xie X, Ju L, Feng X, Sun Y, Zhou R, Liu K, et al. Controlled fabrication of high-quality carbon nanoscrolls from monolayer graphene. Nano Lett 2009;9:2565–70.  Fogler MM, Castro Neto AH, Guinea F. Effect of external conditions on the structure of scrolled graphene edges. Phys Rev B 2010;81:161408-1–4.  Shi X, Cheng Y, Pugno NM, Gao H. Tunable water channels with carbon nanoscrolls. Small 2010;6:739–44.  Hou H, Schaper AK, Weller F, Greiner A. Carbon nanotubes and spheres produced by modified ferrocene pyrolysis. Chem Mater 2002;14:3990–4.  Huang S, Dai L, Mau AWH. Patterned growth and contact transfer of well-aligned carbon nanotube films. J Phys Chem B 1999;103:4223–7.  Li DC, Dai L, Huang S, Mau AWH, Wang ZL. Structure and growth of aligned carbon nanotube films by pyrolysis. Chem Phys Lett 2000;316:349–55.  Lucas AA, Lambin Ph. Diffraction by DNA, carbon nanotubes and other helical nanostructures. Rep Prog Phys 2005;68:1181–249.  Gao M, Zuo JM, Twesten RD, Petrov I, Nagahara LA, Zhang R. Structure determination of individual single-wall carbon nanotubes by nanoarea electron diffraction. Appl Phys Lett 2003;82:2703-1–3.
4 9 ( 2 0 1 1 ) 1 8 2 1 –1 8 2 8
 Qin L-C. Determination of the chiral indices (n, m) of carbon nanotubes by electron diffraction. Phys Chem Chem Phys 2007;9:31–48.  Schaper AK, Hou H, Treutmann W, Phillipp F. Multi-walled carbon nanotubes without and with metal filling. J Metastable Nanocryst Mater 2005;23:301–4.  Nanofactory Instruments AB, see webpage: http:// www.nanofactory.com/.  Wang M, Wang J, Chen Q, Peng L-M. Fabrication and electrical and mechanical properties of carbon nanotube interconnections. Adv Funct Mater 2005;15:1825–31.  Golberg D, Mitome M, Kurashima K, Zhi CY, Tang C, Bando Y, et al. In situ electrical probing and bias-mediated manipulation of dielectric nanotubes in a high-resolution transmission electron microscope. Appl Phys Lett 2006;88:123101-1–3.  Costa PMFJ, Fang X, Wang S, He Y, Bando Y, Mitome M, et al. Two-probe electrical measurements in transmission electron microscopes – behavioral control of tungsten microwires. Microsc Res Technol 2009;72:93–100.  See video document in the Supplementary Data showing the abrupt failure of the nanoscroll displayed in Fig. 2.  Chiu H-Y, Deshpande VV, Postma HW, Lau CN, Miko´ C, Forro´ L, et al. Ballistic phonon thermal transport in multiwalled carbon nanotubes. Phys Rev Lett 2005;95:226101-1–4.  De Pablo PJ, Graugnard E, Walsh B, Andres RP. A simple, reliable technique for making electrical contact to multiwalled carbon nanotubes. Appl Phys Lett 1999;74:3231–3.  Anantram MP. Current-carrying capacity of carbon nanotubes. Phys Rev B 2000;62:R4837–40.  Svizhenko A, Anantram MP. Effect of scattering and contacts on current and electrostatics in carbon nanotubes. Phys Rev B 2005;72:085430-1–085430-10.  Svizhenko A, Anantram MP, Govindan TR. Ballistic transport and electrostatics in metallic carbon nanotubes. IEEE Trans Nanotechnol 2005;4:557–62.  Bourlon B, Glattli DC, Placais B, Berroir JM, Miko C, Forro´ L, et al. Geometrical dependence of high-bias current in multiwalled carbon nanotubes. Phys Rev Lett 2004;92:0268041–4.  Chen Q, Wang A, Peng L-M. Establishing ohmic contacts for in situ current–voltage characteristic measurements on a carbon nanotube inside the scanning electron microscope. Nanotechnology 2006;17:1087–98.  Liang YX, Li QH, Wang TH. Current saturation in multiwalled carbon nanotubes by large bias. Appl Phys Lett 2004;84:33791–3.  Zamkov M, Alnaser AS, Shan B, Chang Z, Richard P. Probing the intrinsic conductivity of multiwalled carbon nanotubes. Appl Phys Lett 2006;89:093111-1–3.  Yao Z, Kane CL, Dekker C. High-field electrical transport in single-wall carbon nanotubes. Phys Rev Lett 2000;84:2941–4.  Tison Y, Giusca CE, Hayashi VSY, Silva SRP. The inner shell influence on the electronic structure of double-walled carbon nanotubes. Adv Mater 2008;20:189–94.  Ghosh S, Bao W, Nika DL, Subrina S, Pokatilov EP, Lau CN, et al. Dimensional crossover of thermal transport in fewlayer graphene. Nat Mater 2010;9:555–8.  Banhart F, Fu¨ller T, Redlich Ph, Ajayan PM. The formation, annealing and self-compression of carbon onions under electron irradiation. Chem Phys Lett 1997;269:349–55.
 Li J, Banhart F. The engineering of hot carbon nanotubes with a focused electron beam. Nano Lett 2004;4:1143–6.  Rodriguez-Manzo JA, Terrones M, Terrones H, Kroto HW, Sun L, Banhart F. In situ nucleation of carbon nanotubes by the injection of carbon atoms into metal particles. Nat Nanotechnol 2007;2:307–11.  Krasheninnikov AV, Lehtinen PO, Foster AS, Nieminen RM. Bending the rules: contrasting vacancy energetics and migration in graphite and carbon nanotubes. Chem Phys Lett 2006;418:132–6.  Amelinckx S, Bernaerts D. The geometry of multishell nanotubes. In: Yoshimura S, Chang RPH, editors. Supercarbon: synthesis, properties and application. Berlin: Springer; 1998. p. 31–80.  Maniva Y, Fujiwara R, Kira H, Tou H, Nishibori E, Takata M, et al. Multiwalled carbon nanotubes grown in hydrogen atmosphere: an X-ray diffraction study. Phys Rev B 2001;64:073105-1–4.  Reznik D, Olk CH, Neumann DA, Copley JRD. X-ray powder diffraction from carbon nanotubes and nanoparticles. Phys Rev B 1995;52:116–24.  Tibbetts K, Doe R, Ceder G. Polygonal model for layered inorganic nanotubes. Phys Rev B 2009;80:014102-1–014102-10.  Charlier J-C, Lambin Ph, Ebbesen TW. Electronic properties of carbon nanotubes with polygonized cross sections. Phys Rev B 1996;54:R8377–80.  Smith BW, Luzzi DE. Electron irradiation effects in single wall carbon nanotubes. J Appl Phys 2001;90:3509-1–7.  Berber S, Toma´nek D. Stability differences and conversion mechanism between nanotubes and scrolls. Phys Rev B 2004;69:233404-1–4.  Sua´rez-Martinez I, Savini G, Heggie M. First principles modelling of scroll-to-nanotube defect: screw-type dislocation. Mater Sci Forum 2006;527–529:1583–6.  Dai H, Wong EW, Lieber CM. Probing electrical transport in nanomaterials: conductivity of individual carbon nanotubes. Science 1996;272:523–6.  Zhou O, Fleming RM, Murphy DW, Chen CH, Haddon RC, Ramirez AP, et al. Defects in carbon nanostructures. Science 1994;263:1744–7.  Amelinckx S, Bernerts D, Zhang SB, Van Tendeloo G, Van Landuyt J. A structure model and growth mechanism for multishell carbon nanotubes. Science 1995;267:1334–8.  Sun X, Li R, Stansfield B, Dodelet J-P, Me´nard G, De´silets S. Controlled synthesis of pointed carbon nanotubes. Carbon 2007;45:732–7.  McEuen PL, Bockrath M, Cobsen DH, Yoon Y-G, Louie SG. Disorder, pseudospins, and backscattering in carbon nanotubes. Phys Rev Lett 1999;83:5098–101.  Martel R, Schmidt T, Shea HR, Hertel T, Avouris Ph. Singleand multi-wall carbon nanotube field-effect transistors. Appl Phys Lett 1998;73:2447-1–3.  Buitelaar MR, Bachtold A, Nussbaumer T, Iqbal M, Scho¨nenberger C. Multiwall carbon nanotubes as quantum dots. Phys Rev Lett 2002;88:156801-1–4.  Tans SJ, Verschueren ARM, Dekker C. Room-temperature transistor based on a single carbon nanotube. Nature 1998;393:49–52.  Javey A, Guo J, Wang Q, Lundstrom M, Dai H. Ballistic carbon nanotube field-effect transistors. Nature 2003;424:654–7.  Nihey F, Hongo H, Yudasaka M, Iijima S. A top-gate carbonnanotube field-effect transistor with a titanium-dioxide insulator. Jpn J Appl Phys 2002;41:L1049–51.