Selectively growing carbon nanotubes requires an atomic scale understanding of the nanotube – catalyst particle interface, since this is the place where the carbon atoms incorporation takes place. The structure of the catalyst surface [1] as well as its chemical composition [2] seem to be of importance. Using Density Functional Theory calculations [3], we calculate the structure of the interface and adhesion energies of carbon tubes on small particles of pure nickel or nickel carbide. The junctions between single-walled carbon nanotubes and nickel clusters are on the cluster surface, and not at subsurface sites, irrespective of the nanotube chirality, temperature, and whether the docking is gentle or forced. Gentle docking helps to preserve the pristine structure of the SWNT at the metal interface, whereas forced docking may partially dissolve the SWNT in the cluster. We then use these results to fine tune a tight binding model [4, 5], which makes it possible to further investigate the effect of the temperature, chirality and particle size. [1] A. R. Harutyunyan et al., Science, 326, 116-20 (2009).[2] S. Hofmann et al., J. Phys. Chem. C, 113, 1648–1656 (2009); R. Sharma et al. Nano Letters 9, 2, 689-94 (2009); H. Yoshida et al. Nano Letters 8, 7, 2082-6 (2008).[3] A. Börjesson, W. Zhu, H. Amara, C. Bichara and K. Bolton, Nano Letters 9, 3, 1117-20 (2009).[4] H. Amara, C. Bichara and F. Ducastelle, Phys. Rev. Lett., 100, 056105, (2008).[5] H. Amara, J. M. Roussel, C. Bichara, J.-P. Gaspard and F. Ducastelle, Phys. Rev. B 79, 014109 (2009).

Growth mechanism of carbon nanotubes (CNT) at the atomic level is far from being completely understood. We still do not know how to fully control CNT growth or even if such control is possible. Which factors and forces decide if the evolving sp²-network of atoms will adhere to the catalyst particle, encapsulating it, or the graphitic cap will lift-off to extend itself into a hollow tube? Such details cannot be studied experimentally, but can be explored by molecular dynamics (MD). We perform a large scale of MD simulations (over 500 simulations) to establish the role of the adhesion strength (W_ad) of the graphitic cap to the catalyst and the temperature (and C diffusion rate) in CNT nucleation¹. To study the competition of tubular structure vs. encapsulated catalyst as a function of temperature and W_ad, we perform MD simulations for a temperature range between 200 K and 1400 K, at 200 K increments. For each temperature we vary W_ad from 0 to 0.3 eV/C. Such systematic observations allow us to build a statistically representative map of CNT nucleation and define the conditions for growth or fullerene-like metal encapsulation (catalyst poisoning). Our simulations show that weak W_ad, sufficient kinetic energy (high temperature), or fast C diffusion favor catalytic CNT growth. An analysis of these results in light of previous models (curvature energy, thermal decohesion, and requirement of fast C diffusion), shows that below 600 K carbon-diffusion on the catalyst surface limits the growth, but at higher temperature it fully depends on cap lift-off. With an informed set of parameters in which CNT growth is expected to happen, we obtain the longest simulated nanotube structures within reasonable simulation × a significant step towards achieving CNT realistic computational modeling. This study reveals a means of designing metal catalysts for better CNT synthesis, potentially at desirably low temperatures, which is especially important for possible in situ growth for nanoelectronics applications.¹ M. A. Ribas, F. Ding, P. B. Balbuena, and B. I. Yakobson J. Chem. Phys., in press.

Chiral symmetry is definitive for almost all essential properties of carbon nanotubes. To understand the origin of chirality distribution or to possibly control this distribution is a key to tantalizing applications. Proposed view of an arbitrary tube as a base-zigzag type but with a screw dislocation along the axis [1] leads to prediction of simple relationship K ~ Sin(C) between the steady-state growth rate K and the chiral angle C. This trend can indeed be clearly seen as the material abundance distribution in several reported experiments. The mechanism of carbon insertion into the kink area is also supported by the observed stepwise rotation of the growing tube [2], although details are awaiting further investigation. Recently, we have explored the stage of nucleation, when the forming cap can also be seen as a structure with a dislocation. Detailed analysis of caps' topology and their relative energies allows us to evaluate the probabilities of different chiralities, leading to a general form P(C) ~ exp[-A*Cos(C+C_o)]. We will discuss how the parameters A and C_o can be tuned by the growth conditions, therefore how the preferred chiral type (angle C) can in principle be controlled, and what are thermodynamic limitations of the intriguing possibilities that follow. [1] F. Ding, A.R. Harutyunyan, and B.I. Yakobson, Proc. Natl. Acad. Sci., 106, 2506 (2009); [2] M. Marchand, C. Journet, D. Guillot, J.-M. Benoit, B.I. Yakobson, and S.T. Purcell, Nano Lett., 9, 2961-2966 (2009).

We study the possibility of chirality selection during the catalytic chemical vapor deposition (CVD) of a Single-Walled Carbon Nanotubes (SWNT) by density functional simulations. When nanotubes grow by root growth, the tube's chirality is fixed by the chirality of the initial cap that nucleates on the catalyst cluster surface. We previously studied the energies of caps of different chiralities on a flat, fixed Ni(111) layer [1]. We now consider different caps binding to various planes and apices of a 55 atom cluster of Fe and Ni. Firstly, we find that the caps bind more strongly to these nucleation sites than to the flat (111) layer. A second point is that the cap causes considerable distortion of the metal cluster, while the caps themselves are relatively undistorted. This is because C-C bonds are stronger than Ni-Ni bonds. Of the three high symmetry nucleation sites on the Fe cluster, the binding of (6,5) and (7,5) caps to the pyramidal corner is 2 eV per cap stronger on average than to other sites. Binding energies also depend on the cap diameter, this explains why the pyramid site is less stable for certain caps. For the Ni cluster, there is a strong preference to the zigzag and chiral caps; binding energies for these two caps are generally greater than for armchair caps by 2.5 eV per cap. Lastly, there is strong segregation of carbon atoms to the outside of the cluster, with little dissolution into cluster.1. S Reich, L Li, J Robertson, Chem Phys Lett 421 469 (2006)

Growth of carbon nanotubes (CNTs) and nanofibers (CNFs) continues to be an enigma. Though a lot is known about the growth conditions for CNTs and CNFs, the reason that a given growth produces one or the other is still not clear. We propose a new mechanism to explain the formation of CNFs over CNTs. Experimentally; we have observed growth of fibers is favored at conditions of higher pressure or thicker catalyst films. Observation of the impact of pressure on catalyst annealing shows that higher pressure results in the formation of larger particles at the expense of smaller particles, indicative of Ostwald ripening. These larger particles favor the growth of fibers over CNTs. All the prominent catalysts for carbon nanotube growth have extended solid and liquid solubility with carbon, and also form an eutectic at low carbon concentrations. The Gibbs Thompson effect predicts size dependent suppression of melting point. Hence smaller particle exhibit a single liquid phase, conducive to the growth of CNTs. However, larger particles, depending on the growth conditions (temperature and pressure and the extent of carbon dissolution) may exist in a dual solid-liquid phase. The relative concentration of liquid phase increases with increasing carbon dissolution. Beyond a threshold carbon concentration, the dual phase becomes energetically unfavorable, causing the particle to revert to a single solid phase regime by discarding excess carbon. As shown by Helveg, et al., [Nature 427, 426 (2004)] the resulting carbon layers replicate the morphology of the catalyst particle, leading to the characteristic stacked-cone or bamboo morphology of the CNF. TEM investigations validating the above observations and a theoretical model describing the same will be presented.

When materials are very thin in one or more dimensions, their equilibrium shapes are often curved/bent. Such shapes commonly represent a compromise between elastic strain energy and other thermodynamic forces (e.g. related to surface stresses, electrostatic interactions, or adsorption). Examples include ZnO and SnO₂ nanobelts, silica/carbonate helicoids, and graphene sheets and nanoribbons. Here, we demonstrate that when the equilibrium shape of a nanomaterial is not flat/straight, important fundamental material properties may be orders of magnitude different from their bulk counterparts. We focus here primarily on the graphene edges. Graphene in three dimensions is a codimension c = 1 material; the codimension is c = D – d = 3 – 2 = 1, where D is the dimensionality of the space in which the material is embedded and d is the dimensionality of the object. By contrast, a flat graphene sheet has c = 2 – 2 = 0. We use the REBO-II interatomic potential to calculate the edge orientation dependence of the edge energy and edge stresses of graphene with c = 0 and c = 1. The edge stress for all edge orientations is compressive with c = 0. Both edge energy and stresses are in reasonable agreement with DFT calculations. The compressive edge stresses in c = 0 lead to edge buckling (out-of-the-plane of the graphene sheet) for all edge orientations (c = 1). The edge buckling in c = 1 reduces all edge energies and dramatically reduces all edge stresses to near zero (more than an order of magnitude drop). We also report the effect of codimension on the free energy and entropy of a graphene sheet and the elastic properties of ZnO nanohelices.

While carbon nanotubes (CNTs) have been produced industrially in ton-scale quantities for nearly two decades, scalable manufacturing processes that precisely control the structure, length, and alignment of CNTs are needed to realize the exceptional properties of CNTs at larger scales. Specifically, vertically aligned CNT forests are a model system for further understanding what limits the growth of indefinite CNTs, and are building blocks for novel microstructures and multifunctional thin films. First, I will review our current understanding of the limiting mechanisms of CNT forest growth, where interactions among the growing population of CNTs govern their collective growth behavior. We elucidate mechanisms of CNT forest self-organization, steady growth, and termination using a holistic approach combining in situ and ex situ X-ray scattering, measurements of forest mass and height, and chemical analysis of the reaction atmosphere. Second, I will present our recent approaches to synthesis of hybrid polymer-CNT and nanowire-CNT forests, which are amenable to large-scale manufacturing via continuous CVD, rolling, and printing schemes.

Production of carbon nanotubes (CNT) with controlled structure has become an important area in the nanotubes field since there are multiple applications in which they could be used, particularly if their structure is known and predictable [1-3]. Uses of CNT range from semiconductors and digital memory to drug delivery in biomedical applications. As well as in polymer applications such as paint, tires, etc. Among the production techniques, CVD (chemical vapor deposition) is not only the least expensive but also the easiest to scale up to be used in industry. In several of our previous studies [4-8] we have shown that the structure of single-walled carbon nanotubes (SWNT), e.g. diameter, chirality, length, can be controlled by varying different conditions either related to the catalyst or to the reaction. However, there are a number of important applications that are more suited for MWNT (multi-walled carbon nanotubes). Therefore, it is important to develop similar degree of control in MWNT growth as that reached with SWNT. In this study, we have investigated the effects of varying in a controlled way the interactions between the active metals and the supports on the resulting CNT morphologies. It is shown that characteristics such as number of walls, diameter, and length, can be changed in a systematic fashion by varying parameters related to the catalyst synthesis and the reaction conditions. Specifically, the catalytic systems investigated are cobalt and cobalt-molybdenum supported on different silicas and aluminas, with special attention to the phenomena that occur on these catalysts during the various steps in the CNT growth, that is pre-calcination and pre-reduction, as well as during contact with the carbonaceous gas feed. References[1] Paradise M., Goswami T. Mater. Design. 2007, 28, 1477. [2] Gojny F.H., Wichmann M.H.G., Fiedler B., Schulte K. Compos. Sci. Technol. 2005, 65, 2300 [3] Merkoçi A., Pumera M., Llopis Xavier, Pérez B., del Valle M., Alegret S. Trac-Trend Anal. Chem. 2005, 24, 9, 826.[4] Resasco D.E., Herrera J.E., Balzano L. J. Nanosci. Nanotechno. 2004, 4, 4, 398.[5] Resasco D.E., Alvarez W.E., Pompeo F., Balzano L., Herrera J.E., Kitiyanan B., Borgna A. J. Nanopart. Res. 2002, 4, 1-2, 131.[6] Alvarez W.E., Pompeo F., Herrera J.E., Balzano L., Resasco D. E. Chem. Mater. 2002, 14, 4, 1853.[7] Monzon, A, Lolli, G. Cosma, S., Sayed-Ali, M. Resasco, D.E., J. Nanosci. Nanotech. 2008, 8, 6141 [8] Irurzun V.M., Tan Y., Resasco D.E., Chem. Mater., 2009, 21, 2238.

While carbon nanotube (CNT) synthesis by thermal chemical vapor deposition (CVD) can be achieved using myriad combinations of carbon-containing precursors and nanoparticle catalysts, there is relatively little mechanistic understanding of how the precursor is incorporated into the CNTs. Identification of efficient CNT growth pathways is essential for cost-effective scaling of CNT synthesis for commercial applications and for understanding of the potential environmental impacts of CNT manufacturing. Using a custom-built apparatus which enables rapid (200 C/s) temperature control of a resistively heated silicon platform, we have successfully decoupled the “preheating” of the gas precursor upstream of the reactor from the local thermal environment of the catalyst. This decoupled technique enables testing of gas species by direct injection without preheating, thus enabling precise quantification of the performance of various hydrocarbons for CNT growth. We identify that thermal decomposition of the typical input reactant mixture (C₂H₄/H₂) creates a mixture of species including various alkanes, alkenes, and alkynes, as well as polycyclic aromatic hydrocarbons (PAHs). These analyses, in concert with real-time measurements of growth kinetics via the forest height, reveal a positive correlation between growth rate and the relative abundance of specific molecules, including acetylene, propyne, and vinylacetylene. By directly delivering these species with the normal growth gases C₂H₄/H₂ in the absence of preheating, we identify that alkynes selectively enhance forest growth (rate and ultimate height) and are thereby a family of efficient precursors for CNT manufacturing. Further, by direct delivery of these molecules to the heated catalyst we achieve efficient CNT growth with a 10-fold reduction in PAH production in the exhausted gas mixture as well as a 55% decrease in energy consumption during synthesis. Finally, we show molecules that have previously been proposed as key precursors to CNT formation, such as methane and benzene, exhibit low conversion efficiencies in our system.

The main obstacle that hinders ubiquitous application of carbon nanotubes is our inability to obtain large amount of reasonably homogeneous material. Despite intense research and noticeable achievements in preferential growth of semiconducting SWCNTs there is only a limited understanding of exactly what determines chirality and thereby the electronic structure of grown SWCNT. Our studies reveal that the variation of the noble gas ambient during thermal conditioning of the catalyst, and in combination with oxidative and reductive species, alters the fraction of tubes with metallic conductivity from about 20% of the population to a maximum of 91%. The tubes have been identified based on Raman, photoluminescence and electrical (field effect transistor performance) characterizations. In the meantime, in situ environmental transmission electron microscopy (ETEM) studies reveal that ambient variation leads to differences in both morphology and coarsening behavior of the nanoparticles used to nucleate nanotubes. The relationship between catalyst morphology rearrangements and resulting nanotube electronic structure, will be presented.

Substrate materials for vertical carbon nanotube (CNT) growth have been mostly limited to brittle and flat dielectric oxide materials, such as alumina and silica. An alternative substrate for diverse applications of vertical CNT arrays would (i) be stable at the moderately high temperatures needed for CNT growth, (ii) be mechanically compliant and stretchable for transfer to the widest array of base structures (flexible polymers, non-planar structures, etc.), (iii) allow for the rapid and large-scale fabrication of complex device architectures, and (iv) be electro-conductive with an ohmic contacts with CNT strands for efficient use of input electrical power. In this presentation we will demonstrate versatile carbon hybrid film composed of vertical CNT arrays grown on reduced graphene film substrate, which satisfies all 4 aforementioned requirements. Thin large-area reduced graphene films have been fabricated by deposition from aqueous colloidal suspensions of graphene oxide platelets, decorated them with patterned catalyst particles, and grown CNTs at exceptionally high growth rates and at temperatures that result in the substrate being converted to an electrically conductive graphene-based film. Such carbon hybrid films have excellent flexibility and stretchability, can be readily transferred to any substrate including non-planar surfaces, and were found to have ohmic electrical contacts throughout the junctions in the CNT/metal catalyst/reduced graphene film system. As examples among many possible applications, the carbon hybrid films were readily integrated into a field-emitting device, which demonstrated an excellent performance. (Related articles published: Adv. Mater. in-press; Nano Lett. 9, 1427-1432, 2009; Adv. Mater. 20, 2480-2485, 2008.)

Deterministic control of carbon nanotube growth remains difficult, in particular for single-walled nanotubes, which limits their widespread applicati