Inside...
- The Meeting Scene
- Exhibitors
- Proceedings
- Outstanding Symposium Papers
- Technical Sessions/Program
- Special Events
- Awards
- Tutorial Sessions
Get Involved...
Resource Center
Send Us Your FeedbackAbstracts - Symposium Q: Nanowires and Carbon Nanotubes -- Science and Applications
SYMPOSIUM Q
November 27 - December 1, 2006
Chairs
| Prabhakar Bandaru Dept. of Mechanical Engineering, Materials Science Program University of California-San Diego MC 0411 9500 Gilman Dr. La Jolla, CA 92093-0411 858-534-5325 |
Morinobu Endo Faculty of Engineering Shinshu University 4-17-1 Wakasato Nagano-shi, 380-8553 Japan 81-26-269-5201 | |
| Ian Kinloch Dept. of Materials Science and Metallurgy University of Cambridge Pembroke St. Cambridge, CB2 3QZ United Kingdom 44-1223-762965 |
Apparao M. Rao Dept. of Physics Clemson University 107 Kinard Laboratory Clemson, SC 29634-0978 864-656-6758 |
Symposium Support
U.S. Army Research Office
(see Proceedings Library at www.mrs.org/publications_library)
as volume 963E
of the Materials Research Society
Symposium Proceedings Series.
This volume may be published in print format after the meeting.
* Invited paper
Chairs: Prabhakar Bandaru, Morinobu Endo and Ian Kinloch
Monday Morning, November 27, 2006
Room 312 (Hynes)
8:00 AM Q1.1
Atomistic Theory and Simulation of Electronic and Molecular Transport in Carbon Nanotubes. Yongqiang Xue, College of Nanoscale Science & Engineering, University at Albany-SUNY, Albany, New York.
Inorganic quasi-one-dimensional (Q-1D) nanostructures including carbon nanotubes have been exploited extensively for applications in solid state electronic and optoelectronic devices. Recent development in functionalization schemes to impart solubility and biological functions to these novel materials have also opened up exciting opportunities on the “wet” side of Q-1D nanostructure science and technology. In this talk, we investigate the application of carbon nanotubes as novel transport channels for electrons and molecules using atomistic simulation. (1) Electronic transport: We present a Green's Function based self-consistent tight-binding study of electron transport through carbon nanotube devices, which takes fully into account the 3D atomistic nature of the electronic processes. We discuss insights obtained from such atomistic study on the contact, length, and diameter dependence of junction conductance and self-consistent study of current transport through carbon nanotube devices. (2) Molecular transport: We present molecular dynamics simulation of molecules transporting through carbon nanotubes in vacuo and in aqueous environment for applications in engineered flow channels. The molecules studied include both rigid and flexible molecules. We show that in the absence of water solvation, the dynamics of molecule transporting through the nanotube channel display interesting features for understanding friction and energy transfer phenomena at the atomic scale. For nanotubes dissolved in water, the dynamics of molecular transport reflects competition between the van der Waals, hydrophobic and hydrogen bonding interactions in the nanotube/water/molecule complex, where the details of water adsorption inside the nanotube channel plays an important role.
8:15 AM Q1.2
Dielectric Response of Carbon and Boron Nitride Nanotubes from First-principles Calculations. Boris Kozinsky and Nicola Marzari; Massachusetts Institute of Technology, Cambridge, Massachusetts.
We present a complete characterization of the electrostatic response of isolated single- and multi-wall carbon (CNT) and boron nitiride nanotubes (BNNT) using first-principles calculations and density-functional theory. The longitudinal polarizability of a single nanotube is sensitive to the band gap and its radius, and in multi-wall nanotubes and bundles it is trivially given by the sum of the polarizabilities of the constituent tubes. The transverse polarizability of both types of nanotubes is insensitive to band gaps and chiralities and depends only on the radius. However, the transverse response and screening properties of BNNTs are qualitatively different from those of both metallic and semiconducting CNTs. Their comparison helps to illuminate the fundamental differences of electronic interactions in the two materials, that are inherited from the corresponding two-dimensional sheets. The screening of the external field in CNTs is much stronger than in BNNTs and has a very different radius dependence. The transverse response in BNNTs is found to be that of an insulator, while in CNTs it is intermediate between metallic and semiconducting. For carbon nanotubes we construct a simple electrostatic model based on a scale-invariance relation that captures accurately the first-principles results and allows calculation of transverse response in any multi-wall CNT. Our results have practical implications for selective growth of different types of nanotubes using aligning electric fields and for Raman characterization of nanotubes.
8:30 AM Q1.3
A First-principles Study of Pd-covered Semiconducting Carbon Nanotubes. Wenguang Zhu1,2 and Efthimios Kaxiras1; 1Department of Physics and Division of Engineering and Applied Sciences, Harvard University, Cambridge, Massachusetts; 2Center for Computational Materials, Institute for Computational Engineering and Sciences, Departments of Physics and Chemical Engineering, University of Texas, Austin, Texas.
Carbon nanotube field-effect transistors (CNFET) are regarded as potential building blocks for future nanoelectronics, in which the interaction between a carbon nanotube and metal contacts and the resulting electronic structure effects are crucial for device properties. In this talk, we present recent results on the properties of semiconducting single wall carbon nanotubes in contact with Pd, in a fully covered geometry that resembles experimental setups. We use first-principles calculations to determine the electronic structure, charge-transfer effects, electrostatic potential and Fermi level alignment at the interfaces between the metal contact and various semiconducting single-wall carbon nanotubes.
8:45 AM Q1.4
Pressure Induced Transformation of Carbon Nanotubes into New Crystalline Polymerized Nanotube Phase: A Molecular Dynamics Study. Susumu Saito and Yuichiro Yamagami; Department of Physics, Tokyo Institute of Technology, Tokyo, Japan.
We study the structural stabilities and the interconversion of carbon nanotubes and various nanostructured material phases of carbon under high pressure using the constant-pressure molecular dynamics combined with the transferable tight-binding Hamiltonian. The model can reproduce the energetics of interlayer interaction for sp2 carbon layers as well as the covalent-bond energies given by the density-functional theory. This interlayer interaction is of essential importance in considering the stability of carbon nanotube solids under external pressure. From the molecular-dynamics study, we find that at lower pressure the nanotube solid transforms into graphitic phases, while under the external pressue of 20 GPa the single-walled (10,10) carbon nanotube solid is found to transform into the new crystalline carbon phase with the body-centered tetragonal cell. All the atoms (four atoms per cell) in this new phase are equivalent and are tetrahedrally coordinated. Interestingly, this new phase can be considered as the polymerized carbon nanotubes having much smaller diameter than the original (10,10) nanotubes. The further first-principles study of this new phase has revealed the structural details and the high stability even without external pressure. The electronic properties of this new anisotropic crystalline carbon will be discussed and compared to that of diamond. In addition to single-walled carbon nanotube solids, we also report the structural transformation of double-wall nanotube solids and peapods under various pressure values. Double-walled carbon nanotube solids as well as peapods studied are found to be more stable than its constituent units, i.e., single-walled carbon nanotube solids and solid C60 at lower pressure, and do not show any structural transformation. On the other hand, at 20 GPa or higher, these materials are found to show interesting structural transformations. Detailed external-pressure dependence will be discussed.
9:00 AM Q1.5
Challenges in Chirality Assignment for Carbon Nanotubes. Francesca Clemente1,2, Olivier Richard1, Thomas Hantschel1 and Wilfried Vandervorst1,2; 1IMEC, Leuven, Belgium; 2K.U.Leuven, INSYS, Leuven, Belgium.
The determination of carbon nanotubes (CNTs) structural and electronic properties is of major importance for the development of CNTs-based science. In case of Single Wall nanotubes (SWNT) research, the most relevant breakthrough is that Raman Spectroscopy (RS) leads to chirality assignment of individual, isolated tubes, through a fast, non-destructive, contactless, in situ characterization. The radial breathing mode frequency, fingerprint of CNTs in the Raman spectrum, is linked to tube’s diameter, which is univocally related to its chirality. Although RS has been widely employed to assign chiral numbers (m, n) of isolated SWNT, through a systematic study of the available literature we found that these attributions are model dependent. For a (10,10) tube, the calculated values of the radial breathing mode frequency span in the range of 40 cm-1. Moreover, when dealing with bundled SWNT, Van der Waals interactions between neighboring tubes need to be taken into account. Modeling bundles is computationally achieved by constraining the tubes in specific symmetries, with strong assumptions on their geometrical arrangements, which often deviate from tube entanglement in real samples: (m,n) values for bundled SWNTs are, thus, harder to be gathered. As the chirality is intimately related to the metallic or semiconducting character of SWNT, the model dependency of its assignment is a concern for electronic applications of nanotubes in the post-CMOS era. We report the chirality indexes extracted on the basis of two the most accredited models proposed in the literature: not only the (m,n) values are different but also the electric character of the tube can be misattributed. The need for an accurate, model independent chirality assignment calls for an improvement of the Raman data analysis. In this contribution, the development of a methodology for diameter, and then chirality, assignment is discussed, for both isolated and bundled SWNT. Tubes with a narrow distribution of diameters are used to “calibrate” the Raman response for chirality determination, through a correlated analysis of SWNT with Transmission Electron Microscopy (TEM). Different steps of the procedure such as CNTs dispersion and the role of substrate are investigated.
9:15 AM Q1.6
Band-gap Modulation and Kohn Anomalies in Two-dimensional Graphite and Single-wall Carbon Nanotubes. Georgii G. Samsonidze1, Eduardo B. Barros2, Hyungbin Son1, Riichiro Saito3, Gene Dresselhaus4 and Mildred S. Dresselhaus1,5; 1Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology, Cambridge, Massachusetts; 2Departamento de Fisica, Universidade Federal do Ceara, Fortaleza, Ceara, Brazil; 3Department of Physics, Tohoku University and CREST JST, Sendai, Japan; 4Francis Bitter Magnet Laboratory, Massachusetts Institute of Technology, Cambridge, Massachusetts; 5Department of Physics, Massachusetts Institute of Technology, Cambridge, Massachusetts.
The electron response to the phonon distortions in a two-dimensional graphite sheet (graphene) and single-wall carbon nanotubes (SWNTs) is analysed from the viewpoint of group theory and within the tight binding approach. The highest frequency phonon mode at the Κ point is shown to open a dynamical electronic band gap in graphene, similar to the previously reported band-gap modulation in metallic SWNTs induced by the highest frequency phonon mode at the Γ point. The impact of electron dynamics on phonon dispersion in graphene and metallic SWNTs is investigated by constructing the dynamical matrix to second-order in perturbation theory. It is shown that the band-gap modulation in graphene softens the frequency of the highest mode at the Κ point forming the Kohn anomaly in the phonon dispersion. The strong electron screening further enhances the Kohn anomalies in metallic SWNTs, resulting in a Peierls transition in small-diameter SWNTs.
9:30 AM Q1.7
Thermodynamic Solubility of Carbon Nanotubes in Common Solvents. Jonathan Coleman, School of Physics, Trinity College Dublin, Ireland, Ireland.
Dispersion of single wall carbon nanotubes in a wide range of high surface tension solvents has been demonstrated. In general SWNT can be satisfactorily dispersed by sonication at concentrations in the range 0.1-1 mg/ml. These dispersions have been characterized by TEM, AFM, spectroscopy and IR photoluminescence. However, as these initial dispersions are diluted with additional sonication the SWNT bundles tend to exfoliate. This results in the decrease in the mean bundle diameter from ~10nm at CNT~0.1 mg/ml to <2nm at CNT<10-2 mg/ml. At the same time the population of individual SWNT increases from ~5% to ~80%. Detailed analysis shows that the concentration dependence of the bundle size is controlled by the presence of an equilibrium bundle number density. More interestingly, in certain special solvents such as NMP, these process occur spontaneously upon dilution, without the need for sonication. The kinetics of this process have been measured. At equilibrium, the mean bundle diameter scales with the square root of concentration again indicating the present of an equilibrium bundle number density. Interestingly this equilibrium number density is exactly that for the system to balance on the dilute-semi-dilute boundary. A model describing the enthalpy of mixing for these systems has been developed and confirmed by experiment. This allows us to understand some of the criteria required for a solvent to disperse nanotubes successfully. This model also shows that in the best solvents the enthalpy of mixing is very close to zero. This means that the free energy of mixing is negative and that the nanotubes are truly dissolved in these systems. Finally, the ability to effectively debundle nanaotubes allows us to demonstrate advanced applications such as extremely low percolation thresholds in conductive composites.
Chairs: Prabhakar Bandaru and Ian Kinloch
Monday Morning, November 27, 2006
Room 312 (Hynes)
10:15 AM *Q2.1
Geometry-Controlled Carbon Nanotubes. Sungho Jin, Materials Science & Engineering, University of California, San Diego, La Jolla, California.
For successful engineering applications of carbon nanotubes and other nanowires, an ability to control their basic configurations is essential, for example, in terms of geometry manipulations and appropriate placements on a substrate. In this talk, various fabrication techniques and microstructural analyses for controlling the nanotube geometry such as diameter, length, alignment, spacing, periodicity, bending, branching, opening, cutting, shortening, and bonding will be discussed. Growth of patterned nanotubes or nanocone arrays, sharp bending of nanotubes for creation of 90o bent or zig-zag nanotubes will be discussed. Additional geometry modifications such as the removal of catalyst metal particles from aligned multiwall nanotubes for purification and uncontaminated electrochemical reactions, tip opening and filling of nanotubes for nanocomposite formation, and coating of nanotubes with high-density Pt catalyst nanoparticles will also be discussed. The implications of such geometry controls for physical, electronic, chemical, mechanical, and bio-related properties will be discussed in relation to potential technical applications such as field emission devices, sensors, nanosprings, nanosolenoids, AFM probes, fuel cell electrodes, electronic circuit nano interconnections, and nanocomposites.
10:45 AM Q2.2
Fast and Position-Controlled Growth of Single-walled Carbon Nanotubes. Zuqin Liu1, David Styers-Barnett1, Alex Puretzky1, Chris Rouleau1, IIia Ivanov1, Dongning Yuan2, Jie Liu2 and David Geohegan1; 1Center for Nanophase Materials Sciences, Oak Ridge National Laboratory, Oak Ridge, Tennessee; 2Department of Chemistry, Duke University, Durham, North Carolina.
A novel pulsed laser-assisted chemical vapor deposition (PLA-CVD) technique is described for the synthesis of single-walled carbon nanotubes (SWNTs) in localized regions at very high growth rates. Pulsed laser heating provides precise temporal and spatial control of heat pulses to investigate the nucleation and growth mechanisms of nanotubes, and the possibility for laser direct-writing of nanotube devices. A pulsed Nd:YAG laser (1.06 micron wavelength) with 10,000 W peak power and variable pulse width (> 0.5 ms) was used to systematically irradiate various catalysts on Si wafers and TEM grids using identical gas mixtures as in normal thermal CVD in order to understand the minimum nucleation time, growth rates, growth kinetics and nucleation efficiency of nanotubes grown by this technique. The resulting time-dependent temperature profiles were measured by in situ pyrometry with 2 ms temporal resolution. Nanotubes grown directly on TEM grids were examined sequentially by SEM, Raman, and HRTEM (Hitachi HF-2000) to measure their length, diameter, and wall number as a result of the pulsed thermal treatments. Growth rate averages were estimated using the measured nanotube lengths and the duration of time for which substrate temperatures exceeded 800°C. All of the nanotubes observed on TEM grids (over 200) resulting from PLA-CVD were SWNTs, in contrast to growth by thermal CVD. Average growth rates resulting from single laser pulses ranged from several microns/second up to 100 microns/second, which exceeds previous reported estimates by roughly an order of magnitude. Such high growth rates enable the growth of micron-long SWNTs on single laser pulses. The possibility of multiple laser pulses to repeatedly stop and restart growth was also assessed. SWNTs were observed to grow incrementally by this procedure up to maximum lengths of several microns. Finally, an in-situ fabrication approach to “write” individual SWNT field-effect transistors by PLA-CVD on pre-defined electrodes will be described. The resulting SWNT transistors exhibited pronounced ambipolar transport characteristics with on/off ratios up to 1000. The PLA-CVD approach appears promising for the realization of in-situ direct-writing of integrated SWNT electronics. This research was conducted in the Functional Nanomaterials Theme at the Center for Nanophase Materials Sciences, which is sponsored at Oak Ridge National Laboratory by the Division of Scientific User Facilities, U.S. Department of Energy.
11:00 AM Q2.3
Combinatorial Development of Catalysts for CVD Synthesis of Carbon Nanotubes. Benjamin Hertzberg, Jonathan R. Petrie and R. Bruce van Dover; Materials Sci. & Eng, Cornell University, Ithaca, New York.
Chemical vapor deposition (CVD) allows synthesis of carbon nanotubes with a high degree of control over tube crystallinity, diameter, morphology, yield, and spatial location. CVD requires the use of metal nanoparticles to catalyze growth, but the mechanisms are not well understood and there is little theoretical guidance for the development of improved catalysts. We have developed a “combinatorial materials science” approach to identify superior catalysts, specifically focussing on multi-element metal compositions that yield nanotubes with minimal defects at relatively high growth rate and relative low deposition temperature. Typical catalyst combinations comprise at least one mid-transition metal (e.g., Fe, Ni, or Co) with a second transition metal such as Mo. We prepare ultrathin film composition spreads of two or three elements by cosputtering followed by a high temperature anneal in Ar/H2, thereby forming a dense array of nanoparticles. CVD nanotube growth is accomplished using a simple tube reactor and an ethylene precursor. Variations in yield and tube morphology are readily evaluated using an SEM. This approach allows a wide range of compositions in a binary or ternary system to be synthesized and evaluated in a single experiment; typical films cover ~10 at.% to ~90 at.% of each elemental constituent. A Ni-Fe composition spread grown at 700 °C was used as a model system; it this system it was found that a composition in the vicinity Fe0.2Ni0.8 yielded the highest degree of crystallinity and lowest degree of segmentation/kinking. Other systems, including Pb-Co and Cu-Ni, yielded distinctly inferior results. This composition-spread approach can yield data-driven insights into the behavior of catalysts. We will discuss the challenges associated with generating robust conclusions by this method as well as the opportunities for employing this approach in the study of catalysts for other applications.
11:15 AM Q2.4
Dynamic Restructuring Of Solid Catalyst Cluster During Carbon Nanotube CVD. Stephan Hofmann1, Renu Sharma2, Gaohui Du2, Mirco Cantoro1, Simone Pisana1, Atlus Parvez1, Caterina Ducati3, Rafal Dunin-Borkowski3 and John Robertson1; 1Engineering, University of Cambridge, Cambridge, United Kingdom; 2Center for Solid State Science, Arizona State University, Tempe, Arizona; 3Materials Science and Metallurgy, University of Cambridge, Cambridge, United Kingdom.
It is commonly assumed that the catalyst metal forms a liquid intermediate and that bulk catalyst effects dominate CNT growth dynamics. Calculations of size-corrected melting points and carbon saturation often indicate minimum CNT synthesis temperatures >500°C, seemingly incompatible with the 400-450°C ceiling temperature of present back-end CMOS technology. We present a detailed in-situ and ex-situ analysis of catalyst reconstruction and sintering upon temperature elevation and in particular show how different atmospheres influence the de-wetting and activation of solid transition metal films on Si/SiO2. In-situ environmental HRTEM shows that Ni films supported on SiOx membranes already start restructuring at temperatures as low as 200°C and that the bulk of the catalyst cluster is crystalline but highly mobile during CNF and SWNT CVD. The constant catalyst cluster reshaping is driven by catalytic carbon deposition and the occurring surface energy minimization. The tubular CNF shape anisotropy results from an elongation/contraction scenario of the catalyst cluster in the tip. For SWNTs, we observe mainly root growth with a carbon layer lifting off a reshaped, crystalline catalyst. We report single wall CNT CVD at temperatures below 450°C from undiluted C2H2 and demonstrate field effects in as-integrated CNT FETs [1]. We suggest that surface[2], rather than bulk catalyst effects, dominate CNT growth kinetics for low temperature, surface-bound CVD and analyse the potential narrowing of the chirality distribution. 1. Cantoro et al., Nano Lett. 6, 1107 (2006) 2. Hofmann et al., Phys. Rev. Lett. 95, 036101 (2005)
11:30 AM Q2.5
In-Situ Studies of Single-Walled Carbon Nanotubes Growth Under Reductive and Oxidative Ambient. Avetik R. Harutyunyan and Elena Mora; Material Science, Honda Research Institute USA Inc., Columbus, Ohio.
There have been intensive studies of the growth of carbon nanotubes in a reductive ambient (hydrogen) and oxidative media, (such as oxygen, carbon dioxide and water) [1-3]. However, recent success of water-assisted synthesis of carbon SWNTs [4], has created more interest farther research in area. We present our results of in-situ parametrical studies of carbon SWNTs growth under different ambient, using enhanced by an attached mass-spectrometer CVD technique. The growth of carbon SWNTs was performed by passing a mixture of methane diluted in Ar gas over the pre-reduced Fe catalyst particles supported on alumina. Hence, we are able to evaluate the evolution of hydrogen concentration, which occurred due to the catalytic decomposition of hydrocarbon gas (methane). An initial rapid increase of hydrogen concentration, corresponding to the growth-active state of catalyst, is then followed by a slow decrease, which is a result of the deactivation of the catalyst. The variation of the ambient affects on the catalyst activity and in this manner on the yield of grown SWNTs. Our studies reveal the slight influence of the hydrogen media, whereas the concentration of the water has significant impact on the yield of growth nanotubes. However, the optimal ratio of hydrogen/water can lead to the increasing of the yield. The mechanism of the observed yield variations is discussed based on formation of nanotube on liquefied metal particles concept [5], where carbon atoms are incorporated into the tube by a diffusion-segregation process. 1. G. Zhang, et. al., PNAS 102, 15141, (2005) 2. S.C. Tsang, et. al., Nature 362, 521 (1993) 3. P.M. Ajayan, et. al., Nature 362, 522 (1993) 4. K. Hata et al., Science 306, 1362 (2004) 5. A.R. Harutyunyan, et al., Appl. Phys. Lett. 87, (2005)
11:45 AM Q2.6
Abstract Withdrawn
Chairs: Prabhakar Bandaru and Ian Kinloch
Monday Afternoon, November 27, 2006
Room 312 (Hynes)
1:30 PM Q3.1
Controlling the Catalyst Shape and Size Controls Nanotube Growth. John Robertson, Stephan Hofmann, Mirco Cantoro, Simone Pisana, Atlus Parvez, Andrea Ferrari and Caterina Ducati; Engineering, Cambridge University, Cambridge, United Kingdom.
It is well known that in the growth of carbon nanotubes by chemical vapour deposition (CVD) and also by plasma enhanced CVD [1,2] that the size of the catalyst nano-particle tends to control the diameter of the resulting nanotube. Unusual catalyst nanoparticles such as ferritin have been used. But how do we control the catalyst size when the particle may sinter and move on the surface. This is a particular problem in PECVD where the plasma can enhance surface motion. In CVD, very immobile supports such as fumed SiO2 or Al2O3 could be used. There have been many catalyst designs which result in catalyst restructuring. Fe catalyst tends to oxidise, and then reduction during growth initiation breaks the metal into nano-particles. The tri-layer Al-Fe-Al type catalyst [3] produces de-wetting on the Al oxide. The CoMoCat catalyst may operate by de-wetting of Co on the Mo surface [4]. Here we study the nano-structuring of catalyst under different process conditions using AFM, SEM and TEM [5]. Single and multiplayer catalyst films are deposited by sputtering or evaporation. The various de-wetting and nano-structuring processes are observed, ex-situ. The thickness dependence is noted. NH3 and plasma conditions have a great effect. It is shown that the re-structuring can occur at much lower temperatures than normally expected for growth. It is this which allows access to much lower growth temperatures than would normally be expected; 350C for SWNTs and 200C for MWNTs [6]. 1. V I Merkulov et al, App Phys Lett 76 3555 (2000) 2. M Chhowalla et al, J App Phys 90 5308 (2001) 3. L Delziet et al, J Phys Chem B 106 5629 (2002) 4. J E Herrera et al, J Cat 204 129 (2001) 5. S Hofmann et al, J App Phys 98 034308 (2005) 6. M Cantoro et al, Nanolett 6 1107 (2006)
1:45 PM Q3.2
Catalyst-Free Growth of Carbon Nanotubes on Nonplanar, Porous, Polycrystalline Silicon Carbide Substrates for Electrochemical and Photochemical Applications. Elmo Arthur Blubaugh, Heidi M. Cross, Mike H. Check and Bill L. Riehl; R&D, Riehl-Check Industries, LLC., Kettering, Ohio.
We have investigated and evaluated the catalyst free growth of carbon nanotubes on nonplanar, polycrystalline silicon carbide foam. The motivation for this work is to fabricate materials with hyper-extended surface areas containing carbon nanotubes that demonstrate stable mechanical, thermal, and electrical properties. We are expanding the properties of the material by chemically derivatizing the carbon nanotube layer with redox catalysts for fuel cell and battery applications. These hyper-extended surface area redox catalysts will allow the fabrication of miniaturized, large power density power sources for unmanned aerial vehicles, satellite and portable electronic devices. Vertically aligned carbon nanotubes are desired for evaluation in fuel cell and battery applications. The growth method primarily used to form dense arrays of vertically aligned carbon nanotubes (CNT’s) is metal catalyzed chemical vapor deposition. There are two consequences from this catalyst based synthesis, which are the inclusion of the catalyst as an impurity and the poor electrical and mechanical attachment of the CNT’s to the typically non-conductive quartz substrate, after the growth procedure. A catalyst-free CNT growth method resulting from heating silicon carbide (SiC) wafers in a vacuum furnace was reported for the carbon face of hexagonal SiC wafers. We have experimented with this technique and extended it to include growth on the silicon face of hexagonal SiC wafers. Metal catalyst free growth of CNT’s by vacuum sublimation of cubic SiC has also been reported by Nagano and Shibata and Derycke et al. Although the growth mechanism and structural and electrical properties of the CNTs by this method are still being explored we have, in this work, extended the growth to large area non-planar surfaces coated with cubic SiC. We have found that CNT’s may be grown on these SiC foam samples and that the CNT’s display a robust mechanical attachment to a very electrically conductive substrate formed from this surface decomposition procedure. We have initiated studies to evaluate these CNT/SiC foam samples for their electrochemical behavior and stability. These electrochemical studies have used both cyclic voltammetry and chronocoulometry techniques to investigate and determine the behavior for the CNT’s and the activity of test model redox couples ( Ferri/Ferrocyanide and Ferri/Ferrocene) at these extended surface area electrodes. We have determined that these test model redox couples behave ideally with no adsorption of either the oxidized or reduced component for the redox couple and that the surface area and an estimate for an iR drop due to an ionic shielding affect on the internal pores of the foam. We wish to report in this paper, the results of our initial studies for the CNT/SiC foam samples and the direction for application of this material high power density power sources for portable devices.
2:00 PM *Q3.3
Tuning Structure and Function of Carbon Nanofiber and Metal Nanoparticles in a Controlled Co-synthesis Process. Anatoli V. Melechko1, Kate L. Klein1,2, Jason Fowlkes1,2, Philip Rack2 and Michael L. Simpson1,2; 1Molecular-Scale Engineering and Nanoscale Technologies, Oak Ridge National Lab, Oak Ridge, Tennessee; 2Materials Science and Engineering, University of Tennessee, Knoxville, Tennessee.
As carbon forms a large variety of bonding arrangements, numerous nanostructured materials, such as fullerenes, carbon nanotubes (CNTs), and carbon nanofibers (CNFs), can be formed from carbon. The controlled synthesis of these materials by methods that permit their assembly into functional nanoscale structures lies at the crux of nanoscale science and nanotechnology. Deterministic synthesis implies a method of growing individual nanostructures with precisely defined characteristics (e.g. size, location, chemical composition, internal crystal structure and surface structure). Recently it has been demonstrated that carbon nanofibers can be synthesized almost deterministically by plasma enhanced chemical vapor deposition (PECVD). In this synthesis process, the location of vertically-aligned carbon nanofibers (VACNFs) is defined by the top-down definition of catalyst material patterns; the size of the catalyst nanoparticle controls the nanofiber diameter; nanofiber length is set by the growth rate and duration; alignment is controlled by the plasma sheath electric field; and sidewall chemical composition is defined by the gas composition, substrate materials, and plasma power. We begin this presentation by describing how control of the synthesis process has progressed to the point that VACNFs serve as functional nanoscale elements for a variety of devices, and we review some of these applications with a particular focus on VACNFs as interfaces to biological systems. The remainder of the presentation will focus on gaining even greater control of the synthesis process so that electric, bio- and electro-chemical, and mechanical properties may be tuned for specific purpose. For example, we discuss the ability to vary the internal structure (i.e. grow either individual free-standing vertically aligned nanofibers or single vertically aligned nanotubes) by selecting appropriate growth conditions. Finally we discuss the emerging paradigm of controlled co-synthesis by considering the structure and dynamics of the catalyst nanoparticle-carbon nanofiber interface, which controls the synthesis of both the carbon nanostructure and the enclosed metal alloy nanoparticle.
Chairs: Prabhakar Bandaru and Ian Kinloch
Monday Afternoon, November 27, 2006
Room 312 (Hynes)
3:30 PM *Q4.1
Controlled Assembly of Carbon Nanotube Architectures. Pulickel M Ajayan, Materials Science and Engineering, Rensselaer Polytechnic Institute, Troy, New York.
The talk will focus on the recent developments in our laboratory on the fabrication of carbon nanotube based architectures tailored for various applications. Various organized architectures of multiwalled and singlewalled carbon nanotubes can be fabricated using relatively simple vapor deposition techniques. The work in attaining control on the directed assembly of nanotubes on various platforms will be highlighted. Our efforts on the strategies of growth and manipulation of nanotube-based structures and in controllably fabricating hierarchically branched nanotube and nanotube-hybrid structures will be discussed. We have pursued several novel applications for these structures, for example, as nanostructured electrodes for sensors, electrical interconnects, unique filters for separation technologies, thermal management systems, multifunctional brushes, and polymer infiltrated thin film and bulk composites. Some of these promising applications of carbon nanotubes and composites will be reviewed from the perspective of what has been accomplished in recent years.
4:00 PM Q4.2
The Growth and Properties of Highly Crystalline Multi-walled Carbon Nanotubes. Krzysztof Koziol1, Caterina Ducati1, Steffi Friedrichs2, Milo Shaffer3, Paul Midgley1 and Alan Windle1; 1Department of Materials Science, University of Cambridge, Cambridge, United Kingdom; 2The Technology Partnership plc, Royston, United Kingdom; 3Department of Chemistry, Imperial College Science Technology & Medicine, London, United Kingdom.
Highly crystalline multi-walled carbon nanotubes (MWNTs) were synthesised by chemical vapour deposition (CVD) technique by adding a nitrogen-containing compound, diazine, to the hydrocarbon feedstock. Ferrocene was used as the metal catalyst precursor [1]. It was found that the resultant MWNTs, which contain about 3% nitrogen, are remarkably straight and show unprecedented degrees of internal order: not only is each of the tubular layers of the same orientation (chirality) but they appear to be in crystallographic register with one another, as demonstrated by electron diffraction and high resolution electron microscopy experiments. The characterization of the nanotubes was performed through a range of electron microscopy techniques [2]. High resolution imaging shows that the walls of the nanotubes consist of truncated stacked cones, instead of perfect cylinders, with a range of apex angles that appears to be related to the nitrogen concentration in the synthesis process. Electron energy loss spectroscopy and energy filtered mapping were used to investigate the location and bonding state of the nitrogen, suggesting that molecular nitrogen is trapped in the hollow core of the tubes. The possibility of having nitrogen intercalated in the web-like membranes across the core, or incorporated in the graphitic lattice is considered. A growth mechanism based on the interplay of base- and tip-growth and role of nitrogen presence is proposed to account for our experimental observations. [1] K. Koziol, M.S.P. Shaffer and A.H. Windle, Adv. Mater., 17, 6, (2005), 760-763 [2] C. Ducati, K. Koziol, S. Friedrichs, T. Yates, M. Shaffer, P. Midgley and A. Windle, Small, 2, 6, (2006), 774-784
4:15 PM Q4.3
High Uniformity Growth of Sub-Nanometer (0.7 nm) Diameter Single-Walled Carbon Nanotubes using Catalyst Particle Templates Produced by Nanosphere Lithography. Noureddine Tayebi1,2 and Joseph W. Lyding1,2; 1Electrical and Computer Engineering, University of Illinois at Urbana-Champaign, Urbana, Illinois; 2Beckman Institute for Advanced Science and Technology, University of Illinois at Urbana-Champaign, Urbana, Illinois.
We report a simple and inexpensive technique based on nanosphere lithography [1], which allows for the fabrication of periodically-spaced and monodispersed metal particles from which the chemical-vapor-deposition (CVD) synthesis of single-walled carbon nanotubes (SWNTs) is achieved. The diameter of these metal particles, and thus that of the SWNTs, is controlled from 1.5 nm down to 0.7 nm, with an interparticle spacing varying from 50 nm down to 5 nm. This method has the potential for precise chirality control for sub-nanometer diameter, large band-gap semiconducting SWNTs. A double-layer of hexagonally close-packed (~50 nm diameter) polystyrene nanospheres is self-assembled onto the substrate (SiO2, Si3N4 and single-crystal quartz), which produces hexagonal six-fold interstices of ~7 nm in diameter size. Thin metal (Fe, Ni, Co) films (1-10 Å thick) are then evaporated onto the surface at a 45 angle with respect to the substrate normal, which limits the number of atoms deposited within the six-fold interstices to ~1 nm2 surface area. After lift-off of the nanosphere mask, samples are thermally annealed (700-900 °C), which allows for the diffusion of the confined metal atoms during which stationary clusters are formed [2]. Since the metal atoms are confined within dimensions smaller than the metal diffusion length, discrete single clusters form within each interstice in a periodic fashion with a controlled inter-particle spacing determined from geometry to vary from 5 to 50 nm. The number of atoms deposited controls the cluster size, which varies from 0.7 to 1.5 nm. This technique is then used for the CVD synthesis of SWNTs grown at the individual particle level, which allows for the control of their location in a periodic fashion at a nominal pitch of 5 nm. The SWNT diameter is determined by the cluster size and varies from 1.5 nm down to 0.7 nm. Since the SWNT band-gap is inversely proportional to the tube diameter for semiconducting SWNTs, large band-gaps are achieved for the 0.7 nm diameter SWNTs, which opens the door to a new generation of electronic and optoelectronic devices. Moreover, the current technique was used to grow SWNTs on single-crystal quartz substrates, which allowed for aligned growth due to the surface dipole present in such substrates [3]. Since the number of atoms deposited is controlled and single clusters are formed through a diffusion process, there is a strong indication that the clusters might be crystallographically identical, which might lead to the growth of SWNTs of identical chiralities. We have characterized the particles and SWNTs grown with this technique using TEM and are currently using UHV scanning tunneling microscopy/spectroscopy and Raman spectroscopy to further elucidate the SWNT electronic and atomic-scale structures. [1] J Hulteen et al., J. Vac. Sci. Tech. A, 13, 1553, 1995. [2] A Javey et al., J. Am. Chem. Soc., 127, 11942, 2005. [3] C Kocabas et al., J. Am. Chem. Soc., 128, 4540, 2006.
4:30 PM Q4.4
Vertically Aligned Carbon Nanotube Growth with Uniform Diameter Distribution by Block Copolymer Micellar Catalyst Templates. Xi Liu, Terry P. Bigioni, Alan M. Cassell and Brett A. Cruden; NASA Ames Research Center, Mountain View, California.
We have successfully grown vertically aligned carbon nanotubes (CNTs) using block copolymer reverse micelles to form and pattern catalyst particles. The amphiphilic block copolymer poly(styrene-block-acrylic acid) (PS16500-PAA4500) was dissolved in toluene to form micelles and then loaded with FeCl3. The metal-loaded micelles were spin-coated on Si and then thermally treated to remove the polymer. Using this process, we produced surfaces patterned with iron oxide catalyst particles with particle densities ranging from 3800/μm2 to 1400/μm2 and a size distribution of (6.9+/-0.8) nm. Samples with particle densities of 3800/μm2 were then transferred into a low pressure thermal CVD system for CNT growth. Scanning electron microscopy showed vertically aligned growth of CNTs about 10 μm in length. Transmission electron microscopy revealed that the CNTs typically had double and triple graphitic layers with a normally distributed diameter distribution of (4.5+/-1.1) nm. CNTs grown from a blanket Fe film, on the other hand, had a wide distribution of diameters between 6-21 nm. Micellar catalyst templating approach therefore achieves a better control of vertical CNT growth than standard approaches by allowing for catalyst density control at a well defined diameter distribution. * Support by the UC Discovery Grant and the Industry-University Cooperative Research Program
4:45 PM Q4.5
Growth and In-Situ Optical Characterization of Aligned Carbon Nanotube Monoliths Using a Desktop Reactor Apparatus with Rapid Thermal Control. Lucas van Laake1,2, A. John Hart1, Don A. Lucca3, Lin Shao4, Matt J. Klopfstein3 and Alexander H. Slocum1; 1Mechanical Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts; 2Mechanical Engineering, Eindhoven University of Technology, Eindhoven, Netherlands; 3School of Mechanical and Aerospace Engineering, Oklahoma State University, Stillwater, Oklahoma; 4Los Alamos National Laboratory, Los Alamos, New Mexico.
We present studies of growth of thick films (monoliths) of aligned carbon nanotubes (CNTs), along with in-situ optical characterization of the growth reaction and ex-situ studies of the catalyst evolution by Rutherford Backscattering Spectrometry (RBS). Growth is performed using a novel desktop reactor apparatus, based on resistive heating of a freely suspended p++ silicon substrate. The substrate is cut from a 6” wafer and clamped between steel blocks, which make electrical contact and mechanically support the substrate, and the assembly is sealed in a quartz enclosure. Due to its unique suspended configuration, the substrate can be heated and cooled rapidly at 100°C/s, using a conventional laboratory power supply (25 V, 4 A). With this, we can study growth of nanostructures under parameter variations that are not achievable in classical tube furnaces; for example, we study the effect of rapid temperature oscillations and thermally “pulse” growth using very short reaction times. We also optically image the film during growth and thereby measure the time evolution of the thickness and topography. Specifically, by heating a catalyst-coated growth substrate (1.2/10nm Fe/Al2O3 on Si [1]) to 800°C, while separately pre-activating the reaction gases (C2H4/H2/CO) by quickly heating to 1100°C followed by subsequent cooling before reaching the substrate, we rapidly grow monoliths of aligned multi-wall CNTs (MWNTs) to over 3 mm thickness in just 20 minutes. In-situ optical measurement of the film thickness reveals an essentially constant growth rate during this period. The method of decoupling the thermal pre-treatment of the reaction gases from the substrate temperature enables optimal generation of active carbon species for CNT growth, which significantly boosts the growth rate. Combining the in-situ measurement results with RBS measurements of the substrate surface after delaminating the CNT film, we compare effects of catalyst migration and reactant diffusion limitation on the terminal film thickness. Further, in-situ imaging from the top view reveals crack-formation in the film during growth, newly demonstrating the significance of mechanical stresses arising from spatial non-uniformities in the growth rate. We have separately shown that applying vertically-compressive stress during growth can generate severe defects in MWNTs [2]. By a different choice of catalyst and gas composition (3/1.5/20nm Mo/Fe/Al2O3, CH4/H2 [3]) we also grow high-quality films of tangled SWNTs using this system. In an aluminum enclosure we make secondary electrical contact to substrates or microsystems placed on the heated substrate, enabling direct growth and in-situ electrical testing of CNTs on microelectronic and micromechanical devices. [1] A.J. Hart, A.H. Slocum, J. Phys. Chem. B, 110:8250-7, 2006. [2] A.J. Hart, A.H. Slocum, Nano Letters, 6:1254-60, 2006. [3] A.J. Hart, A.H. Slocum, L. Royer, Carbon, 44:348-59, 2006.
Chairs: Prabhakar Bandaru, Morinobu Endo, Ian Kinloch and Apparao Rao
Monday Evening, November 27, 2006
8:00 PM
Exhibition Hall D (Hynes)
Q5.1
Lower-pressure Effects on Single-wall Ccarbon Nanotubes Growth in Alcohol Catalytic Chemical Vapor Deposition. Takao Shiokawa1,2, Hiroshi Yoshida1,3, Bao-Ping Zhang4, Masaki Suzuki1,2 and Koji Ishibashi1,2; 1Advanced Device Laboratory, RIKEN, Wako, Sitama, Japan; 2CREST-JST, Kawaguchi, Saitama, Japan; 3Tokyo University of Science, Shinjyuku, Tokyo, Japan; 4Xiamen University, Xiamen, Fujian, China.
The low-pressure growth of single-wall carbon nanotubes (SWNTs) is attractive in terms of an expected precise growth control and a necessary low-temperature growth to realize the SWNTs based nanodevices combined with a two-dimensional GaAs/AlGaAs system[1]. In this paper, we will describe the successful growth of SWNTs at the low pressure down to 0.02Pa in an ultra high vacuum (UHV) chamber, using the alcohol catalytic chemical vapor deposition (ACCVD) with Co catalysts. A sample of Co(0.1nm)/ SiO2/Si was prepared for the substrate. In the low-pressure growth at 0.02Pa, the growth temperature was varied from 550°C to 900 °C with growth time of 500 min. The base pressure before introduction of ethanol was less than 1.0 x10-6 Pa. To investigate the low-pressure effects, SWNTs were grown also with the standard conditions of the pressure of 1.5 kPa and the growth time of 5 min, using the conventional ACCVD system and the same sample. The sample surface was observed with the field-emission scanning electron microscope (FE-SEM), Raman spectra were measured by the argon-ion laser with a wavelength of 514.5 nm. In the samples grown at 0.02 Pa, peaks of the radial breathing mode (RBM) and the splitting of the G-band peak are observed, indicating the existence of SWNTs on the substrate. In our knowledge, 0.02 Pa of the ethanol gas pressure is the lowest one ever tried for the CVD growth. The maximum intensity of the G+ signal was obtained at 550 °C for 0.02Pa, while it was obtained at 850 °C for 1.5 kPa, suggesting that the optimum temperature was reduced with ~300 °C by decreasing the pressure from 1.5 kPa to 0.02 Pa. It is interesting to mention that the RBM spectra of 0.02 Pa measured in the sample with the maximum G+ intensity was similar to that of 1.5kPa, although the two conditions are very different, meaning that the diameter distribution of SWNTs is also similar. This result suggests that it may be defined dominantly by the size distribution of the catalyst. More detailed results will be presented. [1] T. Shiokawa et al., Jpn.J.Appl.Phys., Part 2 45,L605(2006).
Q5.2
Control of The Growth Rate of Carbon Nanotubes by Ion Fluxes in CH4/H2 Plasma. Atsushi Okita1, Yoshiyuki Suda1, Atsushi Ozeki1, Junji Nakamura2, Akinori Oda3, Krishnendu Bhattacharyya1, Hirotake Sugawara1 and Yosuke Sakai1; 1Graduate School of Information Science and Technology, Hokkaido University, Sapporo, Japan; 2Institute of Materials Science, University of Tsukuba, Tsukuba, Japan; 3Graduate School of Engineering, Nagoya Institute of Technology, Nagoya, Japan.
Applications of carbon nanotube (CNT) to LSI require a precise control of CNT growth in terms of the length, diameter, number density, and quality. The goal of this study is to achieve the CNT growth with a high controllability for the applications. We have attempted control of the growth rate of CNTs using plasma-enhanced chemical vapor deposition (PECVD). An advantage of PECVD is effective production of active species (radicals and ions). We believe that the growth rate of CNT is controllable via the supply of these active species. So far, we have studied the CH4 and CH4/H2 plasmas and grown CNTs to analyze the correlation between species produced in the plasmas and catalyst surface on substrate [1-3]. Utilizing these results, we control the growth rate of CNT through fluxes of ions containing carbon atoms. We have performed CNT growth by a CH4/H2 RF plasma. To analyze the ion fluxes in CH4/H2 RF plasma, we have quantitatively simulated its plasma using 1-dimensional fluid model [1]. The PECVD setup and experimental procedure are described in the previous reports [1-3]. In this experiment, we kept total gas pressure at 10 Torr and the substrate temperature at 650oC. We used triple layered catalysts (Al2O3/Fe/Al2O3/ = 1/1/1 nm) on SiO2/Si substrates for the CNT growth. To control the ion fluxes onto the catalyst, we applied positive bias voltage to the substrate. Using a scanning electron microscope and a transmission electron microscope, we evaluated the CNT length, diameter and number density. We also investigated the effect of bias voltage using computer simulation of the CH4/H2 plasma. The amount of carbon atoms supplied as ion fluxes onto the substrate were calculated. In our experimental results, the CNT length became shorter when we applied higher positive bias voltages. The growth rate of CNT ranged from 0.15 to 0.5 μm/min depending on the bias voltage. We consider that the higher positive bias voltage prevents supply of ions onto the substrate. In the simulation results, the carbon amount supplied as ions decreased with increasing bias voltage. This simulation result agreed well with experimental one. We conclude that ions are the main precursor for CNT growth in our experiment. Controlling the bias voltage enables us to progress or to suppress the growth rate of CNT. [1] A. Okita, J. Appl. Phys. 99, 014302 (2006) [2] A. Okita, Jpn. J. Appl. Phys. (2006) in press [3] A. Ozeki, MRS Symposium Proceedings Vol. 901E 0901-Rb24-03 (2006)
Q5.3
Continuous Growth of CNT Forests using Bimetallic Nanoparticles. Amandeep kaur Sra1, Fabian Reyes1, Lawrence Overzet1, Gil Lee1 and Duck Joo Yang2; 1Department of Electrical Engineering, The University of Texas at Dallas, Richardson, Texas; 2Department of Chemistry, The University of Texas at Dallas, Richardson, Texas.
Multiwall carbon nanotubes (MWCNTs) promise many applications of great technological importance as they have excellent behaviors as carriers of functionalized molecules, electron field emitters, conductive wires, bearers of rotational motors, etc. While current efforts have been concentrated on using zerovalent metal ions such as Fe, Ni, Co as catalysts, we have expanded our horizons to using bimetallic nanoparticles as catalyst materials. One of the main advantages of using bimetallic nanoparticles is that both the external (size and shape) and internal composition (atomic ordering) can be well controlled. Here, we report a simple, efficient method of producing bimetallic Fe-Pt alloy nanoparticles that are subsequently used as catalysts for continuous growth of forests of MWCNTs by Atmospheric Pressure Plasma Enhanced CVD process. The synthesis of Fe-Pt nanocrystals was carried out using the bottom-up polyol process using standard air free techniques. TEM analysis of the product indicates that the nanoparticles are monodispersed 2-2.5 nm particles. Subsequent growth of MWCNTs forests was accomplished by an atmospheric pressure plasma jet using acetylene as a precursor gas. TEM and SEM analysis of the sample cross-section grown at substrate temperature of 680°C indicates that the diameters of the CNTs are ~ 20 nm while height of the forest varies from 5-20 μm. The number of walls in the CNTs and the graphitization content could easily be manipulated by varying the temperature (increasing to 760°C) or by pre-treatment of the nanoparticles with oxygen plasma. These grown CNTs could then be easily scraped off from the substrate and again re-grown under the same conditions without any further addition of catalyst nanoparticles. The potential use of this continuous harvesting process of fabricating CNTs is enormous in terms of large scale production capabilities.
Q5.4
Surface Reactions of Metal Catalysts in Ethanol-CVD Ambient at Low-pressure Studied by in-situ Photoelectron Spectroscopy. Fumihiko Maeda1, Satoru Suzuki1, Yoshihiro Kobayashi1, Daisuke Takagi2 and Yoshikazu Homma2; 1NTT Basic Research Laboratories, NTT Corporation, and CREST, JST, Atsugi-shi, Kanagawa, Japan; 2Department of Physics, Tokyo University of Science and CREST, JST, Shinjuku, Tokyo, Japan.
There is a consensus that metal catalysts play an important role in the various proposed growth mechanisms of carbon nanotubes (CNTs). However, a clear understanding of the chemical and physical states of catalysts during CVD has not been established. Therefore, we have been investigating these states in metal catalysts using in situ photoelectron spectroscopy (PES), which is one of the most suitable methods for chemical analyses of surfaces. To understand the catalyst state during CVD growth, a real-time analysis is desirable but is technically difficult. Therefore, in this work, we looked at the reaction process by which form the nanoparticles of catalysts are formed on substrate surface in an ethanol ambient for CNT CVD by analyzing the surface in situ in a stable state in anr ultra-high vacuum. Co was deposited on Si substrates on which a SiO2 thin layer had been formed by thermal oxidation. Using this substrate, we prepared two samples. One was exposed to air and Co was oxidized. On the other sample, Co was deposited in an analysis/growth chamber, which was modified from an existing photoelectron analysis system, and CNTs were grown without exposure to air. The samples were heated in an ultra-high vacuum. Then, after the sample temperature reached the growth temperature, 600°C, ethanol was supplied to a pressure of less than 0.1 Torr. Single-walled CNTs were obtained by these procedure and this as-grown surface was analyzed without exposure to air. Before and after the CNT growth procedures, the Co 2p spectrum of the sample not exposed to air showed an asymmetric single peak without satellite peaks, indicating that the Co was in a metallic state. That is, the Co remained in the metallic state after the CNT growth process. Meanwhile, after the CNT growth, single peaks of asymmetric C 1s were observed, which are attributed to graphite or amorphous carbon. If the eutectic composition of C had been involved in the all of the Co nanoparticles, we would have detected it; however no peak was found at the binding energy corresponding to Co carbide. These results indicate that the eutectic composition of C was not stably involved in the Co particles. For the air-exposed sample, we found that Co was oxidized (Co3O4) when the sample was introduced into the analysis chamber. However, Co oxide was reduced during the annealing process and we obtained the same results for the sample not exposed to air. Because the ratio of the number of CNTs to the number of Co nanoparticles is at most several % at present, almost all photoelectrons were obtained from inactivated catalyst particles. Nevertheless, we can draw some conclusions about the stable state of Co in CNT-growth ambient: The oxide state and the state involving carbon are not stable but that metal is stable for Co under this CNT growth condition. We think that external disturbances on metallic Co nanoparticles stochastically induce CNT growth.
Q5.5
Three-Dimensional Single-Walled Carbon Nanotube Architecture using Suspended Catalyst Particles. Tomoya Ito1 and Toshio Ogino1,2; 1Electrical and Computer Engineering, Yokohama National University, Yokohama, Japan; 2Japan Science And Technology Agency, Kawaguchi, Japan.
Our research goal is to fabricate a three-dimensional architecture of carbon nanotubes (CNTs) by chemical vapor deposition (CVD). To realize a new architecture, we used suspended catalytic metal particles immobilized on the suspended CNTs that were pre-grown on a patterned Si substrate. In this method, we observed that single-walled CNTs (SWNTs) can be grown from the surfaces of the pre-grown CNTs. We have also found that the present method is very useful to investigate CNT growth mechanism because we can observe the CNTs and catalysts in an almost free space using a transmission electron microscope (TEM). In the first CVD, CNTs were grown on a patterned Si substrate by the hydrogen-methane CVD. Fe was used as a catalyst and the growth temperature was 900 degrees. Under these conditions, we can obtain straight suspended SWNTs. After the suspended CNTs were grown, we used a vacuum evaporation technique to form catalytic Co particles on the surfaces of the pre-grown CNTs and then CNTs were grown again using the Co particles as catalysts. The growth temperature of the second CVD was 800 degrees. In this case, both SWNTs and multi-walled CNTs (MWNTs) were grown. We observed the CNT networks after the first and the second CVDs by a scanning electron microscope (SEM) and TEM. After the first CVD, suspended CNTs without catalyst particles were observed on the patterned Si surface in the SEM images. After the second CVD, we observed the CNTs with metal particles on their surfaces in addition to no-particle CNTs. The important result is observation of the CNT growth from the metal particles supported on the CNTs grown at the first CVD. In the TEM images, we observed two types of CNTs grown on the suspended CNTs. The first one is SWNT growth with Co particles. In this growth mode, the SWNTs are grown sliding on the pre-grown CNTs, resulting in a bundle. The other type is wavy-shaped MWNT growth from the pre-grown CNTs. In this case, the MWNTs blanch out from the pre-grown CNTs. These results show that the straight SWNT tend to form a bundle more frequently that the wavy MWNT. In summary, CNT growth from the catalyst particles supported on the pre-grown suspended CNTs are possible. This process is promising to fabricate a complex CNT network. Moreover, this structure is useful to observe the growth process because the CNT and the catalyst particle exist in a free space. This work was partly supported by CREST/JST and Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology.
Q5.6
Abstract Withdrawn
Q5.7
Abstract Withdrawn
Q5.8
Optimization of the Production Single-Walled Carbon Nanotubes with Fe-Mo/MgO as Catalyst. Marcio Dias Lima1,2, Mônica Jung de Andrade1,2, Siegmar Roth1 and Carlos Pérez Bergmann2; 1von Klitzing's department, Max Planck Institute for Solid State Research, Stuttgart, Baden-Württemberg, Germany; 2Materials Department, Federal University of Rio Grande do Sul, Porto Alegre, Rio Grande do Sul, Brazil.
Great attention have ben focused on Thermal Chemical Vapour Deposition (TCVD) due to its great potential for scale-up the synthesis of carbon nanotubes (CNTs) with powder catalysts. The main purpose of this work is to optimize the production of Fe-Mo/MgO Single-Walled Carbon Nanotubes (SWNTs) through TCVD. In order to that, the composition of the catalyst (Fe:Mo precursos ratio), the catalyst preparation route and synthesis conditions were evaluated (temperature, time and carbon source). Ethanol and Hexane were tested as carbon sources. Catalysts were produced through a Solution Combustion Synthesis technique, which allows a fast preparation of single or several oxides. Iron and molybdenum oxides dispersed in magnesium oxide (MgO) matrix form an inexpensive SWNT catalyst and the easily dissolution of the MgO in mild acids facilitates the purification process. In situ characterization by electrical conductivity of the catalysts were also used to evaluated the effect of several synthesis parameters. The diameter of the nanotubes were not significantly affected by the carbon source. Higher yields of SWCTs were obtained using hexane but also more deposition of amorphous carbon was observed. It was found that molybdenum addition in small amount (Fe:Mo≤20) causes an increasing in the G/D ratio and yields SWNTs, but higher amount causes an increase in the number of walls of CNTs and also the production of undesirable carbon nanostructures (nanofibers and carbon onions).
Q5.9
Structural Investigation of Nano-Carbon Produced by Electric Arc-Discharge in Non-Conventional Environments. Emanuela Piscopiello1, Marco Vittori Antisari2, Daniele Mirabile Gattia2, Vittoria Contini2, Maria Rita Mancini2, Amelia Montone2 and Renzo Marazzi2; 1UTS MAT, ENEA-c.r. Brindisi, Brindisi, Italy; 2UTS Materiali e Nuove Tecnologie, ENEA-c.r. Casaccia, Rome, Italy.
The arc-discharge method is the one by which CNTs were first produced and recognized. Among other techniques the arc-discharge is an important method for the mass production of different carbon nanostructures. This work reports the experimental results from the production of siglewalled carbon nanohorns (SWNH) and multiwalled carbon nanotubes (MWCN) synthesized by DC and AC arc-discharge carried out at room pressure in air and Ar-enriched environment, by a specially designed experimental device. No vacuum system is required so that the operation cost has been drastically reduced. Electron microscopy and X-Ray diffraction were used to investigate the different nanostructures originated from the re-condensation in the solid state of the sublimated carbon atoms. Multi Walled Carbon Nano Tubes are mainly found in the hard crust formed at the cathode, while nano-horned particles can be recovered from a cylindrical collector surrounding the discharge. In addition to this, an amorphous structure can be collected in the reaction area. On the basis of our experiments, we conclude that three carbon structures (carbon nanotubes, carbon nanohorns and amorphous carbon) can be successfully obtained from a single carbon arc device which operating parameters can be customized to specific requirements.
Q5.10
Diameter and Metallicity Dependent Dedoping and Separation of Single Wall Carbon Nanotubes. Sang Nyon Kim, Zhengtang Luo and Fotios Papadimitrakopoulos; Polymer Program, IMS, University of Connecticut, Storrs, Connecticut.
Separating and/or enriching certain fractions of single wall carbon nanotubes (SWNTs) according to their metallicity and diameter (dt) have emerged as one of the hottest issues to be resolved for better nanotubes research and application. Exposure of acid-treated SWNTs to long chain alkylamine has resulted in such separation. In this study, the underlying mechanism for this separation is enlightened and better description of the physicochemical properties of charge-stabilized SWNT dispersions in polar aprotic media, such as N,N-dimethylformide (DMF), is given by establishing the reversible nature of the SWNTn+ + n/2 H2 O <=> SWNT + n H+ + n/4 O2 redox chemistry. Variations in H+, H 2O and O2 concentration in DMF are shown to be responsible for differential SWNTs dedoping trends according to dt and metallicity. These dedoping trends result in similar SWNTs separation behavior. Nernst-based Gibbs free energy and charge-loss as it pertains to the (n,m)-dependent SWNT integrated density of states (IDOS) across the corresponding pH-induced, redox jump, have also shown matching trends with dedoping and separation.
Q5.11
Single Wall Carbon Nanotube (SWNT) Production in a Vertical Furnace by the Floating Catalyst Method. Ranadeep Bhowmick1,2, Brett Cruden2 and Bruce Clemens1; 1Materials Sc & Engg, Stanford University, Stanford, California; 2Center for nanotechnology, NASA Ames Research Center, Moffett Field, California.
Carbon nanotubes, particularly SWNTs, are sought for various applications since they have remarkable electrical and mechanical properties. The bottle neck for the assimilation of SWNT into materials/devices is due to limited control over the nature of the SWNT produced and small laboratory scale production rates. There are also many unknowns surrounding the growth mechanisms of SWNTs. Here we report a variation on the floating catalyst method for the production of SWNTs. In this method, catalyst particles are suspended in a flow of a carbon containing gas, both being continuously fed into the reactor. This presents a viable way for continuous production of CNTs and avoids catalyst poisoning issues. Ferrocene is used as the catalyst source and alcohol as the carbon precursor with the furnace operating at ambient pressure. SWNTs are successfully produced using this technique. A detailed investigation of the reaction space of the vertical furnace for SWNT production has been carried out as a function of temperature, carrier gas flow rate and the precursor solution flow rate. The SWNTs are characterized by Raman spectroscopy, transmission electron microscopy (TEM) and UV-VIS spectroscopy. UV-VIS spectroscopy is used to evaluate the purity of the SWNT. The intensity of the radial breathing modes in Raman spectroscopy was used to understand variations in SWNT chirality. Preliminary results describing the dependence of the quality of the SWNTs on growth conditions via in situ studies using a residual gas analyzer probe will be presented. Also, a detailed characterization of the catalyst nano-particles will be made to better understand the role of catalyst particles on the formation of the SWNTs.
Q5.12
A New Laser-heated CVD Setup for Studies of Local Carbon Nanotube Growth. Lucas van Laake, Yves Bellouard and Andreas Dietzel; Mechanical Engineering, Eindhoven University of Technology, Eindhoven, Netherlands.
We report on a new setup for laser-assisted chemical vapor deposition of carbon nanotubes. The basic concept is to use a focused laser beam to locally heat a catalyst layer deposited on a substrate in order to thermally induce nanotube growth. The objective is to introduce nanotubes at selected locations on delicate substrates or to form arbitrary patterns without the need to pre-pattern the catalyst layer. We use two different substrate materials: Si with a SiO2 layer and fused silica. A CW-CO2 laser beam is radiated from the bottom; as silicon is partially transparent at the laser wavelength (10.6μm), most of the energy directly couples to the oxide intermediate layer resulting in local heating of the catalyst. In the case of fused silica, as the material is opaque at 10.6μm, the energy is directly absorbed in the first tenths of microns of the substrate thickness. Specimens are placed inside a vacuum chamber where pressure and gas composition are independently controlled: Gas composition and total mass flow of the reactant mixture are controlled by mass flow controllers and pressure is set by a variable leak valve between chamber and vacuum pump. Pressure ranges between low vacuum (10-3mbar) and twice atmospheric pressure (2000 mbar). A custom-designed self-aligning substrate holder ensures repeatable substrate placement and an integrated gas delivery unit provides repeatable gas supply even at higher flow rates and pressures. Two viewports are provided to illuminate and observe the growth site from the top. Since the heating laser enters from the bottom, it does not interfere with measurements on the top surface. CNT growth is studied in-situ using various optical methods including reflectivity measurements. The effect of substrate temperature distribution, gas composition, pressure and flow-rate are investigated.
Q5.13
Contrast Imaging of Carbon Nanofiber-Substrate Interface using Scanning Electron Microscopy. Makoto Suzuki1, Yusuke Ominami1, Quoc Ngo1,2, Toshishige Yamada2, Alan M. Cassell2, Jun Li2 and Cary Y. Yang1; 1Center for Nanostructures, Santa Clara University, San Jose, California; 2Center for Nanotechnology, NASA Ames Research Center, Moffett Field, California.
Although nanoscale materials have been extensively studied for their applications in high performance electronics, there still exists a critical issue of the controllability of their precise positioning for higher production yield. Recent studies reveal that the thermal and electrical properties of one-dimensional materials such as carbon nanotubes and carbon nanofibers (CNFs) strongly depend on their configuration, especially thermal dissipation via the support material on which they rest. The extent to which the nanofiber is in contact with the support material (or substrate) determines how heat is transported within the entire structure. Thus investigation of the CNF-substrate interface is essential for achieving stability and reliability of any potential CNF device. We present a rapid imaging technique to study the interface nanostructures between CNFs and substrate using scanning electron microscopy (SEM). In our experiment, CNFs grown by plasma-enhanced chemical vapor deposition (PECVD) are dispersed onto a silicon substrate. Because as-grown CNFs are not completely straight, some of them are placed with only part of their sidewalls in contact with the substrate. This partial contiguity between nanofiber and substrate can adversely affect heat dissipation. We perform SEM imaging of these CNFs with diameters ranging from 250 nm down to 25 nm by irradiating the electron beam perpendicular to the substrate. We found that, by using electron beam energy higher than a particular threshold, the SEM image of CNFs exhibits different contrasts depending on whether or not the irradiated part of the CNF is in contact with the substrate. The threshold energy is found to be close to the energy required to penetrate the CNFs, thus the contrast mechanism observed here can be explained by modeling electron penetration into a solid and considering the edge effect of SEM image formation. With this contrast mechanism, the interface nanostructures can be imaged very rapidly without substrate tilting, which can deteriorate image resolution due to the large working distance and can degrade the imaging efficiency due to the small in-focus area within the scanning field of the electron beam. This technique can also be applied to imaging other similar structures such as nanowires resting on support materials.
Q5.14
The Effect of Catalyst Composition on the Growth of Carbon Single-Walled Nanotubes. Elena Mora, Toshio Tokune and Avetik R. Harutyunyan; Honda Research Institute USA Inc., Columbus, Ohio.
The growing interest to use carbon single-walled nanotubes (SWNTs) in different applications demands a reasonably pure and homogeneous (diameter, chirality) material. To achieve this, a full understanding of the effect of the catalyst composition and the synthesis parameters on the growth of the tubes should be acquired. It is well known that the addition of small amounts of certain additives to the metal catalyst significantly improves the SWNT yield when using chemical vapor deposition (CVD) method. In this work, the performance of Fe/Mo/Al2O3 and Fe/Al2O3 catalysts are compared, in order to study the effect of the different composition in the growth of carbon single-walled nanotubes (SWNTs). The use of a Mass Spectrometer attached to the outlet of the CVD apparatus allowed us to study, in situ, the effect on both catalyst activity and lifetime during the growth of SWNTs, by following the hydrogen formed in the decomposition of the hydrocarbon (CH4). We observed that Mo not only improves the Fe catalyst activity and lifetime, but also decreases the activation energy for decomposition of the hydrocarbon (over 350 οC relative to the thermal case). As a result, the growth initiates earlier and is prolonged for longer times resulting in higher SWNTs yields compared to the monometallic catalyst. The results are in agreement with independent experiments to study the evolution of the Raman spectra with both temperature and synthesis time, as well as the evolution of the carbon up-take. In addition, experiments performed using sequential introduction of 12CH4 and 13CH4 not only confirmed the longer catalyst activity for SWNTs growth of the bimetallic catalyst, but also revealed that the poisoning of the catalysts occurs through different mechanism after the addition of Mo. The results obtained in this work also point to a substantial intermetallic interaction between Mo and Fe with possible formation of Fe-Mo alloy, in this manner protecting the solidification of the catalyst due to the formation of stable carbide phases and prolonging the catalyst lifetime.
Q5.15
Synthesis and Structure of Carbon Nanotube Y-junctions. Bimal Pandey and Wenzhi Li; Physics, Florida International University, Miami, Florida.
The effect of catalysts on the growth and structure of carbon nanotube Y-junctions (CNTYs) has been investigated by using three different nitrates (i.e. cobalt-, magnesium-, and calcium-nitrates) and their mixtures as catalyst precursors. CNTYs with straight branches have been synthesized by using mixture of cobalt nitrate/magnesium nitrate or cobalt nitrate/calcium nitrate with different concentrations. Pure cobalt nitrate, magnesium nitrate, and calcium nitrate, or mixture of magnesium nitrate and calcium nitrate will not grow any CNTYs, indicating that only binary catalysts Co/Mg and Co/Ca will facilitate the formation of CNTYs. In addition, the effect of carbon sources on the formation of CNTYs has also been studied. It is found that thiophene (C4H4S) can promote the formation of CNTYs; other sources such as methane (CH4) and acetylene (C2H2) can grow linear CNTs rather than branched Y-junctions. This result shows that sulfur is playing an important role in the formation of CNTYs. Structure examination indicates that the angles between the branches of CNTYs are about 80°, 135°, and 145°, showing fractal growth feature.
Q5.16
Direct Evidence for base-growth Through Aligned Multi-walled Carbon Nanotube Multilayers. Mathieu Pinault1,2, Martine Mayne-L’Hermite2, Cecile Reynaud2, Vincent Pichot3, Pascale Launois3 and Hicham Khodja4; 1Chemical Engineering, Yale University, New Haven, Connecticut; 2Laboratoire Francis Perrin (URA CNRS 2453), CEA Saclay DSM-DRECAM-SPAM, Gif sur Yvette, France; 3Laboratoire de Physique des Solides (UMR CNRS 8502), Université Paris Sud, Orsay, France; 4Laboratoire Pierre Süe, UMR 9956, CEA Saclay, Gif sur Yvette, France.
More than 10 years after their discovery, carbon nanotubes are still attracting much interest for their potential applications, which largely derives from their exceptional structural, mechanical and electronic properties [1]. There is a variety of growth techniques for multi-walled carbon nanotube (MWNT) synthesis but the Catalytic Chemical Vapour Deposition (CCVD) method offers the most commercially viable technique due to the large uniform reaction area and the control over key characteristics such as length and vertical alignment. Samples of well aligned multiwall carbon nanotubes (MWNTs) can be formed by aerosol-assisted Catalytic Chemical Vapor Deposition (CCVD) [2,3]. Growth mechanisms of aligned MWNTs in liquid-assisted CVD methods are still discussed [4-6], and no definitive experimental evidence has been carried out. A thorough understanding of the formation mechanisms for such aligned nanotubular carbon systems is crucial to design procedures for controlling the growth conditions in order to obtain structures which might be directly used in different fields in nanotechnology. In this study, we demonstrate that sequential synthesis (multi step process) is an effective way to evidence the growth mechanisms. In a given step, the aerosol (toluene/ferrocene) injection in the CVD reactor is controlled through parameters such as duration and composition of precursors. Thus, multilayered carpets of aligned MWNTs were obtained and analysed by SEM and X-Ray scattering to identify the chronology of the growth. Our results demonstrate that nanotubes grow through a base-growth mechanism: any new injection sequence leads to the growth of a new layer directly at the substrate surface, under the pre-existing one by lifting it up. It is confirmed by nuclear microprobe analysis of multilayered carpets obtained from alternative injection of benzene enriched in 13C isotope labels [7]. We also demonstrate the absolute need of the catalyst source (ferrocene in our study) for the continuous growth of the carpet as well as for the growth of a new carpet at the base of a pre-existing one [8]. References: [1] A. Loiseau, P. Launois, P. Petit, S. Roche and J.P. Salvetat Understanding Carbon Nanotubes : from science to applications, Lecture Notes in Physics, Springer,vol. 677 (2006), Eds [2] M. Mayne et al., Chem. Phys. Lett. 338, 101 (2001). [3] M. Mayne-L'Hermite et al., Proceeding of the Chemical Vapor Deposition-XVI and EUROCVD-14, edited by M. Allendorf, F. Maury, F. Teyssandier, 2003-08, 549-556. [4] R. Kamalakaran et al, Appl. Phys. Lett. 77 (2000) 3385 [5] Zhang X et al., Chem. Phys. Lett. 362 (2002) 285-290 [6] C. Singh et al., Carbon 41 (2003) 359-368 [7] H. Khodja, M. Pinault, M. Mayne-L’Hermite, C. Reynaud, IBA Conf. Proc., 2005, in press [8] M. Pinault, V. Pichot, H. Khodja, P. Launois, C. Reynaud, M. Mayne-L’Hermite, Nanoletters, 5, 12 (2005) 2394-2398
Q5.17
Isolated Bundles of High Quality, Highly Uniform Single Wall Nanotubes Grown at very Low Substrate Temperature. Sakthi Kumar1,2 and Yasuhiko Yoshida1; 1Bio Nano Electronics Research Center, Toyo University, Kawagoe-shi, Saitama, Japan; 2REDS Group/JST, Saitama Small Enterprise Promotion Corporation, SKIP City, Kawaguchi, Saitama, Japan.
After the invention of carbon nanotubes (CNTs) by S. Ijima[1], many growth methods (arc discharge, laser ablation, chemical vapor deposition etc.) have been introduced in the field to develop CNTs (both multi walled and single walled carbon nanotubes (MWCNTs and SWCNTs)) for the past one decade. However almost all of them produced endless web natured curled (tangled - spaghetti type) arch like structures of nanotubes which are difficult to purify, manipulate and assemble for building addressable nanotube structures.[2] And again intense research work is going on to lower the processing temperature to use nanotubes on glass, plastic and other flexible but heat susceptible substrates by keeping the quality of nanotubes. It is generally accepted that producing CNTs at low temperatures by keeping its quality, is a highly formidable task. This is due to the fact that the relatively low growth temperature does not provide sufficient thermal energy to anneal nanotubes into perfectly crystalline structures.[3] It is suggested that high temperature processes (such as arc discharge and laser evaporation) are the best methods to produce high quality nanotubes [2,4]. In fact we may need to develop high quality nanotubes at substrate temperature much below 450 oC [5] even to imagine the possibility of online processing for the integration of nanotubes in micro and nano electronics circuits [6] with currently existing industrial infrastructure facilities based on CMOS technology [7]. Here we report a new, simple but elegant processing method with which we can develop SWCNTs of short isolated bundles of high quality and highly uniform at low substrate temperatures (~280oC). This method provides solution for almost all the above mentioned problems in a single step itself. We have characterized our SWCNTs with the help of TEM, SEM, ESCA, EDS, Mapping, Line scans analysis, Raman and XRD. The characteristic features of the XRD spectrum of SWCNTs [8] obtained by the XRD experiment clearly and undoubtedly suggest that we have succeeded to produce high quality SWCNTs at very low substrate temperature. References 1. Iijima, S., Helical microtubules of graphitic carbon. Nature, 354 (1991) 56. 2. V. N. Popov, Materials science and Engg. R 43 (2004) 61. 3. Hongjie Dai, Carbon nanotubes synthesis, structure, properties and applications. Dresselhaus, M. S., Dresselhaus, G., Ph.Avouris (Eds) Springer-Verlag Berlin Heidelberg, 80 (2001) 29. 4. M. F. Chisholm, et.al, Science, 300 (2003) 1236b. 5. P. R. Bandaru, et.al.,, Materials Science and Engg.B, 113 (2004) 79. 6. S. Murayama, et.al, Chem. Phys. Lett., 360, (2002) 229. 7. X.M.H. Huang et.al., Nano Letters, 5(7) (2005) 1515. 8. J. E. Fischer, et.al, J. Appl. Phys., 93 (4), (2003) 2157.
Q5.18
Growth of Horizontally Aligned Carbon Nanotubes: A Systematic Study of Growth Mechanisms. Alfonso Reina1 and Jing Kong2; 1Materials Science, MIT, Cambridge, Massachusetts; 2Electrical Engineering and Computer Sciences, MIT, Cambridge, Massachusetts.
Carbon nanotubes (CNTs) are considered excellent candidates for interconnect applications due to their high current carrying capability, low resistance and high mechanical stability. Theoretical studies have compared the performances of current copper interconnects and interconnects made of individual CNTs connected in parallel or CNT bundles. These studies predict a significant enhancement of interconnect performance at the 22 nm node if CNTs are implemented. Also, extensive work has been done to build and test prototypes of CNT-based transistors and electronic devices. However, their reliable synthesis and integration for modern circuits remains a challenge. It is still necessary to have further control over CNT location, orientation and density to build required device architectures. The simultaneous synthesis and horizontal alignment of single-walled carbon nanotubes (SWNTs) on a substrate using the gas flow inside a CVD chamber has been demonstrated, but the mechanism is not yet well understood and the results have encountered non-reproducibility issues. Here, a systematic study of gas flow-assisted synthesis of long and horizontally aligned CNTs from iron-based catalyst nanoparticles on SiO2/Si substrates is presented. From the vast number of process variables, the most critical ones needed to obtain alignment of CNTs with the gas flow are underlined and their effect on the final result discussed. It is found that not only the drag of the gas flow or its non-turbulent nature is necessary for alignment but also catalyst preparation, treatment and in-situ chemical changes are extremely important. Finally, these experimental observations are helpful to understand both chemical and physical mechanisms of the growth of short (few microns) and long (hundreds of microns to centimeters) CNTs from catalyst nanoparticles on SiO2/Si substrates.
Q5.19
Controlling the Morphology of Carbon Nanotube Films by Varying the Areal Density of Catalyst Nanoclusters Using Block Copolymer Micellar Thin Films. Ryan Derek Bennett1, Anastasios John Hart2 and Robert E. Cohen1; 1Chemical Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts; 2Mechanical Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts.
We demonstrate a general route that utilizes block copolymer micelles as a means to create tunable inorganic nanocluster arrays which serve as catalysts for carbon nanotube (CNT) growth. Our approach uses the amphiphilic block copolymer, poly(styrene-block-acrylic acid) (PS-b-PAA), which forms spherical micelles in solution that can be selectively loaded with metal ions and then spin-coated onto a substrate to create quasi-hexagonal arrays of metal-loaded PAA domains within a PS matrix. This catalyst system has significant value compared to commonly prepared thin metal film catalysts because it enables the creation of nanocluster arrays of a chosen metal species, with independent control of the nanocluster diameter and areal density. We use this block copolymer micellar system to create iron oxide nanoclusters, which are catalytically active in the thermal chemical vapor deposition (CVD) growth of CNTs. Through appropriate selection of the substrate, catalyst preparation procedure, and reaction conditions (combination of C2H4/H2/Ar gases), we achieve vertical CNT growth from our catalyst system. Because this catalyst system allows for precise quantification of the nanocluster areal density, we can also estimate the percentage of nanoclusters which nucleate the growth of a CNT. By uniformly varying the areal density of iron oxide nanoclusters on the substrate surface, we manipulate the morphology of the CNT film from a tangled and sparse arrangement of individual CNTs, through a transition region with locally bunched and self-aligned CNTs, to rapid growth of thick vertical CNT films. We also present a microcontact printing approach to create patterned inorganic nanocluster arrays that are utilized to synthesize patterned vertical growth of CNTs. The control afforded by this micellar catalyst system could be promising for both CNT applications as well as for improved understanding of the combination of chemical and mechanical conditions necessary for the growth of uniform CNT films.
Q5.20
Tubulization Mechanism of Amorphous Carbon-nanopillar. Toshinari Ichihashi1,3, Masahiko Ishida1,3 and Jun-ichi Fujita2,3; 1Fundamental & Environmental Research Laboratories, NEC Corporation, Tsukuba, Japan; 2Institute of Applied Physics, University of Tsukuba, Tsukuba, Japan; 3CREST-JST, Kawaguchi, Japan.
Amorphous carbon nanopillars, which were fabricated by electron-beam induced chemical vapor deposition, were tublized by heating at 600oC-650oC in the presence of iron nanoparticles. The tublization process of the amorphous carbon nanopillar was observed in situ by transmission electron microscopy and scanning transmission electron microscopy. A molten catalyst nanoparticle penetrated the amorphous carbon nanopillar, dissolving it and leaving a graphite track behind. The iron-carbon alloy nanoparticle moved through inside the nanopillar with its shape changing like a liquid. These shape changes, between the shrunken shape of the tail and the elongated shape of the top, are repeated like the movement of an earthworm. Increasing of the carbon composition caused the elongation of the top shape of the nanoparticle. The elongated top of the nanoparticle was melted. The carbon atoms were diffused from the top to the tail. The diffused carbon atoms were precipitated at the tail. When the carbon composition of the nanoparticle was decreased, the nanoparticle was solidified and shrunk. The phase of the iron-carbon alloy nanoparticle was fluctuated between solid and liquid in this tubulization process. When the tail of the catalyst particle moved at constant speed, the graphitic tube structure was formed. If the tail of the particle stayed at the same place for a few hundred milliseconds, the graphene sheet closed to make a cap within the nanotube. Transmission electron micrographs of the moving catalyst particle showed sometimes diffraction contrast, which revealed that the particle had crystalline nature. The Z-contrast image of the outer part of the catalyst nanoparticle was darker than its inner part. It was revealed that the carbon-iron ratio of the outside part of the particle is larger than that of its inside. We concluded that the dissolved carbon atoms diffused through the outer part of the catalyst nanoparticle. The tubulization mechanism is a solid-quasiliquid-solid mechanism where the carbon phase transformation is a kind of liquid phase graphitization of amorphous carbon catalyzed by liquefied metal-carbon alloy nanoparticles.
Q5.21
Catalyst Patterning and Location Controlled Growth of Single-Walled Carbon Nanotubes. Ruth Y. Zhang, John Tresek, Islamshah Amlani, Kevin Nordquist, Donald Weston, Ray Tsui and Larry Nagahara; Embedded Systems Research, Motorola Labs, Tempe, Arizona.
Chemical vapor deposition (CVD) is a well recognized approach for location controlled formation of carbon nanotubes (CNTs) directly on substrates. Selective area growth of CNTs in this case usually involves a lithographic step to pattern catalyst materials at predetermined locations on a substrate prior to the CVD process. For the growth of single-walled carbon nanotubes (SWNTs), a layer of transition metal with thickness on the order of 1 nm or less is commonly used as the catalyst. And in some cases, another thin (a few nm) metal or insulating layer beneath the catalyst layer is required to serve as a diffusion barrier and/or catalyst support to prevent the catalyst from agglomerating into large particles during the heating process. Patterning these thin layers to successfully achieve the selective growth of SWNTs is challenging because the catalyst layer can easily be poisoned by solvent and/or resist residuals. In this paper we studied effects of the patterning process on the growth of SWNTs on oxidized Si wafers in a thermal CVD process using various catalyst materials. SWNT yield was analyzed using low voltage field emission scanning electron microscopy, and using electrical measurements after contact fabrication. We compared SWNT growth from a blanket layer of as-deposited catalyst with that from catalyst patterns defined by lift-off and by a plasma etch process. We observed a lower SWNT density and less reproducibility on wafers patterned by lift-off when compared to wafers with the corresponding blanket catalyst film. On the other hand, no density reduction is observed on wafers with catalyst patterns defined by the plasma etch process. We believe that the improved SWNT growth from etched patterns is because the catalyst is deposited on a pristine SiO2 surface while it is difficult to consistently create SiO2 windows with a clean surface using a lift-off process prior to catalyst deposition. Residual contamination can potentially poison the catalyst and hinder the formation of SWNTs.
Q5.22
Synthesis of Very Dense and Vertically Aligned Carbon Nanotubes by Radical CVD. Tsuyoshi Yoshida, Daisuke Yokoyama, Takayuki Iwasaki and Hiroshi Kawarada; School of Science and Engineering, Waseda University, Tokyo, Japan.
As we know plasma assistant chemical vapor deposition (CVD) is good at controlled growth of carbon nanotubes (CNTs). By employing a microwave plasma (MP) CVD system, and preparing the substrate with a sandwich-like coating structure (Al2O3/Fe/Al2O3), high yield, selective growth of millimeter long, vertically aligned single-walled carbon nanotubes (SWNTs) have been successfully synthesized.[1] We can also decrease the growth temperature as low as 390°C to grow CNTs using Co catalyst by this CVD machine. Our CVD has an antenna of microwave plasma which is 50mm above the substrate holder and the plasma is fixed as spherical shape with a diameter of 10mm. There is no plasma etching and ion bombardment effects on catalyst and the subsequently as-grown CNTs so that very long lifetime of catalyst can be achieved. In this system, only carbon radicals reach the catalysts on the substrate to grow CNTs. No current is measured when a bias voltage is applied to the substrate. The concentration of carbon radicals on the substrate can be well controlled by microwave power, substrate distance from the microwave plasma and concentration of CH4 gas. In this sense, we can easily control the length of CNTs and the growth temperature because the amount of carbon radicals is adjustable. We have achieved millimeter long SWNTs with good reproducibility at a temperature of 600°C, 20Torr and the microwave power of 20W. The volume density is 66kg/cm3 which is the highest as reported. Very dense and vertically aligned CNTs are also synthesized at a temperature as low as 390°C by increasing the pressure and the microwave power to 60Torr and 120W, respectively. To use this novel radical CVD, several applications such as supercapacitors and wiring technique of CNTs are hope to realize. [Reference] [1] G. Zhong, T. Iwasaki, K. Honda, Y. Furukawa, I. Ohdomari, H, Kawarada. Chem. Vapor Depos., 11 (2005)127
Q5.23
Determination of the Chiral Indices of Carbon Nanotubes. Lu-Chang Qin1, Zejian Liu1, Qi Zhang1, Gongpu Zhao1, Hakan Deniz1 and Jie Tang2,1; 1Department of Physics and Astronomy, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina; 2Material Physics Division, National Institute for Materials Science, Tsukuba, Ibaraki, Japan.
We present a one-step approach to determine the chiral indices [u,v] of carbon nanotubes indiscriminately using electron diffraction carried in a transmission electron microscope. On the electron diffraction pattern of a carbon nanotube, the chiral indices of the nanotube can be read from the intensity profiles of the layer lines. Speficially, the intensity profile of the first layer line, formed by the two (010) type graphene reflections, correspond to the square of Bessel function of order v. On the other hand, The second layer line, formed by the two (100) type graphene reflections, has an intensity profile corresponding the square of the Bessel function of order u. Therefore, by examining the intensity profiles of the layer lines, we can determine the chiral indices [u,v] unambiguously. Examples of application of this method to map the distribution of chiral indices of carbon nanotubes will be presented. The determination of the handedness of carbon nanotubes using electron diffraction will be illustrated with examples. In addition, an in situ measurement of both the chiral indices and its field electron emission of a carbon nanotube will also be presented and discussed.
Q5.24
Influence of Substrate Morphology and Catalyst-substrate Interaction on Single and Multi-wall Carbon Nanotube Growth Characteristics. Ageeth Bol1, Joseph C Spagnola1,2 and Christopher B Murray1; 1TJ Watson Research Center, IBM, Yorktown Heigths, New York; 2Materials Science & Engineering, North Carolina State University, Raleigh, North Carolina.
The carbon nanotube (CNT) literature is full of CNT growth recipes and growth results. A lot of these results are however not well understood and therefore precise control over the CNT synthesis and thus the CNT characteristics is still lacking. This is a major obstacle for the realization of mass-production of CNT-based devices. One of the most important parameters in controlling carbon nanotube growth is the catalyst employed and the nature of the catalysts-substrate interaction. This paper will focus on the influence of the morphology of the substrate and chemical interaction between the catalyst and substrate on the CNT growth and characteristics. In this study we used a catalytic chemical vapor deposition (CVD) process, using ethanol as the carbon source. Both Fe3O4 nanoparticles and Fe thin films were used as catalysts. In order to vary the catalyst-substrate interaction, SiO2 substrates were given different pre-treatments. We also applied various buffer layers on top of the SiO2 layer. Using a combination of high resolution microscopies, Raman spectroscopy, and chemical characterization of the substrate surface, the influence of various pre-treatments and buffer layers on CNT growth was explored. This was done under a wide variety of synthesis conditions. Our experiments show that depending on the substrate pre-treatment and buffer layers, Fe3O4 nanoparticles (3-4 nm) and Fe thin films can either form SWNT or MWNT, randomly oriented on the substrate surface or vertically aligned in CNT forests. Surface morphology and chemical interaction between the catalyst and substrate surface play an important role in determining these CNT growth characteristics. It is hoped that these experiments will be a step forward towards better control of the CNT synthesis and the fabrication of CNT-based devices.
Q5.25
Synthesis of Large Uniform Arrays of Mono-metallic and Bi-metallic Nanoparticles as Catalysts for SWNTs. Sreekar Bhaviripudi1, Alfonso Reina1, Jifa Qi1, Jing Kong2 and Angela Belcher1,3; 1Materials science and Engineering, MIT, Cambridge, Massachusetts; 2Department of Electrical Engineering and Computer Science, MIT, Cambridge, Massachusetts; 3Biological Engineering Division, MIT, Cambridge, Massachusetts.
Block copolymer micellar templates were used for the controlled synthesis of large uniform arrays of mono-metallic (Fe, Co, Ni, Mo) and bi-metallic (Fe-Mo) nanoparticles with average diameters ranging from 1-4 nm and distance between the nanoparticles ranging from 40-45 nm. XPS data reveal the presence of nanoparticles in their oxidized states. These uniform arrays were used to investigate the growth of single-walled carbon nanotubes with chemical vapor deposition (CVD) process, using both ethanol and methane+ethylene gases as carbon sources. The periodicity and the arrangement of nanoparticles were unaffected even at high growth temperatures indicating that nanoparticle agglomeration does not take place during growth. AFM and SEM results reveal uniform growth of nanotubes with diameters smaller than the initial size of the catalyst nanoparticles. The Fe, Co and Ni nanoparticles all serve as effective catalysts for nanotube growth with different carbon feed stock, and Co and Ni give rise to relatively higher yield than Fe. The catalytic activity of Fe and bi-metallic Fe-Mo nanoparticles of similar size and identical densities using methane CVD would also be presented.
Q5.26
Abstract Withdrawn
Q5.27
Growth of Carbon Nanocoils in Alcohol Chemical Vapor Deposition. Toshio Ogino and Noriaki Tsuchiya; Dept. of Electrical and Computer Engineering, Yokohama National University, Yokohama, Japan.
Catalytic chemical vapor deposition (C-CVD) is a growth technique of carbon nanotubes (CNTs) suitable for electronic applications because the growth position and the density can be controlled by the arrangement of catalyst particles. In the C-CVD using alcohol as a source gas, single-walled carbon nanotubes (SWNTs) and multi-walled carbon nanotubes (MWNTs) can be selectively grown by using appropriate growth temperature and catalyst materials. In this technique, we have found a new growth mode of carbon nanocoils (CNCs). The catalysts for the CNT growth were prepared by dissolving the metal hydrates, such as Fe(NO3)3/9H2O, Co(NO3)2/6H2O, and Ni(NO3)2/6H2O, into ethanol. Patterned Si substrates were fabricated by the optical lithography and reactive ion etching. Mesoporous alumina powder was used as a catalyst supporter. In the CVD process, Ar was used as a carrier gas and C2H5OH was added to the Ar gas by bubbling at room temperature. The growth time and the temperatures were 10 min and 700-1000 degree C, respectively. The shape of the CNTs and CNCs was observed by a scanning electron microscopy (SEM), and the CNT species were characterized by a Raman spectroscopy. When Fe or Co was used as a catalyst, high-density SWNTs were grown at 900 degree C, and high-density MWNTs at 700 degree C. When Ni was used as a catalyst, CNCs with various coil diameters and pitches were grown in addition to the SWNTs (900 degree C) or MWNTs (700 degree C). The smallest coil diameter was as small as 40 nm. Suspended CNCs were also obtained over a trench on the patterned Si surface. We have also found that some CNCs were grown coaxially with a CNT existing at the center of the coils. In order to investigate the growth mechanism, we examined alloyed catalysts of Ni/Fe and Ni/Co. In both cases, the yield of CNCs decreases and the shape becomes elongated as the Fe or Co content increases. These results suggest that Ni is strongly involved in the growth of CNCs though the growth mechanism is still unclear. We have found that SWNTs, MWNTs and CNCs can be selectively grown by using appropriate growth temperature and catalyst materials in the alcohol CVD. The CNCs grow only when the combination of a Ni catalyst and an alcohol source gas is employed. CNCs are expected to be applicable to nano-scaled magnetic sensors, shielding material of electromagnetic waves, and a nano-scaled magnetic field generator. We believe that the present process is promising to fabricate these nano-components. This work was partly supported by CREST/JST and Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology.
Q5.28
SWCNTs for C82 Peapods: Synthesis using Nonmagnetic Catalysts. Miroslav Haluska1, M. Hulman2, J. Cech1, V. Skakalova1, B. Hornbostel1 and S. Roth1; 1von Klitzing, FKF-MPI, Stuttgart, Germany; 2Institute for Material physics, Faculty of Physics, Vienna, Austria.
We demonstrate the way to grow SWCNTs, large enough to accommodate C82 fullerenes, using nonmagnetic metal catalysts. The growth was performed in water-cooled dc arc-discharge Krätschmer generator with horizontally aligned electrodes. Different methods such as Raman and optical spectroscopy, TGA and TEM were used for characterization of the materials produced at various growth conditions. Both, the type and the pressure of the buffer gas strongly affect the yield of SWCNT webs, but they have only a small influence on the diameter of SWCNTs. In fact only a slight increase of diameters was observed with the decrease of the molecular mass of buffer gas and with the decrease of its pressure. The highest yield of SWCNT webs was obtained when helium was used as the buffer gas. The concentration of sulphur promoter strongly influences the yield of SWNTs and affects their diameters. The highest relative concentration of SWCNTs with diameters larger than 1.5 nm was found for 0.9 - 1.3 at. % of sulphur addition to PtRh catalysts.
Q5.29
Synthesis and Characterization of Water-Soluble Phospholipid-Modified Multiwalled Carbon Nanotubes. Peng He, Chemistry, North Carolina State University, Raleigh, North Carolina.
A simple three-step strategy to functionalize multiwalled carbon nanotubes using 1,2-distearoyl-sn-glycero-3-phosphoethanolamine phospholipids has been described. The resulting modified carbon nanotubes were analyzed by TEM, AFM, NMR, IR, UV-Vis, and TGA techniques. The experimental outcome shows that the use of amine-terminated phospholipids results not only in the improved water solubility of multiwalled carbon nanotubes in an aqueous phase, but also in the significant enhancement of biocompatibility. These findings will serve as a future biological platform for new devices ranging from biosensors to nano-detectors.
Q5.30
Systematic Investigation of Electrodeposition Process for Synthesis of Nanowires inside Nanoporous Templates. Adam Friedman and Latika Menon; Physics, Northeastern University, Boston, Massachusetts.
Nanoporous alumina templates are increasingly becoming attractive templates for the synthesis of nanostructures of a variety of materials. These are aluminum oxide templates consisting of nano-sized cylindrical holes arranged parallel to each other in a vertical manner. The pores are open at the top and closed below by means of a U-shaped barrier layer of aluminum oxide. Fabrication of nanowires by means of electrodeposition inside porous alumina templates is a highly effective means of preparing nanowires. AC electrodeposition is commonly used (instead of dc) due to the presence of the barrier aluminum oxide below the pores. This method is highly effective in growing nanostructures and is also cost-effective. However, despite its pervasiveness in nanofabrication, little work has been done to understand the process itself. Here we carry out a systematic study of the electrodeposition process to qualitatively understand the effects of changing parameters such as pH of electrolyte, applied ac voltage and frequency on the average nanowire growth rate. Experiments are done by depositing iron into porous alumina templates under varying electrodeposition parameters. The resulting samples are characterized by means of structural and magnetic property measurements. From these studies we have determined optimum parameters for the fabrication of nanowires with the desired properties.
Q5.31
HRTEM and EELS Analysis of the Functionalized Carbon Nanotubes. Jiri Cech1, Martin Kalbac2,3, Donghui Zhang4,5, Lothar Dunsch3, Seamus A. Curran5 and Siegmar Roth1; 1Synthetic Nanostructures, Max Planck Institute For Solid State Research, Stuttgart, Germany; 2J. Heyrovsky Institute of Physical Chemistry, Prague, Czech Republic; 3IFW, Dresden, Germany; 4Department of Chemistry and Biochemistry, New Mexico State University, Las Cruces, New Mexico; 5Department of Physics, New Mexico State University, Las Cruces, New Mexico.
We successfully visualized the structure of selected, chemically functionalized, Carbon Nanotubes (CNTs) using high resolution transmission electron microscopy (HRTEM) and electron energy loss spectroscopy (EELS). To represents the chemically modified CNT, we selected three systems. First system are oxidized and surface thiolated MWCNT, second system are Dy3N@C80 peapods prepared by depositing trimetal nitride fullerenes into SWCNTs, forming the Dy3N@C80@SWCNT, and third system are the conventional C60@SWCNT fullerene peapods, fluorinated by the Xenon difluoride (XeF2 ) up to 18 % of F. We achieved detection of very low amount (0.6 %) of Sulfur and proven the covalent bonding onto MWCNT surface. We present EELS imaging of the separated metal clusters inside endohedral metallofulerene peapod budles and in the fluorinated C60 peapods we show homogeneous fluorination across whole surface.
Q5.32
Theoretical Analysis of Non-Catalytic Growth of Nanorods on a Substrate. S. Joon Kwon and Jae-Gwan Park; Materials Science and Technology Division, Korea Institute of Science and Technology, Seoul, South Korea.
A theoretical analysis explaining the whole process of growth of nanorods on a substrate without a catalyst is presented. Prior to the growth of the nanorods, the reaction precursors form nuclei on the substrate. The nuclei undergo cluster migration caused by the surface diffusion of adatoms on the substrate, and this migration continues until the mean free time of the adatoms is larger than surface diffusion time. The most probable mechanism by which cluster migration takes place is the one that leads to the minimization of the cluster free energy, namely the migration of six adatoms into one fixed adatom. This cluster migration continues during several (typically smaller than 6) consecutive nuclei growth steps. After the process of cluster migration comes to an end, the nuclei grow in an isotropic manner by collection of the adatoms, until the nucleus reaches the thermodynamic size limit. The one-dimensional growth of nanorods on the nuclei which is associated with the critical radius begins when the reactant dose is smaller than a certain value, which is determined by the thermodynamic size limit and mass transport parameter. The mass transport of the reaction precursors leads to the expansion of the radius and elongation of the height of the nanorods, and the growth rate of the height is greater than that of the radius. This difference in the growth rate causes the aspect ratio to increase with increasing growth time. By comparing the experimental data in the literature (ZnO nanorods), the presented analysis explains well the non-catalytic growth of nanorods on a substrate.
Q5.33
Bundle Dissociation of Single-Walled Carbon Nanotubes in Amide Solvents. Shane D Bergin, Silvia Giordani, Valeria Nicolosi, Werner J. Blau and Jonathan N. Coleman; Physics, Trinity College Dublin, Dublin, Leinster, Ireland.
The implementation of single-walled carbon nanotubes (SWNTs) into a wide array of applications is hindered by the formation of bundles. Many methods have been suggested to de-bundle the SWNTs, including both covalent functionalisation and non-covalent funtionalisation with surfactants, polymers and macromolecules. These methods have their advantages but the ideal situation must be to dissolve and de-bundle the SWNTs in an appropriate solvent at concentrations that are useful for their implementation in applications. In this work, we outline the physical process underlying the de-bundling of the SWNTs in the amide solvent N-methyl pyrrolidone (NMP). This is shown by an extensive atomic force microscopy study of the dispersed SWNTs over a range of concentrations, showing the SWNTs to de-bundle as a function of concentration. Similarly, the photoluminescence of the dispersed SWNTs was measured and indicated the presence of a large population of individual SWNTs. A host of other solvents, that have differing dispersive effects on SWNTs, and an explaination of the physical process underlying their varying dispersive abilities are suggested.
Q5.34
Tailoring Dispersion and Microstructure of Carbon Nanotubes Using Weak Polyelectrolytes. Jaime C. Grunlan1,3,2 and Lei Liu3,2; 1Mechanical Engineering, Texas A&M University, College Station, Texas; 2Polymer Technology Center, Texas A&M University, College Station, Texas; 3Materials Science and Engineering, Texas A&M University, College Station, Texas.
Carbon nanotubes are an exciting material due to their small size, high modulus, and high intrinsic conductivity. As a result, nanotubes hold significant promise for imparting electrical conductivity, mechanical strength, and thermal conductivity to polymeric materials. Despite this potential, the ability to stabilize nanotubes in solution remains a significant hurdle to their widespread use. This has led to significant research efforts on the use of stabilizing agents and chemical modification of the nanotubes to impart solubility. The present work demonstrates a method to control the dispersion of carbon nanotubes in aqueous solution and the microstructure of composite films using poly(acrylic acid) and poly(allylamine hydrochloride). As the pH of an aqueous mixture containing 1 wt% polymer and 0.1 wt% nanotube is altered, a significant change in viscosity is observed that suggests changing dispersion state of the nanotubes. Drying these aqueous suspensions into composite films reveals pH-dependent microstructural differences that influence electrical conductivity. In general, nanotube exfoliation improves with increasing pH for PAA and decreasing pH for PAH. Additionally, these microstructural changes are reversible. This behavior has significant implications for the processing of carbon nanotubes and tailoring of composite properties. Many of the relationships uncovered here could be applied to other types of hydrophobic nanotubes and nanowires.
Q5.35
Fabrication of Carbon Nanotube Monolayer Film at the Liquid-liquid Interface. Jun Matsui, Kohei Yamamoto, Nobuhiro Inokuma, Hironori Orikasa, Takashi Kyotani and Tokuji Miyashita; Institute of Multidisciplinary Research for Advanced Materials, Tohoku University, Sendai, Miyagi, Japan.
Carbon nanotubes (CNTs) represent an important group of nanomaterials with attractive electronic, chemical, and mechanical properties. Several kinds of electric devices, such as chemical sensors, biological sensors, field-effect transistors (FET) and transparent conductive film based on the unique electric properties of the carbon nanotubes have been reported. Implementation of CNTs for these applications requires methods to fabricate carbon-nanotube ultrathin film. Moreover, carbon-nanotube ultrathin film is important for understanding its basic electrical and optical properties. Here, we report fabrication of CNT monolayer using the liquid/liquid interface. A water-dispersible carbon-nanotube synthesized using alumina template was used to fabricate CNT monolayer. To fabricate a CNT monolayer at the liquid-liquid interface the CNT aqueous solution was place into a vessel and toluene was added to the vessel. The vessel was then sonicated 20 min to create dispersion. The oil/water dispersion was stood for 1 days to form the liquid-liquid interface. Careful observation after sonication reveals that a black film is formed at the toluene-water interface. A silicon substrate was dipped vertically into the interface to transfer the assembled film formed at the toluene-water interface onto the solid substrate. Deposition method is the same as the Langmuir-Blodgett technique. The silicon substrate was washed with O3 treatment and made hydrophilic before the dipping process. The substrate was dipped into the interface at the rate of 10 mm/min and withdrawn at the same rate. The deposited film was observed by atomic force microscopy. The AFM image indicates that a densely packed CNT thin film is transferred onto the silicon substrate. Moreover, analysis of the height profile of the small area AFM image (indicated that the height of each CNT of the transferred film was determined as 17-20 nm, which is similar to the diameter of the template nanochannels, indicating that the assembled film that is formed at the liquid-liquid interface is an CNT monolayer and that the monolayer is transferable onto the silicon substrate using LB technique.
Q5.36
Liquid-crystalline Processing of Oriented Carbon Nanotube Array for the Application as Thin Film Transistors. Hyunhyub Ko1,2 and Vladimir V Tsukruk1,2; 1Materials Science and Engineering, Iowa State University, Ames, Iowa; 2Materials Science and Engineering, Georgia Institute of Technology, Atlanta, Georgia.
We introduce a simple solution-based method for the fabrication of highly-oriented carbon nanotube (CNT) array to be used in thin film transistors. We exploit liquid crystalline behavior of CNT solution near the receding contact line during tilted-drop casting and produced long-range nematic-like ordering of carbon nanotube stripes caused by confined micropatterned geometry. We further demonstrate that the performance of thin-film transistors is largely improved compared to TFT with random monolayer of CNTs with near-record career mobility. This approach has great potentials in low-cost, large-scale processing of high-performance electronic devices based on high-density oriented CNT films with record electrical characteristics such as high conductance, low resistivity, and high career mobility.
Q5.37
Patterned Forest-Assembly of Single-Wall Carbon Nanotubes on Gold Using a Non-Thiol Functionalization Technique. Haoyan Wei1, Sejong Kim2, Sang Nyon Kim2, Bryan D. Huey1, Fotios Papadimitrakopoulos2 and Harris L. Marcus1; 1Materials Science and Engineering Program, Department of Chemical, Materials and Biomolecular Engineering, Institute of Materials Science, University of Connecticut, Storrs, Connecticut; 2Nanomaterials Optoelectronics Laboratory, Polymer Program, Institute of Materials Science, Department of Chemistry, University of Connecticut, Storrs, Connecticut.
An approach of non-thiol functionalization of single-wall carbon nanotubes (SWNTs) on gold substrates was demonstrated via Fe3+-assisted self-assembly technique, which eliminated the time-consuming thiol adsorption process in the conventional methods. Upon immersion of gold substrates into pH 2.2 aqueous FeCl3 solution, FeCl3 directly hydrolyzed on the gold surface due to aqua regia effect and as a result films of FeO(OH)/FeOCl crystallites formed. Subsequent immersion into SWNT/dimethylformamide (SWNT/DMF) dispersion led to needle-like forest assemblies of SWNTs based on metal-assisted chelation and electrostatic interactions. Towards fabrication of SWNT patterns, two methods on localizing Fe3+/Au composite pads were investigated by either sputtering Au/Pd though a TEM grid or utilizing conventional photolithography. Such patterned Fe3+-functionalized Au structures provided the basis for the patterned site-specific forest-assembly of SWNTs as demonstrated by atomic force microscopy (AFM) measurement and resonance Raman spectroscopy.
Q5.38
Thermodynamics Modeling of Defective Carbon Nanotubes in the Presence of Adsorbates. Amanda S Barnard, Department of Materials, University of Oxford, Oxford, United Kingdom.
The growth of carbon nanotubes using hydrogen plasmas has received much attention in recent years, but far less attention has been given to the affect of adsorbed hydrogen on the relative stability of the final, possibly defective, structures. Presented here is a general analytical model for describing the energetics and relative stability of defective carbon SWNTs in the presence of adsorbates. The model is parameterized for the case of exohedral hydrogen, using results obtained from first principles simulations, and is constructed around experimentally relevant parameters such as the concentrations and configurations of adsorbates on the nanotube walls.
Q5.39
Mechanism and Control of CNT Adsorption onto Silicon Dioxide Surfaces. Timothy Peter Burgin1, Justin Lewenstein2 and Dennis Werho3; 1Motorola Labs, Motorola, Tempe, Arizona; 2Freescale Semiconductor, Tempe, Arizona; 3Medtronic, Tempe, Arizona.
The factors that control carbon nanotube (CNT) adsorption onto aminopropyl siloxane (APS)-derivatized surfaces have been investigated using two distinct types of well characterized films with significant differences in their detailed structures. Both types of APS films showed a marked increase in CNT adsorption relative to untreated SiO2 surfaces but differed in the amount of CNTs adsorbed. To gain insight into the factors governing adsorption, the surface coverage of the CNTs was monitored as a function of the pH during the deposition, the surfactant used to suspend the CNTs, and the type and amount of salt added to the deposition solution. The adsorption is shown to be governed by electrostatic and VDW forces. In the case of complimentarily charged surfaces, the adsorption occurs through an ion exchange mechanism. Patterned APS functionalized surfaces prepared by either photo-lithography, e-beam lithography, or micro-contact printing can be used to leverage the selectivity in CNT adsorption between bare SiO2 and APS derivatized regions of the surface to selectively place CNTs as either individual CNTs/bundles or CNT mats of controlled density.
Q5.40
A Facile Solution Method to Synthesis of C60 Nanorods. Bingbing Liu1, Lin Wang1, Guangtian Zou1, Agnieszka Iwasiewicz2 and Bertil Sundqvist2; 1Jilin University, National Lab of Superhard Materials, Changchun, China; 2Department of Physics, Umea University, Umea, Sweden.
Synthesis of C60 nanorods is a challenging topic due to its unique structure and properties and their potential application in the nanometer scale field. We report here that a very facile solution evaporation method has been found to synthesize C60 nanorods with widths and thicknesses of the order of nanometers using m-xylene as a shape controller. These unusual nanorods can easily grow on various substrates such as glass and silicon. Nanorods with different diameters and length-to-diameter ratios can be synthesized under different growth conditions. The nanorods obtained are highly crystalline and single phase. A gradual expansion of the lattice constant is also found in the C60 nanorods as their widths decrease.
Q5.41
First-principles Characterization of the Electrical, Vibrational, and Optical Properties of Carbon Nanotubes Functionalized with [2+1] Cycloadditions. Young-Su Lee, Nicola Bonini and Nicola Marzari; Department of Materials Science and Engineering, and Institute for Soldier Nanotechnologies, Massachusetts Institute of Technology, Cambridge, Massachusetts.
Recent theoretical studies have predicted that cycloadditions of carbenes or nitrenes can cleave the bond between two sidewall carbon atoms, disrupting the three-membered ring that is formed with the bridgehead carbon or nitrogen of the functional group. While other common functionalizations (such as hydrogenation, fluorination, or aryl addition) introduce sp3-hybridization for the sidewall carbons, and are accompanied by a rapid decrease of conductance (also observed experimentally), these bond-cleaved functionalized tubes are expected to preserve the number of pi-electrons and the electronic conductance typical of the pristine tubes. In order to provide a link with experimental characterization, we have studied in detail the electrical, vibrational, and optical properties of the bond-cleaved tubes to identify clear signatures that are directly or indirectly related to bond cleaving, and to assess the local bonding structure of such promising materials.
Q5.42
Novel Hybrid Polypropylene Nanocomposites with Montmorillonite and Single-Walled Carbon Nanotubes. Emmanouil Vamvounis, John L Stanford, Arthur N Wilkinson and Robert J Young; School of Materials, University of Manchester, Manchester, Greater Manchester, United Kingdom.
Polymer nanocomposites constitute a new category of composite materials, the combination of polymers and particulate fillers of which at least one dimension is in the nanometer range. The large surface area of the fillers leads to enhanced adhesion with the polymer matrix, providing large increases in desired properties, at low filler contents. Polypropylene (PP) is one of the most widely used polyolefin polymers for nanocomposite preparation. Montmorillonite (MMT) is the most widely used layered silicate in this application and the property improvement of polymer/MMT nanocomposites hinges on the successful exfoliation of the individual clay layers. Single-walled carbon nanotubes (SWNTs) are a cylindrical allomeric form of carbon with an average diameter of 1 nm; the composite reinforcement of their polymer composites improves with better dispersion of the nanotubes in the matrix [3]. Interaction between the MMT platelets and carbon nanotubes has been previously reported [1,2], providing a synergistic effect in polymer composites properties due to the combination of the two filler materials. In this study we report the results from our examination into the combination of MMT and SWNTs in a PP matrix to produce a hybrid nanocomposite. Two methods of mixing the materials were investigated: melt intercalation and solution casting. Formation of the hybrid nanocomposite is obstructed by the small interfacial interaction between the hydrophilic clay and the organic polymer, and the difficulties in nanotubes dispersion in certain solvents. To overcome these hindrances, MMT platelets are organically modified, the carbon nanotubes are functionalized with carboxylic acid, and maleated polypropylene (MAPP) is used as a compatibilizer to promote the interaction between the PP chains and the organically-modified MMT. The nanocomposites were characterised using X-Ray diffraction, transmission and scanning electron microscopy, differential scanning calorimetry, thermogravimetry and Raman microscopy. X-Ray diffraction and Raman microscopy were employed to establish the intercalation of the polymer and the carbon nanotubes between the clay layers [1], and the frequency shift of the characteristic G’ band of the nanotubes Raman spectrum was followed to quantify the composite reinforcement [3]. In some cases we observed interaction between the MMT platelets and the carbon nanotubes in the polymer matrix, which provides a synergistic effect due to the combination of the two filler materials. REFERENCES 1. Georgakilas V., Gournis D., Karakassides M.A., Bakandritsos A., Petridis D.; Carbon, 2004, 42, 865 2. Peeterbroeck S., Alexandre M, Nagy J.B., Pirlot C., Fonseca A., Moreau N., Philippin G., Delhalle J., Mekhalif Z., Sporken R., Beyer G., Dubois Ph.; Compos. Sci. Technol., 2004, 64, 2317 3. Lucas M., Young R.J.; Physical Review B, 2004, 69, 0854505
Q5.43
Heat Transport of Carbon Nanotubes Embedded in a Metal Layer. Yoichi Taira and Kuniaki Sueoka; Tokyo Research Lab, IBM, Yamato, Japan.
Carbon nanotubes (CNT) show high thermal conductivity along the axis of their fiber shape. These materials can be used to realize a conformable and highly thermal conducting material called as a thermal interface material (TIM) for the heat transport between a VLSI and a heat sink. Since a half of the total heat resistance between the VLSI and the final heat exit is occupied by the TIM layer, and the VLSI power density is getting higher and higher, there is always a need for a better TIM having lower heat resistance. The TIM layer is also necessary to connect the VLSI package to the heatsink so that separately manufactured components are well connected thermally but a relative mechanical motion is allowed along with the temperature change when starting or shutting off the system. Since CNTs have very high thermal conductivity along the axis, it is most effective to align the CNTs along with the heat path for the lower heat resistance. Therefore a possible form CNT based TIM is densely packed aligned CNTs where the both ends of the nanotubes are connected to the heat source and the heat sink. To study how much thermal resistance can be reduced, we set up a model consisting of an aligned CNTs in contact with two copper blocks as a heat source and a heat sink. One interesting result of the modeling is that the resulting heat resistance is much higher than expected when the ends of CNT are just in touch with the copper block. When some length of the CNT is embedded in a metal layer, the resulting thermal resistance gets much lower and approaches to the value calculated by the CNT heat conductivity. This result indicates a need of special structure between the copper block and the CNTs so that the CNTs are embedded in metal layer.. The result is also consistent with our experimental measurement using gold as an interfacial material.
Q5.44
Abstract Withdrawn
Q5.45
Multifunctional Macro Architectures of Double-walled Carbon Nanotube Fibers. Lijie Ci1, Saikat Talapatra1, Niramol Punbusayakul1,2, Robert Vajtai1 and Pulickel. M. Ajayan1; 1Department of Materials Science and Engineering, Rensselear Polytechnic Institute, Troy, New York; 2School of Bioresources and Technology, King Mongkut's University of Technology, Bangkok, Thailand.
Many applications are envisioned which will ultimately utilize macro-architectures fabricated using carbon nanotube building blocks. For example mechanically strong and long fibers of CNTs can act as superb electromechanical actuators, filler materials in composites, power cables and electrodes. In this presentation we report on the fabrication, characterization and application of long spun fibers of double walled carbon nanotubes (DWCNT). The DWCNTs have a higher structure stability and oxidation resistance than single walled CNTs (SWCNTs) and have properties such as high strength, low density and admirable thermal and electrical conductivity. A simple and scalable pulling-drying process of spinning pure DWCNT fibers out of nanotube “cotton” is described. Fibers having a wide range of diameters (10 μm -100 μm) and possessing high mechanical strength (tensile strength ~ 299 MPa, and Young’s modulus ~8.3 GPa) can be drawn using this process. Electrochemical tests using the DWCNT fibers as electrodes indicate that their surfaces possess very fast electron transfer kinetics and can be used as excellent working electrode for the electrochemical and bio sensing applications. Field emission tests performed on individual DWCNT fiber revealed a very low turn-on electric field (~0.4 V/mm) and a very high emission current density (~9 A/cm2 at the electric field of 1.0 V/mm). A very high field enhancement factor, b (~ 2.2×104) was also obtained form the nanotubes. Our findings show the possibility of using DWCNTs fibers as excellent components in various device architectures.
Q5.46
Strengthened, Aligned Carbon Nanotube Film via Vapor Phase Infiltration of Carbon. Xuesong Li, Lijie Ci and Pulickel M. Ajayan; Materials Science and Engineering, Rensselaer Polytechnic Institute, Troy, New York.
Due to the low density and high porosity of aligned carbon nanotube (ACNT) films, the as-grown ACNT film has low effective compression modulus and electrical and thermal conductance since the interstitial space between nanotubes is only occupied by air. Filling the space in the films with solid materials could modify the properties of the ACNT films and render them useful for applications that require increased robustness or hardness, protection from oxidation at high temperatures and other damaging ambients, or increased effective electrical and thermal conductance. In this study, the growth of ACNT films was conducted using a vapor phase catalyst delivery method by chemical vapor deposition (CVD) at 770 C using a mixture of xylene and ferrocene vapors. By simply increasing the reaction time for the nanotube growth, deposition of pyrolyzed carbon created carbon infiltrated ACNT/C composite films. Scanning electron microscope (SEM) analysis and microbalance measurements showed that after infiltration, the diameters of nanotubes and bulk density of ACNT film could be increased by an order of magnitude. TEM and Raman scattering analysis showed that the ACNT/C composite structure included a CNT core and a partially graphitized carbon shell. Compression tests showed that the modulus of the ACNT film could be increased by nearly 2000 times. These property enhanced ACNT film could have several potential applications, making up for the low density of as prepared ACNT films.
Q5.47
Stability and Crystallization Behaviors of Regioreguler Poly(3-hexylthiophene) Nanotube Composites. Ananta Raj Adhikari1, Mircea Chipara2, Chang Y. Ryu3, Pulickel M. Ajayan4 and Hassaram Bakhru1; 1College of Nanoscale Science & Engineering, State University of New York, Albany, New York; 2Department of Chemistry, Indiana University, Bloomington, Indiana; 3Department of Chemistry and Chemical Biology, Rensselaer Polytechnic Institute, Troy, New York; 4Department of Material Science and Engineering, Rensselaer Polytechnic Institute, Troy, New York.
The basic step to optimize the properties of filler in composite is the interfacial interaction between the matrix and the filler. Irradiation, a novel technique, was used in this work to introduce a wide range of defects in Single-walled Carbon nanotubes (SWNTs) as filler prior to composite formation. The thermal stability along with phase transition behavior of an organic conducting polymer Poly(3-hexylthiophene) (P3HT) loaded with pristine and ion implanted SWNTs was investigated using Thermogravimetry Analysis (TGA), Differential Scanning Calorimetry (DSC), Raman spectroscopy, and Electron Spin Resonance (ESR). Interestingly, we observed substantial improvement on the thermal stability and the changes in phase transition behavior of the composite with pristine and irradiated fillers.
Q5.48
Work Function of Functionalized Single-wall Carbon Nanotubes. Nicholas E. Miller, Janet S. Ryu and Nicola Marzari; Department of Materials Science and Engineering, Massachusetts Institute of Engineering, Cambridge, Massachusetts.
Engineering the properties of carbon nanotubes is of fundamental importance for many of their intended practical applications. Control of the Fermi level alignment is especially relevant for field-effect devices and for the nanotube contacts with the metallic leads. We focus here on the work function of metallic nanotubes, and on the changes that can be induced by electropositive or electronegative functionalizations. We use density functional theory within the PBE-GGA to study pristine, hydrogenated, and fluorinated (5,5) and (5,0) nanotubes along with more complex organic ligands as functional moieties.
Q5.49
In-situ Raman Study on Lithium Insertion into Double Walled Carbon Nanotubes-derived Bucky Paper. Yoong Ahm KIm1, Hayashi Takuya1, Muramatsu Hiroyuki1, Endo Morinobu1, Terrones Mauricio2 and Dresselhaus Mildred S.3; 1Shinshu University, Nagano, Japan; 2IPICYT, San Luis Potosí, Mexico; 3Massachusetts Institute of Technology, Massachusetts, California.
We carried out the Li+ storage behaviors of highly pure single wall carbon nanotube (SWNT)- and double walled carbon nanotubes (DWNT)-buckypapers (consisting of entangled tube bundles) as an anode material in lithium ion batteries (LIBs) using an in-situ the Raman technique. The fabrication of these SWNT- and DWNT-buckypaper via a filtering process resulted from entangled long nanotube bundles of either SWNTs or DWNTs, in which the nanotubes are packed into hexagonal arrays. These thin, flexible and mechanically tough SWNT- and DWNT-foils exhibit an analogue behavior to hard carbons upon Li ion insertion at different voltages. From our studies, we mainly suggest the interstitial space in bundled nanotubes as the Li+ storage sites in nanotube-foils. In detail, Raman changes as a function of Li+ addition will be discussed.
Q5.50
Subwavelength Microscopy with Elongated Nanostructures. Pavel S. Dorozhkin, Konstantin Mochalov, Sergey Saunin and Victor Bykov; NT-MDT Co., Zelenograd Moscow, Russian Federation.
In this report, we study optical properties of elongated nanostructures with a combined "AFM + confocal microscope" system. We use the NTEGRA Spectra system by NT-MDT that combines SNOM, AFM, confocal microscope and Raman spectroscopy in a single setup. Various types of elongated nanostructures (nanowires, nanotubes, polymer fibers etc.) are studied. Light transport in such subwavelength waveguides is investigated. We study spectra of transmitted light with respect to the injected light wavelength. We also estimate light transmission efficiency for nanowires of different diameters and materials paying attention on the role of defects on the transmission coefficient. Investigated structures are characterized by AFM, SNOM, and confocal Raman/luminescence techniques. Potential use of such structures as SNOM probes is discussed. We also use apex of AFM cantilever for local enhancement of electromagnetic field in attempt to obtain subwavelength resolution in fluorescence and Raman 2D maps.
Q5.51
Binding of Charged Supramolecular Complexes to Carbon Nanotubes. Harsh Chaturvedi2, Andrea Giordano1, Ryan Phillips1, Thomas Younts1 and Jordan Poler1,2; 1Chemistry, UNC Charlotte, Charlotte, North Carolina; 2Center for Optoelectronics and Optical Communications, UNC Charlotte, Charlotte, North Carolina.
Single walled carbon nanotubes (SWNTs) bind strongly to rigid ruthenium metallodendrimers. High valence ions effectively coagulate these nanotubes from stable dispersions in N,N-Dimethylforamide. While ruthenium salts and small [Ru(diimine)3]2+ complexes also coagulate the nanotubes they require much higher concentrations and are easily extracted from the nanotube floc with acetonitrile. Enantiomerically pure ruthenium metallodendrimer [Δ6Λ3Δ-Ru10]20+[PF6-]20 is shown to bind strongly and specifically to the SWNTs. Most of the nanotube’s ends are functionalized with this large (5.8 nm) optically active, rigid supramolecular complex. We study the binding capacity and binding kinetics with UV-VIS and Atomic Absorption spectroscopy and dynamic light scattering. Imaging the functionalized nanotubes with scanning electron microscopy and atomic force microscopy (AFM) yields direct confirmation of end functionalization. Statistical analysis of the AFM images shows the morphology of the functionalized ends is consistent with the nanotubes binding to one of the endo- or exoreceptors around the dendrimer. Implications of these results towards efficient nanomanufacturing of carbon nanotube devices are discussed.
Q5.52
Resonant Raman Scattering Excitation Profiles from a 1D System. Yan Yin1, A. Vamivakas2, A. Walsh1, S. Cronin3, M. S. Unlu2,1, B. B Goldberg1,2 and A. K. Swan2,1; 1Physics Department, Boston University, Boston, Massachusetts; 2Electrical and Computer Engineering Department, Boston University, Boston, Massachusetts; 3Electrical Engineering Department, University of Southern California, Los Angeles, California.
We examine resonant Raman scattering excitation (RRSE) profiles in carbon nanotubes as functions of the excitation energy and phonon wave vector for different kinds of Raman processes (one-phonon first order, one-phonon second order and two-phonon scattering). The different Raman processes results in qualitatively different resonance profiles. Hence, RRSE profile measurements should be helpful to determine the origin of Raman active peaks in carbon nanotubes and other 1D systems. As is well established, a Γ point one-phonon scattering ( e.g. from the radial breathing mode) produces a symmetrical profile of the outgoing and incoming resonant peaks. However, non-Γ point defects assisted one-phonon second order Raman scattering, e.g. the D band mode, causes an asymmetric profile where the outgoing resonant peak is more intense than the incoming resonant peak due to the nonzero phonon momentum. We calculate the D-band RRSE profiles for different carbon nanotubes and find that their line-shapes asymmetry depends on the nanotube Brillouin zone boundary. RRSE profiles generally have four resonance peaks for two-phonon scattering from two Γ point phonons with different energies Eph1 and Eph2; Eii, Eii+Eph1, Eii+Eph2 and Eii+Eph1+Eph2. In the special case of two phonons from the same branch and same energy, e.g. the G’ mode or RBM overtone, the profile has tree peaks. The origin of a Raman active band is not always easily identified; for example the high energy mode around 1700cm-1 (M band) in carbon nanotube Raman has been attributed both to RBM+LO scattering and as the second harmonic of an 850cm-1 mode. We show that these processes yields different profiles, hence resonance profile measurements should be helpful to determine the origin of the M mode.
Q5.53
Study on Double-walled Carbon Nanotube X-junction Formation. Takuya Hayashi, Yuta Shiba, Hiroyuki Muramatsu, Daisuke Shimamoto, Satoru Naokawa, Shusuke Furui, Yoong-Ahm Kim and Morinobu Endo; Shinshu University, Nagano, Japan.
Double-walled carbon nanotubes (DWNTs) are thought to possess remarkable electronic and mechanical properties compared to single-walled carbon nanotubes. The progress in nanoelectronics requires various molecular connections among individual DWNTs. The X-junction nanodevice made by joining coaxial tubes can have great mechanical strength and unique electrical characteristics. This device can form a nanonetwork, can contribute to the simplification of electronic circuits, composites, and more. Therefore, the elucidation of X-junction formation by joining DWNTs is crucial. We studied the formation of X-junction of DWNTs using molecular dynamics (MD) calculation. In the present study, we will discuss the results on MD simulations on the X-junction formation between DWNTs and the effect of defects in the tubes for the coalescence at the junction.
Q5.54
Sodium Chloride-Catalyzed Oxidation of Multi-Walled Carbon Nanotubes for Environmental Benefit. Kenji Takeuchi1, Takeyuki Tajiri1, Ki Chul Park1, Morinobu Endo1 and Mildred S. Dresselhaus2; 1Faculty of Engineering, Shinshu University, Nagano, Japan; 2Massachusetts Institute of Technology, Cambridge, Massachusetts.
The easy availability of multi-walled carbon nanotubes (MWCNTs) in a large scale (approx. 100 ton/year) through the development of a catalytic chemical vapor deposition method has established new areas of chemistry and physics for nanometer-sized carbon materials. We report here the use of sodium chloride (NaCl) as a catalyst for facilitating the oxidation reaction of graphitized MWCNTs. The reaction mechanism of the NaCl-catalyzed oxidation of MWCNTs has been investigated by the kinetic, spectroscopic and microscopic analyses. The results imply that the lowering of the oxidation temperature in the presence of NaCl originates from the introduction of disorder into the MWCNTs, thus increasing the facility of the oxidation reaction of the disorder-induced nanotubes. An accelerated oxidation process using the NaCl catalyst is important for contributing to facilitate the disposal of nanotubes for environmental considerations. This work has clearly demonstrated the effectiveness of NaCl as a catalyst for the oxidation of well-graphitized MWCNTs that exhibit a relatively high oxidation resistance. From the viewpoint of material costs, this approach has the advantage that the NaCl catalyst required for the large-scale treatment of MWCNT-based products could be available from abundant seawater, making the practical use of this approach very appealing.
Q5.55
Imaging and Engineering Strain in Individual Single-Wall Carbon Nanotubes. Hyungbin Son1, Xiaojie Duan2, Jin Zhang2, Georgii G. Samsonidze1, Yingying Zhang2, Ervin Mile1, Mildred S. Dresselhaus1, Zhongfan Liu2 and Jing Kong1; 1Massachusetts Institute of Technology, Cambridge, Massachusetts; 2Peking University, Beijing, China.
We explored two different methods to induce continuously varying strain in individual single-wall carbon nanotubes on a silicon dioxide substrate: van der Waals interaction with the substrate and manipulation with an atomic force microscope tip. We then demonstrate that the strain can be clearly imaged by Raman spectroscopy along the length of SWNTs. These results give insight into the interaction mechanism between the substrate and the SWNT. Our results also provide direct evidence for the modulation of electronic transition energies induced by strain. The method presented is an easy and straightforward way of engineering strain into SWNTs but also can be applied to other nanomaterial systems such as a single layer graphene sheet or other polymer molecules.
Q5.56
Embedding Nanoparticles in the Walls of Carbon Nanotubes. Davide Mattia, Gulya Korneva and Yury Gogotsi; Department of Materials Science and Engineering, Drexel University, Philadelphia, Pennsylvania.
Decoration of the walls of carbon nanotubes (CNT) with functional groups and/or nanoparticles has been performed to add additional functionality to CNT, ranging from increased solubility in water to the possibility of aligning nanotubes in magnetic field. In this work we present a different approach based on embedding nanoparticles in the walls of CVD-carbon nanotubes during the synthesis process. Nanoparticles ranging in size from 5 to 30 nm and composition from Fe3O4 to Au and nanodiamond were embedded in the walls of carbon nanotubes during the chemical vapors deposition (CVD) process. Multifunctionality can, therefore, be added to carbon nanotubes in a single step, avoiding complex and potentially damaging chemical processes to create anchor points for particles on nanotube walls, as done by decoration techniques. Gold and iron oxide nanoparticles can be used for surface enhanced Raman spectroscopy and magnetic manipulation, respectively. Upon annealing, silicon carbide nanoparticles and nanodiamonds transformed into hollow carbon onions embedded in the walls of the nanotubes as anchoring points between the graphene sheets of the tube. When the particles are only partially embedded in the walls of the nanotubes, additional carbon layers grow inside the hollow cavity of the tube covering these particles and producing hillocks inside the nanotubes. These features can be used for fluid mixing or separation at the nanoscale. Iron oxide particles are partially reduced to metallic iron during the CVD process and act as catalyst for small multi-wall carbon nanotube growth inside the CVD nanotubes. These structures can be used for mixing or separation of bio-polymers and nanoparticles from the fluid.
Q5.57
Boron-induced Coalescence in Double Walled Carbon Nanotubes. Hiroyuki Muramatsu, Kim Yoong-Ahm, Takuya Hayashi and Morinobu Endo; Shinshu University, Nagano, Japan.
The interconnection and coalescing of double walled carbon nanotubes (DWNTs) have attracted much attention of various researchers, because it is possible to fabricate complex architectures such as nanotube junction and networks that exhibit novel electronic and mechanical properties. Recently, we demonstrated that the outer shells of two adjacent DWNTs could coalesce via a zipping process by high temperature thermal treatment (ca. 2100oC). In addition, with the help of boron atoms, the temperature of merging in DWNTs is lowered by ca. 600oC, when compared to conventional DWNTs. Here, we have carried out systematic studies on boron-induced changes in DWNTs using Raman, UV, TEM, SEM and TGA analysis.
Q5.58
Abstract Withdrawn
Q5.59
Abstract Withdrawn
Q5.60
Abstract Withdrawn
Q5.61
Novel Metallic Substrate Mediated Growth of Aligned CNT Films and its Thermal Transport on Metal-NT Interface. Sunil Kumar Pal1, Saikat Talapatra2, Swastik Kar2, Robert Vajtai3, Linda S Schadler2,3, Theodorian Borca-Tasciuc1,3 and Pulickel Ajayan2,3; 1Mechanical, Aerospace, & Nuclear Engineering, Rensselaer Polytechnic Institute, Troy, New York; 2Materials Science & Engineering, Rensselaer Polytechnic Institute, Tory, New York; 3Rensselaer Nanotechnology Center, Rensselaer Polytechnic Institute, Troy, New York.
There is a growing interest to develop a method to replace silicon oxide as a substrate for CVD grown carbon nanotube film growth. Direct growth on a metal surface would open new opportunities to employ carbon nanotubes in novel electronic and sensing applications. We report here growth of vertically aligned multi-walled carbon nanotube (CNT) films on a novel metallic substrate. Our study suggests that not a single metal in its pure state favors the growth of nanotubes. However, a Ni based metallic super-alloy has been found suitable for the growth of aligned multi-walled nanotubes. A variety of nanotube architectures has been grown using this substrate. This system shows very good adhesion on the MWNT films providing provides both mechanical support and an electrically conductive backing which are very useful for a wide range of applications. The metal-nanotube films are also flexible and can be bent into different geometrical shapes and designs like cylindrical, cone, triangular and square. Since the interface between the tubes and the substrate can be a bottleneck for transport, our study will report the electrical and thermal transport across the carbon nanotube metal interface.
Q5.62
Synthesis of TCNQ-Cu organic nanowires and directed growth into devices Kai Xiao1, Jing Tao2, Ilia Ivanov1, Alex A Puretzky1, Zuqin Liu1, Stephen J Pennycok2 and David B Geohegan1,2; 1Center for Nanophase Materials Sciences, Oak Ridge National Laboratory, Oak Ridge, Tennessee; 2Materials Science and Technology Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee.
Single-crystal, one-dimensional semiconductor nanostructures are critical building blocks for nanoscale optical and electronic devices. Here we report a simple method for the chemical vapor deposition synthesis of single-crystal nanowires of many organic semiconductor materials. TCNQ-Cu is an organic charge-transfer complex with unique electrical properties which has applications in both optical and electrical recording media. This talk reports the direct synthesis and integration of semiconducting TCNQ-Cu organic nanowires into prefabricated electrode structures at temperatures compatible with growth on plastics. The organic nanowires are single crystalline and semiconducting, and can be reproducibly grown directly between patterned metal electrodes in a scalable process compatible with on-chip microelectronics without the need for post-processing, facilitating the integration of organic nanowire electronics into larger-scale systems. Current-voltage characteristic curves indicate that the semiconducting TCNQ-Cu nanowires have excellent contact with the electrodes and exhibit bistable switching properties which potentially could be used for high-density memory devices. This research was conducted in the Functional Nanomaterials Theme at the Center for Nanophase Materials Sciences, which is sponsored at Oak Ridge National Laboratory by the Division of Scientific User Facilities, U.S. Department of Energy.
Q5.63
Experimental Study of Factors Controlling the Production of Micro- and Nanofibers by Laser Spinning. Felix Quintero1,2, Adrian B Mann2, Juan Pou1, Fernando Lusquiños1 and Antonio Riveiro1; 1Applied Physics, University of Vigo, Vigo, Pontevedra, Spain; 2Materials Science & Engineering, Rutgers, The State University of New Jersey, Piscataway, New Jersey.
New applications of glass fibers with diameters in the nanometric range are quickly evolving. We have developed a new method, Laser Spinning, for the production of glass fibers with diameters in the nanometer to micrometer scale. The technique allows large quantities of nanofibers to be made with specific, controllable chemical compositions. This will potentially open up a whole new range of applications for the fibers. In Laser Spinning a high power laser is employed to melt the precursor solid material, while the melt forms glass fibers as a result of its viscous elongation and cooling by the drag force and convective heat transfer promoted by the gas jet. The fiber size and efficiency of the process are influenced by the operating conditions which must be control to give the desired characteristics. An analysis of the influence of several factors controlling the morphology and composition of the glass fibers has been performed experimentally. The results and conclusions of this study will be presented in this poster. The influence of the gas jet pressure and the laser power on the morphology and distribution of diameters of the fibers has been evaluated by means of electron microscopy analysis. The dependence of the average composition and homogeneity on the composition and microstructure of the precursor material has been studied using XPS and TOF-SIMS. The experimental results are explained based on a theoretical explanation of the process of Laser Spinning. This leads ultimately to the deduction of a set of rules regarding the influence of the factors studied on the production of nanofibers by Laser Spinning.
Q5.64
Novel Organic Nanohelices via Vapor-Solid Phase Transformation Reaction Seok Min Yoon1, Bonghwan Chon1, Wonjung Kim1, Sang Joo Lee2, Taiha Joo1, Seung-Koo Shin1, Jin Yong Lee3 and Hee Cheul Choi1; 1department of chemistry, pohang university of science and technology, Pohang, South Korea; 2Center for Computational Biology and Bioinformatics, Korea Institute of Science and Technology Information, 52 Eoeun-dong, Yuseong-gu, Daejeon, South Korea; 3chemistry, Sungkyunkwan University, 300 Chunchun-Dong, Jangan-Gu, Suwon, South Korea.
Organic nanostructures have been drawing a great deal of attention due to their potential roles in soft and miniaturized cargo systems for drug delivery organic-inorganic nanocomposites, organic nanoelectronics and also as a structure directing template for inorganic materials. So far, organic nanostructures, such as nanotubes, nanobelts and nanohelices have been synthesized mostly by self-assembly reactions in solution phase using complex unit molecules which should be specifically synthesized as designed to provide proper intermolecular interaction forces. In this presentation, we will discuss about our successful application of vapor-solid (VS) phase transformation reaction to the formation of helical nanobelts, i.e. nanohelices of m-aminobenzoic acid (m-ABA). m-ABA is one of the simplest commercially available molecules which contain both phenyl rings and terminal groups for hydrogen bondings. Among the produced nanobelts, more than 65% nanobelts are helical ribbon shapes. The efficiency of the formation of m-ABA nanohelices is critically affected by the chemical properties of substrate surface, where hydrophilic surface produces higher yields of m-ABA nanohelices. Furthermore, m-ABA nanohelices have been successfully applied as a template for metal helical nanostructures by simply evaporating metals. Among various metal nanohelices, gold nanohelices have been studied for a surface plasmon resonance (SPR) mediated optoelectrical device component.
Chair: Prabhakar Bandaru
Tuesday Morning, November 28, 2006
Room 312 (Hynes)
8:00 AM *Q6.1
Emerging Nanodevices made with Branched Nanostructures. Hongqi Xu, Division of Solid State Physics, Lund University, Lund, Sweden.
Recent theoretical and experimental investigations have shown that branched nanostructures exhibit novel electrical properties and can be used as building blocks for development of a new family of functional nanoelectronic devices and circuits (see, for example, Refs. [1-7]). In this talk, I shall first review these inherent properties of branched nanostructures and then recent works on realization of analog and logic devices and circuits with these emerging nano-objects. The emphasis will be given to self-assembled branched nanotube and nanowire devices [8-10]. I shall demonstrate that the devices show distinctly different electrical properties in different regimes of electron transport: ballistic, diffusive, and self-gating regimes. In particular, in the ballistic transport and self-gating regimes, branched nanostructures exhibit interesting rectification and switching characteristics. I shall also discuss possibilities of realizing truly nanoscale, all-nanowire or all-nanotube based functional devices and circuits [11]. References: [1] H.Q. Xu, Electrical properties of three-terminal ballistic junctions, Appl. Phys. Lett. 78, 2064 (2001). [2] I. Shorubalko et al., Nonlinear operation of GaInAs/InP-based three-terminal ballistic junctions, Appl. Phys. Lett. 79, 1384 (2001). [3] H.Q. Xu, Diode and transistor behaviors of three-terminal ballistic junctions, Appl. Phys. Lett. 80, 853 (2002). [4] I. Shorubalko et al., Tunable nonlinear current-voltage characteristics of three-terminal ballistic nanojunctions, Appl. Phys. Lett. 83, 2369 (2003). [5] D. Wallin et al., Nonlinear electrical properties of three-terminal junction, Appl. Phys. Lett. 89 (2006), in press. [6] I. Shorubalko et al., A novel frequency-multiplication device based on three-terminal ballistic junction, IEEE Electron Device Lett. 23, 377 (2002). [7] H.Q. Xu et al., Novel nanoelectronic triodes and logic devices with TBJs, IEEE Electron Device Lett. 25, 164 (2004). [8] P.R. Bandaru et al., Novel electrical switching behaviour and logic in carbon nanotube Y-junctions, Nature Materials 4, 663 (2005). [9] J. Park et al., Three-way electrical gating characteristics of metallic Y-junction carbon nanotubes, Appl. Phys. Lett. 88, 243113 (2006). [10] D. Suyatin et al., Fabrication and electron transport study of three-terminal InAs nanowire junctions, Proceedings of the 14th Int. Symp. On Nanostructures: Physics and Technology (2006). [11] H.Q. Xu, Nanotubes: the logical choice for electronics? Nature Materials 4, 649 (2005).
8:30 AM Q6.2
Electrical Property-Structure Correlation of Y Branched Carbon Nanotubes. Jeongwon Park1, Prabhakar Bandaru1 and Apparao M. Rao2; 1Materials Science and Engineering, University of California, San Diego, La Jolla, California; 2Department of Physics and Astronomy, Clemson University, Clemson, South Carolina.
Y-junction carbon nanotubes (Y-CNTs) have been suggested as building blocks of nanoelectronic devices such as diodes, switches, interconnections and logic devices. We have shown1 that any of the three branches in the Y-CNTs can be used as a control gate to change the current level in the other two branches, providing an additional intrinsic degree of freedom. In this study, Y-junction CNTs were dispersed on 100 nm thick electron transparent silicon nitride membrane, and connected to prepatterned Au electrodes, on the membranes, through Pt deposited by the Focused Ion Beam (FIB) technique. We report on the subsequent investigation of electrical property-structure correlation of Y-CNTs through TEM analyses and electrical measurements. In addition to demonstrating the viability of the technique for in-situ measurement, we will show the effects of the morphology (i.e., role of embedded catalyst particles, Y-CNT geometry and defects) on the novel electrical behavior that were recently observed.2 References 1. Bandaru PR, Daraio C, Jin S, Rao AM. Novel electrical switching behaviour and logic in carbon nanotube Y-junctions, Nat Mater. 4(9), 663 (2005) 2. J. Park, C. Daraio, S. Jin, P. R. Bandaru, J. Gaillard, and A. M. Rao, Three-way electrical gating characteristics of metallic Y-junction carbon nanotubes, Appl. Phys. Lett. 88, 243113 (2006)
8:45 AM Q6.3
Towards an AC Dielectrophoresis Toolkit for Assembling Carbon Nanotube Devices. Sarbajit Banerjee1,2, Brian White1,2, Limin Huang1,2, Blake J Rego1,2, Stephen O'Brien1,2 and Irving P Herman1,2; 1Nanoscale Science and Engineering Center, Columbia University, New York, New York; 2Department of Applied Physics and Applied Mathematics, Columbia University, New York, New York.
The precise placement of carbon nanotubes in device geometries is critical for their large-scale integration into the next generation of electronic devices. To controllably position nanotube interconnects and active elements, we use AC voltages to direct the alignment of nanotubes between microelectrodes. By optimizing the electrode geometry, applied voltage and frequency, load resistance, and type of nanotube sample used, we are able to achieve a substantial degree of control over the positioning of nanotubes. We are able to limit the number of tubes aligned in the gap by using a load resistor in series with the device. Floating potential metal posts are placed in the gap at positions predicted by static field simulations to direct nanotube assembly with very high resolution. The use of appropriate pointed electrode geometries allows the fabrication of crossed nanotube junction structures. We are also able to align nanotubes across deep pits using this technique. By using different nanotube samples we are also able to obtain some control over the electrical transport properties of the dielectrophoretically assembled devices. Further, we will discuss strategies for optimizing the transport properties of the devices, as well as electrochemical reactions that can be carried out at dielectrophoretically assembled nanotube electrodes. This work is primarily supported by the Nanoscale Science and Engineering Center at Columbia University, which is supported by the Nanoscale Science and Engineering Initiative of the National Science Foundation under NSF Award Number CHE-0117752. It is also partially supported by the MRSEC program of the National Science Foundation under Award Number DMR-0213574.
9:00 AM Q6.4
Techniques Toward lLrge-scale Chiral Separation of Single-walled Carbon Nanotubes via Selective Precipitation. Timothy J McDonald1,2, Jeffrey L Blackburn1, Chaiwat Engtrakul1, Garry Rumbles1 and Michael J Heben1; 1Basic Science, National Renewable Energy Lab, Golden, Colorado; 2Applied Physics, Columbia University, New York, New York.
Applications of single-walled nanotubes (SWNTs) require nanotubes of a single structural type; however, current synthesis methods result in a distribution of structures, each with a unique structure. It has become clear that post-synthetic separation processing techniques need to be developed. There have been a few promising separations demonstrated;[1, 2] however, these procedures do not appear to scale to the quantities necessary. In this presentation, we discuss selective precipitation as a technique toward the bulk separation of single-walled carbon nanotubes by type. We discuss these results, optical techniques that allow for rapid nanotube diameter-selective interaction determination[3] and a roadmap to improve the efficiency of such separations. It was previously thought that aqueous surfactants suspend single-walled carbon nanotubes of each type with equal probability; however, we have observed that removal of surfactant results in a reduction in the PL emission from larger-diameter nanotubes. To explore how surfactants bind preferentially to certain SWNTs, we perform photoluminescence (PL) spectroscopy as a function of time on SWNT suspensions diluted quickly below the surfactant concentration required to keep the SWNTs from rebundling. The PL versus time reveals the chirality-dependent binding energy of the various SWNTs to the surfactant molecule, as well as spectral shifts that indicate increased bundling of tubes with each other. This technique reveals that the binding of nanotube types to certain surfactant molecules strongly depends on the nanotube diameter and chirality. This implies that selective precipitation is possible. By searching for the right surfactant concentration, temperature, and utilizing subsequent centrifugation the efficiency of separation of this technique can be increased. In addition to aqueous surfactants, biological materials such as DNA have been shown to interact preferentially with certain nanotube types.1 We observe selective bundling of certain nanotube chiralities suspended in DNA that have been diluted and heated to a critical temperature. The same strand that has been shown to enhance the (6,5) nanotube in ion-exchange chromatography can also be enhanced via selective precipitation, a possibly higher throughput technique. In addition, we have recently been successful in suspending SWNTs in water with proteins and by wrapping with polymers and suspending in various organic solvents. We will use our rapid chiral-dependent binding energy measurement to search for new selective interactions of these novel suspension materials and SWNTs of a particular chirality and diameter. 1. Zheng, M., A. Jagota, M.S. Strano, et al., Science, 302, 1545-1548 (2003). 2. Peng, H., N.T. Alvarez, C. Kittrell, R.H. Hauge, and H.K. Schmidt, J. Am. Chem. Soc., (2006). 3. McDonald, T.J., M. Jones, C. Engtrakul, R.J. Ellingson, G. Rumbles, and M.J. Heben, Review of Scientific Instruments, 77, 053104 (2006).
9:15 AM Q6.5
Diameter and Chirality Dependent Characterization and Separation of Single Wall Carbon Nanotubes. Fotios Papadimitrakopoulos1, Zhengtang Luo1, Sang Nyon Kim1, Stephen K. Doorn2, Lisa D. Pfefferle3 and Gary L. Haller3; 1Polymer Program, Institute of Materials Science, University of Connecticut, Storrs, Connecticut; 2Chemistry Division, Los Alamos National Laboratory, Los Alamos, New Mexico; 3Department of Chemical Engineering, Yale University, New Haven, Connecticut.
The separation of carbon nanotube, over length, diameter, metallicity (metallic vs. semiconducting) and chirality, is the single greatest obstacle to the technological applications. Our initial report on the preferential interaction of surfactant amines with sem-SWNTs, as opposed to met-SWNT fraction, provided the means to alter their respective solubility characteristics and afford separation.[ J. Am. Chem. Soc. 2003, 125(11), 3370-3375] In the past three years, a number of separation methodologies have been proposed with moderate success, while the underlying nature of this separation remain elusive. In this contribution, the differential dedoping characteristics of SWNTs according to diameter and metallicity will be described as the fundamental cause that triggers this separation.[Nano Lett. 2005, 5(12), 2500-2504] Moreover, we indicate that extreme carefulness must be paid when investigating the nanotube abundance using resonance Raman spectroscopy, while not only their aggregation states [Phys. Rev. B, 2004, 70(24), 245429] but also their electron-phonon interaction [ Appl. Phys. Lett. 206, 88, 073110] dominates the Raman intensity of each (n,m)-SWNTs. In addition, we present a methodology to evaluate individual (n,m)- the nanotube abundance using fluorescence and UV-Vis-NIR absorption spectroscopy. Success of this methodologies was confirmed by theoretical prediction the lognormally distributed nanotube diameter distribution and absorption spectrum of a narrowly diameter distributed Co-MCM-41 synthesized SWNT samples.
9:30 AM Q6.6
Lithography-free In Situ Ohmic Contacts to Single-Walled Carbon Nanotubes. Aaron D Franklin1,2, Joshua T. Smith1,2, Matthew R. Maschmann1,3, David B. Janes1,2, Timothy Sands1,2,4 and Timothy S. Fisher1,3; 1Birck Nanotechnology Center, Purdue University, West Lafayette, Indiana; 2Electrical & Computer Engineering, Purdue University, West Lafayette, Indiana; 3Mechanical Engineering, Purdue University, West Lafayette, Indiana; 4Materials Engineering, Purdue University, West Lafayette, Indiana.
Single-walled carbon nanotubes (SWNTs) have been integrated into electronic and chemical sensing devices because of their exceptional electronic transport properties and potential for high-density integration. Devices produced by dispersing or synthesizing SWNTs across lithographically defined metallic contact pads have been limited to two-dimensional planar architectures, dictated in large part by contact metallization techniques. Although horizontal structures have proven invaluable for examination of SWNT transport properties and functional material selection, their flexibility is severely limited and precludes the exploitation of the nanometer-scale diameter of SWNTs as a scaling metric. The evolution of SWNT devices beyond planar, two-dimensional configurations will require contact metallization methods outside the scope of current lithographic techniques. Though plans for more complicated vertical carbon nanotube devices have been proposed in recent years, the means to establish electrical contact and structural support to individual vertical SWNTs has not been directly addressed. In part, the inability to synthesize vertical SWNTs in predefined locations has hindered research into the topic. We present both a robust vehicle to synthesize vertical SWNTs in channels suitable for device integration and a facile means to contact the top and bottom of SWNTs simultaneously without the use of a single lithographic procedure. Using microwave plasma-enhanced chemical vapor deposition (PECVD), SWNTs were synthesized from a Fe catalyst embedded in the pore walls of a PAA template. In this structure, SWNTs originate from the localized embedded catalyst layer and emerge at the top PAA surface, forming vertical channels within the pores. The SWNTs continue to lengthen along the top PAA surface after emerging from the pores. TEM micrographs and micro-Raman spectra confirm the existence of SWNTs with diameters in the range of 1-2 nm. The development of in situ top and bottom contacts to the SWNTs is achieved simultaneously by electrodepositing Pd into the pore bottoms. Using a Ti layer beneath the PAA template as the working electrode, Pd is electrodeposited within the pores to form nanowires that contact the bottom of the SWNTs. Once the Pd nanowires reach the level of the catalyst layer, they contact the bottom of the SWNTs emerging from the catalyst layer. After contact is established, continued deposition yields Pd nanoclusters formed annularly on portions of the SWNTs on the top PAA surface. Variation of the electrodeposition conditions enables control over the size of the Pd nanoclusters, as well as their morphology. Creating the most stable deposition environment yields mono-crystalline Pd nanocubes that form concentrically around the SWNTs. This structure enables fabrication of vertical SWNT devices such as field effect transistors, field emission arrays, and sensing networks.
9:45 AM Q6.7
Focused Ion Beam Induced Growth of Antimony Nanowires. Christoph Schoendorfer1, Alois Lugstein1, Youn-Joo Hyun1, Peter Pongratz2 and Emmerich Bertagnolli1; 1Institute for Solid State Electronics, Vienna University of Technology, Vienna, Austria; 2Institute for Solid State Physics, Vienna University of Technology, Vienna, Austria.
We investigate the impact of high energy Ga ions focused to diameters below 100nm on antimony (Sb) substrates which offers a new approach for the formation of Sb nanowires. In contrast to several well-known processes for bottom-up fabrication of one-dimensional nanostructures, e.g. the vapor-liquid-solid (VLS) growth mechanism in combination with metals as catalyst material [1] or the electrochemical deposition (ECD) using templates such as anodic alumina membranes [2], for this process no additional temperature treatment nor any additional material component is needed. The catalyst is provided by the FIB itself and the sputtered substrate material may act as localized material source for the growth of the nanostructures on ion exposed sites. The energy needed for the growth process is supposed to be supplied by the FIB due to the impact of the high energy Ga ions on the substrate surface. If a pure Sb surface is exposed to the FIB, interaction between the accelerated ions and the substrate atoms and therefore an evolution of the surface takes place. Depending on beam parameters and scanning mode different effects can be observed. Conventional multi pass scanning even at comparable low ion fluences leads to nucleation of seeds that initiate growth of nanofibers with very homogeneous diameters in the range of 20nm. Ongoing milling results in growth of nanofibers outside the rims of the predefined milling area where undisturbed growth can take place. Using single pass mode, the whole ion fluence is deposited in one single scan, which leads to a growth of nanofibers also at the predefined milling area. A porous ramp-like vertically uplifting structure parallel to the scanning direction can be observed. TEM investigation shows that the FIB induced Sb nanofibers are completely amorphous. Recrystallization by moderate thermal annealing can be obtained. For Sb nanostructures crystal nucleation starts already at temperatures around 150°C [3]. This is observed by using a special TEM specimen holder containing a heating coil for in-situ temperature treatment. An annealing at 200°C results in completely recrystallization of the amorphous nanofibers into finely-grained polycrystalline nanostructures. In this work, we present a new FIB based approach, which allows fabrication of predefined areas with high densities of polycrystalline Sb nanowires. [1] R.S. Wagner, W.C. Ellis, Appl. Phys. Lett. 4, 89 (1964). [2] A.J. Yin, J. Li, W. Jian, A.J. Bennett, J.M. Xu, Appl. Phys. Lett. 79, 1039 (2001). [3] M. Barati, J.C.L. Chow, P.K. Ummat, W.R. Datars, J. Phys.: Condens. Matter 13, 2955 (2001).
Chair: Susumu Saito
Tuesday Morning, November 28, 2006
Room 312 (Hynes)
10:30 AM *Q7.1
Phonon-Phonon Interactions in Suspended Carbon Nanotubes. Jose Menendez, Department of Physics, Arizona State University, Tempe, Arizona.
The combined dynamics of electrons and phonons determines the transport properties of materials. When the electron-phonon coupling is very strong for a particular vibrational mode (for example, LO phonons in polar semiconductors), a large non-equilibrium population is established for this mode under device operating conditions. In such cases, the purely anharmonic decay rate of the mode is of particular technological importance, since the non-equilibrium phonons that are not annihilated through phonon-phonon interactions increase the electron-phonon scattering rates.<br><br> Recent experiments indicate that electrons tunneling into a metallic carbon nanotube induce the establishment of a non-equilibrium population for the radial breathing mode (RBM).1 Thus the anharmonic lifetime of this mode is of interest for transport phenomena in nanotubes. The study of anharmonic interactions in carbon nanotubes is also of fundamental interest. While the tunneling experiments suggest that the RBM anharmonic lifetime is as long as 10 ns, the room temperature linewidth of this Raman-active active mode has been reported to be around 3 cm-1, which corresponds to a lifetime of approximately 2 ps. A discrepancy of more than three orders of magnitude suggests that the theoretical framework connecting time domain and frequency domain experiments needs revision in the case of carbon nanotubes.<br><br> We present new Raman scattering results from suspended carbon nanotubes showing that the natural FWHM at room temperature can be as small as 0.7 cm-1. This width, while much smaller than previously reported, corresponds to a lifetime of 7.6 ps, still three orders of magnitude shorter than the tunneling results. We propose a model to explain the discrepancy based on the existence of a non-equilibrium population for the RBM and its decay products. This has already been observed in a few cases in three-dimensional crystals, but we show that non-equilibrium secondary phonons are much more likely to be found in one-dimensional systems. This scenario can lead to time-domain experiments showing lifetimes approaching the nanosecond range, while at the same time Raman experiments would detect linewidths on the order of 1 cm-1, in the picosecond range. These results may have wide-ranging implications for the use of carbon nanotubes as electrical interconnects.<br><br> This work was done in collaboration with R. Rao, A.M. Rao, and C.D. Poweleit and supported by NSF under grant DMR-0244290.<br><br> 1B.J. LeRoy, S.G. Lemay, J. Kong, and C. Dekker, Nature 432, 371 (2004).
11:00 AM Q7.2
Nanowires Enabling Signal Enhanced Nano-Raman-Spectroscopy. Michael Becker1, Vladimir Sivakov2,3, Ulrich Gösele2, Gudrun Andrä3, Ruth Geiger4, Hans-Juergen Reich4, Samuel Hoffmann5, Johann Michler5 and Silke Christiansen1,3; 1Martin-Luther University, Halle, Germany; 2Max-Planck-Institute for Microstructure Physics, Halle, Germany; 3Institute for Physical Hightechnology e.V., Jena, Germany; 4Horiba Jobin Yvon GmBH, Bensheim, Germany; 5EMPA, Thun, Germany.
Silicon nanowires grown by the vapor-liquid-solid (VLS) growth mechanism, catalyzed by gold show gold caps of ~20 nm - 500 nm in diameter with an almost ideal half-spherical shape when the electron-beam evaporation growth technique is applied. Nanowires with lots of particularly small and particularly large nanowires/gold caps form at a growth temperature of 650°C and an evaporation current of 80 mA. The gold caps with smaller diameters are extremely well suited to exploit the tip- or surface enhanced Raman (SERS/TERS) effects. Our contribution gives examples of SERS-measurements using a nanowire based SERS-substrate, where we show that a single nanowire gold cap can produce a strong signal enhancement. As a model substance for Raman analysis we use malachite green that shows a characteristic Raman signature between 200 cm-1 and 1800 cm-1. Laser excitation is carried out by the 633nm wavelength of the HeNe laser.
11:15 AM Q7.3
One and Two Phonon Resonant Raman Scattering from Single Wall Carbon Nanotubes. Anthony N Vamivakas1, Yan Yin2, Andrew Walsh2, Selim Unlu1, Bennett Goldberg2 and Anna Swan1; 1Electrical and Computer Engineering, Boston University, Boston, Massachusetts; 2Physics, Boston University, Boston, Massachusetts.
Single wall carbon nanotubes (SWNTs) represent prototypical one-dimensional systems that are under intense study to understand their basic physical properties. The one dimensional character of SWNTs results in sharp van-Hove singularities in the density of states for both the electronic and vibronic tube excitations. Resonant Raman scattering is an optical technique that allows one to simultaneously probe the vibrational and electronic properties of SWNTs. In one phonon resonant Raman scattering (1phRRS), the shift between incoming and Raman scattered photon provides information about zone-center phonon energies. In addition, the scattering cross-section reflects singularities in the tube’s electronic structure and therefore permits determination of electronic excitation energies. Recent theoretical and experimental work has concluded that the tube’s electronic excitations are excitonic in nature. We have modeled the 1phRRS cross-section considering the effect of both free electron and excitonic intermediate states on the Raman process and found that for the range of exciton binding energies in the tubes we measure, there is no appreciable change in the resonance excitation profile (the cross-section as a function of laser energy). In two phonon Resonant Raman scattering (2phRRS), the scattering cross-section not only has features associated with electronic transition energies in the tube, but also is modulated by the density of states of the two phonons that participate in the scattering event. We have extended earlier work on achiral tubes to calculate both the phonon dispersion and density of states for chiral SWNTs. We have been able to use the density of phonon states to calculate the two phonon scattering cross-section for tubes of arbitrary chirality.
11:30 AM Q7.4
Raman Antenna Effect of Semiconducting Nanowires. Qihua Xiong1, G. Chen1, M. E. Pellen2, J. S. Petko2, D. Werner2 and Peter C. Eklund1; 1Department of Physics, The Pennsylvania State University, University Park, Pennsylvania; 2Department of Electrical Engineering, The Pennsylvania State University, University Park, Pennsylvania.
We demonstrate the first observation of nanoscale antenna effect in semiconducting nanowires probed by polarization dependent Raman scattering in individual GaP nanowires. We found that below a “threshold” diameter λ/4 (~ 120 nm) both longitudinal optic (LO) and transverse optic (TO) intensities exhibit a dipolar antenna pattern as the angle between laser polarization and nanowire axis was varied, where λ is the wavelength of the excitation field. Beyond this “threshold” diameter, the dipolar behavior was strongly suppressed; instead a quadruple or multipole pattern was developed. We understand this nano-antenna effect by calculating the electric distribution dependence on polarization for a finite length cylindrical GaP nanowire using a discrete dipole approximation (DDA) approach. This calculation was also supported by analytical solutions to Maxwell’s equations for infinite cylindrical wires.
11:45 AM Q7.5
Counting Graphene Layers by Raman Spectroscopy Andrea C. Ferrari1, Jannik C Meyer2, Vittorio Scardaci1, Cinzia Casiraghi1, Michele Lazzeri3, Francesco Mauri3, Stefano Piscanec1, Da Jiang4, Kostya Novoselov4 and Andre Geim4; 1Engineering, University of Cambridge, Cambridge, United Kingdom; 2Max Plank Institute for Solid State Research, Stuttgart, Germany; 3Institut de Mineralogie et de Physique des Milieux Condenses, Paris, France; 4Department of Physics and Astronomy, University of Manchester, Manchester, United Kingdom.
Graphene, a single, one-atom-thick sheet of carbon atoms in a honeycomb lattice, is the two-dimensional (2D) building block for carbon allotropes of every other dimensionality. It can be stacked into 3D graphite, rolled into 1D nanotubes, or wrapped into 0D buckyballs. Only very recently graphene has been produced in its free state [1]. This has fuelled research in 2D carbons and highlighted their remarkable electronic properties [2,3]. Graphene is a ballistic conductor in which electrons mimic the behaviour of massless, relativistic particles [2,3]. Electron transport is governed by the (relativistic) Dirac equation (rather than the Schrödinger equation) and this allows access to the rich and subtle physics of quantum electrodynamics in a condensed matter experiment Here we present the Raman fingerprint of an isolated graphene layer and its evolution with the number of layers [4]. This is supported by the definitive identification of free-standing single and bi-layers by transmission electron microscopy and electron diffraction. We show that graphene’s electronic structure is uniquely captured in its Raman spectrum. We identify the unique features of its Raman spectrum, which fingerprints graphene amongst all other carbon allotropes. We compare its spectrum to that of n graphene layers having the same stacking as graphite, with n=2 to 28. We demonstrate that the Raman spectrum evolution with increasing number of layers uniquely reflects the evolution of the electronic structure and electron-phonon interactions. This makes Raman spectroscopy is a quick, high-throughput, non-destructive technique for the unambiguous identification of graphene layers, which is critically lacking in this emerging research area. Finally we discuss the implications for the interpretation of the Raman spectra of single and double wall nanotubes. 1 K. S. Novoselov et al. Proc. Natl. Acad. Sci. USA 102, 10451 (2005) 2 K S Novoselov et al, Science 306, 666 (2004); Nature 438, 197 (2005) 3 Y Zhang et al Nature 438, 201 (2005) 4 A. C. Ferrari et al cond-mat/0606248 (2006)
Chair: Mahendra Sunkara
Tuesday Afternoon, November 28, 2006
Room 312 (Hynes)
1:30 PM Q8.1
Controlled Synthesis of Millimeter-Long Silicon Nanowires with Uniform Electronic Properties. Won Il Park1, Gengfeng Zheng1, Xiaocheng Jiang1, Bozhi Tian1 and Charles M. Lieber1,2; 1Department of Chemistry and Chemical Biology, Harvard University, Cambridg, Massachusetts; 2Division of Engineering and Applied Science, Harvard University, Cambridge, Massachusetts.
Ultralong one-dimensional (1D) nanostructures can serve as the unique building blocks that interlink the nanometer-scale world with the real macroscopic world. In particular, 1D structures whose lengths extend over macroscopic length scales and reach the size of integrated systems, could be exploited as the basis for complex electronic circuits and functional nanosystems by the direct integration of a number of device elements into a single nanostructure. To this end we report the controlled synthesis of millimeter-long silicon nanowires (SiNWs) that exhibit highly uniform and controllable electronic properties. Quantitative studies show that the average length of SiNWs after one hour growth is 1.8 mm, and that the diameters are uniform along the nanowires lengths. Transmission electron microscopy measurements further revealed that the millimeter-long SiNWs grow along the <110> direction without a diameter-dependent preference for growth direction, in contrast to previous studies carried out under much slower growth conditions. The origin of these differences will be discussed. In addition, millimeter-long p-type SiNWs were synthesized, and transport measurements demonstrated uniform field-effect transistor properties over the entire lengths of these nanowires. The electronic uniformity of millimeter-long SiNWs was exploited for multiplexed electrical sensing of cancer marker proteins, and highlights the unique opportunities opened up by this new nanowire building block.
1:45 PM Q8.2
Hetero-epitaxy of Germanium Nanowires on Silicon Substrates from Gold Colloids and Subsequent Gold Catalyst Removal. Jacob Huffman Woodruff, Joshua Ratchford and Christopher Chidsey; Chemistry Department, Stanford University, Stanford, California.
Gold catalyzed, chemical-vapor-deposition grown semiconductor nanowires are being investigated as promising materials for active device elements in 3D electronics, solar cells, biosensors, and many more applications. In particular germanium nanowires are interesting due to their sub-400°C growth temperatures. The ability to control the nanowire’s orientation and diameter and the ability to remove the gold catalyst after growth are crucial, however, in order for most of these applications to be realized. One technique that we have investigated to gain orientational control is to induce hetero-epitaxy with an underlying silicon crystalline substrate during growth. Previous studies have focused on nanowires grown from evaporated films of gold on silicon, which result in a wide range of nanowire diameters. Because the nanowire’s diameter is determined by the starting gold catalyst size, well-controlled homogeneous nanowire diameters can be obtained by using commercially available gold colloid solutions with narrow size distributions. However, gold colloid deposition techniques most used in literature that rely on “linkers” such as aminopropyltriethoxysilane (APTES) or poly-L-lysine, are shown to inhibit epitaxial growth of nanowires on silicon. It is hypothesized that this is due to either oxide formation or chemical modification of the surface by the linker during the colloid deposition procedure. We have developed a new, linker free, method for depositing gold colloids onto silicon that inhibits silicon oxide formation and results in hetero-epitaxy of germanium nanowires. Studies of hetero-epitaxy of germanium nanowires on silicon substrates using this method are presented. In order to remove the gold catalysts from the tip and surface of the nanowire after growth we have developed novel chemical etching techniques. It was found that iodine based gold etches commercially made for silicon quickly etch away germanium nanowires. We have developed a novel technique to passivate the germanium nanowires and keep them passivated during etching to allow removal of gold without destroying the nanowires. We present gold catalyst etch studies from germanium nanowires using these techniques.
2:00 PM Q8.3
Abstract Withdrawn
2:15 PM Q8.4
Surface and Electrical Characterization of InAs Nanowires. Qingling Hang1, Fudong Wang2, Dmitry Zemlyanov1, William E Buhro2 and David B Janes1; 1Purdue University, West Lafayette, Indiana; 2Washington University, St. Louis, Missouri.
InAs is an excellent candidate for high speed circuits and optoelectronic devices for its extremely high electron mobility and small, direct band gap. InAs nanowire transistors represent a potential geometry for high speed electronics in flexible or large area applications. It has been claimed that it is almost impossible to achieve p type InAs nanowire field effect transistors (FETs) due to the observation of Fermi level pinning above the conduction band of InAs due to the surface states. Appropriate control of these surface states is essential in realizing such devices. In this study, a solution-liquid-solid (SLS) method is used to synthesize InAs nanowires with p type doping. Both Cd and Zn dopants were studied, at high concentrations. Initial mass spectroscopy shows that there are ligands from the synthetic process, including hexaldecylamine (HDA) molecules, on the nanowire surface; these ligands probably bind to In surface atoms. X-ray photoelectron spectroscopy (XPS) of the nanowires provides additional evidence of ligands binding to In, including peaks associated with N from the HDA. The XPS also provides information about the valence states of the In and As atoms near the surface. Source and drain contacts were subsequently formed on InAs nanowires using evaporated Ni, followed by gold deposition, with patterning by electron-beam lithography. A heavily doped P+ Si substrate acts as a back gate, with a 40-nm thick SiO2 layer serving as the gate insulator. The Cd-doped devices exhibited ambipolar conductivity, while the Zn-doped devices exhibited strong n-channel conductivity, but no significant p-channel. The observation of unipolar versus ambipolar conductivity in these devices, controlled by the doping species, indicates that it is possible to vary the position of the Fermi level in the InAs nanowires. Variable temperature measurements show that there are minimal barriers existing at the contacts between Ni and Zn doped InAs nanowires; in contrast, there are Shottky barriers between Ni and Cd doped InAs nanowires. The nearly symmetrical Schottky barriers for n and p channels indicate that the Fermi level of Ni may fall near midgap of the Cd doped InAs nanowires. The observation of p channels in Cd-doped InAs nanowire transistors, along with the XPS study of the nanowire surface, indicate that the Fermi level is unpinned at the surface of the ligand-encapsulated InAs nanowires. The ligands are believed to passivate the surface states, resulting in the same Fermi level as that of nanowire body. This study provides a way to find building blocks to achieve high speed circuits and optoelectronic devices out of InAs nanowires
2:30 PM Q8.5
Shape Control Growth of Crystalline Lanthanum Hexaboride Nanostructures. Joseph Reese Brewer1, Nirmalendu Deo1 and Chin Li Cheung1,2; 1Chemistry, University of Nebraska-Lincoln, Lincoln, Nebraska; 2Nebraska Center for Materials and Nanoscience, University of Nebraska-Lincoln, Lincoln, Nebraska.
Materials of low work function and high aspect ratios are of great importance for the development of new field-induced electron emitters using low applied voltage. Lanthanum hexaboride (LaB6) is among one of the known materials with the lowest work function. Though the synthesis of LaB6 nanowires was reported in the literature, the shape control synthesis of robust LaB6 nanostructures is lacking. Here, we report our development of a low temperature CVD process for the growth of single crystalline LaB6 nano-obelisks and nanowires. Growth of these obelisks and wires was controlled by the substrate lattice orientation, catalyst particles, and the temperature of the reaction. Diameters of the nano-obelisk tips range from 20 to 50 nm. The nanowire diameters range from 20 to 100 nm with a preferred (100) growth orientation. The shape control growth mechanism of these different nanostructures and its application for fabrication of field emitter arrays will also be discussed.
Chair: Sungho Jin
Tuesday Afternoon, November 28, 2006
Room 312 (Hynes)
3:15 PM *Q9.1
Inorganic Nanowires: Novel Synthesis Approaches and Applications Mahendra Kumar Sunkara, Chemical Engineering, University of Louisville, Louisville, Kentucky.
The synthesis of nanowires in bulk amounts with control on growth direction and size is important, but has not been fully accomplished. Toward this goal, our group has been working on several new synthesis strategies for bulk production of inorganic nanowires. These strategies are non-traditional compared to the well accepted metal-cluster templated vapor-liquid-solid route, and they can be generalized as (a) bulk nucleation and growth of nanowires from quasi-immiscible solvents such as low-melting metal melts (b) direct nucleation and growth of oxide nanowires from metal foils and (c) nucleation and growth of nanowires directly from the vapor phase. In all of the above cases, the nucleation step controls the size of the resulting nanowires. A rational basis for nanowire synthesis using low-melting point metal melts is developed and is verified using the nucleation and growth kinetics of Ge nanowires. The experimental demonstrations for nanowire synthesis were however performed for a wide-range of materials systems involving nitrides, oxides, sulfides, metals, semiconductors and several compound semiconductors. Kinetic Monte-Carlo (KMC) simulations were also performed to understand the relationship between the growth kinetics and the resulting growth directions for nanowires. The simulation results suggest that the fast growth kinetics lead to <100> growth direction compared to <111> growth direction under slow growth kinetics for diamond-cubic structured material systems. The experimental results from our group on Ge nanowire growth kinetics from Ga melts and other groups on epitaxial nanowires consistently validate the KMC simulation findings. Finally, this presentation will highlight our studies on the behavior of nanowires in applications requiring bulk quantities such as (a) dispersions, (b) gas sensing, (c) composites and (d) solar cells. The results show that the nanowires with their high aspect ratios have inherent advantages and perform better than their nanoparticle counterparts.
3:45 PM Q9.2
Nucleation Mechanism for Catalyst-free GaN Nanowires. Kris A. Bertness1, A. Roshko1, L. M. Mansfield1, I. Levin2 and N. A. Sanford1; 1Mail Stop 815.04, NIST, Boulder, Colorado; 2NIST, Gaithersburg, Maryland.
We have examined the initial steps for catalyst-free growth of GaN nanowires by molecular beam epitaxy on Si (111) substrates with AlN buffer layers. We find that the nanowires nucleate in the center of small, dense hexagonal pits that form in the GaN layer immediately above the AlN buffer. Both field emission scanning electron microscopy (FESEM) and atomic form microscopy (AFM) indicated that the sidewalls of the pits were tilted approximately 43 to 50 ° from the substrate normal, which is near the 43.2° angle expected for (1 0 -1 2) planes. This orientation resulted pit sidewalls with the same azimuthal alignment as the (1 0 -1 0) planes that formed the sidewalls of the nanowires. We have shown previously that the nanowires grow along the [0 0 0 1] axis with regular hexagonal cross-section. The nanowires were 20 to 500 nm in diameter and had lengths up to 15 μm. These wires form spontaneously under high N-to-Ga ratios for a growth temperature range of about 810 to 830 °C, without the use of a catalyst. Recent results on the polarity of GaN nanowires will also be presented. In contrast to other catalyst-free GaN nanowire growth with MBE,[1,2] our nucleation studies show that the use of an AlN buffer layer is essential to the regular formation of the nanowires and GaN underlayers under our growth conditions. Our typical AlN buffer layer was 40 nm to 50 nm thick. We also observed that higher atomic nitrogen concentrations promoted a higher density of nanowires, and higher concentrations of excited molecular species increased the growth rate of the nanowires and thick underlayers. Based on several observations concerning the growth of the GaN nanowires, including the absence of any observation of Ga droplets at the end facets, we conclude that the nucleation mechanism for nanowires is based on variations in sticking coefficients for Ga on the various planes formed by GaN, and not on Ga droplet formation. [1] J. Ristić, M. A. Sánchez-García, E. Calleja, J. Sanchez-Paramo, J. M. Calleja, U. Jahn, and K. H. Ploog, Phys. Status Solidi A-Appl. Res. 192, 60-66 (2002). [2] A. Kikuchi, M. Kawai, M. Tada, and K. Kishino, Jpn. J. Appl. Phys. Part 2 43, L1524-L1526 (2004).
4:00 PM Q9.3
Electronic Structure of Donor and Acceptor Impurities in Semiconductor Nanowires. Mamadou Diarra1, Yann-Michel Niquet2, Christophe Delerue1 and Guy Allan1; 1ISEN, IEMN (UMR CNRS 8520), LILLE, France; 2DRFMC/SP2M/L_Sim, CEA, Grenoble, France.
Recent breakthroughs in the growth of semiconductor nanowires (SNWs) have opened up great opportunities for nanoscale device applications. SNWs remain semiconducting independent of their diameter and orientation, giving the ability to control their properties by doping. Therefore a large number of experimental works have addressed the problem of doping and of its modulation in SNWs. While there is no doubt that p- and n-type SNWs can be produced, the question of how their electrical conductivity depends on the doping level remains largely open. In fact, most of the works showing good transport properties concern SNWs doped with high impurity concentration, near or above the Mott density. In order to investigate the doping efficiency in SNWs, we present calculations of the electronic structure of donor and acceptor impurities in Si nanowires. We show that their ionization energy increases due to the confinement, the quantum confinement at small sizes (diameter < 5 nm) and above all the so-called dielectric confinement which occurs when there is an important dielectric mismatch between the wire and its surrounding. For SNWs embedded in a material with a low dielectric constant, we obtain that the impurities cannot be ionized at room temperature even for diameters up to several tens of nanometers. We explain the origin of this behavior by considering the effect of the impurity potential and of the self-energy of the carrier, and we make predictions for the ionization energy in different configurations. These results allow us to conclude on the necessity to use heavy doping to obtain good electrical properties in SNWs.
4:15 PM Q9.4
Study of Individual Gallium Nitride Nanowires Grown by HVPE and CVD. Joonah Yoon1,2, Ilan Shalish2, George Seryogin2 and Venkatesh Narayanamurti2; 1Physics, Massachusetts Institute of Technology, Cambridge, Massachusetts; 2Division of Engineering and Applied Sciences, Harvard University, Cambridge, Massachusetts.
Gallium Nitride (GaN) is a technologically important wide bandgap semiconductor material. Further more, the current development in nanowire field can lead to a better performance and efficiency for existing devices as well as a new breed of devices. The focus of our work is to shed light on the unique properties of nanowires differing from the bulk characteristics. In our study, GaN nanowires are grown catalytically using either HVPE (Hydride Vapor Phase Epitaxy) or CVD (Chemical Vapor Deposition) method. After the growth electrical and optical characterizations are done on individual nanowires free of unwanted byproducts from the growth. Electron-beam lithography and photolithography are used to make metal contacts on individual GaN nanowires with various lengths and diameters. Resitivity is measured from room to cryogenic temperatures to illuminate the transport mechanism in GaN nanowires. Transconductance measurement is done in Field Effect Transistor (FET) device geometry to find carrier type, density, and mobility. The measurement shows our nanowires to be highly conductive and n-type regardless of the growth method. Challenges of lowering the unintentional doping level are discussed. An observed hysteretic behavior in transconductance measurement is reported and possible causes are discussed. Photoluminescence measurements are done on individual nanowires and results are discussed. TEM and STEM/EDS x-ray mappings are carried out to characterize the crystal structure and chemical makeup of nanowires.
4:30 PM Q9.5
Fabrication of Horizontal GaN Nanowires and Interconnected Nanobridges for Sensing Applications. Tania Henry1, Kyungkon Kim1, George Cui1, Jung Han1, Y. K. Song2, A. V. Nurmikko2 and Hong Tang3; 1Department of Electrical Engineering, Yale University, New Haven, Connecticut; 2Division of Engineering, Brown University, Providence, Rhode Island; 3Department of Physics, California Institue of Technology, Pasadena, California.
Nanowires are typically processed into nanodevices through solution dispersion and electrically accessed after contact metallization. This process presents inherent challenges in controlling the spatial ordering of final devices and avoiding contact resistance. We will demonstrate in this paper the feasibility of using crystallographic alignment of nanowire arrays, in conjunction with the concept of selective area growth, to achieve spatial alignment of nanowires with enhanced functionality. With the realization of horizontally aligned GaN nanowires, preliminary testing of their nanomechanical behavior has been conducted; high frequency resonance with a very high quality factor (Q~13,000) was observed. Crystallographic information forms the basis in designing and selecting an epitaxial system. The employment of multi-facetted GaN mesas enables a combinatorial survey of the tendency and selectivity of nanowire growth along the different planes. The Ni catalyst deposited rough GaN substrate was transferred to a CVD reactor. After the VLS growth, we observed hexagonally aligned horizontal nanowires from the c-plane and pyramidal m-plane of individual GaN mesas; TEM confirmed that the preferential nanowire growth direction is along the m-axis. In contemporary III-nitride research the use of epitaxial lateral overgrowth (ELO) is known to form well-defined stripes or hexagons bound by either prism or pyramidal m-planes, providing a natural and deterministic support for aligned nanowire growth. The integration of top-down lithographically-defined ELO patterning with bottom-up VLS synthesis enables precise control of the placement of nanowires and their spatial registry. The ELO mesas serve the dual purposes as 1) single-crystalline support for nanowire epitaxy, and 2) conducting electrodes that will facilitate carrier injection across the epitaxial, barrier-free interface. GaN ELO mesas in stripe and hexagon forms were prepared on insulating AlN epilayers. Horizontally aligned GaN nanowires, in comb-like arrays and/or hexagonal network of nanowires, interconnecting ELO mesas were observed with high spatial ordering. The density and diameter of the nanowires are correlated with the thickness of catalyst thin films (Ni). Preliminary results show that the GaN ELO mesas are conducting (resistivity~1e-4 ohm-cm) yet electrically insulated. Metal contacts defined by conventional photolithography are made to connect ELO mesas supporting nanowire arrays. Further measurement of contacted GaN nanowire bridges will be presented. The nanomechanical resonance of both connected GaN nanobridges (doubly clamped) and nanowire cantilever (singly clamped) was tested using optical scattering in a modified SEM setup. Preliminary data of both doubly and singly clamped nanowire reveal a very high quality factor (>13,000). Combination of ELO with nanowire synthesis is expected to provide a new paradigm for nano- electronic and electromechanical devices.
Chairs: Prabhakar Bandaru, Morinobu Endo, Ian Kinloch and Apparao Rao
Tuesday Evening, November 28, 2006
8:00 PM
Exhibition Hall D (Hynes)
Q10.1
Synthesis and Characterization of Pyrochlore-type Bi2Ti2O7 Nanotubes. Hongjun Zhou1, Tae-Jin Park1 and Stanislaus Wong1,2; 1Chemistry, SUNY Stony Brook, Stony Brook, New York; 2Brookhaven National Lab, Upton, New York.
Bismuth titanate (Bi2Ti2O7) nanotubes were successfully synthesized with an alumina template-based sol-gel technique. As-synthesized nanotubes are smooth and uniform with diameter ranging from 180-330 nm and lengths varying from 7 to 12 μm. Extensive characterization of as-prepared samples has been performed using X-ray diffraction (XRD), scanning electron microscopy (SEM), transmission electron microscopy (TEM), high-resolution TEM (HRTEM), energy-dispersive X-ray spectroscopy (EDS), and selected area electron diffraction (SAED).
Q10.2
Sonoelectrochemical Synthesis of Single Crystalline Semiconducting Copper Sulfide Nanowires. Krishna Veer Singh1, Alfredo Martinez2, Omar Faruk Yilmaz2 and Mihri Ozkan2; 1Chemical Engineering, UC Riverside, Riverside, California; 2Electrical Engineering, UC Riverside, Riverside, California.
Structural qualities of nanowires are the key for their adaptation to future electronic or optoelectronic applications. Single crystalline structure is a must to obtain repeatable device performance. Here we report synthesis of copper sulfide nanowires with “sonoelectrochemical” method to achieve single crystal nanowires with single stoichiometric composition (Cu 2.0:S 1.0). The use of sonoelectrochemical method increased the deposition rate by about 39% and 82% after the first one hour of deposition as compared to with stirring-assisted and regular electrochemical deposition, respectively. Observed increase in the bulk electrolyte temperature, high acoustic pressure and shock waves generated from the collapse of bubbles could explain improved mass transport and reaction rate, which results in the formation of single crystal monoclinic and orthorhombic phase nanowires. Their structural properties are confirmed with selected area diffraction analysis. While for the first time an orthorhombic phase copper sulfide nanowires are reported these nanowires are analyzed as p-type semiconductors. Nanowires in the range of 50-200nm in diameter were synthesized and optically characterized. A strong blue shift in optical absorption is observed for nanowires with smaller diameter due to size quantization effect. Excellent structural and repeatable electrical properties of these nanowires make them desirable for applications such as photovoltaics.
Q10.3
The Effect of Manganese on VLS Growth of Germanium Nanowires. Jessica L. Lensch, Eric R. Hemesath and Lincoln J. Lauhon; Department of Materials Science and Engineering, Northwestern University, Evanston, Illinois.
One-dimensional semiconducting materials have shown considerable promise in electronic and optical applications. In order to introduce magnetic functionality into semiconductor nanowires, we have attempted to incorporate high concentrations of Mn atoms into Ge nanowires grown by the vapor-liquid-solid (VLS) mechanism. Ge nanowires were synthesized by the VLS mechanism in a hot wall chemical vapor deposition system using Au catalyst particles and GeH4 gas. During Ge nanowire growth, the Mn source tricarbonyl methylcyclopentadienyl manganese (TCMn) was introduced into the reactor using H2 as a carrier gas to produce molar ratios of 0.001-0.02 TCMn:GeH4. Scanning electron microscopy (SEM) and transmission electron microscopy (TEM) were used to investigate the effects of varying TCMn concentrations on the morphology of the Ge nanowires. In the absence of TCMn, a high density of straight Ge nanowires could be grown, whereas low concentrations of TCMn resulted in branched nanowire structures. For molar ratios of 0.001-0.01 TCMn:GeH4, small branches grew from the primary nanowire, and the density increased with TCM concentration. Previously, TCMn was found to have a similar effect on InAs nanowire growth (May et al, Adv. Mater. 17, 598 (2005)), but for the present case the side branches were not as long nor as straight. At higher TCMn mole ratios (0.02), the catalyst became unstable, resulting in a disruption of unidirectional growth and highly curved and kinked nanowires. Branching was also observed. Analytical and high-resolution TEM studies are being undertaken to understand the role of Mn in the evolution of the Ge nanowire morphology, and to relate basic defect structures to the overall nanowire morphology. This investigation raises the more general question of how of high concentrations of impurities, intentional or otherwise, may affect the VLS growth of nanowires.
Q10.4
General Synthesis and Properties of Novel Transition Metal Silicide Nanowires. Andrew L Schmitt, Lei Zhu and Song Jin; Chemistry, University of Wisconsin-Madison, Madison, Wisconsin.
We report the chemical synthesis of free standing single-crystal nanowires of FeSi, the only transition metal Kondo insulator, and isostructural CoSi, a silicon based metallic structure. These compounds form the host structure for the ferromagnetic semiconductor FexCo1-xSi. Transition metal silicide nanowires are produced on silicon substrates covered with a thin layer of silicon oxide through decomposition of the single source organometallic precursors trans-Fe(SiCl3)2(CO)4 and Co(SiCl3)(CO)4, respectively, in a simple chemical vapor deposition process. Unlike typical vapor-liquid-solid nanowires growth, silicide nanowires form without the addition of metal catalysts, have no catalyst tips, and depend strongly on the surface employed. X-ray spectroscopy verifies the identity and the room temperature state of silicide nanowires, and room temperature and temperature dependent transport properties have been determined. Patterning using standard photolithographic techniques shows the possibility for selectively growing nanowires in place for future bottom-up assembly of devices. This general approach may lead to practical routes to other transition metal silicides and their alloys.
Q10.5
A Novel Growth Method of Single-Crystalline Bi Nanowires. Wooyoung Shim1, Jinhee Ham1, Kyoung-il Lee1, Joonyeon Chang2, Sukhee Han2, Wonyoung Jeoung2, Mark Johnson3 and Wooyoung Lee1; 1Department of Materials Science and Engineering, Yonsei University, Seoul, South Korea; 2Korea Institute of Science and Technology, Seoul, South Korea; 3Naval Research Laboratory, Washington, District of Columbia.
Semimetallic bismuth (Bi) has been extensively investigated over the last decade since it exhibits very intriguing transport properties due to their highly anisotropic Fermi surface, low carrier concentration, long carrier mean free path l, and small effective carrier mass m*. In particular, the great interest in Bi nanowires lies in the development of nanowire fabrication methods and the opportunity for exploring novel low-dimensional phenomena. In the present work, we report a novel method to grow high-quality, single-crystalline Bi nanowires and magneto-transport properties of an individual Bi nanowire. Bi thin films were grown on a thermally oxidized Si substrate in a radio frequency (rf)-sputtering system with a Bi target of 99.9%. The deposition of Bi was carried out in a vacuum chamber with a base pressure of 5.0 × 10-7 Torr. Rf power of 100 W and an Ar working pressure were utilized, yielding a growth rate of 27.5 Å/sec. For growth of the Bi nanowires, the sputtered-Bi thin films were transferred to a furnace for heat treatment at 270 °C for 10 hours. A combination of electron-beam lithography and a lift-off process was utilized to fabricate an individual Bi nanowire device. Interestingly, after heat treatment at 270 °C for 10 hours, uniform and straight Bi nanowires with high aspect ratios were found to be extruded from the surface of the as-grown films. The growth of Bi nanowires on the films is attributable to the relaxation of stress, originating from a thermal expansion mismatch between the film and the substrate. This mismatch is due to the large difference in the coefficient of thermal expansion of Bi (13.4 × 10-6/°C) and Si (3.0 × 10-6/°C). The grains of a Bi film grown at 100 W (27.5 Å/sec) and annealed at 270 °C was found to have preferred orientation, i.e., (003) and (006). Under compressive stress, the flow of Bi atoms to the preferred orientations is substantial so that the grains having the preferred orientation play a role as seeds for the growth of Bi nanowires. The correlations between the thickness and mean grain size of the Bi films deposited at 100W rf power has been investigated. The diameter of the Bi nanowires was found to decrease in proportion to the thickness and the grain size of the films, indicating that the diameter of the Bi nanowires can be controlled by manipulating the thickness of as-grown films. The largest magnetoresistance (MR) reported in the literatures for an individual 400-nm-diameter Bi nanowire, 2496 % at T = 110 K and 286 % at T = 300 K were obtained, indicating the longest mean free path (l) and relaxation time (τ). Our results demonstrate that high-quality, single-crystalline Bi nanowires can be grown by the stress-induced method, providing motivation for exploring novel physics, e.g., the magneto-transport and thermoelectric properties of single-crystalline Bi nanowires.
Q10.6
Abstract Withdrawn
Q10.7
Molecular Beam Epitaxy III-V Nanowires by Using Group III Metals as a Catalyst. Dance Spirkoska, Emanuele Uccelli, Max Bichler, Gerhard Abstreiter and Anna Fontcuberta i Morral; Walter Schottky Institut, Technische Universität München, Garching, Germany.
One-dimensional semiconductor nanowires (NWs) are potential building blocks for future nanoscale electronic and photonic devices. In particular, NWs based on the direct bandgap III-V materials system are very attractive due to their potential for new kind of LEDs, lasers, high power transistors and sensors. The usual method of producing NWs is through a metallic nanoparticle which catalyzes a solid NW from a vapor phase precursor molecule, in the so called catalytic vapor-liquid-solid (VLS) technique. Normally, the catalytic metal differs from the material systems of the grown NWs, i.e. Gold catalyst for Si and GaAs NWs. Here we present an alternative method for growing NWs through Molecular Beam Epitaxy (MBE) technique, where catalyst and NWs are from the same materials system. Different from typical III-V materials MBE synthesis, we start by depositing group III metals (In, Al, Ga) without As supply, and afterwards we proceed with opening the As flux. The formation of In, Al and Ga nanodroplets was studied separately, also as a function of substrate temperature (from 150°C to 500°C), substrate orientations ((001) and (110)), and the nominal thickness (0.5 to 5nm). Atomic Force Microscopy (AFM) and Scanning Electron Microscopy (SEM) measurements showed In- and Al-droplets formation for all the studied scenarios, different for shape and density. For the first time to our knowledge, we obtain also experimentally estimations of the surface energy between In (Al) droplets and GaAs surfaces, thanks to considerations from the contact angle of the droplets on the surface as well as from the substrate surface energy. As evidenced by SEM and AFM measurements, NWs were observed after supplying As immediately after the deposition of In droplets, that acted as metallic catalyst. Our measurements highlight the nucleation of vertically elongated nanostructures, which shape simply follows that of the original In droplets. Additionally, around each nanocrystals a large and low ring is also visible, forming together with the NW a “coral” type structure. Further separated deposition of In and As permit us to go inside this new growth method for vertically elongated wires, emphasizing the crystallization on InAs thanks to As diffusion inside the In droplet for NWs nucleation and In migration outside the droplet for ring nucleation.
Q10.8
Synthesis of Single-walled Carbon Nanotubes on Silicon Nanowires. Hideto Yoshida1, Tetsuya Uchiyama1, Jun Kikkawa2, Seiji Takeda1,3 and Yoshikazu Homma4,3; 1Graduate School of Science, Osaka University, Osaka, Japan; 2Research Institute for Ubiquitous Energy Devices, National Institute of Advanced Industrial Science and Technology (AIST), Osaka, Japan; 3CREST, JST, Saitama, Japan; 4Department of Physics, Tokyo University of Science, Tokyo, Japan.
Single-walled carbon nanotubes (SWNTs) and semiconductor nanowires have many extraordinary properties to impact future science and applications. Moreover, a junction structure of them is a potential nanomaterial as a Schottky diode and a light emitting diode. Therefore, we have synthesized SWNTs on silicon nanowires (SiNWs). We synthesized SiNWs by chemical vapor deposition as in the previous studies [J. Kikkawa et al., Appl. Phys. Lett. 86 (2005) 123109]. Au thin film about 1.0 nm thick was deposited on a Si substrate. The substrate was transfered to a quartz reaction tube set in a furnace, and then annealed at 773 K for 30 minutes. After that, monosilane gas, which was diluted to 1 % in Ar, was introduced into the quartz reaction tube at a flow rate of 1500 sccm and the pressure in the quartz reaction tube was kept at 98 kPa for 5 minutes. Consequently, SiNWs were synthesized on the substrate. SiNWs were dispersed in ethanol, and then the suspension containing SiNWs was deposited on a micro grid with a carbon supporting film. Next, for the SWNTs growth, Co thin film about 0.1 nm thick was deposited on the micro grid with SiNWs. The micro grid with SiNWs was placed in a quartz reaction tube set in a furnace and heated to 1073 K in Ar gas at a flow rate of 100 sccm with the pressure of about 5 kPa. After the temperature reached 1073 K, the Ar gas flow was replaced by ethanol gas as a carbon source at a flow rate of 150 sccm. The pressure in the quartz reaction tube was kept at about 5 kPa for 10 minutes. After growth, we observed an as-grown sample by means of a TEM. Many SWNTs are grown on SiNWs. In other words, heterostructures of SWNTs and SiNWs are synthesized. Among them, we could observed SWNTs with a catalyst nanoparticle at the root. The size relationship between catalyst nanoparticles and SWNTs has been investigated. We have also examined the role of surface oxide layer of SiNWs in SWNT growth. The detail of results will be presented at the congress.
Q10.9
A Novel Tin-rich Phase of the Sn-O System Observed at Layered Nanobelts. Marcelo Ornaghi Orlandi1, Matti Mäki-Jaskari2, Elson Longo3,4 and Edson Roberto Leite4; 1Department of Physics and Chemistry, São Paulo State University - UNESP, Ilha Solteira, São Paulo, Brazil; 2Institute of Physics, Tampere University of Technology, Tampere, Finland; 3Institute of Chemistry, São Paulo State University, Araraquara, São Paulo, Brazil; 4Department of Chemistry, Federal University of São Carlos, São Carlos, São Paulo, Brazil.
Tin oxide is an interesting material due to its unique properties, and the growth of tin oxide nanostructures was reported using several approaches. However, the stabilization of nonstoichiometric phases is still a challenge. In this work we present the growth of tin oxide nanobelts on a novel tin-rich phase. The synthesis was employed by a controlled carbothermal reduction process in a tube furnace, and the results show that this growth method should be a key for stabilization of this novel phase. The collected material was characterized by X-ray diffraction (XRD), Rietveld refinements, field emission scanning electron microscopy (FE-SEM), high resolution transmission electron microscopy (HRTEM) and energy disperssive X-ray spectroscopy (EDX). The results showed that the material is composed by nanobelts with layered growth, and by the Rietvetd refinements of X-ray data as found that the belts grow in a triclinic (P-1) structure with cell parameters a = 4.63 Å, b = 8.22 Å, c = 4.04 Å, α = 90.6°, β = 113.2° and γ = 96.1°. HRTEM results showed that epitaxial growth occurs between the layers and the belts have low concentration of defects. By semi-quantitative EDX analysis it was found that belts grow up in a tin-rich phase, with a stoichiometry close to Sn4O2. First principles theoretical calculations showed that tin-rich phases can be relatively stable in tin oxide crystals if layered growth occurs, which is in agreement with the experimental results. A growth mechanism of the nanobelts is proposed based on crystallization from vapor phase, in which the kinetics aspects of growth and the atmosphere of synthesis are fundamental to stabilize the layered and tin-rich nanobelts.
Q10.10
Mn-incorporated ZnSe and CdSe 1-dimensional Nanostructures. Jinyoung Lee, Dae Sung Kim, Hye Jin Chun, Ja Hee Kang, Shin Young Kim, Sang Won Yoon and Jeunghee Park; Material Chemistry, Korea University, Seoul, South Korea.
Novel Mn-incorporated ZnSe and CdSe 1-dimensionl nanostructures; straight nanowires, zigzagged nanobelts, and nanohooks, were first synthesized using chemical vapor deposition method. The Mn content reaches up to 40%. They all consisted of single-crystalline wurtzite structure for all Mn content. The structure has been thoroughly investigated by high-resolution transmission electron microscopy images as well as energy-dispersive X-ray fluorescence spectroscopy. The X-ray diffraction pattern confirms the formation of the wurtzite structure, even for 40% Mn incorporation. In the case of ZnSe 1-D nanostructures, the lattice constants are unusually reduced by the Mn doping. The Mn2+ emission at 2.1 eV, originating from the d-d (4T1→6A1) transition, confirms the effective paramagnetic Mn2+ doping at tetrahedral coordinate sites. These Mn-incorporated nanostructures exhibit a paramagnetic behavior with antiferromagnetic interactions.
Q10.11
Synthesis of Pure Boron Single-Wall Nanotubes. Dragos Ciuparu2,1, Mathieu Pinault1 and Lisa Pfefferle1; 1Department of Chemical Engineering, Yale University, New Haven, Connecticut; 2Department of Petroleum Processing and Petrochemistry, Petrol-Gaze University of Ploiesti, Ploiesti, Romania.
Boron nanostructures have recently attracted much attention because they are predicted to possess special properties that differentiate them from other one-dimensional nanomaterials. Several theoretical studies anticipated that boron nanotubes are stable [1] and, independent of structural helicity, have metallic conductivities exceeding that of carbon nanotubes [2] and show promise for exhibiting 1-D high temperature super conductivity when doped [3]. Several studies have focused on the synthesis of boron nanowires by differents methods producing both amorphous or cristalline materials [4-8]. Our group has developed a technique for nanotube synthesis based on nm pore silica templates and successfully applied it to grow uniform diameter carbon single-walled nanotubes [9]. Based on this technique we developed a process and became the first group to synthesize pure boron single-walled nanotubes with narrow diameter distribution controlled by the template pore diameter [10]. Here we report the synthesis of boron nanostructures including single wall boron nanotubes by reaction of BCl3 with H2 at temperatures ranging from 800 to 950C over different Mg-MCM-41 catalyst materials with parallel, uniform diameter cylindrical pores in the 1.5 to 3.5 nm range. Our results have shown that selectivity to a particular structure is a strong function of reaction conditions (T, BCl3/H2 ratio, catalyst template etc.). The boron nanotubes produced have diameters comparable to that of the catalytic template pore diameter suggesting the nanotube growth is physically constrained by the pores of the mesoporous molecular sieve. This is consistent with observations that template structure must be excellent for nanotubes to be produced. The morphology of the nanotubes was observed by transmission electron microscopy and the tubular structure was confirmed by the presence of the characteristic spectral features in the Raman breathing mode area. The chemical composition was confirmed by electron energy loss spectroscopy. Preliminary results on a new, ambient temperature sonication synthesis route will also be discussed. Acknowledgment. The authors thank Dr Robert Klie from the Center for Functional Materials at Brookhaven National Laboratory for the first EELS measurements and Professor Gary Haller and Dr. Sangyun Lim for providing the Mg-MCM-41 samples. References: [1] I. Boustani et al., Europhys. Lett., 39, (1997) 527 [2] I. Boustani et al., J. Chem. Phys., 110, (1999), 3176. [3] K. Soga et al., J. of Solid State Chemistry, 177, (2004), 498 [4] L. Cao et al., Appl. Phys. Lett., 80, (2002), 4226 [5] Y. Wang et al., Chem. Phys. Lett., 359, (2002), 273 [6] X. Meng et al., Chem. Phys. Lett., 370, (2003), 825 [7] Y. Zhang et al., S. Chem. Commun. (Cambridge, UK), (2002), 2806 [8] C. Otten et al., J. Am. Chem. Soc., 124, (2002), 4564. [9] D. Ciuparu et al., J. of Physical Chemistry B, 108, (2004), 503 [10] D. Ciuparu et al., J. of Physical Chemistry B, 108, (2004), 3967.
Q10.12
Synthesis, Optical, and Magnetic Properties of Cd1-xMnxS Nanowires. Dae Sung Kim, JinYoung Lee, Hye Jin Chun and Jeunghee Park; Korea University, Seoul, South Korea.
Cd1-xMnxS nanowires were synthesized using chemical vapor deposition method. In the range x ≤ 0.3, they all consist of single-crystalline wurtzite CdS structure with a [010] or [011] growth direction. X-ray diffraction pattern reveals the contraction of the lattice constants due to the incorporation of Mn. The Mn2+ emission at ~2.15 eV, originated from the d-d (4T1 → 6A1) transition, appears below 50-80 K. Its decay time is in the range of 0.55-1 ms, showing a decrease with increasing Mn content. Upon applying magnetic field (up to 7 T), the Mn2+ emission is suppressed and donor-acceptor pair emission becomes dominant, suggesting the energy transfer from the band electrons to the Mn2+ ions. In the range x ≥ 0.9, Cd-doped α-MnS nanowires are consisted of single-crystalline rock-salt MnS structures with the [111] growth direction. The X-ray diffraction pattern analysis indicates that 10 % Cd doping would expand the lattice constants by 0.3 %. As the content of Cd increases, the band edge emission band (2.9 eV) becomes broader in the lower energy region, and the Mn2+ emission band (1.6 eV), which emerged at temperatures below 150 K, decreases in intensity. The decay time of the 1.6 eV band decreases from 40 to 30 μs when the Cd doping is increased from 0 to 10 %. In contrast to the bulk (TN=150 K), he have been found to be paramagnetic with antiferromagnetic interactions.
Q10.13
HVPE and CVD Growth of InN Nanowires. George Seryogin, Ilan Shalish, Venkatesh Narayanamurti, Jiming Bao and Federico Capasso; Division of Engineering and Applied Sciences, Harvard University, Cambridge, Massachusetts.
InN is one of the key components of group III nitride compound system, enabling bandgap engineering to cover the visible and infrared portions of the spectrum. Despite extensive efforts, InN growth remains very challenging. We report on hydride vapor phase epitaxy (HVPE) and chemical vapor deposition (CVD) growth of InN nanowires on Si and sapphire substrates. The wires were grown in a home-built HVPE and CVD systems in temperatures ranging from 500 to 600 C. InN is particularly difficult to grow by HVPE, because HCl gas is present in the process and tends to etch the formed InN. To avoid the HCl reaction, we used indium chlorides formed in a separate process step. To facilitate the vapor-liquid-solid (VLS) mode of growth, the substrates were predeposited with nickel catalyst. In the CVD growth, indium vapor and ammonia reacted directly. Gold or nickel were used as the catalyst and the InN growth was confined to catalyst deposited features. Regardless of the growth method, the low growth temperature of InN results in a low efficiency of ammonia decomposition and almost an order of magnitude longer growth process time as compared to GaN. The use of nickel as a catalyst is commonly believed to also catalyze ammonia decomposition. However, it is not clear whether the nickel particles indeed catalyze the VLS growth mechanism or rather promote the growth through the local enhancement of ammonia decomposition, as nickel is not unambiguously detected at the tip of the wire, as is commonly the case with gold. Scanning electron microscopy (SEM) and electron dispersive spectroscopy (EDS) show the structure and identify the wire elements. X-ray diffraction (XRD) was used to verify the crystalline structure and quality. The thickness of the wire ranged from 200 to 400 nm for HVPE and 100 to 200 nm for CVD. Photoluminescence shows a broad structure of several features ranging from 1000 to 1600 nm.
Q10.14
Straightforward Synthesis of Single Crystalline Rutile TiO2 Nanowires. Syed Amin and Terry Xu; Department of Mechanical Engineering & Engineering Science, The University of North Carolina at Charlotte, Charlotte, North Carolina.
Single crystalline rutile TiO2 nanowires were synthesized by direct heating of Ti(II)O powders in argon at 800-900 °C and 1 atmosphere in a tube furnace. The nanowires were of 20-40nm in diameter and several micrometers in length. These nanowires were characterized by a range of methods including scanning electron microscopy (SEM), transmission electron microscopy (TEM), energy dispersive X-ray spectroscopy (EDX) and Raman spectroscopy. To explore the growth mechanisms, a series of control experiments including varying particle size, introducing catalytic materials, adjusting reaction pressure were performed. The possible growth mechanisms will be discussed. The optical properties of the nanowires were also studied and will be reported. The as-synthesized nanowires have potential applications in UV-protections, gases and humidity sensing, and others.
Q10.15
Solution Based Semiconductor Nanowires. Masaru K. Kuno, Darren Peterson and Vladimir Protasenko; Chemistry and Biochemistry, University of Notre Dame, Notre Dame, Indiana.
Recent efforts at developing solution routes to semiconductor nanowires (NWs) are presented. Specifically, the development of both straight and branched confined CdSe, CdTe and PbSe NWs is described. Using a low melting Au/Bi bimetallic nanoparticle “catalyst” straight and branched NWs with diameters below twice the bulk exciton Bohr radius of the corresponding material are made. Typical NW diameters range from 5 to 10 nm. Corresponding size distributions range from 15-30% and intrawire diameter variations are on the order of 5%. The NWs have lengths ranging from 1-10 microns in both straight and branched configurations. Their surfaces are passivated with organic ligands, enabling subsequent surface functionalization chemistries through ligand exchange. The solution linear absorption spectra of the wires exhibit confinement effects with blueshifts relative to the bulk band gap and structure at higher energies. Band edge emission is also seen in the case of CdSe although overall fluorescence quantum yields are on the order of 0.1%. In addition to the development of straight and branched NW morphologies, the feasibility of substitutionally doping them with impurities is discussed specifically within the context of CdSe and CdTe NWs.
Q10.16
High Yield Growth of Amorphous Free CdS Nanowires, Nanobelts, and Turf-like Structure at Low Temperatures over a Quartz Substrate. Juno Lawrance1, Lifeng Dong2, Josh Green2 and Jun Jiao2; 1Electrical and Computer Engineering, Portland State University, Portland, Oregon; 2Department Of Physics, Portland State University, Portland, Oregon.
CdS nanostructures, like nanowires and nanobelts, exhibiting novel electronic and optical properties owing to their unique structural, one-dimensionality,have drawn great attention in the last several years. Nanostructures are considered to be the critical component in a wide range of potential nanoscale device applications. Yet a procedure to fabricate them with both controllable results and in bulk quantities has to be developed in order to achieve their commercialization at a reduced cost. In this report, we introduce an improved vapor-liquid-solid method that can prepare high yield, high quality CdS nanowires and nanobelts in a turf-like configuration. The reported nanostructures were synthesized over a quartz substrate, rather than a tungsten or silicon substrate. During our procedures, a ceramic boat containing CdS powder was kept in the center of a quartz tube furnace that was pumped down to 300 torr at 630C. The quartz substrates, which were sputter-coated with about a 30nm thickness of gold, were placed downstream of the carrier gas at a distance of 20cm from the ceramic boat containing CdS powder. The flow of argon was maintained at 90sccm. In order to further increase the yield, we placed the gold-coated substrates in a ceramic boat, which was partially covered with a glass slide. Only a small opening was provided to allow the CdS vapor to enter the arrangement. The CdS vapor flow was partially blocked by the compartments at a specific distance and at a consistent temperature over a relatively long period of time. This allowed the CdS vapor to deposit uniformly over the substrate. The CdS structure topography was characterized by an FEI Sirion XL30 field emission scanning electron microscope (FESEM) equipped with an Oxford Inca EDX system and a FEI Tecnai F-20 high-resolution field emission transmission electron microscope (FETEM). Along with the high yield growth of CdS nano wires and belts, a turf-like structure was observed. The TEM characterization suggests that these nanostructures are single crystalline and amorphous free. In addition, a relatively low growth temperature of 450C was investigated; similar structural morphologies of CdS nanostructures were formed, but the uniformity of the nanostructures were not superior to those obtained at 630C. Optimizations of other parameters at 450C, as well as lower temperatures, are under investigation. A study of the electronic and optical properties, including electron field emission and photoemission inrelation to the different morphologies, will also be presented.
Q10.17
Growth of Gallium Nitride Nanowires and Nanospirals. Goutam Koley and Zhihua Cai; Electrical Engineering, University of South Carolina, Columbia, South Carolina.
GaN nanostructure synthesis has been done in a quartz tube furnace using ammonia and liquid Ga as precursors, and hydrogen as the carrier gas. Thin film of Ni, which turned into nanoparticles upon annealing, has been used as the catalyst layer, facilitating vapor-liquid-solid growth of the nanostructures. In the growth process, two types of structures were formed - straight nanowires and nanospirals. Growth using uniform distribution of catalyst over the entire surface always surface always resulted in growth of straight nanowires. However, growth performed on catalyst patterned surface resulted in growth of nanospirals. The growth of the spirals has also been found to be dependent on the distance from the gaseous species inlet, and on pressure and temperature conditions. The diameter of the nanowires varies from 10 - 50 nm, while for spirals the cross-sectional diameters are usually in the range of 100 nm - 1 micron, and spiral diameters in the range of several microns. Using the present growth system and gas flow set-up, it was possible to synthesize ultra-long nanowires and spirals, with overall lengths exceeding 70 microns. The nanowires were found to have a circular cross-section, while the cross-section of spirals was found to be roughly hexagonal. Some of the spirals also changed into straight nanowires, grown at right angles to the original growth direction, as their cross-sectional diameter reduced with length. While more investigation is required to establish their actual structure, we speculate that the nanowires are grown along the c-direction [0001], while the spirals are grown in one of the non-polar directions. The formation of spirals may be related to the polarization properties of GaN, similar to those predicted for ZnO nanosprings and nanoribbons. Detailed investigations are being carried out at present to better understand the growth mechanisms, and structural and electrical properties of these nanostructures.
Q10.18
Ambient Template-directed Synthesis of Single-crystalline Alkaline Earth Metal Fluoride Nanowires. Fen Zhang1, Yuanbing Mao1 and Stanislaus S. Wong1,2; 1Chemistry, State University of New York at Stony Brook, Stony Brook, New York; 2Materials and Chemical Sciences, Brookhaven National Laboratory, Upton, New York.
One-dimensional nanostructures, such as nanowires, have attracted considerable attention because of their singular properties associated with quantum confinement and low-dimensionality, as well as their potential applications as building blocks for the assembly of nanoscale electronic, optoelectronic, and sensing devices. Fluorides have been widely used in optics and components in semiconductor-on-insulator structures. In addition, fluorides doped with rare-earth ions have also been reported to display unique luminescence properties. It is reasonable to expect that nanoscale fluorides will play an important role in technological applications including as high-density optical storage devices, nanosensors, and color displays. Herein, we demonstrate that a family of single-crystalline alkaline earth metal fluoride nanowires, as well as their rare-earth ion doped analogues, of varying controllable sizes can be successfully prepared using a modified template-directed method at ambient room temperature conditions, without the use of either sophisticated experimental setups or high-temperature annealing. Moreover, the diameters of the as-fabricated nanowires could be controlled by choosing commercially available, track-etch polycarbonate membranes with predictable pore sizes. Resulting nanowires have been extensively characterized by microscopy and spectroscopy data. The luminescent properties of lanthanide-doped binary fluoride nanowires imply the possible incorporation into nanoscale devices via a more thorough investigation of their optical and optoelectronic properties.
Q10.19
Growth of Si Nanowires and Nanowire Arrays by Molecular Beam Epitaxy. Peter Werner, Alexey P Milenin, Nikolai D Zakharov and Ulrich Goesele; MPI of Microstructure Physics, Halle (Saale), Germany.
We demonstrate the growth of silicon nanowires (NWs) by molecular-beam epitaxy (MBE) under ultra-high vacuum conditions (UHV). This growth process has similarities to the conventional vapor-liquid-solid process (VLS mechanism), but also has significant differences. The main differences concern i) surface diffusion processes of adatoms and ii) a strain-driven mechanism of the NW formation. Small gold droplets are applied as seeds for the growth of the NWs, and they determine NW diameters. For a technological application there is the necessity to tune size and position of NWs including the formation of large-area NW pattern on a substrate/wafer surface. The geometry of the pattern having 2D-structures in the sub-micrometer range influences locally the surface diffusion and the lattice strain. In our case we combined e-beam lithography, plasma etching, and RCA surface cleaning to get different periodically modified substrate surfaces. On them NW arrays were generated by MBE under UHV conditions. Following surface structures have been analyzed for the NW growth: holes, pillars and trenches.
Q10.20
Vertically Aligned Growth of Sub-30 nm ZnO Nanowires Using Single-ion Nanolithography. Daniel Lorscheitter Baptista1,2, Sharvari H. Dalal2, Ricardo M. Papaleo3, Fernando C. Zawislak1, Paulo F. P. Fichtner4, Ken B. K. Teo2 and Willian I. Milne2; 1Instituto de Física, Universidade Federal do Rio Grande do Sul, Porto Alegre, Brazil; 2Engineering Department, University of Cambridge, Cambridge, United Kingdom; 3Faculdade de Física, Pontifícia Universidade Católica do Rio Grande do Sul, Porto Alegre, Brazil; 4Departamento de Metalurgia, Universidade Federal do Rio Grande do Sul, Porto Alegre, Brazil.
We report on large area growth of thin and aligned ZnO nanowires through the use of single ion nanolithography as a sub-30 nm patterning tool. The ZnO nanowires are grown in a tube furnace by evaporation of a ZnO:graphite mixture onto sapphire substrates patterned with Au catalyst. The substrate patterning was performed using random ion beam irradiation with Au ions at 20 MeV with a fluence range from 10^8 to 10^10 cm^-1. A thin PMMA film of 50 nm was spin coated on the sapphire substrates and then subjected to the ion irradiation. Although the substrate areas were approximately 1 cm^2, the irradiation covered an area of 5x5 cm in a fast scan of approximately 2 seconds. The low fluence range isolates the ion-sample interactions, reducing the probability of spatial superpositions. Each ion interacts individually with the PMMA film forming a cylindrical track of latent chemical damage, which is then developed using MBIK:IPA solution. After development and the evaporation of the Au catalyst, the PMMA film is removed by lift-off and small gold islands of 25 nm remain on the sapphire substrates. The irradiation fluence defines the mean distance between each island and therefore their density. Isolated ZnO wires with diameters of 25-28 nm were produced by the above method, which is not limited by the interdependence of pattern diameter and density typical of self assembling techniques and is much faster than e-beam lithography to draw sub-100 nm features over large areas.
Q10.21
Fabrication of Small MgO Nanowires by Pulsed Laser Deposition. Takeshi Yanagida, Kazuki Nagashima, Hidekazu Tanaka and Tomoji Kawai; ISIR-Sanken, Osaka University, Osaka, Japan.
One-dimensional nanowires have gained attention recently because of their potential applications in diverse areas including opto-electronics, quantum computing, sensors, memories and others. One of desired functionalities in developing these nanowire applications is the control of the diameter size, allowing the fabrication of smaller nanowire-based devices. Especially the ranges less than 10nm are fascinating not only in terms of the miniaturization but also in terms of applying the sensor devices to the detection of nanometer biomolecules. However, there are very few reports concerning with the size reduction of oxide nanowires in such small ranges. For instance, the minimum average diameter sizes of MgO nanowires synthesized by CVD technique in literature were reported to be 10-20nm. Here we demonstrate the first synthesis of ultra small MgO nanowires, whose diameters are less than 10 nm, by laser MBE technique. The MgO nanowires were synthesized on MgO (001) single crystal by laser MBE via gold catalysts. SEM and HRTEM-EDX analysis were performed to evaluate the morphology, the crystallinity and the composition of the nanowires. By controlling precisely the amount of gold catalysts and the ambient atmospheres, the ultra small MgO nanowires, in which the size was more less 3-5nm, were achieved. In addition, the epitaxy of the nanowires and the VLS mechanism for the wire growth were identified by HRTEM-EDX analysis.
Q10.22
Vanadium Dioxides Nanorods: From B Phase to M Phase. Liqiang Mai1,2, Wangli Guo1, Tao Hu1, Wen Chen1, Wei Jin1, Bo Hu1 and Ch.V.Subba Reddy1; 1Institute of Materials Science and Engineering, Wuhan University of Technology, Wuhan, Hubei, China; 2School of Materials Science and Engineering, Georgia Institute of Technology, Atlanta, Georgia.
Vanadium dioxides have attracted much attention because they can be widely used in temperature sensing devices, optical switching devices, energy-conserving coating for windows and so on. VO2 (B) nanords were prepared by rheological phase reaction followed by self-assembling process, which were treated by H2O2 and CTAB solution and VO2 (M) nanorods were attained. The attained VO2 (B) nanords with length of 1~2μm, nanorod diameter of 30~60nm, and the diameter of the nanorod bundles of 100~300nm are formed by“face landing” self-assembling process. VO2 (M) nanorods are attained through phase transfer process of VO2(B)→VO2 (R)→VO2(M). The 10th discharge capacity of VO2(B) nanorods before and after Mo doping is 224 and 258 mAh/g, respectively, and the corresponding capacity conservation rate is 88.2% and 93.5%, respectively. Mo doping reduces the phase separation during deep discharge and supports structure, resulting in improved electrochemical performance. For VO2(M) nanorods, the transition temperature is 65 degree and the hysteresis loop width is 8 degree . The active energy of low temperature semiconducting phase is 0.2eV, which indicates that its Fermi energy level situates on the middle energy level of the forbidden-band. After Mo doping, Tc decreases to 59 degree. Mo doping reduces forbidden band’s width of VO2(M) nanorods as the donor and change its electrical property. Acknowledgments This work was supported by the National Natural Science Foundation of China (Grant No. 50372046), the Key Project of Chinese Ministry of Education (Grant No. 104207), the Foundation for Innovation Research Team of Hubei Province (No. 2005ABC004), Nippon Sheet Glass Foundation for Materials Science and Engineering (2005), the Teaching and Research Award Program for Outstanding Young Professors in Higher Education Institute, MOE, P. R. China and the Wuhan Youth Chenguang Project (2006).
Q10.23
Selective Patterning of Carbon Nanotube added V_2O_5 Nanowire Channel via BOE Treated PDMS Stamping Technique. Yong-Kwan Kim1, So Jeong Park2, Gyu Tae Kim2 and Jeong Sook Ha1; 1Department of Chemical and Biological Engineering, Korea University, Seoul, South Korea; 2Department of Electrical Engineering, Korea University, Seoul, South Korea.
Poly dimethylsilioxane (PDMS) has been widely used in various fabrication processes including soft lithography and MEMS, owing to its useful properties such as hydrophobicity, contamination resistance, and long-term endurance. The hydrophobicity and inertness of PDMS guaranteed the pattered thin film as an efficient passivation layer over reactive ion etching and atomic layer deposition process in a similar way as octadecyltrichlorosilane self-assembled monolayers (SAMs). In this study, we successfully fabricated carbon nanotube (CNT) added V_2O_5 nanowire channel patterns on the desired site of SiO_2 substrate by using transferred PDMS as an passivation layer. Various line and checker-board patterns of PDMS with a width of a few micrometers to 300 nm and a thickness of 30 nm could be transferred by stamping the BOE (buffered oxide etcher)-treated PDMS stamp on SiO_2 substrates for 5 minutes. In the similar way, we also manufactured the PDMS patterns on the -NH2 terminated SAM by printing of BOE treated PDMS stamp on 3-aminopropyl- triethoxysilane (APS) surfaces. This patterned surface of PDMS/APS was immersed into aqueous V_2O_5 nanowire solution for the selective adsorption of the nanowires onto APS area owing to attractive interaction between negatively charged V_2O_5 nanowires and the -NH_2 terminated APS SAM. Successively, the substrate was immersed into CNT solution for the additional adsorption of CNTs onto V_2O_5 patterns due to mutual attractive polar interaction between CNTs and V_2O_5 nanowires. As a result, we could make patterns of CNT added V_2O_5 nanowire channels on APS using transferred PDMS pattern as passivation layer. Electrical measurements showed that the conductivity of the CNT/V_2O_5 hybrid channel increased noticeably compared to that of the original V_2O_5 channel since the co-adsorbed CNTs worked as current bridges in the V_2O_5 nanowire channel. We suggest that the patterned CNT/V_2O_5 nanowire hybrid channel would enhance the sensitivity of V_2O_5 nanowires, which have relatively high electrical resistance, as functional devices such as chemical sensors.
Q10.24
Growth Study of CVD Au Catalyzed Silicon Nanowires and Integration in an Organized Nanoporous Alumina Template. Pascal Gentile1, Thomas David1, Denis Buttard1, Martien Den Hertog1, Jean-Luc Rouvière1, Pierre Ferret2, Florian Dhalluin3 and Thierry Baron3; 1DRFMC/SP2M, CEA, Grenoble, France; 2DRT/Leti/DOPT, CEA, Grenoble, France; 3LTM, CNRS, Grenoble, France.
The growth of semi-conductors nanowires is now of a great interest for nanoelectronic technology, which leads to new electrical or optical properties. The case of silicon nanowires is especially interesting since the silicon is a key material for the nanoelectronic technology. Silicon nanowires (SiNWs) are grown on silicon (111) and (100) substrates in a Low Pressure Chemical Vapor Deposition (LPCVD) reactor, using vapor-liquid-solid (VLS) mechanism, in which the nanowire grows from a gold catalyst droplet. Silane is used as a reactive gas diluted in hydrogen. Catalysts obtained by gold film dewetting present large size dispersion whereas direct deposition from a colloidal solution offers the possibility to grow NWs with uniform diameter. We present a morphological (growth orientation, crystalline defect, faceting), comparison of the NW growth conditions (temperature, SiH4 flux, total pressure) for the two kinds of catalyst by using SEM, STM, TEM characterization techniques. For the colloids case, the efficiency of the catalyst is strongly dependent on the growth conditions, and also varies with their diameter (5 - 200 nm). The surface atomic structure of individual NW will be addressed by STM. For the realization of devices like sensors nanowires must be manufactured with several technological steps like encapsulation and planarization owing to their low mechanical behavior. This is difficult to realize after the growth of nanowires on a bulk substrate. We propose another way which consists to grow nanowires directly in a nanoporous cylindrical template made of porous alumina. Porous alumina layers are composed of uniform porous structure with cylindrical pores parallel to the current lines. They are obtained by direct electrochemical process under a constant voltage value in an aqueous acid solution. We used this nanoholes array as substrate for the growth of crystalline SiNWS matrix. We show the realization of the anodic oxidation of aluminium and we adjust parameters like the cell size, the pores diameter, the layer thickness and the lateral distribution. The silicon nanowires growth with chemical vapor deposition conditions is then performed in the particular vapor-liquid-solid (VLS) mode.
Q10.25
Laser Interference Lithography Tailored for Ordered Arranged ZnO Nanowire Arrays. Dongsik Kim1, Hongjin Fan1, Ran Ji1, Roland Scholz1, Kornelius Nielsch1, Ulrich Goesele1 and Margit Zacharias2,1; 1Dep. II, MPI of Microstructure Physics, Halle, Germany; 2FWIM, FZ Rossendorf, Dresden, Germany.
The controlled growth of one-dimensional semiconductor nanowires (NWs) has been in the focus of attention not only for the reason of understanding the fundamental growth mechanism but also due to possible applications in nanoscale electronics. Especially, the ZnO material, as important functional oxide semiconductor, is considered as useful elements in a wide variety of application including nanolasers, light-emitting devices, sensors, and solar cells. So far, concerning the patterned growth of ZnO NWs, little attention was paid to the reproducibility and quality of the nanostructured material. Based on our recent review [1] various growth processes, especially the vapor-liquid-solid process offers an opportunity for the control of spatial positioning of nanowires. Strategies for position-controlled and nano-patterned growth of nanowire arrays are reviewed and demonstrated by selected examples based on ZnO nanowires as well as discussed in terms of larger scale realization and future prospects. We demonstrate the regular arrangement of gold nanodots (NDs) on macroscopic large areas using laser interference lithography (LIL) which is a reliable, cost-effective, and high-throughput method. Base on this, we will also demonstrate ordered arrays of vertically aligned ZnO NW, synthesized via a chemical vapor transport and condensation process. Individual ZnO NWs are grown on each Au ND (i.e., one-to-one synthesis). ZnO NW arrays are established over cm2 area with a density of 9 nanowires/micometer2. The highly symmetrical arrangement of the vertically aligned ZnO NWs with periods of 230 nm can be clearly distinguished. Structural and spatially resolved optical characterization of the NW arrays will be demonstrated. [1] H.J. Fan, P. Werner, M. Zacharias, Semiconductor nanowires: from self organization to patterned growth, Small 2 (2006) 700, (invited review).
Q10.26
A Novel Technique to Synthesize Alumina Nanofibers/Nanotubes. Anup Pancholi1, Copeland David Kell2 and Valeria Gabriela Stoleru1; 1Materials Science and Engineering, University of Delaware, Newark, Delaware; 2Chemical Engineering, University of Delaware, Newark, Delaware.
Nanoporous alumina membranes have been widely used in recent years, for example, as templates in nanowire fabrication through electrodeposition. Extensive research has been carried out to optimize the growth process of nanoporous alumina membranes by anodization of thin Al films. Very uniformly ordered nanopores have been achieved by using oxalic, phosphoric, and sulfuric acid solutions as electrolytes. As reported in the literature, each electrolyte solution has characteristic optimum anodization conditions (i.e., time and voltage) that provide uniform and high-quality nanopores. By slightly changing the anodization conditions, we observed the formation of dense alumina nanofibers/nanotubes. Alumina nanofibers/nanotubes are of great interest in major areas applications, such as high temperature insulating materials, catalysts or absorbents for scavenging noble or heavy metals, and as reinforcing composites when embedded in either ceramic or metallic matrices. The technological potential of alumina nanofibers/nanotubes is attributed to properties such as high strength, corrosion resistance, chemical stability, low thermal conductivity, and good electrical insulation. The fact that the properties of such oxide fibers depend on the size and homogeneity of the fibrous structure makes this material even more interesting for both fundamental studies and practical applications, especially when the fibers are scaled down to the nanosize regime. In this article we report a novel method to synthesize dense alumina nanofibers/nanotubes with diameters in the 20-50 nanometers range, by slightly changing the well-established anodization parameters which have been previously used for synthesizing highly uniform nanoporous templates. A high yield of alumina nanofibers has been obtained by a two-step anodization process, with an intermediate step of etching. For phosphoric acid solution (10% w/w) electrolyte, the anodization voltage has been varied from 80 V to 160 V in steps of 20 V, while the anodization time was 30 to 60 minutes. For oxalic acid solution (0.3M) electrolyte, the anodization voltage has been varied from 20 V to 80 V in steps of 20 V, while the anodization time was 4 to 7 hours. The intermediate etching was conducted for 30 minutes when the phosphoric acid solution was used, and for 4 hours when the electrolyte was a solution of oxalic acid. A mixture of phosphoric acid (6% w/w) and chromic acid (1.8% w/w) was used for etching, and the process temperature was 60 °C. We studied the effects of the anodization voltage, time, and the electrolyte on the formation and properties of the nanofibers/nanotubes. Characterization of alumina nanofibers/nanotubes was performed by using scanning electron microscopy (SEM) and transmission electron microscopy (TEM). The composition of nanofibers/nanotubes was verified by using energy dispersive analysis of X-ray (EDAX). A possible mechanism for the nucleation and growth of nanofibers/nanotubes is provided.
Q10.27
Synthesis and Characterization of Branched Nanowire Heterostructures. Yeonwoong Jung, Dong-Kyun Ko and Ritesh Agarwal; Materials Science & Engineering, University of Pennsylvania, Philadelphia, Pennsylvania.
Low dimensional semiconductor nanowires are emerging as attractive building blocks for the assembly of electronic and optoelectronic device systems. Fabrication of such nanowire-based hierarchical systems is only meaningful if control over the morphology, position and composition of nanowires is achieved. Single crystalline CdS and ZnS nanowires have demonstrated excellent optical properties, and have been configured as optical and electrical injection lasers, waveguides, and avalanche photodetectors. However, more complex morphologies and compositional modulation of these materials are required to assemble nanowire-based device systems with unique and diverse functionalities. In the present study, we report controlled synthesis of single-crystalline branched nanowire heterostructures, where the backbones and branches are assembled with ZnS and CdS respectively. The nanowires are grown by the metalorganic chemical vapor deposition (MOCVD) process using single molecular precursors via the vapor-liquid-solid (VLS) process. Sequential growth of branch and backbone nanowires with control over the diameter and compositions is enabled via sequential seeding of gold nanocluster catalysts. The structure and the composition of the branched nanowire heterostructures are characterized by various electron microscopy techniques. Elemental mapping and x-ray energy dispersive spectroscope (EDS) in scanning transmission electron microscopy (STEM) study confirms that branched nanowire heterostructures are synthesized as intended, with ZnS as backbone and CdS as branch. High resolution TEM (HRTEM) study indicates that the growth of heterostructure branches occurs epitaxially from the backbone while maintaining single crystalline structure with no structural defects at the junction. Branched nanowire homostructures, where the branches and backbones are compositionally uniform are also synthesized. The structures of ZnS and CdS branched homostructures are studied and compared to the branched heterostructures. Investigation of the novel optical and electrical properties of these unique branched structures will be presented. Our attempts of using branched nanowire heterostructures for three-dimensional electronic and photonic circuits with control over the generation and propagation of carriers and photons will also be discussed.
Q10.28
III-V Nanowires Grown by Metal-organic Chemical Vapour Deposition for Optoelectronic Applications. Yong Kim1, Hannah J Joyce1, Qiang Gao1, Hoe H Tan1, Chennupati Jagadish1, Mohanchand Paladugu2 and Jin Zou2; 1Electronic Materials Enginering, The Australian National University, Canberra, Australian Capital Territory, Australia; 2Centre for Microscopy and Microanalysis, The University of Queensland, Brisbane, Queensland, Australia.
The anisotropic growth of silicon whiskers via the vapour-liquid-solid growth mechanism was discovered by Wagner and Ellis [1]. Recently, axial and radial heterostructure nanowires, scaled down in dimensions from these micrometer-sized whiskers, have been proposed as nano-building blocks for future optoelectronic devices [2]. Despite the recent successes particularly in III-V nanowires, there are many remaining issues that needed to be explored. In this talk, the optical and structural properties of binary and ternary III-V nanowires including GaAs, InGaAs, AlGaAs, and GaAsSb nanowires grown by metal-organic chemical vapor deposition will be presented. Au colloidal nanoparticles (10~50 nm diameter) are employed to catalyze nanowire growth. The growth behavior of GaAs nanowires depending on the growth conditions, substrate orientation, and nanoparticle size will be discussed, in particular GaAs nanowires on (111)B substrate consisting of several truncated triangular segments. A 2-step growth technique is also proposed for achieving nanowires with minimal tapering. For device applications, complex radial/axial heterostructures involving ternary or even quaternary nanowires are required. The large difference of diffusion lengths between constituent reaction species on growing surface could result in compositional non-uniformity along nanowires. Also this effect results in nanowire height and composition instability depending nanowire density. This issue will be discussed with the emphasis on the case of InGaAs nanowires where there is a large difference of surface diffusion lengths between In and Ga reaction species [3]. In addition, various interesting ideas such as GaAs/AlGaAs core/shell structures, quantum shell intermixing, GaAsSb nanowires, and kinked InAs/GaAs heterostructure nanowires and their possible applications for optoelectronic devices will be presented. The Australian Research Council is acknowledged for its financial support. [1] R. S. Wagner and W. C. Ellis, Appl. Phys. Lett. 4 (1964), 89 [2] C. M. Lieber, MRS Bull. 28 (2003), 486 [3] Y. Kim et al., Nano Lett., 6 (2006), 599.
Q10.29
Metallothermic Synthesis of Nanowires and Related Nanostructures of Al2O3 and MgO. Boris Bokhonov, Mikhail Korchagin and Yurii Yukhin; Institute of solid state chemistry SB RAS, Novosibirsk, Russian Federation.
Aluminum and magnesium oxide (Al2O3 and MgO) nanowires, nanotubes and nanorods have been generated by heating mixture of metal oxide and Al or Mg powders under O2 according to the reactions: Me2O3+2AlO2=2Me+Al2O3 or MeO3+3Mg=Me+3MgO (Me2O3=Fe2O3,Bi2O3. MeO=MoO3, WO3). The nanowires and nanotube-like of Al2O3 and MgO ranging from 20 to 100 nm in diameter and up to 1-2 µm in length. Single metal or oxide particles are often found attached to nanowire tips. Because the reaction between oxides and Al (Mg) is extremely exothermic we proposed that a vapor-liquid- solid (V-L-S) mechanism appears to be responsible for the aluminum and magnesium oxide nanowires growth.
Q10.30
Controlled Growth of GaN Nanowires by means of Chemical Vapor Deposition on Patterned Substrates. Zhen Wu1, Yung Joon Jung2 and Latika Menon1; 1Physics, Northeastern University, Boston, Massachusetts; 2Mechanical Engineering, Northeastern University, Boston, Massachusetts.
Semiconductor nanowires are expected to have several applications in advanced nanoelectronic and spintronic devices. For example, GaN is a wide bandgap semiconductor material and is a promising candidate for optoelectronic devices in the blue and ultraviolet region. They are also model systems to study fundamental properties in 1-d nanomaterials. We have developed nanofabrication methods utilizing nanoporous alumina membranes for the growth of GaN nanowires using chemical vapor deposition. The main function of the nanoporous alumina template is in the patterning of nanosized high density Ni nanodots which act as catalyst for the controlled growth of GaN nanowires on a substrate. We will report the optical and electrical properties of these nanowires as a function of wire dimension and investigate their potential applications in nanodevices.
Q10.31
Large-Scale, Multifunctional, 2D and 3D Assemblies of Catalytic Nanowires. Z. Ryan Tian1,2,3, Wenjun Dong1, Michelle McDonald1, Michael Jackson1, Andrew Cogbill1 and Carmen Padilla2; 1Chemistry/Biochemistry, University of Arkansas, Fayetteville, Arkansas; 2Cell and Molecular Biology, University of Arkansas, Fayetteville, Arkansas; 3Microelectronics and Photonics, University of Arkansas, Fayetteville, Arkansas.
One-dimensional nanomaterials including carbon nanotubes and ceramic (V2O5, MnOx) nanowires have been recently fabricated into paper-like free standing membranes (FSM) for important applications. However, making large-scale, multifunctional FSM and the FSM-based 3D macroscopic devices purely from long inorganic nanowires is challenging in many nanomaterials systems. Here we report the synthesis of TiO2-based catalytic long nanowires for direct fabrications of the FSM and the FSM-based 3D macrodevices, which is proved by SEM and TEM as well as XRD studies. The new forms of the assembled catalytic nanowires are among the first to show unusual potentials in important applications including photocatalysis, write-erase-rewrite, microfiltration, tissue regeneration, controlled release, sensing, etc.
Q10.32
Position-controlled Formation of Quantum Dots and Nanowires at Iron Silicide Nanowires. Zhenyu Liu1, Judith C. Yang1 and Kyeongjae Cho2; 1Department of Materials Sci. & Eng., University of Pittsburgh, Pittsburgh, Pennsylvania; 2Department of Mechanical Engineering, Stanford University, Stanford, California.
We demonstrate that organized perpendicular iron silicide nanowire patterns can be synthesized in a self-assembly manner, further T junctions and other complex junctions can be synthesized by fusing individual iron silicide nanowires inclined toward each other. The structure of the perpendicular patterns and different junctions may allow construction of three- or multiterminal nanowire devices directly on Si-based readout circuits through controlled nanowire formation. Some child quantum dots and nanowires can be formed from the parent nanowires after energy dispersive X-ray detector (EDX) irradiation, the diameter and length of the child nanowires are related to the spot size and irradiation time. That implies that it is possible to construct some more complex structure by controllable fashion. By this well controlled nanowire growth, it will enable these nanostructures to efficiently and economically incorporate into nanodevices.
Q10.33
Ultranarrow Bi2S3 and Sb2S3 Nanowires. Reihaneh Malakooti1, Ludovico Cademartiri1, Srebri Petrov1, Andrea Migliori2 and Geoffrey A. Ozin1; 1Department of Chemistry, University of Toronto, Toronto, Ontario, Canada; 2CNR-IMM, Bologna, Italy.
Ultranarrow (1.5-2.0-nm-wide and several microns long) Bi2S3 and Sb2S3 nanowires have been synthesized from solution with a facile reaction scheme, involving a fast injection in a coordinating solvent at low temperatures (130-170 degrees). Crystallinity has been demonstrated via XRD, HRTEM and SAED investigations. The extreme thinness of such nanowires allowed the observation of their spontaneous organization in nanofibrous networks. Such nanowires are interesting for thermoelectric applications where thinner nanowires are expected to be more perfomant, especially when considering that Bi and Sb chalcogenides have high thermoelectric figures of merit.
Q10.34
Abstract Withdrawn
Q10.35
Coaxial Homostructure Gallium Nitride Nanowires. Benjamin Jacobs1, Virginia M. Ayres1, Martin A. Crimp2, Jiaming Zhang2, MaoQi He3 and Joshua B. Halpern3; 1Electrical and Computer Engineering, Michigan State University, East Lansing, Michigan; 2Chemical Engineering & Materials Science, Michigan State University, East Lansing, Michigan; 3Department of Chemistry, Howard University, Washington, District of Columbia.
Development of methods for control of one-dimensional growth of semiconducting nanowires depends in part upon recognizing and taking advantage of inherent energy transfer mechanisms within these nanosystems. Here we present recent research on the growth and analysis of Gallium Nitride nanowires with a self-organized two-phase coaxial crystalline homostructure. Our work complements previously reported work by another group of a self-organized two-phase coaxial crystalline Gallium Nitride-Aluminum Gallium Nitride heterostructure. Detailed high resolution transmission electron microscopy (HRTEM) studies were conducted, including cross sectional analysis of the two-phase homostructure junction made possible using focused ion beam (FIB) techniques. Analytical electron energy loss spectroscopy (EELS) and energy dispersive x-ray spectroscopy (EDS) were used to determine the stoichiometry of the coaxial homostructure. Coaxial nanowires may represent a new class of electron waveguide nanostructures with important applications in quantum transport and in electron and photon confinement.
Q10.36
Wafer Scale Fabrication of E-beam Written Suspended TiW Nanowires. Lauge Gammelgaard, Rudy J Wojtecki and Peter Bøggild; Department of Micro- and Nanotechnology, Technical University of Denmark, Lyngby, Denmark.
Nanowires of various materials have shown a great potential as the sensing part in many different types of sensors, spanning as different areas as bio, chemical, electrical and mechanical sensors [1]. Therefore a large effort to make fundamental studies of nanowires and integrate these in microsystems is ongoing. In this work, we have realized wafer scale fabrication of arrays of suspended TiW nanowires with dimensions comparable to those of very long multi walled carbon nanotubes (50nm x 60nm x 8micron). The nanowires are fabricated by e-beam lithography and lift-off of a low-stress, sputtered TiW thin film, followed by a wet etch in BHF to remove the under laying 1 micron SiO2 layer. Initial studies indicate that stiction can be minimized by etching the oxide in HF vapours instead of performing a wet oxide etch, but for double clamped wires shorter than 10 micron no stiction problems are observed regardless of the oxide etch used. Since the dimensions of these wires can be easily tailored down to nanometer size using state-of-art e-beam lithography, the wires can for example be used to investigate how the mechanical properties scale with size [2]. AFM measurements has shown that the surface roughness of the TiW (10/90 w.t %) in an 10 x10 μm area is around 4-7 nm peak-to-valley, whereas electrical 4 point probe measurements on bond pads have shown a resistivity of around 40 µΩcm. Both of these properties seem to depend on the power and pressure during deposition. Among other important properties of the TiW material used for the wires are the high Young’s modulus of approximately 450 GPa, a non-oxidizing surface, extreme mechanical wear resistance, and a melting point well above 1670 degree, making them ideal for applications such as nanoresonators or heaters. Further studies are needed to assess whether these properties hold at the nanoscale. Finally, TiW is rather inert, and has a very low etch rate in most commonly used substances used in silicon micromashining processes such as; BHF, HF, KOH, Phosphoric acid, Acetone, IPA, 7-up etc (below 1 nm/min, slightly higher for KOH though). Only Piranha seems to etch the TiW with 20 nm/min. All in all, we expect wafer scale TiW nanowire fabrication to be a fast and easy route to a highly versatile NEMS system. References [1] M. S. Gudiksen, L. J. Lauhon, J. Wang, D. C. Smith and C. M. Lieber, Nature 415, 617-620 (7 February 2002) [2] B. Wu, A. Heidelberg, J. J. Boland, Nature Materials 4, 525-529 (Jul 2005) Letters
Q10.37
Carrier Doping of Silicon Nanowires Synthesized by Laser Ablation. Naoki Fukata1,3, Naoya Okada2, Satoshi Matsushita2, Takao Tsurui4, Jun Chen1, Takashi Sekiguchi1, Noriyuki Uchida2,3 and Kouichi Murakami2,3; 1Advanced Electronic Materials Center, National Institute for Materials Science, Tsukuba, Japan; 2Institute of Applied Physics, University of Tsukuba, Tsukuba, Japan; 3Special Research Project on Nanoscience, University of Tsukuba, Tsukuba, Japan; 4Institute for Materials Reseach, Tohoku University, Sendai, Japan.
One-dimensional silicon nanowires (SiNWs) are of great interest in the fields of both fundamental and application research. Various demonstrations have been reported for future application of SiNWs such as electronic and optical devices. To realize such devices, the characterization and impurity-doping of SiNWs are of important subjects. Raman scattering measurements are sensitive for Si optical phonon, which gives valuable information about crystallinity, stress, and diameter of SiNWs. The bonding structures and the site of dopant atoms in SiNWs can be also investigated by Raman scattering measurements. Electron spin resonance (ESR) measurements are also sensitive for defects and electronic states of dopant atoms. In the present study, we synthesized boron (B)- or phosphorus (P)-doped SiNWs by laser ablation and investigated the site, bonding structures, and electronic structures of dopant atoms in them by Raman and ESR measurements. SiNWs were synthesized by laser ablation of a Si target with nickel as a metal catalyst [1] and B or P as dopant impurities which was placed in a quartz tube heated at 1473 K in a flowing Ar gas. The growth mechanism of SiNWs is VLS (Vapor-Liquid-Solid) growth. Micro-Raman scattering measurements were performed at room temperature (RT) with a 100x objective and a 532-nm excitation light at a power of 0.02 mW to avoid the local heating effect due to the excitation laser. ESR measurements were carried out at 4.2 K using an X-band ESR spectrometer with a magnetic field modulation of 100 kHz. A Raman peak was observed at about 618 cm-1 for SiNWs synthesized by using a target with B dopant. The peak frequency is in good agreement with that of a local vibrational mode of B in Si crystal. The Fano broadening due to a coupling between the discrete optical phonon and a continuum of interband hole excitations was also observed in the optical phonon peak, which indicates heavily B doping during VLS growth. These results prove that B atoms were doped in substitutional sites of Si in SiNWs. ESR measurements were also performed to investigate defects and P donor/conduction electrons in SiNWs. An ESR signal due to conduction electrons were observed for SiNWs synthesized by using a target with P, suggesting that P atoms were doped in SiNWs. Several signals related to defects were also observed. The results show that it is necessary to reduce or passivate them for the realization of nano-devices using SiNWs. [1] N. Fukata, T. Oshima, K. Murakami, T. Kizuka, T. Tsurui, and S. Ito, Appl. Phys. Lett. 86, 213112 (2005).
Q10.38
Large Scale Hierarchical Organization of Nanowires for Integrated Nanosystems Mathias Steinmair, Erwin Auer, Alois Lugstein, Christoph Schöndorfer and Emmerich Bertagnolli; Institute of Solid State Electronics, TU Vienna, Vienna, Austria.
The chemical synthesis of 1D silicon nanostructures has provided a new route to manufacture complex vertical semiconductor nanostructures with precision on the atomic scale. A major problem inhibiting the utilization of these structures is finding a way to incorporate them into circuits. The assembly of nanowires on multiple length scales is critical to the realization of integrated electronic and photonic nanotechnologies. As the planar miniaturization era draws to a close, we will need to use a combination of top-down and bottom up-approach. Ideally, it may become necessary to rely on guided self-assembly for the fabrication of active nanoscale components. To exploit these technologies, we developed an approach, which can use nanowire-based architectures as a bridge between lithographic and atomic-scale feature sizes on a scale far beyond that of the individual or small numbers of devices produced previously. The bottom layer consist of an array of nano pillars produced by conventional lithography and an anisotropic chemical etching process. On top of this, pillars with well-defined lengths and diameters, we synthesize 1-D nanostructures at specific sites by a partly self aligned process. The growth of these nanostructures is determined by metal catalysis growth. The control over the 1D nanostructure growth (so called vapor-liquid-solid growth) is mediated through the usage of nanoscopic metal templates. In addition, electrical transport studies show that reliable electrical contacts can be made to the functional nanowire arrays prepared by this method.
Q10.39
ALD Synthesis of Metal and Metal Oxide Nanotubes in AL2O3 Membranes. Kornelius Nielsch, Mato Knez, Mihaela Daub and Ulrich Gösele; Max Planck Institute of Microstructure Physics, Halle, Germany.
Inorganic Nanotubes are highly desirable for many applications in magnetic data storages, sensing, nanoelectronics and for biomedical applications ranging from diagnostics of cancer cells to drug delivery. Self-ordered alumina membranes are extensively used as a template system for the synthesis of 1D nanoobjects, e.g. nanowires and nanotubes. By introducing imprint lithography in the fabrication process of the Al2O3 membranes, long-range ordered pore arrays on a cm2-scale are obtained with a very mondisperse pore diameter and ideal circular cross-section. On the other hand Atomic Layer Deposition (ALD) is a very versatile technology for the conformal coating of Al2O3 membranes with pore diameters ranging from 20 to 400 nm and aspect ratios from 10 to 10,000. This presentation will cover the following aspects on the ALD synthesis of inorganic nanotubes in self-ordered and long-range ordered Al2O3 membranes: (1) Ruthenium nanotubes with diameters from 40 to 180 nm based on the chemical reaction of RuCp2 with hydrogen or oxygen. (2) Ferromagnetic nanotubes based on Ni-ALD and the characterization of their magnetic properties. (3) Arrays of core-shell nanotubes based on multilayer ALD of Al2O3 and TiO2 in chemically widened Al2O3 Membranes, which can be applied as a novel master with sub-20 nm features for imprint lithography. We gratefully thank the German Ministry of Education and Research (BMBF) for financial support (FKZ: 03N8701).
Q10.40
Abstract Withdrawn
Q10.41
High Performance and Durability of AFM Tips Based on a Carbon-carbon (a-C, CNT) Composite. Denise B. Nakabayashi1,2, Alberto L.D. Moreau1, Paulo C. Silva2, Vitor R. Coluci1, Douglas S. Galvão1, Monica A. Cotta1 and Daniel Mario Ugarte1,2; 1Física Aplicada, IFGW-UNICAMP, Campinas, SP, Brazil; 2Lab. Microscopia Eletrônica, Lab. Nac. Luz Síncrotron, Campinas, SP, Brazil.
Atomic Force Microscopy (AFM) represents a key experimental tool to measure the surface topography. This method is based on the vertical deflection of a cantilever, whose extreme contains a very sharp tip that scans the sample surface. Conventional tips are usually made of silicon and tip size is the main factor limiting resolution. A possible method to improve the image resolution is to use multi-walled (MW) or single-walled (SW) carbon nanotubes (CNTs) as imaging tip. CNTs have a a high aspect-ratio because of their cylindrical shape with a diameter of about 10 nm for MWNTs and few microns in length. AFM tips containing carbon nanotubes have already been shown to improve image resolution and, due to the CNT resilience, they can last much more than high resolution Si tips (HRTs). However, the tip fabrication process is difficult to control, showing low reproducibility; for example if the CNT is very long, its vibration generates noisy AFM images. Here, we report a fabrication method of high performance and high durability AFM tips based on CNTs by nanomanipulation inside a scanning electron microscope. Firstly, we glue a CNT on the AFM tip; secondly an amorphous carbon contamination is used to reinforce the CNT and Si Tip contact region. Finally, a contamination shell is deposited on some length of the protruding CNTs. The CNT enclosed in carbon contamination could be considered as a carbon-carbon (C-C) composite; this high-aspect ratio composite nanostructure is more rigid and supports much larger bending efforts. It can be compared to a reinforced “nanoconcrete” beam, where the CNTs become “rebars”, and a-C plays the role of concrete. The CNT tip remains uncovered to maintain high resolution capabilities and resilience property. MWNT AFM tips were tested for image acquisition and easily generated high resolution images, equivalent to those obtained using commercial HRTs. However, these latter usually maintain the high imaging quality for the first few images. In contrast, the C-C composite tips have shown a surprising long life time, yielding equivalent performance for over 400 images, when the fatigue test was interrupted. We have also carried out molecular dynamics simulations of the interaction of carbon tips with a rigid obstacle. To model this process we have used three types of tips: a) a double walled carbon nanotube; b) the inner tube of (a) and covering the top part with a cylindrical diamond-like layer; c) a pure diamond-like structure. Short-range interactions between carbon atoms were modeled by reactive empirical bond-order potential with parameterizations due to Brenner; long-range interactions were described by the Lennard-Jones potential. The presence of diamond-like structure in the second tip increases the frequency but reduces the amplitude of oscillations after releasing contact with the substrat. From these results, it was possible to qualitatively understand the increase in tip performance.
Q10.42
Engineering Carbon Nanotube Atomic Force Microscopy Nanoprobes. Haoyan Wei1, Sang Nyon Kim2, Minhua Zhao1, Sang-Yong Ju2, Bryan D. Huey1, Fotios Papadimitrakopoulos2 and Harris L. Marcus1; 1Materials Science and Engineering Program, Department of Chemical, Materials and Biomolecular Engineering, Institute of Materials Science, University of Connecticut, Storrs, Connecticut; 2Nanomaterials Optoelectronics Laboratory, Polymer Program, Institute of Materials Science, Department of Chemistry, University of Connecticut, Storrs, Connecticut.
Carbon nanotubes (CNTs) were assembled onto conductive atomic force microscopy (AFM) probes using dielectrophoresis (DEP). This process involved the application of a 10V, 2 MHz, AC bias between a metal-coated AFM probe and a dilute suspension of single-wall carbon nanotubes (SWNTs). The high electric field exerted at the tip of these AFM probes (positive dielectrophoresis) aligned the nanotubes and caused them to precipitate out of their suspension and onto the probe. The gradual displacement of the AFM probe away from the SWNT suspension consolidated nanotubes into nanofibrils with a high degree of alignment as demonstrated with polarization Raman experiments. By varying the pulling speed, immersion time and concentration of the CNT suspension, one can tailor the diameter of these probes, and according to nanomechanical measurements also their corresponding stiffness. Precise length trimming of these nanofibrils was demonstrated by their partial re-immersion into a liquid that strongly interacted with the exposed precipitated nanotubes, such as sodium dodecyl sulfate (SDS) solution. Such dissolution was feasible because SDS micellarized SWNTs and re-dispersed them. Vacuum annealing these nanoprobes at temperature up to 450°C further increased their stiffness and rendered them insoluble to SDS and all other aqueous media. Finally, re-growth of a new CNT fibril on top or at the end of a previously grown nanoprobe was also demonstrated by re-immersing the previously grown fibril to the desired depth in a new CNT suspension and repeating the dielectrophoretic assembly. The resulting high-aspect-ratio CNT AFM probes have been tested and proven electrically conductive and mechanically robust to be used as mechanical nano-needles and electrochemical nanoprobes.
Q10.43
Single-Walled Carbon Nanotube Tip for Scanning Tunneling Microscope. Winadda Wongwiriyapan1, Takafumi Ohmori1, Yuya Murata1, Kenji Motoyoshi1, Hirofumi Konishi1, Masaru Kishida1, Kenji Kisoda2, Hiroshi Harima3, Jung-Goo Lee4, Hirotaro Mori4, Shin-ichi Honda1 and Mitsuhiro Katayama1; 1Electrical, Electronic and Information Engineering, Osaka University, Osaka, Japan; 2Wakayama University, Wakayama, Japan; 3Kyoto Institute of Technology, Kyoto, Japan; 4Research Center for Ultrahigh Voltage Electron Microscopy, Osaka University, Osaka, Japan.
Carbon nanotube (CNT) is expected to be an ideal material for use as a tip in scanning probe microscopy (SPM) due to its intriguing properties such as small diameter, high aspect ratio, excellent electrical conductivity and mechanical robustness. Since the pioneering work by Dai et al.[1], efforts have been focused on CNT tip for atomic force microscopy (AFM). However, the method for synthesizing the CNT tip for scanning tunneling microscopy/spectroscopy (STM/STS) has not been fully established. To date, most of the CNT tips for STM have been fabricated using multi-walled carbon nanotubes (MWNTs) [2], while there are no reports on single-walled carbon nanotube (SWNT) tips. In this study, we present the method for fabrication of SWNT tip by thermal chemical vapor deposition (CVD) and its potentiality for STM/STS application. W and PtIr tips were used as a supporting tip for growth of SWNT. In the case of W tip, Fe/Al catalyst thin film was directly deposited on the W tip by electron-beam evaporation at room temperature, and CVD of SWNTs was conducted using pure CH4 gas (300 sccm, 400 Torr) at 850°C for 5 min. In the case of PtIr tip, a solution of Fe, Mo and alumina catalytic nanoparticles was dropped onto the PtIr tip, and CVD of SWNTs was conducted using a mixture of CH4 gas (1,000 sccm) and H2 gas (250 sccm) under a pressure of 760 Torr at 900°C for 10 min. A high-yield growth of SWNT networks with an approximately uniform diameter was obtained. The density of the SWNT networks gradually decreased with proximity to the tip apex. A small number of SWNTs were formed on the tip apex. The SWNTs that protruded from the tip apex were relatively straight with a length of approximately 400 nm. Thus, the direct growth of SWNTs on the metal tip apex was achieved. TEM observation provides direct evidence that the product was an isolated SWNT and free of obvious defects. Catalyst particles were rarely found at the tip of the SWNTs, indicating base growth mode. Using the SWNT-PtIr tip, the observation of highly orientated pyrolytic graphite (HOPG) surface demonstrated stable atomic imaging. Thus, our fabrication method provides a potential application of SWNTs to a novel tip for STM. This work was partly supported by SENTAN Program of Japan Science and Technology Agency and Grant-in-Aids for Scientific Research from Japanese Society for the Promotion of Science. One of the authors (W.W.) acknowledges the support of the Japan Society for the Promotion of Science. [1] H. Dai et al., Nature 384, 147 (1996). [2] T. Ikuno et al., Jpn. J. Appl. Phys. 43, L644 (2004). [3] W. Wongwiriyapan et al., Jpn. J. Appl. Phys. 45, 1880 (2006).
Q10.44
Carbon Nanotubes (CNTs) Characterization and Quality Control. Mohammad A Al-khedher1, Chuck Pezeshki1 and Jeanne L. McHale2; 1School of Mechanical and Materials Engineering, Washington State University, Pullman, Washington; 2Department of Chemistry, Washington State University, Pullman, Washington.
In order to commercialize carbon nanotube technology, advances must be made in production of nanotubes. One of the major obstacles for successful mass-production that have been identified is performing quick and precise characterization of the properties of a given batch of nanotubes. In this research, we have identified a set of intermediate steps that will lead to the goal of a comprehensive, scalable set of procedures for analyzing nanotubes. The proposed methodology is originated with data processing for Raman spectrum of CNT turfs and image analysis of SEM images. The morphology of CNT structure is extracted using image processing techniques of SEM images and the sample structural properties (i.e. CNTs alignment, curvature, thickness and orientation) are estimated using Artificial Neural Networks (ANN) classifier. The mechanical properties of CNT specimens will be characterized along with a Raman spectroscopy study. The correlation between the mechanical behavior of Multi-Wall carbon nanotubes (MWNTs) turfs, Raman spectra of CNTs and the morphology of the turf structure will be investigated through the use of Adaptive Neuro-Fuzzy Artificial phenomenological modeling. This system is built to model and estimate CNTs mechanical properties based on Raman spectroscopy, morphology information extracted from SEM images using image analysis, and nano-indentation results. This research will open the doors to understand CNT turfs mechanical behavior and relate it to the morphology of the structure. We believe this novel methodology, relating the material physical behavior with the structure morphology, will ultimately improve our capability to control the quality of the produced nanotubes through the supervised growth of CNTs and studying the InSitu nucleation of the turf. This scalable process will dramatically improve nanotubes design, the use of CNTs in Nanotechnology and MEMS, and it will facilitate full scale production.
Q10.45
Abstract Withdrawn
Q10.46
Organic Doped Carbon Nanotube: Structural, Electronic and Transport Properties. Rodion V. Belosludov1, Amir A Farajian2,3, Sang Uck Lee3, Hiroshi Mizuseki3, Taishi Takenobu3, Yoshihiro Iwasa3 and Yoshiyuki Kawazoe1,3; 1ARCMG, Institute for Materials Research, Tohoku University, Sendai, Miyagi, Japan; 2Department of Mechanical Engineering and Materials Science, Rice University, Houston, Texas; 3Institute for Materials Research, Tohoku University, Sendai, Miyagi, Japan.
The carbon nanotubes have the potential to provide new classes of sensors, electronic devices, gas storage materials as well as materials for many other applications. They can be used as framework in order to stabilize novel one-dimensional systems of different atoms or molecules and this feature is very attractive for molecular electronics. The possibility of single-wall carbon nanotubes (SWNT) doping with various organic molecules has been reported experimentally [1], exemplifying importance of estimating the electronic and conductance properties of organics/carbon nanotube complexes. The first-principles calculations were performed within density functional theory (DFT), using the pseudopotential method and a plane-wave basis set. Exchange correlation was included using the generalized gradient corrected functional. In order to compare with experimental date, the intercalation density of organic molecules (molecule/C140) in the case of metallic (12,12), (10,10) and density (molecule/C136) in the case of semiconducting (17,0) SWNT have been selected. The transport properties of organics/SWNT complexes have been investigated using the non-equilibrium Green’s function formalism for quantum transport and DFT of electronic structures using local orbital basis sets [2]. The encapsulation properties of several organic molecules such as tetracyano-p-quinodimethane (TCNQ), anthracene, tetrathiafulvalene (TTF) and ferrocene, have been investigated. First, we determine the stable configuration of various organic molecules inside SWNT. It has been found that the configurations of organic molecules are sensitive to changes in diameter of SWNT from 12.53Å (16,0) to 13.56Å (10,10). In the case of anthracene doped SWNT, there is no charge transfer between inserted molecules and wall of nanotube and the interaction of anthracene molecule with SWNT is van der Waals one. This result is well correlated with experimental observation of optical adsorption measurements [1]. The situation is changed in the case of doping (10,10) SWNT by TCNQ. In this case, the interaction between TCNQ and SWNT is much stronger than that in the anthracene/SWNT case. The value of intercalation energy of TCNQ is equal to -0.549 eV in the case of most stable configuration. The favorable intercalation is associated with the charge transfer from SWNT to organic molecule followed by delocalization of a negative charge on the N atoms. The sliding ability of TCNQ inside (10,10) and (17,0) nanotubes is also in agreement with experimental XRD measurements which did not indicate any extra XRD peaks attributable to the regular one-dimensional configuration of organic molecules in SWNT [1]. REFERENCES 1. T. Takenobu, T. Takano, M. Shiraishi, Y. Murakami, M. Ata, H. Kataura, Y. Achiba and Y. Iwasa, Nature Materials 2 (2003) 683. 2. A. A. Farajian, R. V. Belosludov, H. Mizuseki, and Y. Kawazoe, Thin Solid Films, 499 (2006) 269.
Q10.47
Development of Multi Walled Carbon Nanotube p-n Junction Arrays via Micro Contact Printing. Yamini Yadav, Prasanna Padigi, Xiaoyu Song and Shalini Prasad; Electical Engineering, Portland State University, Portland, Oregon.
The limitation of shrinking silicon has lead to the development of nanosytems. As the size of integrated circuit decreases, bottom-up and self - assembly approaches offer the capability of developing multi functional devices. Over the past few years formation of functional assemblies using these nanostructures has become a prime focus of research in the field of nanotechnology, which concentrates on symmetrical alignment and patterning of nanomaterial for integrated nanosystems. Here we present a rapid prototyping technique based on micro contact printing to arrange carbon nanotubes (NT) in symmetric array pattern for alignment of nanodevices. This forms the building blocks for designing a nanoscale circuit. This is achieved by employing, a poly (dimethylsiloxane) PDMS mold associated with trench like structures with microfluidic channels. This is used to pattern the multi walled carbon on to a silicon substrate with metallic micro electrodes that is formed using micro-fabrication techniques. The PDMS mold is used to pattern the carbon NTs through micro contact printing. The hierarchical assembly and change in stamping direction of these NTs result in the formation of cross-bar structures. Each of these cross point functions as the addressable device. The width and the spacing can be changed depending to array structure. Thus with this technique spacing between two parallel NTs is controlled. The functionality of this circuit is demonstrated and studied by I-V characteristics. The long term goal is to form high density cross-array circuit pattern to develop an array of “devices” with an array of gold electrodes, fabricated on silicon substrate.
Q10.48
Low Resistance Ohmic Contact Formation Mechanism for Carbon Nanotube Via Determined by Hard X-ray Photoemission Spectroscopy. Daiyu Kondo1,2, Mizuhisa Nihei1,2, Shintaro Sato1,2, Akira Kawabata1,2, E. Ikenaga3, M. Kobata3, J. -J. Kim3, K. Kobayashi3, S. Komiya3 and Yuji Awano1,2; 1Nanotechnology Research Center, Fujitsu Laboratories Ltd., Atsugi, Japan; 2Fujitsu Limited, Atsugi, Japan; 3Japan Syncrotron Radiation Research Institute/SPring-8, Hyogo, Japan.
Carbon nanotubes (CNTs) have been expected for use as wiring materials to solve a several problems in future ULSI interconnects. Recently, we have succeeded in forming CNT via whose resistance was in the same order as that of W plugs [1,2]. To improve via resistance, we have investigated the electronic structures at the interface between CNTs and metal electrodes using by hard x-ray photoemission spectroscopy (PES). The PES measurements were performed at the BL47XU in the SPring-8 with the approval of the Japan Synchrotron Radiation Research Institute (JASRI). We measured Ti 2p core-level spectra after CVD growth of CNTs of the following three sample structures: (a) Co 1-nm/Ti 2-nm film, (b) Co 2.5-nm/Ti 6-nm film and (c) as-deposited 4.5-nm Co particles/TiN 5-nm film on a silicon substrate. The growth temperature was 450-510°C. As the carbon source, a mixture of acetylene and argon gases was used. The resistance of 2 µm-diameter-via was about 1MΩ, 0.7Ω [1] and 0.59Ω [2], respectively. In the spectra (b) and (c), it has been found one feature originated from the titanium carbide (TiC) and titanium nitride (TiN), respectively. On the other hand, only a titanium oxide layer was observed in the sample (a), resulting in the high resistance. These results indicate that the existence of TiC or TiN contact layers should be important to realize such a low resistance CNT vias. Acknowledgement: The authors thank Dr. N. Yokoyama of Fujitsu Laboratories Ltd. for their support and useful suggestions. This work was supported by the Advanced Nanocarbon Application Project, which was consigned to JFCC by New Energy and Industrial Technology Development Organization (NEDO) of Japan. References: [1] M. Nihei et al., IITC 2005. [2] S. Sato et al., IITC 2006.
Q10.49
Abstract Withdrawn
Q10.50
Improvement of X-ray Current Density Using CNT Cold Cathode. Choi Haeyoung, Won Suk Chang and Joung-uk Kim; Applied Imaging Research Group, Korea Electrotechnology Research Institute, Sa-1dong, Sangrok-gu, Ansan-shi, Gyeonggi-do, South Korea.
Studies on the electronic structure of carbon nanotube (CNT) are of much importance because of its efficient utilization in electronic devices [1]. These CNTs have many applications, such as field emission display (FED) and backlight unit and so on. One of these applications is the electron emitter for X-ray source. In order to obtain X-ray images of hard instruments or components, such as PCB board or machine tools, high quantity of X-ray current is generally required. In this study, we report that the current density of X-ray source can be greatly enhanced by using the CNT cathode. In general, the emission current of CNT cathode is very sensitive to gaps between CNT and grid metal mesh. And also, the current density appeared to be different with respect to the kinds of metal meshes and their sizes. And partial results of these were shown in our recently report works [2]. For example, as the distance between CNT and grid metal mesh was getting shorter, the current density of the triode was getting larger. Detailed parameters and corresponding results were presented and discussed in this study. 1. K. Tanaka, T. Yamabe, K. Fukui; the Science and Technology of Carbon Nanotubes, 1999 Ed., p. 40. 2. H. Y. Choi, W. S. Chang, H. S. Kim, Y. H. Park, J. U. Kim; Physics Letters A, 2006, In Press. *E-Mail: jukim@keri.re.kr
Q10.51
Controlled Carbon Nanotube Networks and its Corresponding Channel Effect at High Bias. Jun Huang, Harindra Vedala, Bangalore Kiran Rao, Do-Hyun Kim and WonBong Choi; Mechanical and Materials Engineering, Florida International University, Miami, Florida.
Geometrically controlled carbon nanotube networks were fabricated by a novel width-wise confinement technique to characterize their electrical behavior. The results demonstrate a non-linear decrease of total resistance which is dependent on the number of channels. The current-voltage characteristics at high field were also studied until the electrical breakdown takes place. Large current (~2mA), low resistance (~5KΩ) and current densities exceeding ~1.4 ×10^9 A/cm^2 were demonstrated at room temperature. Additionally, chronological SEM imaging along with simultaneous acquisition of the total current and voltage drop were used to identify the agents initiating breakdown sequences in the carbon nanotube networks. The results elucidate complicated sequences of breakdowns, often switching between individual nanotubes in the networks, depending on the actual path of the current.
Q10.52
Ultra-Large-Scale Fabrication of Individually Contacted Single-Wall Carbon Nanotube Devices by A/C-Dielectrophoretic Deposition. Aravind Vijayaraghavan, Matti Oron-Carl, Simone Dehm, Frank Hennrich, Horst Hahn and Ralph Krupke; Institut für Nanotechnologie, Forschungszentrum Karlsruhe, Karlsruhe, Germany.
Single-Wall Carbon Nanotube (SWNT) devices are conventionally fabricated either by e-beam lithography or focused ion beam fabrication of metal contacts on randomly grown nanotubes, or the deposition of nanotubes from suspension randomly on a predefined array of electrodes. Both these approaches are tedious and results in very low yields and poor reproducibility. The recent demonstration of A/C-Dielectrophoretic deposition of SWNTs on prefabricated electrodes [1,2] provides the best possibility yet for scaled-up fabrication of a large number of SWNT devices. En-route to achieving this goal, we will explore a number of parameters that govern the deposition. These include conductivity and pH of the suspension; electrode design, density and spacing; deposition voltage and frequency; and the extent of capacitive coupling between the electrodes and the substrate. We will also present a model of the entire system as a network of competing impedances and discuss the regimes of the deposition in terms of these impedances. We will also discuss an equivalent picture in terms of electric field lines and gradients. Interestingly, this analysis and model is not specifically limited to SWNTs, and can be used to explore similar fabrication of other nanotube and nanowire devices. Having optimized all the parameters, we will present the successful fabrication of large parallel arrays of individually contacted SWNT devices on a single chip, with numbers exceeding tens of thousands of such devices per square centimeter, which is orders of magnitude greater then anything that is possible with other techniques. These arrays can be fabricated in a single, short, deposition step. The nanotubes in these devices enjoy the advantage of being contacted all around with metal using a two-step lithography process, to ensure minimum contact resistance. Additionally, by tuning the deposition condition, it also provides the possibility of making entire such arrays of only metallic nanotube devices, or a mixture containing predominantly semiconducting nanotube devices. This can be viewed as the first clear sign that it is in fact possible to successfully integrate SWNTs as transistors or interconnects in practically large numbers, as a part of existing silicon-based device architectures. (1) Krupke, R.; Hennrich, F.; Lohneysen, H. v.; Kappes, M. M.; Science, 2003, 301, 344-347. (2) Krupke, R.; Hennrich, F.; Weber, H. B.; Beckmann, D.; Hampe, O.; Malik, S.; Kappes, M. M.; Lohneysen, H. v.; App. Phys. A, 2003, 76, 397-400.
Q10.53
Fabrication of Multi-Intramolecular Junctions Array of Single-Walled Carbon Nanotubes on Surface by Temperature Oscillated CVD. Jin Zhang, Yagang Yao and Zhongfan Liu; College of Chemistry and Molecular Engineering, Peking University, Beijing, China.
Since the discovery of carbon nanotube in 1991, it was regarded as an ideal building block for nanoelectronics device, because they can function both as device and as the wire[1-2]. However, the use of nanotubes as building blocks is limited, because the selective controlled growth of semi-conducting or metallic nanotubes is not currently possible. The ideal building blocks for nanoelectronics device will require its electronic properties and structures should be defined and controlled. Recently, many efforts have been made to fabricate nanojunction as building blocks for nanoelectronics devices, such as intermolecular metal-semiconductor nanojunction and nanotube crossed nanojunction. However, the controlled growth of this sort of nanojunctions is still in its infancy. In the present work, multi-intramolecular junctions array of single-walled carbon nanotubes (SWCNTs) has been fabricated by temperature oscillated chemical vapor deposition (CVD). Micro-resonance Raman spectroscopy was used to detect the metallic-metallic, metallic-semiconductor and semiconductor-semiconductor SWCNTs junctions. References: 1.H. W. C. Postma, T. Teepen, Z. Yao, M. Grifoni, C. Dekker, Science, 293, 76, 2001 2.P. G. Collins, A. Zettl, H. Bando, A. Thess, R. E. Smalley, Science, 278, 100, 1997.
Q10.54
First-Principles Electronic Properties of Graphene Nanostrips. John W. Mintmire1, Junwen Li1 and Carter T. White2; 1Department of Physics, Oklahoma State University, Stillwater, Oklahoma; 2Code 6189, Naval Research Laboratory, Washington, District of Columbia.
The electronic structure of graphene nanostrips with zigzag edges is known to exhibit localized edge states in the vicinity of the Fermi level, with work by Fujita and others suggesting the possibility of ferrimagnetic behavior of these edge states. We present first-principles results for zigzag-edge graphene nanostrips including both closed shell and ferrimagnetic states as a function of width and substituent. Within the mean-field approach used, we find that the ferrimagnetic states examined are lower in energy than the closed shell states. We also discuss the implications of the first-principles band structures on empirical Hamiltonians. This work was supported by the DoD HPCMO CHSSI program through the Naval Research Laboratory.
Q10.55
Geometry Dependent Resistivity in Single-walled Carbon Nanotube Films Patterned Down to Submicron Dimensions. Ashkan Behnam1, Leila Noriega1, Yongho Choi1, Zhuangchun Wu2, Andrew G Rinzler2 and Ant Ural1; 1Electrical & Computer Engineering, University of Florida, Gainesville, Florida; 2Physics, University of Florida, Gainesville, Florida.
Single-walled carbon nanotubes (SWNTs) have attracted significant research attention in the last decade because of their remarkable physical and electronic properties, such as high mobility and current density. Despite these outstanding properties, however, controlling the diameter, chirality, location, and direction of individual nanotubes has proven very challenging. Recently, there has been a growing interest in using 2D nanotube networks and 3D nanotube films as a new class of materials, in which individual variations in diameter and chirality are ensemble averaged to yield uniform physical and electronic properties. Several applications of SWNT networks and films have recently been demonstrated, such as thin film transistors, flexible electronics, sensors, and transparent and conductive electrodes for optoelectronics. All of these potential applications require patterning of the nanotube films and understanding their electrical properties as a function of device geometry. Although 2D nanotube networks have been patterned recently by a variety of techniques, reproducible patterning of thicker nanotube films, particularly down to sub-micron linewidths, has not been demonstrated previously. Furthermore, how electrical properties of nanotube films scale as a function of device geometry, particularly device width, at submicron dimensions remains unexplored. In this talk, we demonstrate patterning of SWNT films down to 200 nm lateral dimensions using e-beam lithography and reactive ion etching with good selectivity and directionality. We then use this etch capability to fabricate standard four-point-probe structures to characterize the resistivity of these films as a function of device geometry. The resistivity of nanotube films are found to be independent of device length for a given width and thickness, while increasing over three orders of magnitude compared to bulk films, as the width and the thickness of the films shrink. In particular, we find that the resistivity of SWNT films starts to increase with decreasing device width below 20 μm, exhibiting an inverse power law dependence on device width at sub-micron dimensions with a critical exponent of about 1.5. We show that this behavior can be explained by a purely geometrical argument. The “top-down” patterning of these transparent, conductive nanotube films allows for their use in sub-micron device structures, and perhaps for their integration into standard silicon microfabrication technology. However, since the resistivity of very thin and narrow SWNT lines is over three orders of magnitude greater than that for thick and wide films, the resistivity scaling is an important effect that needs to be taken into account when fabricating small devices in which nanotube film transport characteristics play a vital role.
Q10.56
Contact Transfer of Aligned Carbon Nanotube Arrays onto Conducting Substrates: Low temperature Process for Electronic Applications. Ashavani Kumar1, Rajashree Baskaran2, Victor Pushparaj1, Swastik Kar1, Omkaram Nalamasu1 and Pulickel M. Ajayan1; 1Materials Science and Engineering, Rensselaer Polytechnic Institute, Troy, New York; 2Components Research, Intel Corporation, Chandler, Arizona.
Carbon nanotubes (CNTs) have been considered a potential candidate for future interconnects technology because of various advantages such as higher carrying current density (~109 A/cm2) and high thermal conductivity over traditional copper based interconnects. In spite of these excellent properties, the main challenges of using CNT based interconnects are the incompatibility of high temperature of growth of CNTs, imperfect contact with metal, and poor control over the process to preferentially grow metallic or semiconducting nanotubes. We believe growth in a temporary substrate followed by low temperate solder attach process could be a feasible process to integrate CNTs as interconnects.To this end, We demonstrate the fabrication of different architectures of carbon nanotubes on conducting substrate via contact transfer of nanotubes using low temperature solders. Lithographically patterned multiwalled carbon nanotube arrays grown on silica substrates by CVD are transferred onto solder coated substrates. Both negative and positive patterns can be obtained by changing the printing parameter. Good wetting and electrical contacts are confirmed by measuring their field emission properties. This method can be used to construct nanotube structures of different shapes and dimensions over large areas on substrates of choice and could be a feasible process to integrate nanotubes into various devices.
Q10.57
Patterning Carbon Nanotube Mat Transistors using Transfer-Printing. Vinod K Sangwan1,2, Dan R. Hines1,2, Vincent W. Ballarotto2, Gokhan Esen1, Michael Fuhrer1 and Ellen D. Williams1,2; 1Physics, University of Maryland, College Park, Maryland; 2Laboratory for Physical Sciences, College Park, Maryland.
Carbon nanotube mats (CNTM) are a promising electronic material. However, patterning using standard lithographic techniques can degrade their properties. We have developed a chemical-free method to pattern CNTM using differential adhesion. CNTM, grown via a CVD method on SiO2, were selectively removed by transfer printing against a polyethylene terephthalate (PET) template. After patterning, further transfer printing steps were used to fabricated thin-film transistor (TFT) devices by sequentially printing the patterned CNTM and the electrode subassembly components onto a PET device substrate1. The electrode subassembly consisted of gold (Au) gate and source/drain electrodes separated by a poly(methyl-methacrylate) (PMMA) dielectric layer. Top gate - top source/drain electrode (TE) TFTs were fabricated by first printing the patterned CNTM onto the PET device substrate and then printing each component of the electrode subassembly. Bottom gate - bottom source/drain electrode (BE) TFTs were fabricated by first printing each component of the electrode subassembly and then printing the CNTM onto the PMMA/Au surface of the resulting device substrate. Both TE and BE devices were studied as a function of channel length for L = 3 to 100 μm. Bottom electrode devices showed on/off ratios of the order of 103, while TE devices exhibited ambipolar behavior with no hysteresis. Output and transfer characteristics of TE and BE CNTM TFTs as a function of tube length and mat density are currently being studied. 1. D. R. Hines et al., Appl. Phys. Lett. 86, 163101 (2005). *Work supported by the Laboratory for Physical Sciences, College Park, MD
Q10.58
Enhancement of Field Emission of Aligned Carbon Nanotubes by Thermal Oxidation. Baoqing Zeng1,2, Guangyong Xiong1, Shuo Chen1, WenZhong Wang1, Dezhi Wang1 and Zhifeng Ren1; 1Department of Physics, Boston College, Chestnut Hill, Massachusetts; 2School of Physical Electronics, University of Electronic Science and Technology of China, Chengdu, China.
To improve the field emission behavior of the aligned carbon nanotubes grown by thermal chemical deposition, post thermal annealing in vacuum and air was conducted. The largest field emission current density of the aligned carbon nanotubes arrays were considerably improved from19 mA/cm2 to 79 mA/cm2 with a decrease of the threshold field from 7.2 V/μm to 5.4 V/μm after annealing in vacuum and air. SEM and TEM were employed to elucidate the reason for such a significant improvement of the field emission when annealing in vacuum and air. Those analyses suggested that air annealing can reduce the density of the aligned carbon nanotubes, improved the crystal quality. This work is important for applications of aligned carbon nanotubes as high brightness electron sources, flat panel displays, and microwave devices.
Q10.59
Abstract Withdrawn
Q10.60
An Electrically-tunable Ultra Small Liquid Crystal Micro-lens Array based on the Carbon Nanotubes Xiaozhi Wang, Xin Ji, Timothy Wilkinson, Ian Bu, Kenneth Teo and William Milne; Center for Advanced Photonics and Electronics, Dept. of Engineering, University of Cambridge, Cambridge, United Kingdom.
The liquid crystal (LC) micro-lens array is valuable for in information collection, integrated optics and optical communications. The general design of the micro-lens array includes the zone patterned structure, hole or hybrid-patterned electrode, and surface relief profile. Here we propose a new design and fabrication process for a LC mirco-lens, 5μm in size with an array which is based on a Carbon Nanotube (CNT) array grown by Plasma-enhanced Chemical Vapor Deposition (PECVD).<br><br> When single CNTs are grown by combined nickel micro-dot lithography and PECVD it is possible to produce a sparse array of single CNTs of the order of a few microns in height and 5-10μm pitch that is shown in Figure 1. When these nanotubes are subjected to an electrical field, they form a field profile around the tube tip to the ground plane. If the sparse array were to be immersed in a nematic LC, then the molecules of the LC would follow this electric field profile and form the same profile in refractive index in three dimensions as shown in Figure 2. This is in effect a micro-lens capable of focussing an applied light wave. <br><br> The theoretical results are simulated using a Finite Element method. From Figure 3 to Figure 5, various CNT arrays are modelled between Indium Tin Oxide (ITO) coated quartz plates. The voltage is fixed at 10v on the top plane and 0v on the bottom. When three and four CNTs arrays are investigated, it is found that the field lines are centred at the point above the CNTs rather than at the tips of each CNT. This means that the combined effect is to produce a larger micro-lens which develops further when the CNTs are close enough. <br><br> Experimental results demonstrate the impact on the LC from the CNT under applied voltage. This effect is even greater than that predicted from the simulation as shown in Figure 6. The raw diameter of the micro-lens on the single CNT which is about 120nm in width can be nearly 5μm under 10v as seen using a polarized microscope. This is more than 20 times the diameter of a single CNT, and can be varied by the voltage. The focal properties are thus tuneable using the applied electrical field. The CNT arrays are fabricated using E-beam lithography.
Q10.61
Directed Linking of Carbon Nanotubes with Single CdSe Quantum Dots. Kathryn Elizabeth Leach and Todd D. Krauss; Chemistry, University of Rochester, Rochester, New York.
As circuit miniaturization continues, the demand for smaller and more efficient component parts has increased. Metallic single-walled carbon nanotubes (SWNTs) are the ideal nanometer-scale wire, as they can withstand current densities up to 2 to 3 orders of magnitude higher than copper currently used in electronic chips. These conductive nanotubes can therefore be utilized as “nano-electrodes” to efficiently electrically contact another nanoscale object, such as a single semiconductor quantum dot (QD), thus creating macroscopic integrated systems based on nanometer-scale components. Although semiconductor QDs have been previously attached to NTs, the attachment scheme was uncontrolled; direct and defined attachment of QDs to SWNTs remains elusive.<br>We have designed a strategy for directed assembly of fabricated QD-SWNT devices. NTs were grown across patterned catalyst islands on a silicon wafer followed by electrode deposition. NTs spanning pairs of electrodes have been cut in half (leaving nanometer scale gaps) by applying voltage pulses to a metal-coated atomic force microscope (AFM) tip in very close proximity to a grounded nanotube. The resulting carboxylic group moieties found at the cut NT edges were used to covalently attach CdSe QDs with amine-terminated surface ligands using standard bioconjugation techniques. Electrostatic force microscopy (EFM) was used to monitor NT conductivity before and after cutting, as well as after QD attachment. The photoelectrical transport properties of a typical hybrid QD-SWNT device will be discussed.
Q10.62
Influence of chemical dopant technique to reduce Schottky barriers of Pd-contacted CNTFETs Damian Casterman and Maria Merlyne De Souza; Emerging Technologies Research Centre, De Montfort University, Leicester, United Kingdom.
Pd-contacted devices have been shown to produce both ohmic as well as Schottky contacts in CNTs [1,2]. In particular, the variation of the Schottky barrier height depending upon processing conditions hampers the commercial viability of the technology. Pd has the ability to adsorb various types of atoms, leading to complexities which require to be well understood [3]. IBM research recently demonstrated subthreshold slopes of about 70mV/dec using Hexachloroantimonate, SbCl6-, in regions near Pd/CNT contacts [1]. In this work, ab initio investigations of the effect of SbCl6- on both a (8,0)-semiconducting carbon nanotube and a (100)-Palladium contact are reported for two situations (A) - Interaction of SbCl6- with (100)-Pd surface (B) - Interaction of SbCl6- with CNT Methodology: The ab initio calculations have been carried out using Density Functional Theory (DFT) implemented in Vienna Ab initio Simulation Package (VASP) within the Local Density Approximation (LDA). Ultra Soft pseudo-potential were used with plane wave method. Monkhorst-Pack mesh of (4x4x4) for Pd and (1x1x15) for CNT were used for integration of Brillouin zone. Cut-off energies were 219 eV for Pd and 358 eV for CNT. Results: A - SbCl6- is placed above a hollow site of a (100)-Pd surface. One Cl adsorption modifies the electronic properties of Pd surface due to charge transfer from Pd to Cl and leaves a neutral rehybridized SbCl5 close to surface. Our calculation reveals a potential lowering localized at interface which consequently reduces Schottky barrier height for holes. B - SbCl6- relaxed above a 6-hexagonal-ring of two unit cells of a (8,0) zigzag CNT. Charge transfer from CNT to SbCl6- degeneratly p-dopes the semiconducting nanotube by shifting down the Fermi level into the valence band. The averaged local potential variation is just less than half-midgap. Conclusion: Interactions between SbCl6- with CNT as well as Pd favours injection of majority carriers leading to better electronic properties. This work gives numerical insight into the influence of the charge transfer technique on Schottky barrier heights using ab initio pseudopotential theory. References: [1] Self-aligned carbon nanotube transistors with charge transfer doping, J. Chen, C. Klinke, A. Afzali and Ph. Avouris, Condensed Matter, Vol. 1, 0511039, November 2005. [2] Ballistic carbon nanotube field-effect transistors, Ali Javey et al., Nature (London), Vol. 424, p 654-657, August 2003. [3] Effect of hydrogen on the surface relaxation of Pd(100), Rh(100), and Ag(100), S. C. Jung and M. H. Kang, Phys. Rev. B 72, 205419, 2005.
Chair: Prabhakar Bandaru
Wednesday Morning, November 29, 2006
Room 312 (Hynes)
8:00 AM *Q11.1
Inorganic Nanotubes and Fullerene-like Structures (IF): A Progress Report. Reshef Tenne, Materials and Interfaces, Weizmann Institute, Rehovot, Israel.
In this presentation a progress report, focused mainly on the results obtained in our lab will be presented. While the synthesis and study of IF materials from layered metal dichalcogenides, like WS2 and MoS2 remain a major challenge, some progress with the synthesis of IF structures from other compounds, like metal oxides and metal halides have been realized. The synthesis of some new IF materials, like Cs2O, NiBr2 and others will be described. The study of the mechanical properties of individual WS2 nanotubes will be discussed in some detail. The agreement between theory and experiment suggests that the nanotubes are of high crystalline order and their mechanical properties are predictable. The study of MoS2 nanooctahedra 2-5 nm in size, which can be considered to be the true inorganic fullerenes of these and many other layered structures, will be discussed. The agreement between the calculated and experimentally observed structures indicate that the nanooctahedra are indeed the stable structures in this size range, beyond this size the quasi-spherical nested MoS2 structures become stable. Some new potential applications for these and related materials will be discussed in the fields of friction reduction of various objects; catalysis; rechargeable batteries, coatings, etc. will be discussed as well.
8:30 AM Q11.2
Molecular Beam Epitaxy Growth of CdTe over Carbon Nanotubes. Rodolfo Camacho, Stephan Turano and W. Jud Ready; GTRI-EOSL, Georgia Tech, Atlanta, Georgia.
Cadmium Telluride is deposited via molecular beam epitaxy (MBE) over Carbon Nanotubes (CNTs) grown on Silicon wafers. Characterization through scanning electron microscopy (SEM), energy dispersive x-ray spectroscopy (EDS) and x-ray diffraction (XRD) is performed. The motivation behind the use of CdTe-coated CNTs is for use in light-trapping “Generation-IV” photovoltaic systems. The properties and orientation of the CdTe are strongly correlated to the MBE growth conditions. These properties can be readily modified through an annealing process.
8:45 AM Q11.3
SiC Nanowires by Silicon Carburization. Loucas Tsakalakos, Jody Fronheiser, Larry Rowland, Mohamed Rahmane, Michael Larsen and Yan Gao; General Electric - Global Research Center, Niskayuna, New York.
Nanowires and related one-dimensional nanostructures have recently been shown to be important building blocks for nanosystems with significantly improved or completely new performance capabilities compared to micro-devices. SiC nanowires have been the subject of particular interest due to their wide-bandgap (Eg = 2.2 eV), high electron mobility (1000 cm2/V-s), high breakdown field, and high temperature stability, making them promising candidates for nanoscale UV LEDs, harsh environment sensors, and electronics. While most SiC nanowires materials studied are nominally single crystals, polycrystalline SiC nanowires may also be of interest as toughening elements in mechanical load bearing nanocomposites or as field emitters. Numerous methods for synthesizing SiC nanowires have been described in the literature. These include metalorganic chemical vapor deposition (MOCVD), microwave plasma CVD, direct heating of substrates, nanotube confined reactions, thermal evaporation of powders, hot filament CVD, and sputtering combined with rapid thermal annealing. Recently we proposed and demonstrated a method to synthesize transition metal carbide nanowires using a two-step process in which metal or metal oxide nanowires are first formed on a Si substrate and subsequently carburized in situ by flowing methane and hydrogen [L. Tsakalakos, et al., J. Appl. Phys. 98, 044317 (2005)]. Mo2C nanowires and nanoribbons were grown on Si and the growth mechanism was shown to be due to a complex interaction of MoxOy vapor species with Au catalyst on the Si surface, leading to an autocatalytic growth. Here we describe a similar, yet mechanistically less complicated, method for making SiC nanowires in which Si nanowires are grown using the VLS mechanism, followed by carburization in methane or propane. This method allows us to synthesize nanowires that are fully polycrystalline 3C SiC or composites of Si nanowires covered with nano-sized SiC grains. Si nanowires are grown at low temperature (550-650 C) and subsequently carburized at 1100-1200 C in a methane/hydrogen or propane/hydrogen environment. Thermochemical calculations showed that the Si carburization is thermodynamically favorable over a wide tempareture range, whereas our studies showed that the Si nanowire carburization is kinetically limited below 1100 °C. Partially carburized nanowires contained distinct SiC nanosized grains on the Si nanowire surface, whereas fully carburized nanowires were polycrystalline 3C SiC with grain sizes of ~ 50-100 nm.
9:00 AM *Q11.4
Inorganic Nanowires: Growth and Applications in Electronics and Optoelectronics. Meyya Meyyappan, Center for Nanotechnology, NASA Ames Research Center, Moffett Field, California.
Investigation of one dimensional nanowires has receieved much attention recently with reports of semiconducting, metallic and dielectric nanowires for various applications. In general, bandgap engineering is a possibility with the inverse dependence of bandgap on radius. Also, quantum confinement at small radius provides interesting physics for device applications. Single crystal wires with well-defined surface properties and the amenability for bottom up integration make nanowires an attractive candidate for future electronics and optoelectronics devices. We have used a vapor-liquid-solid (VLS) technique to grow Ge, phase change materials, gallium and indium antimonides, indium nitride and various oxides. Gold and other low melting metals have been used as catalysts. Growth results will be discussed along with device fabrication efforts including surround gate transistors, phase change memory devices and characterization results for optoelectronic applications.
9:30 AM Q11.5
Fabrication of Spinel Nanotubes Based on Kirkendall Effect. Hongjin Fan1, Mato Knez1, Roland Scholz1, Kornelius Nielsch1, Eckhard Pippel1, Dietrich Hesse1, Margit Zacharias1,2 and Ulrich Goesele1; 1Dep. II, MPI of Microstructure Physics, Halle, Germany; 2FWIM, FZ Rossendorf, Dresden, Germany.
There has been increasing interest in intentional synthesis of nanowires and nanotubes based on a large variety of materials. Where normally the growth of nanowires is done by vapor liquid solid growth we want to demonstrate here a way of fabrication ultra long nanotubes by using the Kirkendall effect. The formation of the voids is attributed to the Kikendall effect associated with the different diffusion rates of the atoms moving in and out. A fabrication route for hollow nanoparticles based on binary compounds (CoS, CoO, CoSe, CdS) was recently demonstrated by Yin et al. in solution starting from solid nanoparticles as of one of the constituents [1]. While the experimental demonstration and theoretical treatment of the Kirkendall effect mainly concern planar interfaces or nanoscale spherical interfaces, the nanoscale Kirkendall effect should provide a general fabrication route to hollow nanostructures, including high aspect ratio cylinders (i.e., nanotubes). In this paper, we report one-dimensional free-standing spinel nanotubes which are transformed from nanowires via the Kirkendal effect in solid-state reaction [2]. Our nanotubes are 30-40 nm in diameter, smooth with excellent monocrystallinity, and up to 20 micrometer long. As an example of quasi one-dimensional ternary compound system, these spinel nanotubes differ from the commonly discussed binary compound and quasi zero-dimensional system. Hence, our results corroborate the versatility of applying the Kirkendall effect in producing hollow nano-objects. In our study, the starting object is a core-shell nanowire in which the crystalline ZnO core and the amorphous Al2O3 shell have an initially smooth interface and can form compounds (rather than solid solutions) through interfacial solid-state reactions. ZnO can react with a variety of other oxide materials to form spinels, particular ZnAl2O4 as demonstrated here. In an ideal case, the nanoscale Kirkendall effect leads to a nanotube with a complete hollow interior and single phase spinel walls. This requires a suitable thickness relationship between the core and shell so that no excess material is remained after the reaction. Atomic layer deposition (ALD) was used which assures a uniform thickness of the Al2O3 shell. As a result of the conformal characteristics of ALD, a uniform coating of the amorphous alumina surrounding the single-crystal ZnO core is evident. After annealing, the structure transforms into hollow tubes. Although we demonstrate here randomly oriented ultralong nanotubes, it is in principle possible to obtain vertically-aligned and regularly-ordered nanotubes following the same approach. Such ordered arrays of spinel nanotubes may possess similar application potentials as carbon nanotubes. [1] Y. Yin et al. Science 304, 711 (2004). [2] H.J. Fan et al. Nature Materials, in press (2006).
9:45 AM Q11.6
Self-organized TiO2 Nanotubes and Their Applications. Jan Macak, Hiroaki Tsuchiya, Andrei Ghicov, Eugeniu Balaur and Patrik Schmuki; Dep. of Materials Science, Chair for Surface Science, Erlangen, Bavaria, Germany.
The presentation deals with self-organized high aspect ratio titania nanotubes that can be produced by tailored electrochemical anodization. Tubes can be grown to the length of several micrometers with single tube diameter of several tens of nm and with the possibility to modify the geometry (1-4). This strategy to form self-organized structures can be used also for the whole range of other valve metals such as Hf (5), Zr (6), Nb (7), Ta (8) or Titanium alloys (9,10). We show that self-organized highly ordered nanotubes are formed under a competition of TiO2 formation and its dissolution. Using various growth conditions the pore geometry can be varied significantly. Particularly TiO2 nanotubes have found considerable interest due to a wide range of different functional properties. The tubes can be dye-sensitized or (11) or N-doped (12, 13) and be used for solar energy conversion, or for the catalytical electrooxidation of methanol (14). Further, interactions with hydroxyapatite (15) and changing of wettability (16) have been investigated for biomedical purposes. These issues will be discussed in detail. References: 1. V. Zwilling, E. Darque-Ceretti, A. Boutry-Forveille, D. David, M.Y. Perrin, M. Aucouturier, Surf. Interface Anal. 27, 629 (1999). 2. J. M. Macak, K. Sirotna and P. Schmuki, Electrochim. Acta, 50, 3679 (2005). 3. J. M. Macak, H. Tsuchiya and P. Schmuki, Angew. Chem., 44, 2100 (2005). 4. J. M. Macak, H. Tsuchiya, L.V. Taveira, S. Aldabergerova, P. Schmuki, Angew. Chem. 44, 7463 (2005) 5. H. Tsuchiya and P. Schmuki, Electrochem. Commun., 7, 49 (2005). 6. H. Tsuchiya and P. Schmuki, Electrochem. Commun., 6, 1131 (2004). 7. I. Sieber, H. Hildebrand, A. Friedrich and P. Schmuki, Electrochem. Commun., 7, 97 (2005). 8. I. Sieber, B.Kannan, P. Schmuki, Electrochem. Solid-State Lett., 8, J10 (2005). 9. J. M. Macak, H. Tsuchiya, L.V. Taveira, A. Ghicov, P.Schmuki, J. Biomed. Mater. Res. 75A, 928 (2005). 10. H. Tsuchiya, J.M. Macak, A.Ghicov, P. Schmuki, Small 2, 888 (2006). 11. J.M. Macak, H. Tsuchiya, A. Ghicov, P.Schmuki, Electrochem. Comm. 7, 1133 (2005). 12. R.P. Vitiello, J. M. Macak, A. Ghicov, H. Tsuchiya, L.F.P. Dick, P. Schmuki, Electrochem. Commun. 8, 544 (2006). 13. A. Ghicov, J. M. Macak, H. Tsuchiya, J. Kunze, V. Haeublein, L. Frey, P. Schmuki, NanoLetters 6, 1080 (2006). 14. J. M. Macak et al., Electrochem. Commun. 7, 1417 (2005). 15. H. Tsuchiya et al., J. Biomed. Mater. Res. 77A, 534 (2006). 16. E. Balaur, J. M. Macak, H. Tsuchiya, P. Schmuki, J. Mater. Chem 15, 4488 (2005).
Chairs: Prabhakar Bandaru and Reshef Tenne
Wednesday Morning, November 29, 2006
Room 312 (Hynes)
10:30 AM *Q12.1
Carbon Nanotubes, Innovations to Applications. Wonbong Choi, Department of Engineering and Computing, Florida International University, Miami, Florida.
Carbon nanotube (CNT) based device applications have been studied and developed to meet with future nano devices’ requirement. In 1999 we reported the first fully sealed field emission display based on carbon nanotube through innovative large area deposition and CNTs’ vertical aligning techniques. The turn-on field was 1 V/μm with brightness of 1800 cd/m2 at 3.7 V/μm. This work has led to a recent demonstration by SAMSUNG SDI of over 40-inch field emission flat panel display. In 2002, a concept of ultra-high density transistor was demonstrated based on the vertical-CNT array grown on the predefined nano template. Each device element is formed on a vertical CNT attached to bottom and upper electrodes and a gate electrode, which can be integrated in large arrays with the potential for tera-level density (1012/cm2). A nonvolatile memory based on single-wall CNT with oxide-nitride-oxide (ONO) charge trap was presented. The stored charges increase the threshold voltage with a quantized increment of 60 mV, suggesting that the ONO has traps with quantized energy state. Recently we proposed a nanoscale biomolecule recognition platform for direct, specific bioelectrical sensing by tuning nanotube electrodes. Novel applications of functionalized carbon nanotubes in the field of display, high power generator, transistor, memory and bio-sensor will be presented.
11:00 AM Q12.2
Synthesis of Heteroepitaxially Aligned Ge Nanowires and Ge/Si Core/Shell Nanowires. Irene A. Goldthorpe and Paul C. McIntyre; Materials Science and Engineering, Stanford University, Stanford, California.
Ge nanowires (NWs) have been less extensively studied than Si NWs, however, they offer advantages such as a lower synthesis temperature and higher intrinsic electronic carrier mobilities. In this work, Ge NWs with diameters of 50 nm and less have been heteroepitaxially grown on Si (111) substrates in a cold-wall CVD reactor. To obtain arrays of wires with similar lengths and diameters, monodisperse Au nanoparticles were used as the catalysts. Because these nanoparticles are generally suspended in an aqueous solution, this presents challenges for obtaining epitaxial NW growth on Si. We will describe how sample preparations dictate the quality of the epitaxial relationship between the wires and the substrate. The epitaxial relationship between the GeNWs and the Si substrate was quantitatively studied by x-ray diffraction, including detailed symmetric and asymmetric diffraction scans, and x-ray pole figures. The pole figures indicated that the majority of wires are heteroepitaxial, however, there exists a low density of twinning defects at the Ge/Si interface at the bases of both vertically oriented wires and wires oriented along inclined <111> directions. X-ray diffraction was also used to characterize the strain state of the Ge NWs. Even though the wires have a thin native oxide layer, no axial component of strain in the Ge NWs was measured by symmetric x-ray diffraction from vertical NW arrays. Heteroepitaxially depositing a Si shell around the Ge NWs may allow for the engineering of strain inside the NWs and reduce the influence of surface defects in carrier scattering by confining carriers to the NW interior. We will compare different strategies of Ge/Si core/shell NW synthesis and report the conditions which dictate whether three-dimensional Si islands or a continuous amorphous, polycrystalline, or crystalline Si film forms around the Ge cores.
11:15 AM Q12.3
Feasibility of Si Nanowire Integration: CVD Growth, Characterization and Comparison of Au vs PtSi Catalysts. Baron Thierry1, Michael Gordon1, Florian Dhalluin1, Martien Den Hertog3, Pierre Ferret2, Pascal Gentile3, Celine Ternon1, Karim Aissou1 and Jean-Luc Rouviere3; 1LTM, CNRS, Grenoble, France; 2CEA/DRT, Grenoble, France; 3CEA/DRFMC, Grenoble, France.
Silicon nanowires (Si NWs) are promising materials for some of the basic building blocks in microelectronics (interconnects, transistor channels, nanoelectrodes, etc.) and emerging application areas of photonics, chemical sensing, and solar cells. We present a study on the growth of Si NWs combining a metal catalyzer and CVD processing with silane precursor gas. Experimental parameters such as growth temperature, SiH4 partial pressure, and substrate orientation were investigated. For the gold catalyst, TEM and EDX experiments clearly show the presence of Au at the sidewalls of NWs. Because Au is not CMOS-compatible, several other microelectronics-friendly metals were studied. In particular, we show that PtSi can be used as an efficient solid catalyst with particles size varying between 20 nm and 100 nm. For the PtSi case, NW growth proceeds via solid-phase epitaxy which differs significantly from the traditional vapor-liquid-solid process using Au. Additionally, PtSi has a long history in the microelectronics industry (bipolar and MOS contacts)—as such, we believe that PtSi-based NW growth is an encouraging route for direct integration of NWs into CMOS processing. This talk will compare Au and PtSi-catalyzed NW growth in terms of wire morphology, orientation, and crystalline quality. For instance, changing the catalyst can modify the NW growth direction: (100) Si NWs can be obtained directly on a (100)-Si substrate with PtSi, whereas, this not the case with Au (i.e., wires are 110 or 111, but never 100). We will also discuss the incorporation of Pt in NWs and the origin of the “tapering effect” for both catalysts (i.e., metal inclusions vs. uncatalyzed Si growth at sidewalls). Finally, the electrical and mechanical properties of single NWs were studied using scanning probe microscopy. For instance, PtSi-based NWs show intrinsic resistivities of 40-500 MOhm and Young moduli in the 80-200 GPa range for diameters of 30-100 nm. A comparison of the electrical properties of single Au and PtSi-catalyzed NWs will be given. For these cases, the I(V) characteristics are qualitatively explained by an MIMS metal/ insulator/metal/semiconductor model originating from the AFM tip/insulator (native oxide shell)/PtSi/Si NW + substrate system.
11:30 AM Q12.4
Theoretical and Experimental Studies of Au-Ge Liquidus Behavior in Germanium Nanowire Nucleation and Growth. Hemant Adhikari1, Ann F Marshall2, Christopher E.D. Chidsey3 and Paul C McIntyre1; 1Materials Science and Engineering, Stanford University, Stanford, California; 2Geballe Laboratory for Advanced Materials, Stanford University, Stanford, California; 3Chemistry, Stanford University, Stanford, California.
In 3-dimensional nanoelectronics, vertically aligned nanowires have been proposed to provide a solution to attain ultra high density nanoscale device arrays. In this paper, we present results of growth of vertically aligned single-crystal germanium nanowires (GeNWs) at temperatures of 350°C or less by metal nanoparticle-catalyzed chemical vapor deposition. We have observed homoepitaxial growth of GeNWs from planar Ge crystals along <111> and <110> directions. As shown by others, we also observed single crystal nanowire growth far below the bulk eutectic temperature, but intriguingly find that temperatures close to bulk eutectic are required for efficient nucleation of epitaxial nanowires on Ge substrates. Our results indicate that the range of process conditions for growth of single crystal GeNWs and those for efficient nucleation of epitaxial GeNWs on Ge surfaces are not identical. Because wires grown at higher temperatures are tapered, a two-temperature growth procedure was used to obtain epitaxial GeNWs of constant diameter. To understand the nucleation of nanowires from gold catalyst particles and to test whether the Vapor-Liquid-Solid mechanism is actually responsible for the growth of nanowires, it is important to understand the phase equilibrium between Au nanoparticles and germanium. Capillary effects, often represented by the Gibbs-Thomson pressure, increase the free energy of the nanoparticle catalyst and the nanowire relative to their bulk values and hence lower the eutectic temperature. NW nucleation, where the catalyst nanoparticle is initially in contact with a flat Ge surface, and NW growth, where it is in contact with a GeNW, are very different situations. We have calculated the equilibrium phase diagrams both for the case of a Au-rich liquid nanoparticle in contact with flat Ge (nucleation) and for the case of a Au-rich liquid nanoparticle in contact with the nanowire (growth). We have also derived limiting expressions for the metastable liquidus for the Au-Ge binary system when a nano-scale liquid droplet is supersaturated with Ge during GeNW growth. Quantitative results obtained from these calculations are consistent with our experimental observations.
11:45 AM Q12.5
Silicon Nanowire Synthesis on Metal Implanted Silicon Substrates. Thomas Stelzner1,2, Gudrun Andrae2, Elke Wendler1, Werner Wesch1, Danilo Zschech3, Martin Steinhart3, Alexey Milenin3, Ulrich Goesele3 and Silke Christiansen3,4; 1Institute of Solid State Physics, Friedrich Schiller University, Jena, Germany; 2Laser Technology, Institute of Physical High Technology, Jena, Germany; 3Max Planck Institute of Microstructure Physics, Halle, Germany; 4Physics, Martin Luther University Halle-Wittenberg, Halle, Germany.
Various methods have been developed to synthesize one-dimensional nanostructures including vapor-liquid-solid (VLS) growth. VLS growth using chemical vapor deposition (CVD) is based on the cracking of a gaseous precursor, such as silane (SiH4), and the supersaturation of nano-sized liquid droplets of a catalyst metal, such as gold, forming a low temperature eutectic with silicon. The gold-silicon alloy droplet becomes supersaturated with silicon and nanowire growth starts to occur at the solid-liquid interface. We introduced gold catalyst by ion implantation, a few 10 nm below the silicon wafer surface and studied how gold droplets of a few 10 nm in diameter reach the substrate surface upon annealing and how these droplets yield the growth of silicon nanowires (SiNWs) by CVD. Gold implantation with high ion fluences leads to the amorphization of a surface near layer (~30nm) of the silicon wafer. In this amorphized silicon the gold diffusion is eased by diffusion coefficients four orders of magnitude higher than in crystalline silicon. After annealing at 350°C for 1 hour, amorphous or nanocrystalline gold nanoparticles of a few nanometers in diameter form throughout the entire amorphous silicon (a-Si) layer. At higher annealing temperatures (>600°C, 1 hour) the gold particles crystallize, grow and reside essentially at the wafer surface. In addition, the a-Si re-crystallizes, essentially epitaxially to the underlying wafer. The gold particles at the wafer surface can initialize VLS growth of nanowires in a CVD process. The approach to deposit the catalyst for VLS growth by using ion implantation has the potential to implant through wafer level stencil masks thereby obtaining regular catalyst nanopatterns on wafer-level. First results will be presented using a diblock-copolymer mask for patterning. Moreover, spatially resolved ion implantation and nanowire growth can be obtained by using a focused ion beam to insert the metal catalyst. Furthermore, the new implantation based templating method is less prone to oxide formation that alters the epitaxial growth of silicon nanowires, since the metal droplets reach the surface only upon annealing, i.e. during the CVD process if desired. This might be especially important when using metals other than gold with a better compatibility to semiconductor technology such as Ga, In, or Al, which oxidize easily.
Chair: Prabhakar Bandaru
Wednesday Afternoon, November 29, 2006
Room 312 (Hynes)
1:30 PM *Q13.1
Nanogenerators - Working Principle and Potential Applications. Zhong L. Wang, School of Materials Science and Engineering, Georgia Institute of Technology, Atlanta, Georgia.
Developing novel technologies for wireless nanodevices and nanosystems are of critical importance for in-situ, real-time and implantable biosensing, biomedical monitoring and biodetection. An implanted wireless biosensor requires a power source, which may be provided directly or indirectly by charging of a battery. It is highly desired for wireless devices and even required for implanted biomedical devices to be self-powered without using battery. Therefore, it is essential to explore innovative nanotechnologies for converting mechanical energy (such as body movement, muscle stretching), vibration energy (such as acoustic/ultrasonic wave), and hydraulic energy (such as body fluid and blood flow) into electric energy that will be used to power nanodevices without using battery. It also has a huge impact to miniaturizing the size of the integrated nanosystems by reducing the size of the power generator and improving its efficiency and power density. We have demonstrated an innovative approach for converting nano-scale mechanical energy into electric energy by piezoelectric zinc oxide nanowire (NW) arrays [1, 2]. By deflecting the aligned NWs using a conductive atomic force microscopy (AFM) tip in contact mode, the energy that was first created by the deflection force and later converted into electricity by piezoelectric effect has been measured for demonstrating nano-scale power generator. The operation mechanism of the electric generator relies on the unique coupling of piezoelectric and semiconducting dual properties of ZnO as well as the elegant rectifying function of the Schottky barrier formed between the metal tip and the NW. The efficiency of the NW based piezo-electric power generator is ~ 17-30%. [1] Zhong Lin Wang and Jinhui Song, Science, 312 (2006) 242-246. [2] X.D. Wang, J.H. Song, C.J. Summers, J.H. Ryou, P. Li, R.D. Dupuis, and Z.L. Wang, J. Phys. Chem. B, 110 (2006) 7720-7724. [3] http://www.nanoscience.gatech.edu/zlwang/
2:00 PM Q13.2
Dynamic Response of Carbon Nnanotube Field-effect Transistors in the GHz Range Analysed by S-parameters Measurements. Jean-Marc Bethoux2, Henri Happy2, Gilles Dambrine2, Vincent Derycke1, Marcelo Goffman1 and Jean-Philippe Bourgoin1; 1SPEC, CEA Saclay, Gif sur Yvette, France; 2IEMN, Villeneuve d'Ascq, France.
Carbon nanotube field-effect-transistors (CNTFETs) are expected to operate at extremely high frequencies up to the THz range [1]. Yet, their high impedance makes difficult the evaluation of their dynamic behavior due to the large mismatch with conventional 50 Ohms equipment. We use a chemical self-assembly technique [2] to selectively place nanotubes in predefined areas of test structures specifically designed and optimized for HF operation [3,4]. By assembling a number of carbon nanotubes in a dense network above efficient metallic gate electrodes (using a thin Al2O3 layer as the gate dielectric), we achieve driving currents in the mA range and appropriate transconductances. This allows us to measure the scattering parameters (S-parameters) of the device with a good accuracy up to some GHz. A current gain cut-off frequency (ft) of 8 GHz, and a Maximum Stable Gain (MSG) value of 10 dB at 1 GHz have been obtained [4] after de-embedding the access pads using 'open structures' (prepared in the exact same conditions and on the same wafers but without nanotubes). The reported ft and MSG are, to the best of our knowledge, the highest reported values for nanotube based transistors. From these measurements, a small signal equivalent circuit is also proposed. Considering such results, nanotube-based circuits with GHz performance are now conceivable. [1] Burke, Solid-State Electronics 48, 1981 (2004). [2] Choi et al, Surf. Sci. 462, 195 (2000); E. Valentin et al, Microelec. Eng. 61, 491 (2002); Auvray et al, Nano Lett. 5, 451 (2005). [3] Bethoux et al, IEEE Trans. on Nanotech. (2006) in press [4] Bethoux et al, Elec. Dev. Lett. (2006) in press
2:15 PM Q13.3
Radio Frequency Conductivity Measurements of Silicon Nanowire Networks. Wenchong Hu1, Alexey Kovalev1, Sarah M Dilts2, Yanfeng Wang1, Bangzhi Liu2, Suzanne E Mohney1, Joan M Redwing1 and Theresa S. Mayer1; 1Department of Electrical Engineering, The Pennsylvania State University, University Park, State College, Pennsylvania; 2Department of Materials Science and Engineering, The Pennsylvania State University, State College, Pennsylvania.
Considerable effort is being invested into characterizing the field-dependent DC electrical properties of individual and 2-D arrays of silicon nanowires (SiNWs) for on-chip NW-based electronic and sensing devices. The large surface-to-volume ratio of SiNWs also makes them attractive for wireless sensing applications that involve changes in their RF dielectric properties. In this talk, we will discuss the measurement and analysis of the RF conductivity of dense, interconnected networks of intentionally-doped SiNWs synthesized by vapor-liquid-solid (VLS) growth. Intentional n- or p-type doping was introduced during VLS growth by the addition of phosphine (PH3) or trimethylboron (TMB) to the inlet gas stream using SiH4 as the source of silicon. The networks were fabricated by casting a suspension of as-grown SiNWs with nominal 60-nm diameter and 30 μm length to form a 0.1 - 1.0 μm thick layer on a quartz disk. A cavity perturbation technique was used to measure the RF conductivity of SiNW networks and to compare to the four-point DC conductivity of individual SiNWs grown under the same conditions. The TE011 mode of a cylindrical cavity resonator with a frequency 7.7 GHz was used because it provided a high Q-factor (>20,000), which allowed measurements to be conducted using a small sample volume. We found that the sheet conductivity of the SiNW networks increased with increasing n- and p-type doping concentration, which was consistent with trends observed in DC conductivity. Data will be presented to show how the RF sheet conductivity depends on nanowire length, network connectivity, and density.
2:30 PM Q13.4
Frequency-Dependent Capacitance of Carbon Nanotube Thin Film Transistors Fabricated on SiO2 for Chemical Sensing Applications. Gokhan Esen1,2, Michael S. Fuhrer1,2, Jian Hao Chen1,3, Masa Ishigami1,3 and Ellen D. Williams1,3; 1Physics, University of Maryland at College Park, College Park, Maryland; 2Center for Superconductivity Research, University of Maryland at College Park, College Park, Maryland; 3Laboratory for Physical Sciences, University of Maryland at College Park, College Park, Maryland.
Recently [1] the capacitive response of CNT-TFT has been shown to be a good candidate for various chemical sensing applications. Here, we report measurements of the frequency-dependent gate capacitance of carbon nanotube thin film transistors (CNT-TFTs) at various frequencies between 50 Hz and 20 KHz, in various gaseous environments. CNT-TFTs are fabricated from nanotube thin films grown by chemical vapor deposition onto Si/SiO2 substrates. The transfer characteristics and the frequency dependence of the capacitance have been measured under various gaseous environments and in ultra-high vacuum. The change of the capacitance upon introduction of gases is highly dependent on frequency and DC gate voltage, and may thus give additional information about the adsorbed chemical species. The effect of the molecular dipole moment of the analyte molecules and the change in the capacitance with partial pressure of analyte molecules are currently under study. This work was supported by the Laboratory for Physical Sciences, the U.S. Army Research Laboratory Power and Energy Electronics Research Program, and the NSF-UMD-MRSEC Shared Equipment Facility. MI is supported by a Director of Central Intelligence Postdoctoral Fellowship. 1. Snow, E.S., et al., Chemical Detection with a Single-Walled Carbon Nanotube Capacitor. Science, 2005. 307(5717): p. 1942-1945.
2:45 PM Q13.5
Effects of Current-Induced Annealing on Charge and Spin Transport in Carbon Nanotube Devices. Chen-Wei Liang, Serhat Sahakalkan and Siegmar Roth; Max-Planck-Institute, Stuttgart, Germany.
Current-induced annealing has strong influence on charge and spin transport in single wall carbon nanotube field effect transistor (SWNTFET). After applying high current through carbon nanotube devices, both contact resistance and threshold voltage decreased dramatically. A progress of Coulomb-Blockade-to-Fabry-Pérot transition can be observed when annealing samples several times. On the other hand, the peaks of Column-Blockade, which were originally asymmetrical to the zero gate voltage, became symmetrical. This indicates that the annealing does not only reduce the contact resistance but also clean the body of carbon nanotube. To study how the annealing affects the spin transport, we measured the magnetoresistance of SWNTFET with Co contacts before and after annealing. Results show that the magnetoresistance either became larger or inversed from positive to negative.
Chair: Andrew Rinzler
Wednesday Afternoon, November 29, 2006
Room 312 (Hynes)
3:30 PM *Q14.1
Cold Cathode Applications of Carbon Nanotubes. William Milne, Electrical Engineering Department, University of Cambridge, Cambridge, United Kingdom.
Micro and nano-structurally rich carbon materials are alternatives to conventional metal/silicon tips for cold cathode (field emission) sources. In particular, carbon nanotubes exhibit extraordinary field emission properties because of their remarkable electrical conductivity, their high aspect ratio "whisker-like'' shape for optimum geometrical field enhancement, and excellent thermal stability. In this talk I will describe the growth of CNTs and their application in a range of cold cathode field emission based systems including their use in SEM sources, emitters for use in microwave amplifiers and in Field Emission Based displays (1). 1 “Carbon nanotubes as field emission sources” W.I.Milne, K.B.K. Teo, G.A.J. Amaratunga, P.Legagneux, L,Gangloff, J.P. Schnell, V.Semet, V.T Binh and O Grtoening, Journal of Materials Chemistry 14 (6): 933-943 2004
4:00 PM Q14.2
In search of Potential Cold Cathodes: Are Carbon Nanotubes the Candidate? Sanju Gupta, Physics and Materials Science, Missouri State University, Springfield, Missouri.
Materials science is playing a dramatic role in discovering new materials with tailored physical properties. Cold cathodes/field emitters are one of the examples. Electron field emitting materials are of vital importance for a variety of vacuum microelectronic devices including field emission displays for flat panel displays, electron microscopes, X-ray generators, and vacuum lamps. This is the driving force to investigate the advanced nanostructured carbons as cold cathodes as one of the potential candidates. Recently, they are also being proposed for thermionic power generators. The rationale is that reducing one or more dimensions of a system below some critical length changes the systems’ physical properties, where carbon nanotubes (CNTs) in the class of carbon nanostructures serve as a model example [1, 2]. We will present the synthesis and characterization of vertically aligned multiwall and single-/double-wall carbon nanotube films using a microwave plasma-assisted chemical vapor deposition technique. Recent advances in their synthesis, processing, and characterization indicate that the above mentioned potential is slowly being realized. Experiments showed that by continuous reduction in the thickness of the catalyst film produces hollow concentric tubes in contrast to bamboo-like multiwalled tubes with larger thickness. In relation to synthesis, the proposed growth model in terms of thermodynamic parameter will be discussed. To assess the electron field emission properties, besides the traditional field emission (I-V) properties, temperature dependent field electron emission microscopy (T-FEEM) enabling real-time imaging of electron emission providing information on emission site density, temporal variation of the emission intensity, and insight into the role of adsorbates from nanotube films will be discussed. Physics based models (such as negative or low electron affinity, geometric enhancement, surface dipole, tunneling due to adsorbates, structure modification due to doping etc.) will be described to support the experimental observations in addition to weak thermionic contribution. These findings indeed provided a great insight into the field emission mechanism and a contrasting comparison between small and large diameter carbon nanotubes. [1] Gupta et. al. Appl. Phys. Lett. 86, 063109 (2005). [Virtual Journal of Nanoscale Science and Technology, October 11 Issue, 2004]; [2] Gupta, et. al. J. Appl. Phys. 95, 8314 (2005). [Virtual Journal of Nanoscale Science and Technology, July 14 Issue, 2004].
4:15 PM Q14.3
Quantum Transport and Trap Effects in Tunneling Rate Measurements of Metal Nanocrystal Based Carbon Nanotube Memory. Udayan Ganguly1,2, Tuo-Hung Hou3 and Edwin Chichuan Kan3; 1Nanotechnology, NASA Ames Research, Moffet Field, California; 2Dept. of Materials Science and Engineering, Cornell University, Ithaca, New York; 3School of Electrical and Computer Engineering, Cornell University, Ithaca, New York.
The device structure with metal nanocrystal (NC) charge storage and carbon nanotube (CNT) 1D channel has been demonstrated [1]. Prototype devices show improved retention characteristics with composite high-k tunneling dielectric integration. For the read operation, CNT transport in the presence of charges in the vicinity has been theoretically addressed to confirm the experimental observation of single electron sensitivity at room temperature [2]. The quantum transport during programming and retention involves tunneling charge between a 1D CNT to either a 0D trap or a 3D NC through a thin dielectric. The electrostatics of the device which is highly 3 dimensional shows significant advantage over the planar nanocrystal memory structure [3] and requires a 3D quantum transport formulation. In this report, we present the room temperature and cryogenic tunneling rate measurement and the corresponding quantum transport modeling. Programming rates measured at room temperature for NC-CNT memory devices and control CNT-FETs without NC give similar performance because of the large trap density in the tunneling dielectric and the resultant hopping transport. To separate the contribution of traps and NC, low temperature (10K) programming measurements were done with pulse widths of 1μs to 1s for biases ranging from 3V to 15V. The measurements show three clear regimes of quantum transport. The direct tunneling regime occurs that low biases with the NC samples showing faster tunneling, which can be ascribed to the large density of states of the metal NC and electrostatic field focusing [4] compared to a trap. Following the direct tunneling regime, the FN tunneling regime dominates at higher biases producing larger tunneling rates. The density of states of traps in the dielectric plays an important role which is the same for devices with and without NCs. However, the metal NCs still produce higher electric fields from potential contour repulsion that enhance tunneling. The FN tunneling regime is responsible for charge injection until the charging of the NC or traps causes sufficient electric field reduction so that the direct tunneling dominates again. The bias dependant charging rates for structure with tunneling dielectric of single-layer SiO2 and composite SiO2/Al2O3 bi-layer dielectrics have been characterized. These measurements provide the assessment of tunneling rates in different regimes of multi-dimensional tunneling, which provides a basis for quantum transport models from the fundamental perspective and guides to practical design rules for the nanoscale memory devices from the technological perspective. References: [1] U. Ganguly et al, Appl. Phys. Lett., vol. 87, 43108, 2005. [2] J. Guo et al, J. Appl. Phys. 99, 084301, 2006. [3] U. Ganguly et al, IEEE Transactions on Nanotechnology (under review). [4] C. Lee et al, IEEE Elec. Dev. Lett., vol. 26, pp. 879, 2005
4:30 PM Q14.4
A Compact Single-Walled Carbon Nanotube Transistor Integrated with Silicon MOSFET Using a Single Common Gate. Hao Lin1, Yong Wook Park2 and Sandip Tiwari3; 1Applied and Engineering Physics, Cornell University, Ithaca, New York; 2Namseoul University, Chonan-si, South Korea; 3Electrical and Computer Engineering, Cornell University, Ithaca, New York.
Carbon nanotubes potentially provide high current drive and high transconductance in transistors because of superior electron and hole transport, and are also potentially useful as sensors. However, integration of carbon nanotube, control of placement, good contacts, and suppression of instabilities and drift continue to be a problem. Compactness of structures is essential in most electronic and sensing applications. In this work, we have implemented a compact integrated structure with a silicon MOSFET on the bottom and a carbon nanotube MOSFET on the top, both sharing a common gate. Such a structure is an inverter and a prototype for a sensor where the nanotube is exposed to the ambient. Its uniqueness is in sharing the gate - a buried gate. Difficulties in making local back gate device comes from the need for planarization of the back gate to the surrounding insulating material - usually silicon dioxide. In our novel approach of fabricating single-walled carbon nanotube local back gate, devices with good planarization and thin gate dielectric film are achieved. Damacene process was employed to obtain the planarized gate structure. Both polysilicon and tungsten are employed as gates using chemical mechanical polishing. The control oxide for the nanotube is either a grown oxide from polysilicon or a deposited oxide for tungsten. A variety of structures, with oxide thickness of 8-10 nm have been fabricated using an approach that is compatible with silicon technology and allows integration of carbon nanotube devices on top of CMOS circuits. In addition to the devices and their chracteristics, we also demonstrate an inverter with a p-type CNTFET placed directly on top of an n-type MOSFET, both of which are controlled by the same polysilicon gate in between. The approach provides high density and potentially provides high local gain for ultra-sensitive detection using CMOS circuits underneath.
4:45 PM Q14.5
Synthesis of Metal-Oxide-Metal (MOM) Heterojunction Nanowires for Chemical Sensing, Ferroelectric, and Piezoelectric Functions. Edward D Herderick, Jason S. Tresback, Alexander L. Vasiliev and Nitin P. Padture; Materials Science and Engineering, The Ohio State University, Columbus, Ohio.
There is growing interest in the field of nanoelectronic devices, where nanoscale building blocks, such as nanowires (metals, semiconductors, oxides), are fabricated in isolation and assembled into nanocircuits. In the case of functional oxides, currently all-oxide nanowires are assembled across metal contact-pad electrodes to create devices, where the oxide active region is determined by the distance the spanning the electrodes. In this context, metal-oxide-metal (MOM) heterojunction nanowires, where two Au nanowires (50-100 nm diameter and several microns in length) are separated by a nanoscale segment (50-100 nm diameter and length) of a functional oxide, offer several advantages over all-oxide nanowires. Here the metal interconnects are integrated within the building block making them more suitable for large-scale assembly, as well as providing Schottky junctions and cataylsis sites for sensing applications. In addition, the small interfacial area of the metal interconnects and the oxide offer the possibility of measuring coupled electronic properties of these nanoscale oxides without substrate effects. We have developed a novel, template-based method for the synthesis of Au-ZnO-Au, Au-TiO2-Au, and Au-BaxSr(1-x)TiO3-Au nanowires. Results from the synthesis and characterization of these MOM nanowire structures will be presented, as well as a discussion of relevant properties and device applications.
Chairs: Prabhakar Bandaru, Morinobu Endo, Ian Kinloch and Apparao Rao
Wednesday Evening, November 29, 2006
8:00 PM
Exhibition Hall D (Hynes)
Q15.1
Length Characterization of Single Walled Carbon Nanotubes using Resonance Raman Spectroscopy. Shin Grace Chou1, H. B. Son2, M. Zheng3, Ado Jorio4, R. Saito5, Gene Dresselhaus6 and Mildred Dresselhaus2,7; 1Chemistry, MIT, Cambridge, Massachusetts; 2EECS, MIT, Cambridge, Massachusetts; 3DuPont Central Research and Development, Experimental Station,, Wilmington, Delaware; 4Physics, Universidade Federal de Minas Gerais, Belo Horizonte, Brazil; 5Physics, Tohoku University, Sendai, Japan; 6Francis Bitter Magnet Laboratory, MIT, Cambridge, Massachusetts; 7Physics, MIT, Cambridge, Massachusetts.
Using different excitation laser energies, a systematic study is presented on the ratio between the disorder-induced D band and the G band for single walled carbon nanotubes samples with different average nanotube length. The study draws the analogy between the crystallite size of nanographite samples and single walled carbon nanotubes of finite length. The second order G' band has shown little dependence on nanotube length, whereas a direct correlation can be drawn between the D/G band ratio to nanotube length. The correlation observed between the D to G-band intensity ratio and nanotube length indicates that the D to G-band intensity ratio can be used as a qualitative gauge for estimating the average nanotube length. The contributions of the resonant metallic nanotubes to D/G ratio are also discussed. The MIT and Dupont authors acknowledge support under the Dupont-MIT Alliance, NSF Grants, DMR 04-05538, and INT 00-00408. The Brazilian authors acknowledge support from CNPq and the Instituto de Nanosciencias, Brazil. R.S. acknowledges a Grant-in-Aid (No. 13440091) from the Ministry of Education, Japan
Q15.2
Raman Spectroscopy Studies on Pure Carbon Nanotube Fibres. Anna Moisala, Marcelo Motta, Ian A. Kinloch and Alan H. Windle; Dept. of Materials Science and Metallurgy, University of Cambridge, Cambridge, Cambs., United Kingdom.
We have previously introduced a catalytic gas-phase process for carbon nanotube (CNT) formation and collection in bulk quantities (Li et al. Science 304, 2004, 276-278). The major benefit of this process is the ability to collect the CNTs continuously as pure fibres or films. The method can be used to produce either multi- or single/double-walled nanotube fibres depending on the process conditions and reaction chemistry. Raman spectroscopy has been widely used for the characterisation of carbon nanotubes (e.g. Dresselhaus and Eklund Adv. Phys. 49, 2000, 705-814). This work discusses the use of Raman spectroscopy to study various aspects of the nanotube fibres. Polarised Raman measurements were carried out to study the nanotube and nanotube bundle alignment within the fibres. In situ Raman spectroscopy was used during mechanical testing of the fibres to study the deformation of the nanotubes upon stretching. The nanotube types were studied at various laser wavelengths (514, 633, 785 and 830 nm) and applying either holographic (notch) or near excitation tuneable (NExT) filters, which allowed the measurement down to approximately 10 cm-1 Raman shift. Characteristics of both single- and multiwalled nanotubes were observed in a single spectrum. These unique microstructures were also studied by scanning and transmission electron microscopy and X-ray diffraction.
Q15.3
EDLC Properties Depend on the Diameter of Mass Produced Multi-walled Carbon Nanotubes. Yong-Jung Kim1, Yusuke Abe2, Takashi Yanagiura2, Masaaki Kitani2, Tsuyoshi Kodama2, Keita Higuchi2 and Morinobu Endo1,2; 1Institute of Carbon Science & Technology, Shinshu University, Nagano, Nagano, Japan; 2Eletrical and electronics engineering, Shinshu University, Nagano, Nagano, Japan.
Two types of mass produced multi-walled carbon nanotubes (MWNTs, which is generally known as a VGCF) with different diameters (the trade mark of the thicker one is VGCF® and of the thinner one, VGNF®) have been investigated for their potential for use in electric double layer capacitors (EDLCs). The variation aspects of the two MWNTs by KOH activation depend on their diameters. The capacitance enhancement and specific surface area (SSA) on KOH activation is more drastic for the thicker MWNT (VGCF®). The VGCF-KOH 500 exhibits a capacitance enhancement as much as 30 times greater (37.2 F/g) than that of the as-grown materials (1.2 F/g), under the conditions of charging up to 3.5V and discharging at a current density of 10mA/cm2. Interestingly, only for the case of the thinner MWNT (VGNF®), selective attack on its amorphous carbon impurity has also been observed, but only the case of thinner MWNT (VGNF®) as demonstrated from both SEM observations and Raman spectra. Consequently, the results of this study will provide the insight into the potentiality of using MWNTs for EDLC electrodes, which would enable cheapest production cost among the various types of carbon nanotubes.
Q15.4
Four-Terminal Conductivity Measurement Using PtIr-Coated Carbon Nanotube STM Tips. Shinya Yoshimoto1, K. Kubo1, H. Okino1, R. Hobara1, I. Matsuda1, Y. Murata2, H. Konishi2, M. Kishida2, S. Honda2, M. Katayama2 and S. Hasegawa1; 1Graduate School of Science, the University of Tokyo, Tokyo, Japan; 2Graduate School of Engineering, Osaka University, Osaka, Japan.
A multi-probe scanning tunneling microscope (STM) is expected to be a very powerful tool for investigating electronic transport properties of nano-materials and -structures [1]. But conventional STM tips, typically 100 nm in radius of curvature at the tip end, are not small enough for the nano-scale measurements because the minimum probe spacing is limited by the size of a tip apex. It was about 200 nm by using conventional STM tips. A carbon nanotube (CNT) tip can be a breakthrough for it, because CNTs have smaller radii and higher aspect ratio. It makes the minimum probe spacing ten times smaller than that by the conventional metal tips. However, it is difficult to control the contact resistance between a CNT and a support metal tip. It scatters from 10 kΩ to 10 MΩ, which prevents stable electrical measurements. We have developed metal-coated CNT tips as conductive probes for transport measurements [2], and investigated their electrical and mechanical characteristics [3]. We have found that PtIr-coated CNT tips have high conductivity (typically less than 10 kΩ) and high mechanical stability. CNT tips used here were fabricated by following procedure. 1) Multi-walled carbon nanotubes, typically 10 nm in diameter, were connected to an electrochemically etched tungsten tip by AC dielectrophoresis [4]. 2) Amorphous carbon was deposited around the contact between the CNT and the tungsten tip by electron beam irradiation in a scanning electron microscope (SEM). 3) The CNT tips were heated in a vacuum chamber. 4) They were wholly coated with a 6-nm thick PtIr thin film by using pulsed laser deposition (PLD) [5]. In this research, we have conducted multi-terminal conductivity measurements using these PtIr-coated CNT tips in an independently driven four-tip STM system [1]. We performed four-terminal conductivity measurements on individual Co Silicide nanowires at room temperature [6]. The four-terminal resistance was proportional to the probe spacing between voltage probes, meaning diffusive conduction. The resistivity was 57Ω per μm. Four-terminal resistance at the minimum probe spacing was 1.2Ω, which indicate that the probe spacing was surely reduced to around 20 nm. References: [1] S. Hasegawa, et al., Current Appl. Physics 2, 465 (2002). [2] T. Ikuno, et al., Jpn. J. Appl. Phys. 43, L644 (2003). [3] S. Yoshimoto, et al., Jpn. J. Appl. Phys. 44, L1563 (2005). [4] J. Tang, et al., Adv. Mater 15, 1352 (2003). [5] T. Ikuno, et al., Jpn. J. Appl. Phys. 42, L1356 (2003). [6] H. Okino et al., Appl. Phys. Lett. 86, 233108 (2005).
Q15.5
Abstract Withdrawn
Q15.6
Tunneling Electron Transport of Silicon Nanochains Studied by in-situ Scanning Electron Microscopy. Hideo Kohno and Seiji Takeda; Osaka University, Toyonaka, Osaka, Japan.
Electron transport and field emission properties of one dimensional materials such as nanowires and nanotubes have been studied intensively. We report electron transport properties of silicon nanochains, in which silicon nanoparticles are covered with and connected alternately by oxide forming periodically-heterostructured nanowires [Kohno and Takeda, Appl. Phys. Lett. 73, 3144 (1998)., Kohno and Takeda, e-J. Surf. Sci. Nanotech. 3, 131 (2005)], measured at high bias voltages up to 120 V. I-V curves are measured in-situ using a micro-manipulator in a scanning electron microscope (SEM). Thus, the distance between silicon nanochains and the microprobe can be controlled under SEM observation. The molybdenum substrate on which silicon nanochains are grown is grounded and the tungsten microprobe is positively biased. We observe significant current increase at the bias voltages of several tens of volts when the microprobe is both separated by about 1 micrometer and attached to silicon nanochains. Fowler-Nordheim (FN) plots of the I-V data above 100 V show that the transport properties of silicon nanochains at such high bias voltages can be described well by the FN law. Accordingly, we conclude that the field-induced tunneling current is dominant even when the microprobe is attached to silicon nanochains. Our results suggest a possible application of Si nanochains to cold electron emitters.
Q15.7
Electronic and Mechanical Properties of Super Carbon Nanotube Networks. Vitor R. Coluci1, Socrates O. Dantas2, Ado Jorio3 and Douglas S. Galvao1; 1Applied Physics, State University of Campinas, Campinas, Sao Paulo, Brazil; 2Physics Department, Federal University of Juiz de Fora, Juiz de Fora, Minas Gerais, Brazil; 3Physics Department, Federal University of Minas Gerais, Belo Horizonte, Minas Gerais, Brazil.
Carbon based materials present an enormous variety of forms and properties. Among these structures we can mention graphite, diamond, fullerenes and nanotubes. Carbon nanotubes represent potential building blocks for different areas in nanotechnology. They can be formed either by one single (single walled carbon nanotubes) or by multiple graphite sheets (multi walled carbon nanotubes). In particular, single walled carbon nanotubes (SWNTs) can be interconnected by junctions. These interconnections can present different shapes such as X-, T-, and Y-like arrangements, directly associated with the respective junctions which form them. In this work we use different types of junctions to build single walled carbon nanotubes networks. For instance, the use of Y-like junctions allows the construction of a hexagonal network, named super-graphene (SG), which is heuristically generated replacing the carbon-carbon bonds of the graphene architecture by SWNTs, and the carbon atoms by junctions. From the SG a new tubular carbon structure was proposed. We named this tubular structure super-nanotube which represents a carbon nanotube made of carbon nanotubes. Such procedure can be repeated several times, generating a fractal structure. This procedure is not limited to carbon nanotubes and can be easily modified to be applied to other materials and to other network arrangements through the use of different types of junctions. We investigated the electronic and mechanical properties of hexagonal and squared SWNTs networks. Tight-binding total energy and density of states calculations showed that the ‘super’- graphene sheets and the respective tubes are thermodynamically stable and predicted to present metallic or semiconducting behaviour. Impact molecular dynamics simulations for nets using Brenner-Tersoff type potentials showed that these structures exhibit remarkable impact-absorbing properties and could be the basis to new classes of elastic materials. Using classical molecular mechanics calculations we predicted bulk modulus of the order of 10 GPa for the investigated networks.
Q15.8
Three-way Electrical Gating Characteristics of Metallic Y-junction Carbon Nanotubes. Jeongwon Park1, Prabhakar Bandaru1 and Apparao M. Rao2; 1Materials Science and Engineering Program, University of California, San Diego, La Jolla, California; 2Department of Physics and Astronomy, Clemson University, Clemson, South Carolina.
Y-junction based carbon nanotube (CNT) transistors exhibit novel switching behavior and have the structural advantage that the electrical gate for current modulation can be formed by any of the three constituent branches. The gating characteristics of metallic Y-CNT morphologies were studied1 by measuring the output conductance (gd=(∂Ids/∂Vds)Vgs) and transconductance (gm=(∂Ids/∂Vgs)Vds). The voltage gain is also calculated by considering the voltage change (ΔVds) for a given increment in gate voltage (ΔVg), at a constant value of the source-drain current. The voltage gain when the stem is used for electrical gating is ~6, while the value drops to ~0.3 when a branch is used for a gate. By analyzing the gd and gm of metallic CNT Y-junctions, we put forward the idea that the switching efficiency and gain depend on the branch diameter and is electric field controlled. Based on these principles, we will talk about a design for a Y-junction based CNT switching device, with tunable electrical properties. In addition monitoring switching behavior will be presented. References 1. J. Park, C. Daraio, S. Jin, P. R. Bandaru, J. Gaillard, and A. M. Rao, Three-way electrical gating characteristics of metallic Y-junction carbon nanotubes, Appl. Phys. Lett. 88, 243113 (2006)
Q15.9
Defect Induced Enhancement of Carrier Transport in Individual Multiwalled Carbon Nanotubes and their Networks. Saurabh Agrawal1, Makala S Raghuveer1, Rampi Ramprasad2 and Ganapathiraman Ramanath1; 1Materials Science and Engineering, Rensselaer Polytechnic Institute, Troy, New York; 2Department of Materials Science and Engineering & Institute of Materials Science, University of Connecticut, Storrs, Connecticut.
Ballistic carrier transport in oriented bundles of metallic singlewalled carbon nanotubes (CNTs) is attractive for nanodevice wiring applications. The-state-of-the-art, however, does not yet allow scalable separation or selective growth of metallic singlewalled CNTs in exclusion to semiconducting ones. Large diameter multiwalled CNTs (30-50 nm) could potentially be an alternative solution because their outershells, which provide the primary carrier transport pathway, are either metallic or quasi-metallic with similar and reproducible electrical characteristics. Here we demonstrate that the conductance of individual CNTs and their networks can be enhanced by controlled defect creation and increasing the electric field. In particular we show that defect induced cross-linking of adjacent shells promotes thermally activated carrier injection opening up new conduction pathways. At low nominal electric fields (ξ). carrier injection in individual CNTs, films with highly oriented CNTs or randomly dispersed CNTs, is characterized by a ~ 60-250 meV energy barrier that is weakly dependent on ξ. This barrier is in good agreement with that for radial transport of carriers between metallic shells, computed using density functional theory (DFT) calculations on (5,5)-(nout,nout) double-walled CNTs with nout = 8-12. Comparing electrical characteristics of individual CNTs and CNT films in different configurations reveals that carrier transport occurs by field enhanced thermal hopping between adjacent outer shells in each CNT, and across outermost shells of overlapping CNTs. On increasing ξ, inter-shell transport increases non-linearly due to field-enhanced thermionic carrier injection across the outermost shells. These findings helped us enhance multishell transport in CNTs by cross-linking the shells using controlled defect creation. Cross-linking led to ~ 150% improvement in conductivity and upto 50% reduction in the energy barrier for transport reinforcing our theory. The results were reproducible over individual CNTs, and their random and aligned assemblies. A phenomenological model capturing the salient aspects of trans-shell carrier injection and multi-shell carrier transport in multiwalled nanotubes is presented. Our findings open up the possibility of utilizing the inner shells in multiwalled CNTs as multiple parallel channels for efficient carrier transport for device applications.
Q15.10
Wide Bandgap Semiconductors - Nanowires of p- and n-type Silicon Carbide Bettina Friedel and Siegmund Greulich-Weber; Department of Physics, University of Paderborn, Paderborn, Germany.
Monocrystalline nanowires of cubic silicon carbide were synthesized using a combined sol-gel and carbothermal reduction process in which tetraethoxysilane was used as primary silicon and sucrose as carbon source. The diameters of the as-grown nanowires varied depending on process parameters from several tens to several hundreds nanometers, whereas the length of the wires was located in the millimetre region. By precisely controlling the atomic ratio of Si / C, silicon carbide nanowires were synthesized exclusively and pure without the presence of residual carbon or unwanted silica, thus leads to semi-insulating behaviour. Supported by their consistence the silicon carbide nanowires can be processed to textile or felt structures and are therefore usable for many applications such as for fireproof clothing, high temperature or chemical filters and composite materials. Additionally during sol-gel synthesis the silicon carbide nanowires were easily doped to achieve p- or n-conduction, guiding to new applications in the field of wide bandgap semiconductors. The structure of 3C-SiC nanowires was determined using scanning electron microscopy, X-ray diffraction, nuclear magnetic resonance and fourier transform infrared spectroscopy. The electronic properties were studied using electron paramagnetic resonance spectroscopy and current-voltage measurements.
Q15.11
Electric Field Dependence of the Raman Spectra in Single-wall Carbon Nanotube Thin Film Networks. Giovanni Fanchini, Husnu Emrah Unalan and Manish Chhowalla; Rutgers University, Piscataway, New Jersey.
We report multiwavelength electrical Raman measurements in transparent and conducting single-wall carbon nanotube (SWNT) thin film networks. Application of constant external electric fields results in downshifts of the Raman D and G modes, associated with a reduction of their intensity. The intensity of the radial breathing modes increases at increasing electric fields in metallic (m-) SWNTs, while it decreases in semiconducting (s-) SWNTs. We propose a simple model explaining these phenomena in terms of four features typical of transparent and conducting SWNT thin films: 1) Strong Joule heating of some m-SWNTs in the film, 2) linear electronic band edges and electronic density of states at the Fermi level tending to zero in m-SWNTs, 3) anomalous electron screening enhanced by the1-D nature in m-SWNTs and 4) dependence of the electron relaxation times on the network density, implying that the electron relaxation processes are mainly inter-tube processes. Our findings rule out that pre-conditioning SWNT thin film transistors by high field pulses destroys large amounts of m-SWNTs, as often believed, and corroborate the attribution [2] to Kohn anomalies of the strong electron-phonon coupling in the Raman-active optical branches of metallic graphitic materials. 1. H E Unalan, G Fanchini, A Kanwal, A Du Pasquier and M Chhowalla, Nano Lett, 6 (2006) 677 2. S Piscanec, M Lazzeri, F Mauri, AC Ferrari, J Robertson, Phys Rev Lett 93 (2004) 185503
Q15.12
Measurement of Minority Carrier Diffusion Lengths in Semiconductor Nanowires. Jonathan E. Allen, Yi Gu, Eric R. Hemesath and Lincoln J. Lauhon; Materials Science and Engineering, Northwestern University, Evanston, Illinois.
Semiconductor nanowires have been identified as potential candidates to replace traditional CMOS technology, but quantitative measurements of device figures of merit are not abundant. This is due in part to the challenge of identifying and/or adapting characterization techniques that are suitable for very small devices. The realization of reliable nanowire device technology can therefore be facilitated by the development of techniques to quantify fundamental metrics including carrier mobility, diffusion, and lifetime. Here we report the first quantitative measurement of minority carrier diffusion lengths in semiconductor nanowires using an electron beam induced current (EBIC) technique. Two-terminal Schottky diode devices were fabricated using n-type Si nanowires grown via the vapor-liquid-solid growth mechanism. Ohmic contacts were made using Ni while Au was used for Schottky contacts. Scanning the devices with an electron beam provided a highly localized source of excess carriers whose effects were monitored as changes in device current. In Schottky diode devices under reverse bias, this response was localized near the Schottky contact and decayed exponentially along the device channel with characteristic decay constant, L, the minority carrier diffusion length. In heavily doped (Si:P ~ 500:1) n-type silicon nanowires of diameter 20 nm, L was found to be ~30 nm, and was insensitive to changes in source-drain bias and electron beam power. L was also measured as a function of nanowire diameter and doping concentration in order to investigate the role of surface effects and carrier-carrier interactions. Analysis of the generation region suggests that secondary electrons play a negligible role in the measured response. The measurement resolution is sufficient to provide a reliable measurement of the 30 nm diffusion lengths reported here and is significantly higher than our previously reported resolution using scanning photocurrent microscopy[Y. Gu et al, Nano Letters 6, 948 (2006)]. Minority carrier diffusion length is a key parameter in determining semiconductor device performance. The measurements described above can therefore contribute to the development and optimization high-performance devices based on semiconductor nanowires.
Q15.13
Absorption Cross Section and Quantum Efficiency of Pristine and Boron-doped Single-walled Carbon Nanotubes. Jeffrey L Blackburn1, Timothy J McDonald1,2, Thomas Gennett1,3, Yanfa Yan1, Kim M. Jones1, Chaiwat Engtrakul1, Kelly Knutson1, Randy J Ellingson1, Garry Rumbles1 and Michael J Heben1; 1Basic Science, National Renewable Energy Lab, Golden, Colorado; 2Applied Physics, Columbia University, New York, New York; 3Chemistry, Rochester Institute of Technology, Rochester, New York.
In interpreting single-walled carbon nanotube (SWNT) optical measurements, the absorption cross section is an important parameter that connects the measured absorption to the quantity of nanotubes in the sample. In addition, the quantum efficiency is important in evaluating the potential for SWNTs in photoconversion applications. We have recently developed a synthetic procedure for the high yield synthesis of B-SWNTs by pulsed laser vaporization. We perform photoluminescence excitation spectroscopy (PLE) and absorption spectroscopy, and note significant differences between undoped laser-generated SWNTs and boron-doped SWNTs. We note a diameter dependence of the relative PL quantum yield for the boron-doped sample versus the undoped SWNTs, and absorption spectroscopy reveals a higher optical density for the boron-doped SWNTs. These differences may have several origins, including a higher affinity of the b-doped nanotubes for the surfactant molecules, different length distributions for the doped and undoped suspended SWNTs, and/or a change in the oscillator strength of interband transitions due to doping. In order to assess these issues, it is essential to accurately measure the total weight of nanotubes in each suspension. The total mass of nanotubes in suspension has so far only been estimated, and the absorbance of each nanotube type is difficult to obtain because of the overlap of other tube species present in a sample. In order to obtain an independent measurement, we have developed a technique to determine the mass of carbon attributed to nanotubes in suspension. This technique involves surfactant removal followed by high-temperature oxidation and mass spectrometry measurements of the oxidation products. We will couple this measurement with a novel technique to determine the absorbance of each nanotube species by utilizing PLE spectra to guide the fitting of each nanotube’s absorption feature. The end result is an accurate value for the absorption cross section of each nanotube and an accurate comparison for the cross section between a variety of nanotube samples.
Q15.14
aF Resolution C-V Characterization of Nano-scale FETs Using Ambient Noise and Non-linearities. Ali Gokirmak and Hazer Inaltekin; Electrical and Computer Engineering, Cornell University, Ithaca, New York.
There has been considerable interest in quasi-1-D structures due to the nature of eigen-states, eigen-energies, and density of states in the recent years. Scattering mechanisms are expected to be suppressed in quasi-1-D devices, channel width close to 10 nm or less, and mobilities as high as 107 cm2/V.s have been predicted. Accurate observation of characteristics of quasi-1-D structures, such as nanotube and nanowire FETs, require very sensitive capacitance measurements. The main difficulties in achieving aF resolution C-V characteristics have been the equipment resolution, large parasitic capacitances, system fluctuations and the ambient noise. The limitation due to the equipment resolution can be overcome by making use of the noise in the system. We will be presenting a noise enhanced C-V characterization technique that enables sub-0.1 aF resolution Cg-sd-Vgcharacterization of stable FETs using a commercial RLC meter for sufficient noise levels. The inversion layer capacitance is extracted from the ΔCg-sd-Vg information making use of the non-linear C-V characteristic of FETs, avoiding the difficulties of accurately accounting for parasitic capacitances in the measurement system. The optimum noise level, known as stochastic resonance, and the maximum resolution achievable depend on the number of measurements. The affect of noise in the resolution of measurements, optimum noise level -stochastic resonance- and the limitations of the technique will be presented through numerical calculations, simulation results and measurements performed on small-scale FETs with inversion layer capacitances less than 60 aF. Electron mobility in these small scale FETs are extracted from the experimental I-V and C-V information.
Q15.15
Ternary PtRuNi Nanocatalysts Dispersed on Multiwall Carbon Nanotubes for Methanol Electro-oxidation in Acid Medium. Yu-Kuei Hsu1, Wei-Horng Su1,2, Yan-Gu Lin3,4, Ju-Lan Yang3, Chia-Liang Sun1, San-Yuan Chen3, Chii-Ruey Lin2, Kuei-Hsien Chen1,4 and Li-Chyong Lin4; 1Institute of Atomic and Molecular Sciences, Academia Sinica, Taipei, Taiwan; 2Institite of Mechatronic Engineering, National Taipei University of Technology, Taipei, Taiwan; 3Los Alamos National Lab, Hsinchu, Taiwan; 4Center for Condensed Matter Sciences, National Taiwan University, Taipei, Taiwan.
The methanol oxidation reaction of nitrogen-containing carbon nanotube (CNx NTs)-supported PtRuNi ternary catalysts was investigated. The CNx nanotubes were grown by microwave-enhanced chemical vapor deposition on the carbon cloth and then acted as electrode and support for PtRuNi dispersion in the subsequent sputtering process. By controlling the sputtering condition, the well-separated ternary composite catalysts containing different compositions of Ni while keeping 1:1 atomic ratio of Pt/Ru with an average diameter of 3-6 nm have been successfully fabricated on the CNx nanotube. The PtRuNi alloys showed excellent catalytic activities compared to that of PtRu which was synthesized following the same way. The role of Ni as a catalytically enhancing agent in the oxidation process was examined using X-ray diffraction, scanning electron microscope, X-ray photon electron spectroscopy, cyclic voltammetry, and chronoamperometry. X-ray diffraction data showed that the inclusion of Ni in the PtRu system has effectively reduced the particle size and promoted alloy formation. X-ray photoelectron spectroscopies confirmed that the chemical states of Pt were exclusively metallic. In contrast, presence of metallic Ni, NiO, metallic Ru, and RuO2 were observed. Cyclicvoltammetry results demonstrated that the PtRuNi catalyst presented lower onset potential and larger oxidation current density compared with PtRu (1:1). The optimum Ni concentration has been determined by the combination method. Finally, temperature-dependent methanol oxidation behavior of the PtRuNi catalyst will also be discussed.
Q15.16
FET Properties of Surface Silylated Single Wall Carbon Nanotubes. Ryotaro Kumashiro1,7, Nobuya Hiroshiba1, Hirotaka Ohashi1, Takeshi Akasaka2, Yutaka Maeda3, Shinzo Suzuki4, Yohji Achiba5, Rikizo Hatakeyama6 and Katsumi Tanigaki1,7; 1Department of Physics, Graduate School of Science, Tohoku University, Sendai, Japan; 2Center for Tsukuba, Advanced Research Alliance, University of Tsukuba, Tsukuba, Japan; 3Department of Chemistry, Tokyo Gakugei University, Tokyo, Japan; 4Faculty of Science, Kyoto Sangyo University, Kyoto, Japan; 5Graduate School of Science and Engineering, Tokyo Metropolitan University, Tokyo, Japan; 6Graduate School of Engineering, Tohoku University, Sendai, Japan; 7CREST, Japan Science and Technology Agency, Kawaguchi, Japan.
Single wall carbon nanotubes (SWNTs) having semiconducting properties are promising as electronic materials for nano-scale devices in the future, and the electrical properties of SWNTs are of significantly fundamental and practical interests. It is well known that the field effect transistors (FETs) fabricated using semiconducting SWNTs show high performance in terms of the mobility. However, carriers in pristine SWNTs are mostly holes and, therefore, SWNTs -FETs usually show p-type properties. As for SWNTs, chemical carrier doping has been reported so far for controlling carrier concentration like graphite intercalations. Two major techniques in SWNTs are generally possible; one is endohedral doping and the other is the exohedral chemical modifications. It has been exemplified that doping with alkali metals can introduce electron carriers into SWNTs. Furthermore, the electrical transport properties of SWNTs were reported to be controlled by the endohedral insertion of organic molecules inside the SWNTs. A similar carrier doping could exohedrally be possible when the SWNTs surface is chemically modified. With such chemical modifications, the charge transfer from the substituent groups to SWNTs will be expected and this could modify the electronic states of SWNTs. We reported the FET properties of individual SWNTs exohedrally modified by Si-containing organic moieties, and demonstrated that p-type nanotubes can be converted to n-type ones. However, because of ununiformity of the surface-chemical modifications of SWNTs, the true effects of the exohedral modifications on FET properties were extremely difficult to be evaluated. In this meeting, we will present comparison of the FET properties of the exohedrally silylated SWNTs between separated individual and spread-sheet samples. For evaluating the FET properties, the chemically modified SWNTs have been dispersed on a FET substrate, and the measurements have been carried out at ambient temperature using a conventional method for a separated SWNTs and a spread-sheet SWNTs film. As a reference, the experiments were also made in the same manner on chemically non-modified CNTs. From the experimental results, it will be demonstrated that an n-type property can be enhanced by the exohedral modifications both in the case of the spread-sheet samples and in the case of the individual ones. We will discuss the effects of surface silylation on the electronic states of these SWNTs.
Q15.17
Performance of Thin Film Transistors from Brominated Single Wall Carbon Nanotube Networks. Giovanni Fanchini, Husnu Emrah Unalan and Manish Chhowalla; Rutgers University, Piscataway, New Jersey.
We describe the properties of transparent and conducting brominated single-wall carbon nanotube (SWNT) thin film transistors. Unlike conventional doping of SWNTs with physisorbed species (oxygen, Br2), our process leads to a tunable amount of chemically bonded bromine in the form of acyl bromide (COBr).Our experimental and theoretical results indicate that Br functionalization of SWNTs not only shifts the Fermi level, as expected from a doping agent, but also leads to the reduction of the density of electronic states at the Fermi level in metallic SWNTs. We demonstrate that these effects manifest as an increase in mobility and the on/off ratio in the Br functionalized SWNT thin film transistors (TFTs) in comparison to devices fabricated from unmodified SWNTs. Furthermore, in contrast to reports in the literature, the on/off ratio and mobility of Br doped SWNT TFTs increase with nanotube density (tubes/mm2). This indicates that Br doping leads to a change in the conduction mechanism in metallic SWNTs. Finally, we show that the Br-doped SWNT TFTs are stable in various atmospheres, in contrast to devices from unmodified SWNTs.
Q15.18
Ba6Mn24O48 Whiskers: A New Material for Batteries, Sensors and Catalysis. Ekaterina Pomerantseva1, Eugene Goodilin1, Marina Kozlova1, Valery Krivetsky2, Lyudmila Leonova2, Yuri Dobrovolsky2 and Yuri Tretyakov1; 1Department of Materials Science, Lomonosov Moscow State University, Moscow, Russian Federation; 2Institute of Problems of Chemical Physics, Chernogolovka, Russian Federation.
One-dimensional structures such as nanofibers, nanowires, nanotubes, nanobelts, nanorods and whiskers offer an extremely broad range of potential applications such as reinforced materials, molecular and nanoelectronics. Manganese-containing oxides with tunnel crystal structures hold much promise due to their cation-exchange and molecular sieve properties, catalytic activity, low cost and ecological safety. In this work, Ba6Mn24O48 whiskers (50-100 nm thick and 0.3-3 mm long) were grown from chip and non-toxic alkali - chloride fluxes at 950oC reproducibly and with a high yield. The structural analysis data confirm that the Ba6Mn24O48 phase has a composite unit cell with a tetragonal symmetry (space group I4/m, a = 18.2963(0)Å, c = 2.8509(7)Å). The tunnels of three types are formed in the Ba6Mn24O48 crystal structure including barium-free rutile-like tunnels, hollandite-like tunnels containing one row of barium ions and the tunnels with a complex shape containing two rows of Ba2+. According to SAED data, the crystals grow along the c axis and therefore the structural tunnels run along the whisker growth direction. Fast development of hydrogen energy and power source fields makes it highly promising to investigate proton and lithium insertion into the Ba6Mn24O48 whiskers is of the most interest. Proton-inserted whiskers (H-form of Ba6Mn24O48) were obtained after being immersed into the concentrated HNO3 for one week. Such a treatment results in delamination of initial whiskers and formation of fibrous nanocrystals with the thickness of 30-50 nm. The H-form of Ba6Mn24O48 whiskers is characterized by improved sorption kinetics of heavy metal cations. The electrochemical study has shown that both electronic and ionic transport contribute to the total conductivity of the material, which makes it a potential protonic conductor. Lithium can be effectively introduced into the tunnels of the Ba6Mn24O48 whiskers by interaction with LiNO3 melt at 300oC for ~ 1 hr. Alternatively, Li-inserted whiskers can be obtained by immersion into 1M LiOH/1M LiNO3 at room temperature for ~ 2 weeks. The parameter a of the lithiated Ba6Mn24O48 phase decreases significantly (a = 18.15 - 18.06 Å) while the c lattice constant remains almost unchanged (c = 2.83 - 2.84 Å). Current-voltage characteristics of the lithium-intercalated Ba6Mn24O48 whiskers reveal significant contribution of the ionic part into the total conductivity of the material. Flexible composite electrodes were prepared by ultrasonic polymerization of Ba6Mn24O48 whisker suspension in polyisovanadates. Such insertion electrodes demonstrate 50% higher tensile strength compared to V2O5 xerogel. Fabrics of Ba6Mn24O48 whiskers is a unique form of self-supporting catalyst demonstrating simultaneously new features of sensor activity.
Q15.19
Rare Earth Implanted Nanowires. Sebastian Geburt1, Daniel Stichtenoth1, Sven Müller1, Wilma Dewald1, Quan Li2 and Carsten Ronning1; 1II. Institute of Physics, University of Göttingen, Göttingen, Germany; 2Department of Physics, Chinese University of Hong Kong, Shatin, Hong Kong.
Rare earth elements embedded in suitable matrices show optical active and sharp intra-4f-transitions with long life-times. E.g. such states are necessary for the realization of Nd:YAG-lasers. It is known that the geometry of semiconductor nanowires could act as cavity; therefore, rare earth doped semiconductor nanowires may be suitable for lasing issues. Silicon, ZnO and ZnS nanowires were grown according the VLS mechanism using the vapour transport technique. The resulting material was investigated by SEM, XRD, and TEM. After growth, the nanowires were brought into solution and dispersed onto clean silicon substrates. Rare earth elements were implanted into the nanowires with different ion fluences and ion energies. The ion energy, which determines the ion range, was set to match the diameter of the nanowires. The radiation defects were healed by annealing in vacuum. The structural changes of the nanowires upon implantation and annealing were investigated in detail by TEM as well as the optical properties using PL, time resolved PL and µ-PL. In this presentation, we will present the obtained results and discuss the feasibility of rare earth based nanowire-lasers.
Q15.20
Morphology and Optical Properties of ZnO Nanorods Grown by Catalyst-assisted Vapor Transport on Various Substrates. Vitaliy Avrutin1, U. Ozgur1, N. Izyumskaya1, S. Chevtchenko1, J. Leach1, J. C. Moore2, A. A. Baski2,1, H. O. Everitt3, K. T. Tsen4, M. Abouzaid5, P. Ruterana5 and H. Morkoc1,2; 1Department of Electrical and Compputer Engineering, Virginia Commonwealth University, Richmond, Virginia; 2Department of Physics, Virginia Commonwealth University, Richmond, Virginia; 3Army Aviation & Missile RDEC, Redstone Arsenal, Alabama; 4Department of Physics and Astronomy, Arizona State University, Tempe, Arizona; 5SIFCOM, UMR 6176 CNRS-ENSICAEN, Caen, France.
ZnO-based nanostructures, including nanorods, nanobelts etc., have attracted a great deal of attention primarily due to their potential applications in light-emitting devices. Therefore, the optical properties of such nanostructures are of crucial importance. In spite of this interest, properties such as carrier relaxation in ZnO nanorods have not been studied in sufficient detail. In this study, ZnO nanorods were grown by catalyst-assisted vapor phase transport on different substrates: Si, c-sapphire, GaN, and bulk ZnO. A mixture of powdered bulk ZnO and C powder was used as a material source and placed in the high temperature zone of a three-zone horizontal furnace, and 6N Ar was employed as a carrier gas. The substrates were located downstream of the carrier gas in the lower temperature zone, and the growth temperature was 600-650 C. A thin gold film (2-5 nm thick) served as a catalyst, and the substrates were prepared with continuous and microstructured gold films. In the latter case, gold islands of 100x300 nm2 size were fabricated using nano-imprinting lithography. The morphology of the prepared structures was studied by scanning electron microscopy (SEM) and atomic force microscopy (AFM). The morphologies obtained were dependent on the substrate used. For example, SEM and tapping-mode AFM studies revealed that tightly packed ZnO nanorods ~70-100 nm in diameter grew mostly perpendicularly to the GaN substrate surface (along c-axis of the substrate), whereas a more random distribution was found for nanorods on Si. Localized conduction behavior was examined using conductive AFM (C-AFM) via indium ohmic contacts. Transmission electron microscopy (TEM) was employed to investigate the crystalline quality of the nanorods vs. the deposition conditions and substrate material. Steady-state photoluminescence (PL) recorded both at 15 and 300 K from the prepared nanorod structures showed very intense emission that was comparable to emission from high quality bulk ZnO. Salient features of PL spectra were correlated with various types of morphologies obtained on different substrates. Near field optical microscopy was employed to identify emission from individual nanorods. Time-resolved and excitation-intensity dependent PL data were obtained to investigate carrier dynamics and natural cavity formation, respectively. Time-resolved Raman and micro-Raman spectroscopy were used to study transient carrier transport and phonon modes in the ZnO nanorods. Characteristics of the ZnO nanostructures grown on different substrates are correlated to their microstructure and specifics of growth mechanisms involved.
Q15.21
Experimentally Determined Absorption Cross-sections of Solution-based CdSe and CdTe Nanowires. Vladimir Protasenko, Daniel Bacinello and Masaru Kuno; Chemistry, University of Notre Dame, South Bend, Indiana.
The absorption cross-section is an important intrinsic parameter of various low dimensional systems, such as colloidal quantum dots and semiconductor nanowires (NWs). Specifically, the cross-section enables one to estimate the number of carriers generated in a nanostructure upon the absorption of light. This has relevance to basic questions regarding material parameters such as fluorescence quantum yields. It also has practical implications, allowing one to determine the photocurrent generation efficiency of potential nanowire-based photovoltaic or photodetector devices. Systems of specific interest in the current study are CdSe and CdTe NWs made via solution-liquid-solid growth. Resulting NWs have typical diameters ranging from 5 to 10 nm in diameter and have lengths greater than 1 micron. Size distributions of as made wires range from 15-30% and intrawire diameter distributions are on the order of 5%. The wires are crystalline as determined through high resolution TEM images. Corresponding NW absorption cross-sections are estimated using a combination of high/low resolution TEM imaging, UV-visible absorption spectroscopy and inductively couple plasma atomic emission spectroscopy. The linear “absorption cross-sectional density” of CdSe and CdTe nanowires is determined to be approximately 4x10^(-11) cm^2/micron and 3x10^(-11) cm^2/micron at 488 nm (2.54 eV). When multiplied by the average length of a typical NW (1-10 micron), lower limits to actual NW cross sections are obtained. These values agree well with prior theoretical estimates for CdSe and CdTe NWs at 488 nm (2x10^(-11) cm^2 and 1x10^(-11) cm^2 respectively). The experimentally determined NW cross sections are nearly 5 orders of magnitude larger than those of comparable colloidal CdSe and CdTe QDs.
Q15.22
Ultraviolet GaN Single Nanowire Light Emitting Diode. Mariano Adolfo Zimmler1, Jiming Bao1, Joonah Yoon1,2, George Seryogin1, Ilan Shalish1, Venkatesh Narayanamurti1 and Federico Capasso1; 1Division of Engineering and Applied Sciences, Harvard University, Cambridge, Massachusetts; 2Department of Physics, Massachusetts Institute of Technology, Cambridge, Massachusetts.
Semiconductor nanowires have been shown to exhibit many of the properties needed for future integrated nanoscale electronics and optoelectronics. [1] Central to their usefulness, however, is the ability to fabricate reliable electrical contacts for nanoscale light emitting diodes and lasers. Here we use a new and general strategy for efficient injection along the length of a semiconductor nanowire [2] to fabricate a GaN single nanowire light emitting diode, which may be used as a compact ultraviolet light source. GaN nanowires catalytically grown by hybrid vapor phase epitaxy [3] were placed between a heavily doped p-type silicon wafer and a metallic thin film, serving as hole—and electron—injecting contacts, respectively. A high-resolution negative electron beam resist is used as an insulating layer between the two contacts. [2] The single nanowire electroluminescence spectrum was measured as a function of temperature and applied voltage bias. The room temperature spectrum exhibits a distinct peak at 355 nm as well as a broad luminescence centered at ~720 nm. With decreasing temperature, both features are blue-shifted (340 nm and ~700 nm, respectively, at 5 K), and their relative magnitudes change, making the ultraviolet contribution sharply dominant. [1] Huang, Y.; Lieber, C. M. Pure Appl. Chem. 2004, 76, 2051. [2] Bao, J.; Zimmler, M. A.; Capasso, F.; Wang, X.; Ren, Z. F. To appear in Nano Lett. [3] Seryogin, G.; Shalish, I.; Moberlychan, W.; Narayanamurti, V. Nanotechnology 2005, 16, 2342.
Q15.23
Active Semiconductor Nanowires for Functional Photonic Circuit Elements Carl Julien Barrelet1, Hong Gyu Park1, Yongning Wu1 and Charles M. Lieber1,2; 1Chemistry, Harvard University, Cambridge, Massachusetts; 2Division of Engineering and Applied Science, Harvard University, Cambridge, Massachusetts.
The high quantum efficiency, large refractive index and nonlinear optical coefficients make direct-gap semiconductors nanowire a unique class of material for nanophotonic devices. Nanowires with diameters smaller than the optical wavelength can radially confine optical modes and function as subwavelength waveguide, and moreover, nanowires can serve as efficient light-emitting diodes and lasers. Progress in this exciting area of nanophotonics has, however, been limited by the lack of control on the coupling of light into and out of functional nanowire elements. Here, we demonstrate a new approach to efficiently input and extract light from nanowires that that combines chemically-synthesized single nanowires coupled to lithographically-defined photonic crystal structures. Finite-difference time-domain (FDTD) calculations were used to design nanowire photonic crystal structures where the photonic band gap overlaps with the electronic band gap of cadmium sulfide nanowires. Photoluminescence imaging and spectroscopy of nanowire photonic crystal structures demonstrate strong light suppression as well as localized emission in engineered artificial defects. Moreover, we have extended this work on photonic crystal to couple light emitted from single-nanowires into photonic crystal waveguides and photonic diodes. Finally, recent work on nonlinear mixing, which takes advantage of the large nonlinear coefficient in nanowire structures, demonstrates the potential for all-optical switching at the single nanowire level. These hybrid structures exploit unique strengths of bottom-up and top-down approaches and thereby open new opportunities in nanophotonics from efficient and localized light sources to integrated optical processing.
Q15.24
Raman Scattering Studies on Br2-doped Double Wall Carbon Nanotubes. Antonio Gomes Souza1, Morinobu Endo3, H. Muramatsu3, T. Hayashi4, Y. A Kim3, Eduardo B Barros1, Riichiro Saito5, Ge G Samsonidze2 and Mildred S Dresselhaus2; 1Física, Universidade Federal do Ceara, Fortaleza, Ceara, Brazil; 2Electrical Engineering, MIT, Cambridge, Massachusetts; 3Faculty of Engineering, Shinshu University,, Nagano, Japan; 4Tokyo National College, Tokyo, Japan; 5Physics, Tohoku University, Sendai, Tohoku, Japan.
In this work we report a resonant Raman scattering study of Br2-adsorbed double-wall carbon nanotubes (DWNTs) using different laser energies. Spectra are observed for both the Br-Br molecular vibration, as modified by the intercalation, and the DWNTs vibrations that were observed after intercalation by bromine. The analysis of the resonant Raman profiles associated with the DWNTs allowed us to identify the (n,m) tubes that were contributing to the Raman spectra. The analysis of the Raman spectra show that the Br2 molecules decorate the outer tube surface of the individual DWNT tubes and the spectra further show that the adsorbed bromine molecules behave as an electron acceptor. We have also shown that the Br2 adsorption in SWNTs is completely reversible upon thermal annealing to 600 oC. Furthermore,we observe that metallic tubes are extremely sensitive to doping and the presence of Br2 molecules affect their Raman spectra even when the metallic nanotubes are the inner tubes of the DWNTs.
Q15.25
Optical Properties of 55Mn Implanted ZnS Nanowires. Daniel Stichtenoth1, Sebastian Geburt1, Carsten Ronning1, Tobias Niebling2 and Peter J. Klar2; 1II. Institute of Physics, University of Göttingen, Göttingen, Germany; 2Physics and Material Sciences Center, Philipps-University of Marburg, Marburg, Germany.
Doping of semiconductor nanostructures via ion implantation offers the advantage of precise control of the doping concentration in both lateral and depth direction beyond any solubility limit. In this study, single crystal ZnS nanowires with varying diameter were synthesized according to the VLS mechanism and subsequently dispersed on top of Si substrates. The nanowires were implanted with 55Mn choosing varying ion fluences resulting in different Mn concentrations. The range of the ions, set by the implantation energy, matched the diameter of the nanowires, and post-implantation annealing procedures were done under vacuum conditions in order to remove the introduced damage. Electron spin resonance measurements showed that after these procedures the Mn substitute Zn sites in the lattice. The treated nanowires were investigated by time resolved PL measurements, where the well known long-living 4T1 → 6A1 intra 3d transition of 2+Mn was observed. Corellations between the life time of this intra 3d transition and the Mn concentration as well as the diameter of the wires will be discussed in detail.
Q15.26
Abstract Withdrawn
Q15.27
Investigation of the Epitaxial Growth of Zinc Oxide Nanowires on Sapphire by Grazing Incident Synchrotron X-ray Diffraction. Rodrigo Gribel Lacerda1, Leonardo Cristiano Campos1, Roberto Magalhães-Paniago1, Andre Santarosa Ferlauto1, Sharvari H. Dalal2, Daniel L. Batista2, Sergio Oliveira1, W. Milne2 and Luiz Orlando Ladeira1; 1Departamento de Física, ICEX, Universidade Federal de Minas Gerais, Belo Horizonte, Minas Gerais, Brazil; 2Engineering Department, Cambridge University, Cambridge, CB2 1PZ, UK, United Kingdom.
The controlled synthesis of quasi one-dimensional materials, like nanowires (NW) and nanotubes, is one of the key research areas of nanotechnology. If these materials are to have widespread usage in electronics, biology and medicine, full control of their structural and electronic properties is needed in the growth processes. To achieve this goal, a complete picture of the growth kinetics is fundamental, especially for nanomaterials grown via catalyst-assisted methods. There have been numerous reports that show preferential alignment of NWs by choosing appropriate substrates using physical vapor deposition (PVD) methods. The alignment has been attributed extensively to an epitaxial relation between the material and the substrate. However, although epitaxial mechanism exists in growth process with well-controlled atmospheres, the achievement of such phenomenon is surprising in CVD systems at sub-atmospheric pressures without any further control. Such epitaxial growth occurs when Zinc Oxide (ZnO) NWs are grown on sapphire substrates by CVD using a thin layer of gold as metal catalyst. Epitaxial growth of ZnO can occur in two different surfaces of sapphire, the so called c-surface (0001) and the a-surface (11-20). Usually, for growth on the a-surface the ZnO wires are oriented in the [0001] direction which is parallel to the sapphire [11-20]. In addition, for ZnO films, the in-plane orientational relationship has been found to be [11-20]ZnO//[0001]sapphire, and although it is believed that the same relationship occurs for the ZnO NWs growth, the in-plane measurement of the epitaxial relation between the ZnO wires and the sapphire substrate has not been realized so far. In this work, we present a careful study on the vertically aligned growth of ZnO NWs on sapphire using synchrotron x-ray diffraction (XRD) in the National Laboratory of Synchrotron Light (LNLS) in Brazil. Firstly, conventional (theta-2theta) measurements were performed to confirm that the vertically aligned ZnO wires were oriented in the [0001] ZnO direction parallel to the [11-20] sapphire. Secondly, to directly probe the in-plane lattice match between the sapphire and the NWs, we performed an XRD experiment at grazing angles (GID) where the crystallographic planes perpendicular to the substrate surface (and responsible for the epitaxy) can be probed. Firstly, the detector was positioned at the (11-20) peak of the ZnO wires and the sample was rotated by 360o. The (11-20) peak showed a C6 symmetry indicating a global orientation of the lateral faces of the wires with respect each other. Secondly, to further prove the epitaxy, theta-2theta measurements in GID configuration was performed and we observed simultaneously a lattice match between the (0001) plane of the sapphire and the (11-20) plane of the ZnO. Those results demonstrate, in spite of growth conditions outside equilibrium, that an epitaxial growth occurs and it determines the high homogeneity and orientation of the NWs.
Q15.28
Investigation of the Optical Gap in Ge Nano-wires. Scott P Beckman1, Jiaxin Han2 and James R Chelikowsky1,2,3; 1Center for Computational Materials, Institute for Computational Engineering Sciences, University of Texas at Austin, Austin, Texas; 2Department of Physics, University of Texas at Austin, Austin, Texas; 3Department of Chemical Engineering, University of Texas at Austin, Austin, Texas.
We investigate the role of quantum confinement for the optical and electronic properties of Ge [110] nano-wires ranging up to 2.8 nm in diameter. The electronic structure is investigated using "density functional theory - pseudopotential" methods. By placing an H molecule in the computational cell we align the bands of the wires and predict the energies of the valence band maximum and conduction band minimum independent of one another. We observer that the valence band maximum does not change energy once the nano-wire reaches a diameter larger than 2.0 nm. We also observe that the shape of the bands are approximately that of the bulk, once the wire reaches a diameter of greater than 3.0 nm.
Q15.29
Comparative Raman Spectroscopy Study of Single-Wall and Double-Wall Carbon Nanotube Systems Doped with H2SO4. Eduardo B Barros1, Antonio G. Souza Filho1, HyungBin Son3, Yoong-Ahm Kim2, Hiroyuki Muramatsu2, Takuya Hayashi2, Morinobu Endo2 and Mildred S. Dresselhaus3; 1Physics, Universidade Federal do Ceara, Fortaleza, Brazil; 2Electrical And Electronic Engineering, Shinshu University, Nagano, Japan; 3Electrical Engeneering and Computer Science, Massachusetts Institute of Technology, Cambridge, Massachusetts.
In this work we performed Raman experiments on a mixture of Single-wall and Double-wall carbon nanotubes for different relative concentrations and using different laser energies. Two sets of samples were analyzed, one which was exposed to H2SO4 for 5s and one which is pristine. The H2SO4 is known to act as an acceptor for the electrons of graphitic materials. The effect of the hole doping on the vibrational and electronic properties of the double and single-wall carbon nanotubes is probed using Resonant Raman scattering with different excitation energies probing different nanotubes. The effect of charge transfer to the relative intensities of RBM Raman peaks was found to be more drastic in small diameter nanotubes. The inner and outer walls of double-wall nanotubes can also be studied at the same time for selected excitation energies. The lineshape of the G and G' bands can be used to probe the relative concentration of single and double wall nanotubes in the sample and also to characterize the effect of charge transfer on the vibrational and electronic properties of carbon nanotubes.
Q15.30
Atomically Resolved Imaging on an Epitaxially Grown Silver Nanorod Surface. Saw Wai Hla1, Albert Prodan2 and Herman J.P. van Midden2; 1Physics & Astronomy, Ohio University, Athens, Ohio; 2Solid State Physics, J Stefan Institute, Ljubljana, Slovenia.
Engineered nanoparticles represent a new class of materials whose structures and physical properties often differ from those in the bulk. An important property of anisotropic particles, like nanowires/nanorods, is the surface layer bending. A theoretical report [1] suggested that a stress fluctuation should occur near the surface region of a small gold slab to accommodate the surface layer bending. To confirm this phenomenon and to check whether such behavior is valid in other cases as well, we have investigated structural peculiarities occurring at surfaces and edges of epitaxially grown silver nanorods on a (001) van der Waals surface of β-MoTe2 [2]. The β-MoTe2 surface layer consists of troughs formed by tellurium (Te) atoms with a 0.06 nm height corrugation. Silver nanorods grow on (001) β-MoTe2 surface with its (211) surface plane parallel to the substrate. At the initial stages of the growth, silver forms small wires composed of chains of atoms along the β-MoTe2 [010] direction due to a preferential diffusion. At a nominally 2 ML thick deposit, 20 to 50 nm wide, up to 300 nm long and four to eight layer high silver nanorods are formed. The upper-most surface layers of most nanorods examined here are only partially filled exposing the underlying layer structures. Thus, the structural changes at both, the top and second top-most layers of the nanorods, can be probed with the scanning tunneling microscope (STM). The atomically resolved STM images reveal intimate details of atomic-scale surface bending and surface-reconstruction at the nanorod surfaces and edges. In addition, we discover novel surface reconstruction on a second top-most layer of the silver nanorod surface. In agreement with a theoretical prediction a stress fluctuation is observed in the two upper-most atomic layers of nanorods. Due to a tensile stress the surface adlayers are occasionally bent while the first subsurface layer experiences a compressive stress resulting in a novel surface reconstruction. This work is financially supported by US-DOE DE-FG02-02ER46012 grants. [1] Tartaglino, U.; Tosatti, E.; Passerone, D.; Ercolessi, F. Phys. Rev. B 2002, 65, 241406-1-241406-4 [2] S.-W. Hla, A. Prodan, H.J.P. van Midden, “Atomistic Stress Fluctuation at Surfaces and Edges of Epitaxially Grown Silver Nanorods”, Nano Lett. 4 (2004) 1221-1224.1997-2001.
Q15.31
Determination of Electron-Phonon Coupling in Carbon Nanotubes by Resonant Raman Scattering. Yan Yin1, A. Vamivakas2, A. Walsh1, S. Cronin3, M. S. Unlu2,1, B. B Goldberg1,2 and A. K. Swan2,1; 1Physics Department, Boston University, Boston, Massachusetts; 2Electrical and Computer Engineering Department, Boston University, Boston, Massachusetts; 3Electrical Engineering Department, University of Southern California, Los Angeles, California.
We report on an optical method to directly measure electron-phonon coupling in carbon nanotubes by correlating the first and second harmonic of the resonant Raman excitation profile. We investigate a series of singly optically resonant single wall carbon nanotubes (SWNTs) in small ropes grown by chemical vapor deposition over 1-1.5μm wide trenches etched in quartz substrates. We extract an absolute value of the radial breathing mode e-ph coupling for nanotubes of 5 different chiralities under E22 resonance by correlating the full resonant excitation profiles of the first and second harmonic Raman peaks of the radial breathing mode and accounting for the phonon density of states [1]. The measured strength of the e-ph coupling varies from 3-11 meV, depending on chirality. Recent calculations of the deformation potential based on the tight-binding approximation predict a strong chirality and family dependence of the e-ph coupling of the radial breathing mode (RBM) in carbon nanotubes [2]. We find that our measured data follow the calculated functional dependence on chiral angle, tube diameter and chiral index ν suggested in [2]. These results confirm that the e-ph coupling is well described by the deformation potential interaction, derived from the empirical tight binding model. The results demonstrate that the interference of the chirality-dependent and chirality-independent terms indeed lead to stronger interactions for ν=−1 carbon nanotubes at E22, in agreement with the many experimental studies that observe stronger Raman cross-sections for ν=−1 for E22 transitions. Practically, this means that ensemble Raman measurements must be appropriately scaled to extract the chirality distribution. [1] Y. Yin et al., arXiv:cond-mat/0605670. [2] S. V. Goupalov, B. C. Satishkumar, and S. K. Doorn, Physical Review B 73, 115401 (2006).
Q15.32
Analysis of Carbon Nanomaterials Using Tip-Enhanced Near-Field Raman Spectroscopy. Prabhat Verma1, Taka-aki Yano1, Yasushi Inouye2 and Satoshi Kawata1; 1Applied Physica, Osaka University, Suita, Osaka, Japan; 2Dept. of Frontier Biosciences, Osaka University, Suita, Osaka, Japan.
The spatial resolution in the optical microscopy of carbon nanostructures, such as carbon 60 (C-60) or carbon nanotube (CNT), is limited to the diffraction limits of the probing light, which is not small enough to characterize and image these nanostructures. The basic idea to overcome this diffraction-limited resolution in Raman scattering leaded to the discovery of tip-enhanced near-field Raman scattering (TERS). In this phenomenon, when light is incident on a nano-sized metallic tip, a nano-sized enhanced electric field is generated at the tip apex due to localized surface plasmon polariton excitation, which enhances Raman scattering from molecules existing only in the vicinity of the tip. Therefore, TERS can provide an enhancement with extremely high spatial resolution, far beyond the diffraction limits of the probing light. In the present study, TERS has been applied to analyze the nanoscale vibratioanal properties of C-60 molecules and single-wall CNT (SWCNT). A silver coated AFM cantilever tip with a tip-apex of 40 nm was approached to the C-60 molecules and individual SWCNT bundles, which were dispersed on glass coverslips, and the samples were illuminated by a green laser (wavelength: 532 nm). Compared to the far-field, huge enhancements in Raman scattering were observed in the near-field configuration. Further, super-resolved near-field Raman imaging of isolated SWCNT bundles was performed with a spatial resolution of 40nm by setting the spectrometer to a frequency corresponding to the radial breathing mode, and then raster-scanning the bundle under the silver-coated tip. In addition to the conventional topographic-type imaging, this nano-Raman image also showed a spatial distribution of the tube diameters within the bundles.
Q15.33
Fabrication of Omega-shaped-gate ZnO Nanowire Field Effect Transistors. Kihyun Keem1, Dong-Young Jeong2, Sangsig Kim1, Moon-Sook Lee3, In-Seok Yeo3, U-In Chung3 and Joo-Tae Moon3; 1Electrical Engineering, Korea University, Seoul, South Korea; 2Institue for Nano Science, Korea University, Seoul, South Korea; 3Memory Division R&D Center, Samsung Electronics Co., Yongin-City, South Korea.
Omega-shaped-gate (OSG) nanowire-based field effect transistors (FETs) have attracted a great deal of attention recently, because theoretical simulations predicted that they should have a higher device performance than nanowire-based FETs with other gate geometries. OSG FETs with channels composed of ZnO nanowires were successfully fabricated in this study using photolithographic processes. In the OSG FETs fabricated on oxidized Si substrates, the channels composed of ZnO nanowires are coated with Al2O3 using atomic layer deposition, which surrounds the channels and acts as a gate dielectric. About 80 % of the surfaces of the nanowires coated with Al2O3 are covered with the gate metal to form OSG FETs. A representative OSG FET fabricated in this study exhibits a mobility of 30.2 cm2/Vs, a peak transconductance of 0.4 μS (Vg=-2.2 V), and an Ion/Ioff ratio of 10^7. To the best of our knowledge, the value of the Ion/Ioff ratio obtained from this OSG FET is higher than that of any of the previously reported nanowire-based FETs. Its mobility, peak transconductance, and Ion/Ioff ratio are remarkably enhanced by 3.5, 32, and 10^7 times, respectively, compared with a back-gate FET with the same ZnO nanowire channel as utilized in the OSG FET.
Q15.34
Electrical Properties of Metal Doped Vanadium Pentoxide Nanofibers. Pil Soo Kang1, Hyeyeon Ryu1, Seongmin Yee1, Gyu-Tae Kim1, G. s. Zakharova2 and V. l. Volkov2; 1School of Electrical Engineering, Korea University, Seoul, South Korea; 2Institute of Solid State Chemistry, Ural Division Russian Academy of Sciences, Yekaterinburg, Russian Federation.
Electrical properties of molybdenum doped vanadium pentoxide nanofibers were investigated. Samples for electrical transport measurements were prepared by dispersing V1.66Mo0.33O5 n H2O nanofibers in a network configuration with Ti/Au electrodes. At room temperature, the nonlinear current/voltage(I/V) characteristics with significant hysteresis were observed, which may originate from the existence of the dipole layer due to the ionic conduction through the filamentary structures of vanadium pentoxide and the water molecules around V2O5 layers. To explore the influence of water molecules, we annealed the samples to remove the water. After annealing, the hysteresis disappeared, confirming the role of water molecules in the hysterical I/V curves. In addition the tenfold conductance increase can be explained by the enhanced hopping conduction with the change of the relative concentration ratio of V4+/(V4++V5+).
Q15.35
Electrical Property of Single Crystalline Dilute Magnetic GaMnN Nanowires. Choi Sung-Churl1, Byeun Yun-Ki1,2, Lee Jin-Seok1 and Han Kyoung-Sop2; 1ceramic engineering, hanyang univeristy, Seoul, South Korea; 2Material science, Korea Institute Science and Tehcnology, Seoul, South Korea.
Theoretical studies indicate that transition metal doped gallium nitride (GaN) possesses a ferromagnetic transition temperature higher than room temperature by hole mediated ferromagnetism, which would be advantageous for many proposed spintronic applications Recently, we reported on single crystalline GaMnN nanowires with Curie temperature above room temperature, magnetoresistance near room temperature, and spin-dependent transport. So, we report on the electrical properties of single crystalline dilute magnetic semiconductor Ga1-xMnxN nanowires. The nanowires were fabricated by a chemical vapor transport process, and had diameters of < 100 nm and length of several micrometers. Controlled doping of manganese in the range of x = 0.08 was successfully achieved by varying the processing conditions. Electrical property of GaMnN nanowires was measured using muli-pad electrode by photolithography.
Q15.36
Dielectrophoretically Aligned GaN Nanowire p-n Junction Diodes. Sang-Kwon Lee1, Tae-Hong Kim1, Seung-Yong Lee1, Duk-Il Suh1, Wook Bahng2 and Nam-Kyun Kim2; 1Dept. of Semiconductor Science and Technology, Chonbuk National University, Jeonju, South Korea; 2Power Semiconductor Group, Korea Electrotechnology Research Insitute, Changwon, South Korea.
One dimension semiconductor nanowire offers a good system to investigate the dependence of the electrical and thermal transport or mechanical properties on dimensionality and size reduction or quantum confinement effect. Here, we first demonstrated the hybrid p-n junction diodes by coupling the n-type GaN nanowires with p-type Si substrates using dielectrophoresis (DEP) techniques to align and manipulate the nanowires on silicon substrate. The cathode electrodes for contact to n-type GaN nanowires were patterned on the oxidized p-Si (100) wafe by the standard photolithography and lift-off process while the anode contacts were defined on backside of p-Si substrate. The DEP for these p-n junction structures was performed at the frequency of 10 kHz with 15 Vp-p. From the observation of FE-SEM, GaN nanowires are well-aligned around the electrodes after DEP process as expected above DEP experiment. Current-voltage (I-V) measurements show well-defined current rectifying behavior. Specially, little leakage current and no breakdown are observed in reverse bias up to -25V for these p-n junction diode structures. The observed reverse leakage current was to be ~ 3xE-4 A at 25 V of reverse bias voltage at room temperature. For the reproducibility of the n-GaN nanowire/p-Si substrate junction structures, we have fabricated several samples under the same condition of DEP and found that all of the samples have a similar rectifying behavior and clearly working with low leakage current.. The sheet resistances (RS) for these DEP prepared p-n junction diodes were determined to be 10 kΩ~ 30 kΩ from the inverse of the slope in I-V curve. The high sheet resistance is due to the pure metal contacts to Si and n-GaN NWs as well as high resistivity of the n-GaN NWs. We also demonstrate the GaN nanowire current rectifier using these GaN nanowire p-n diodes. A more detailed study of nanowire p-n junction structures is being initiated for well-aligned large scale photonic as well as electronic device applications.
Q15.37
Electrical Transport in Vanadium Oxide Nanowires. Jitae Park, Eunmo Lee, Kyu Won Lee and Cheol Eui Lee; Physics, Korea University, Seoul, South Korea.
The electrical DC conductivity (σ≈10-9-10-6) and photoconductivity measurements were carried out on semiconducting vanadium oxide nanowires network, which were synthesized through polycondensation method of vanadic acid, over temperature range 60-300K. At higher temperature (T > 180K), Mott’s optical multiphonon assisted hopping model described well the temperature dependence of the conductivity. On the other hand, at low temperature, the variable-range hopping mechanism is dominant and the activation energy for hopping conduction also changed. The photoconductivity showed some distinct behaviors other than the dark conductivity.
Q15.38
Dielectrophoretic Alignment and Fabrication Technique for Realizing GaN Nanowire Devices. Abhishek Motayed1,2, Albert Davydov1, Maoqi He3, John Melngailis2 and S. N Mohammad4,3; 1MSEL, Metallurgy, NIST, GAITHERSBURG, Maryland; 2Dept. of Electrical and Computer Engineering, University of Maryland, College Park, Maryland; 3Dept. of Electrical and Computer Engineering, Howard University, Washington, District of Columbia; 4Dept. of Material Science and Engineering, University of Maryland, College Park, Maryland.
The Group IIIA nitrides (binary and ternary alloys of AlN, GaN, and InN) have unique properties such as a direct bandgap spanning the whole solar spectrum (from 0.7 eV for InN to 6.2 eV for AlN), high saturation velocity, and high breakdown electric field. As a result nanostructures and nanodevices made from GaN and related nitrides have great potential for realizing next generation efficient nanoscale UV/visible light emitters, detectors, and gas sensors. We have demonstrated a technique for assembling long (50 µm - 200 µm) GaN nanowires for device applications. These catalyst free nanowires were grown by direct reaction of NH3 and Ga, which resulted in free-standing nanowires along with GaN microplatelets. GaN nanowires were suspended in a solvent using sonication, and using dielectrophoretic forces, nanowires were assembled on prepatterned substrates (SiO2 coated Si and Sapphire) followed by a fabrication sequence to form stable nanowire device structures. The present technique is potentially compatible with CMOS technology and integrating nanodevices with conventional Si microelectronics on the same chip can be made possible with this technique. Only batch fabrication processes like standard photolithography, etching and oxide deposition are used. Calculations have been carried out to reveal the effect of different solvents on the dielectric force factor to better understand the alignment process. Effect of processing conditions on the device yield will be discussed. Present fabrication technique results in the nanowire devices embedded in a passivation layer (SiO2), which have shown to minimize the surface charge effects. Reliable GaN nanowire field effect transistors (FET), with Si substrate as the backgate, have been routinely achieved using this technique. These nanowire FETs do not show any measurable degradation or variation in their electrical characteristics even after extended period of use and/or storage. Field effect mobilities as high as 250 cm2V-1s-1 at 300 K has been measured in these devices. Low temperature transport measurements of these nanowires revealed a correlation between the structural characteristics and electrical properties in these nanowires.
Q15.39
Theoretical and Experimental Study on Transport Properties of the One-dimensional Electron Gas in an InAlAs/InGaAs Quantum-wire System. Ilho Ahn1, Sung Geun Kim1, Jhang W. Lee2, Yong Tak Lee1 and G. Hugh Song1; 1Information and communication, GIST, Gwangju, Geon-nam, South Korea; 2Kowon Technology, Yongin, Geonggi-do, South Korea.
We investigated theoretically and experimentally on the mobility of the quasi-one-dimensional electron gas in the InAlAs/InGaAs HEMT quantum-wire structures. The wire was fabricated on the MBE grown InAlAs/InGaAs HEMT layers using laser hologram or electron beam lithography, and inductively-coupled-plasma reactive-ion-etching (ICP-RIE) technique. In order to measure the mobility and the drift velocity in a one-dimensional quantum wire, we carried out current-voltage measurements with conductive scanning-probe-microscope (SPM) tip and four-point resistivity measurements. Hall measurement were done also for comparison purpose and interpreting the calculation data. Photoluminescence (PL) measurements were performed at 10 K to confirm the lateral quantum confine-ment effect of the quantum-wire structures. The PL spectrum showed a blue-shift of the luminescence peak as the lateral size of InGaAs quantum wires decreases and showed a shoulder structure attributed to laterally quantized subbands. Four-point probe measurements on resistivity of the fabricated single quantum wire, measured at 2 K, displayed the well-known features of a quasi-one-dimensional electron-gas system. Mobility calculations were done using the self-consistent method with the exchange correlation potential. Analyzed data comprises the various kinds of mobility dependencies on the wire length, wire width, interface roughness, applied voltage, and temperature. We will present the characteristics of ridge-type InAlAs/InGaAs quantum-wire field effect transistors (QWR-FET) exhibiting the enhanced mobility. This research was supported by the Ministry of Information and Communication, Korea, under the ITRC support program supervised by the IITA-2005-C1090-0502-0029
Q15.40
In-situ Characterization of Electronic Properties of a Single ZnO Nanowire Xudong Wang, Jun Zhou and Zhong Lin Wang; Materials Science and Engineering, Georgia Institute of Technology, Atlanta, Georgia.
Quasi-one-dimensional (1D) ZnO semiconducting nanostructures, such as nanowires (NWs) and nanobelts (NBs), are considered as an important multi-functional building block for fabricating various nanodevices. Owing to their unique electronic, optical and piezoelectric properties, ZnO NWs/NBs have been successfully applied in FETs, LEDs, laser diodes, sensors, resonators, and piezoelectric devices. Most importantly, ZnO exhibits both semiconducting and piezoelectric properties. The coupling of these two properties endows it with very unique advantages and novel applications [1]. In order to characterize their electronic properties and apply them to nanodevices, we designed an in-situ measuring system inside an SEM chamber, where their I-V characteristics can be examined with a simultaneous observation of their locations and geometrics. In this system, ZnO nanowires were connected to one electrode and located on the SEM stage; while the counter electrode was attached on another independently controlled x-y mechanical stage. Using this set-up, the field emission property of single ZnO NW/NB has been measured under different emission distance. Furthermore, the relationship between the conductance of a single ZnO NW and its bending curvature was identified by connecting the ZnO NW to the counter electrode surface. A decrease of the conductance was observed with increasing the bending curvature by in-situ manipulation and measurements in the SEM chamber. This novel phenomenon is attributed to the unique semiconducting and piezoelectric coupling effect of ZnO, which is different from other semiconductors or carbon nanotubes. These mechanisms realized a new type of nano-device — piezoelectric-field effect transistor (P-FET) composed of a ZnO NW bridging two ohmic contacts, in which the source to drain current is controlled by the bending of the NW. The P-FET has been demonstrated as a force sensor for measuring forces in nano-Newton range and even smaller with the use of smaller NWs. <br>[1] Z.L. Wang and J.H. Song, Science, 312 (2006) 242-246. <br>[2] for details: http://www.nanoscience.gatech.edu/zlwang/
Q15.41
Size-dependent Conductance Behavior of [110] Gold Nanowire. Kurui Yoshihiko1, Oshima Yoshifumi2,3, Okamoto Masakuni4 and Takayanagi Kunio1,3; 1Condenced Matter Physics, Tokyo Institute of Technology, Meguro, Tokyo, Japan; 2Matterials Science and Engineerings, Tokyo Institute of Technology, Yokohama, Kanagawa, Japan; 3CREST, Japan Science and Technology Corporation, Kawaguchi, Saitama, Japan; 4Mechanical Engineering Research Laboratory, Hitachi Ltd., Hitachinaka, Ibaraki, Japan.
Atomic size metal nanowire(NW) attracts much interest because it shows quantized conductance even at room temperature. Quantized conductance of thin metal NW have been investigated mainly by mechanical cotrollable breaking junction (MCBJ) or scanning tunneling microscope (STM) methods. It has been confirmed that a single atomic strand of gold has a quantized unit of conductance, G0 (G0=2e2/h) and a double strand, two[1]. For gold NWs with conductance bolow 13 - 16G0, the conductance histgram was found to have a periodic peak structure, which was explained by electric shell model. In this model, NW was supposed to be cylindrical wire. On the other hand, for gold NWs above 14-16G0, the conductance histgram had different periodic peak structure, whichi was explained by atomic shell closings model. In this model, the NW was supposed to be crystalline with axis parallel to the [110] direction [2-4]. However, such a stable structure have not been confirmed yet. In this study, we report stable structure of gold NW in the range of conductance from 0 to 50 G0. Using ultra-high-vacuum transmission electron microscope (UHV-TEM) combined with a scanning tunneling microscope (STM), We investigated structure of gold NW simultaneously with conductance. Conductance of NW was measured at 3kHz TEM images were recorded at 30Hz. The stable structure of gold [110]NW was determined by TEM images viewed from two different directions. We found that the stable structure had hexagonal cross-section with wire axis parallel to the [110] direction. We also found that the cross-section had two different symmetries, which has not been point out previously. Such a stable structure was characterized by the number of atomic rows, N, along the wire axis: N=1,3,7,12,19,27,37,48,61,75. Conductance g(N) for each stable NW was determined. The conductance behavior for gold NW above N=37 was explained by atomic shell closing model [3]. In this region, conductance per atomic row was constant (g/N=0.54±0.02). However,conductance behavior below N=37 was not explained by electric shell model [2]. Conductance per atomic row had a tendency to increase with reducing the size (g/N increase from 0.54 to 1.00). Reference: [1] H.Ohnishi, Y.Kondo, and K.Takayanagi, Nature 395, 780 (1998) [2] A.I.Yanson, K.I.Yanson and J.M.van Ruitenbeek Nature 400(London) 144 (1999) [3]A.I.Yanson, K.I.Yanson and J.M.van Ruitenbeek Phys.Rev.Lett 87 216805 (2001) [4]A.I.Mares, A.F.Otte, L.G..Soukiassian, R.H.M.Smit, and J.M.van Ruitenbeek Phys.Rev.B 70 073401 (2001)
Q15.42
Electrical Conductivity and Optical Properties of Ultra-Thin Gold Nanowires. Aditi Halder, Bratindranath Mukherjee and Ravishankar Narayanan; Materials Research Centre, Indian Institute of Science, Bangalore, India.
The bottom-up approach to nanostructures relies on the ability to synthesize building blocks in a controlled way. Metal nanowires constitute one of the most important building blocks owing to the interesting optical and transport properties that they exhibit that makes them useful in nanoelectronic device applications. In spite of their importance, there is no simple method to synthesize ultra-thin (~ 2nm) metal nanowires. We have synthesized ultra-thin gold nanowires with large aspect ratios (~ 1000) by a wet-chemical method exploiting the oriented-attachment mechanism. Detailed structural and microstructural characterization has been carried out using X-ray diffraction and high-resolution electron microscopy. The highly tunable optical properties of the nanowires have been studied by UV-Vis-NIR spectroscopy. Electrical conductivity of the gold nanowires is measured by scanning tunneling microscopy (STM).
Q15.43
Determination of the Electrical Properties of the Well-Aligned ZnO Nanorods on Si Substrates using AC Impedance Analysis. Jih-Jen Wu and Daniel Kwan-Pang Wong; Department of Chemical Engineering, National Cheng Kung University, Tainan, Taiwan.
The electric properties of the well-aligned ZnO nanorods grown on p-Si substrate under various molar ratios of Zn to O sources (MR) were investigated using the AC impedance analysis. A device composed of p++ Si/n-ZnO nanorod junctions in parallel with air variable resistances was fabricated directly on the as-deposited sample. The junction capacitances of the p++ Si/n-ZnO nanorod junctions and the resistances of the ZnO nanorods can be extracted from the fitting results of the impedance data. The carrier concentration of the ZnO nanorods is obtained from the slope of the linear relationship of the inverse capacitance squared and the reverse-biased voltage. The mobility is further calculated using the carrier concentration and the resistance of the ZnO nanorods. The carrier concentrations of the ZnO nanorods grown using metalorganic chemical vapor deposition in the present work are measured to be in the range of 1017~1018 cm-3. The carrier concentration and the mobility of the ZnO nanorods increase with MR. The room-temperature photoluminescence (PL) spectra of the ZnO nanorods grown at various MRs show uv the emission peaks at 378 nm and the absence of the green band emission. The intensity of the uv emission peak decreases when the MR is increased.
Q15.44
ZnO Nanowire Field-effect Transistor Based on the Edge Defined Lithography. Hyun-Wook Ra1, Sang Hoon Kim2, Song Yi Jo2, Kwang Sung Choi2, Yoon Bong Hahn2 and Yeon Ho Im2; 1Semiconductor Physics Research Center, Chonbuk National University, Jeonju, Jeonbuk, South Korea; 2School of Chemical Engineering and Technology and Nanomaterials Processing Research Centre, Chonbuk National University, Jeonju, Jeonbuk, South Korea.
ZnO nanowires are being widely investigated for applications in electronics, optoelectronics, photovoltaics and sensors. For these applications, the control of the size and location of the nanowires is one of the current challenges to fabricate nanosized patterns. We demonstrate a photolithography-based method for fabricating sub-50 nm ZnO nanowires arrays on a wafer scale instead of the bottom up method. For this work, achieving conformal ZnO ultrathin film by atomic layer deposition will be a key to fabrication of ZnO nanowires by size reduction lithography. The width and height of the nanowires are controlled with nanometer precision, as chip manufacturers now do. The resolution of this method is not limited by photolithography but by the thickness of the material deposited. A further lift-off process is applied to define source-drain electrodes of ZnO nanowire FETs. This work will show the possibility of nanostructured applications composed of ZnO nanowires fabricated by means of lithographic technique.
Q15.45
Lateral and Vertical Silicon Nanowire Field-Effect Transistors. Mikael T Bjork1, Oliver Hayden1, Heike Riel1, Heinz Schmid1, Siegfried Karg1, Ute Drechsler1, Walter Riess1, Volker Schmidt2, Stephan Senz2 and Ulrich Goesele2; 1IBM Research GmbH, Ruschlikon, Switzerland; 2Max Planck Institute, Halle, Germany.
Semiconducting nanowires have recently attracted considerable attention as the ongoing miniaturization in microelectronics demands new, innovative fabrication techniques and device concepts. Owing to their potential compatibility with existing CMOS technology, silicon (Si) nanowires (NWs) and in particular epitaxially grown Si NWs are considered to be one of the most promising candidates for future logic and memory elements. In the first part of this paper we present our results on lateral Si NW field effect transistors (FETs) based on doped and intrinsic NWs. Our results substantiate the importance of the electrical contacts to the NW FET performance and by using implanted n-contacts we were able to realize individual lateral n-i-n NW FETs which could be operated in inversion mode. In the second part of the paper a generic process flow for fabricating vertical surround-gate field-effect transistors (VS-FET) from epitaxially grown Si nanowires is described and device characteristics are presented. The catalyst for the nanowire growth was patterned by electron beam lithography resulting in well defined arrays of vertical Si nanowires grown on Si (111) substrates. We demonstrate the fabrication processes using n-type silicon nanowires grown on a p-type substrate in ultra-high vacuum using gold as catalyst and silane as precursor gas. The VS-FET fabrication consists of various deposition and etching steps, and has the advantage that no chemical mechanical polishing is required. Moreover, the process can be used to fabricate individual as well as arrays of nanowire VS-FETs. Electrical characterization was carried out on vertical ungated n-doped and p-doped two-terminal devices and gated NW FETs. Single nanowires as well as arrays of about 10^4 nanowires were tested. The measured transistor output- and transfer-characteristics indicate the behavior of an inversion mode driven FET similar to a conventional p-channel MOSFET.
Q15.46
Carbon/Boron Nitride Nanotube Superlattices. Serena Povia and Stephanie Reich; Department of Materials Science and Engineering (DMSE), Massachussets Institute of Technology, Cambridge, Massachusetts.
We describe superlattices made of carbon and boron nitride nanotubes. Nanotubes composed of short BN and C segments were recently reported experimentally[1]. We studied the structure and electronic properties of C-BN nanotube superlattices by first-principles calculations. Structurally, BN-C nanotubes are a straightforward composition of boron nitride and carbon segments; the carbon-carbon and boron-nitride bond lengths change byless than 5%. The C-B distance at the BN-C interface is 1.51 Å; the C-N distance is 1.37 Å.
The BN segments open an electronic band gap of ~1eV in metallic (5,5) and (9,0) carbon tubes. Similar band gaps are also found for the semiconducting (8,0) carbon tube when introducing BN segments. Interestingly, the mixing between the carbon and BN wave functions is already quite strong at small electronic energies (1-2 eV from the Fermi level) despite the large band gap of pristine BN tubes (4 eV). The valence wave functions of C-BN superlattices are localized on C and N, the conduction wave functions on C and B. This is in good agreement with the shorter bond length of C-N compared to C-B.
In zigzag C-BN tubes we observe valence and conductions bands that are free of dispersion[2]. Their wave functions are strictly localized on the C-BN interface. In the (5,5) tube these electrons can propagate along the tube and the band has a finite dispersion (width ~ 0.5 eV). We discuss the dependency of the band structure of C-BN superlattices on the thickness of the carbon and boron nitride segments. Longer segments will lead to a localization of the wave functions in one of the two tube types[3]. This would allow us to realize nanotube quantum dots without the need for external gating.
[1] A. Loiseau "C-N and C-BN single wall nanotubes: growth and structural properties", IWEPNM 2006, Mar.4-11, 2006, in Kirchberg, Austria
[2] J. Choi et al., Phys. Rev. B, vol.67, 125421 (2003)
[3] V. Meunier et al., Appl. Phys. Lett., vol.81, 46 (2002)
Q15.47
Chalcogenide Nanowire Devices for Highly Scalable Phase Transition Memory. Se-Ho Lee, Yeonwoong Jung, Dong-Kyun Ko and Ritesh Agarwal; Mat. Sci. & Eng., University of Pennsylvania, Philadelphia, Pennsylvania.
Chalcogenide materials (i.e., Ge-Sb-Te alloys) have been dominant in the field of optical and electrical storage applications such as DVD, CD-RW, and Phase transition random accessing memory (PRAM) because of their reversible crystalline to amprphous phase transitions, which is signified by change in the optical reflectivity and electrical resistivity. In particular, the realization of these advantages in memory device applications is, however, still limited due to requirements of high scalability and low power consumption. These challenges motivate novel memory device schemes with sub-lithographic features based on bottom-up approach where high density integration can be enabled using nanowires that can be grown with arbitrarily small diameters. In this study, we report the synthesis of GeTe and Ge2Sb2Te5 nanowires and their device characteristics for highly scalable nanowire memory cells. The nanowires are synthesized via vapor transport method assisted by Au nanoparticle catalysts. The structure and composition of the nanowires are studied by scanning electron microscopy (SEM), high resolution TEM (HRTEM) and X-ray energy dispersive spectroscopy (EDS). The analysis indicates that the GeTe and Ge2Sb2Te5 nanowires possess single-crystalline structure with diameters range from 20-200 nm and lengths up to 50 m. We demonstrate the feasibility of the GeTe nanowire memory devices by studying their current-voltage (I-V) and programming curves. The nanowire devices were fabricated on oxidized Si substrate with Pt electrodes defined by Focused Ion Beam (FIB) technique. Upon applying a voltage pulse (1.3mA; 100ns), the nanowire undergoes a reversible phase-switching behavior with high resistive “RESET” state (4.0 x106 at 0.2 V) assigned as data “1” and low resistive “SET” state (3.0 x103 at 0.2 V) for data “0”. The threshold voltage from amorphous to crystalline state is 1.0V, which depends on the programmed state. The nanowire based memory devices are observed to be free of SET resistance variations with repeat cycles, which is a serious drawback in thin film memory devices. Significantly, the writing currents are reduced drastically by scaling-down the nanowire diameter, to as low as 0.42 mA achieved for a nanowire of diameter 28 nm. This study indicates that nanowire-based approach to construct memory cells is very promising for reducing RESET currents, which is a critical issue to realize memory chips with high density and low power consumption. Detailed experimental results on the phase-transition time scales upon cyclic application of electrical pulses will be presented. We also propose nanowire-based novel device architectures with the potential of achieving memory cells of density as high as ~80 Gb/in2. Our results on Ge2Sb2Te5 ternary nanowires, which are more promising for memory devices will be also presented.
Q15.48
Electronic and Optical Properties of SiONWs Grown from a Patterned Reagent: An EELS Study. Feng Wang1,2, Marek Malac1,2 and Ray F Egerton2,1; 1National Instiute for Nanotechnology, Edmonton, Alberta, Canada; 2Department of Physics, University of Alberta, Edmonton, Alberta, Canada.
Silicon-based nanowires, including crystalline and amorphous silicon and silicon oxide nanowires, have shown promising applications in nanoelectronics [1]. We have achieved size and position control of the silicon oxide nanowires (SiONWs) by growing them from a patterned reagent[2]. Metallic iron nanoparticles embedded in SiO2 matrix were used as high-activity catalyst and electron-beam (EB) patterned hydrogen silsesquioxane was used as the feedstock for nanowire growth. Microstructure and morphology of the SiONWs were investigated in an analytical transmission electron microscopy; electronic and optical properties were studied by near edge fine structures (ELNES) and valence electron energy loss spectrum (VEELS). The low oxidation state of the iron catalyst is implied by the absence of a characteristic postpeak [3], and low white line ratio in the iron L23 edge. The structural order of the amorphous SiONWs was studied using ELNES of the oxygen K edge; broadening or absence of ELNES peaks, their relative intensities compared to the calculated ELNES of a crystalline phase (a-quartz) suggest amorphous nature of the SiONWs. Redshift of the bulk plasmon peak and appearance of additional features in VEELS suggest that the optical properties of the SiONWs are different from silicon oxide (SiO2) film; the origin of the difference will be discussed. Complex permittivity e(E)obtained by Kramers-Kronig analysis [4], and photoluminescence of regular arrays of SiONWs provide additional insight to their optical properties. 1. B. Zheng, Y. Wu, P. Yang, and J. Liu, Adv. Mater. 14, 122 (2002). 2. F. Wang et al., Appl. Phys. Lett., in preparation. 3. F. Wang, R.F. Egeton and M. Malac, Ultramicroscopy (in press) 4. R.F. Egerton, Electron Energy-Loss Spectroscopy in the Electron Microscope, 2nd edition,Plenum Press, New York (1996).
Q15.49
Fabrication and Electrical Characterization of InGaAs Nanowires by Selective-area MOVPE. Jinichiro Noboriska, Takuya Sato, Junichi Motohisa, Shinjiro Hara and Takashi Fukui; Graduate school of Information Science and Technology and Research Center for Integrated Quantum Electronics, Hokkaido University, Sapporo, Japan.
Semiconductor nanowires (NWs) are one of promising building blocks for nanoelectronics because of their potential in broad range of applications, such as nanoscale-FETs, LEDs, and chemical sensors. To date, most of the NWs are formed using catalyst-assisted vapor-liquid-solid (VLS) growth. We have recently reported catalyst-free approach for GaAs based NWs using selective-area (SA) metal-organic vapor-phase epitaxy (MOVPE) technique. Here, we report on the fabrication of InGaAs NWs by SA-MOVPE and their electrical characterizations. The fabrication process started from a preparation of patterned InP (111) B substrates partially covered with SiO2 mask for SA-MOVPE. SiO2 mask pattern was formed using electron beam lithography and wet chemical etching, and designed to have periodic array of circular holes with diameter ranging from 50 to 100 nm. SA-MOVPE growth of nominally undoped InGaAs was carried out in a horizontal system working at 0.1atm. Trimethylgallium (TMG), Trimethylindium (TMI) and Arsine (AsH3) were used as source materials. The growth temperature was 625 °C. The typical InGaAs growth rate on InP (100) planar substrate was 0.43 nm/sec. After the growth, free-standing InGaAs NWs were formed in the circular hole region of the mask. The thickness of the NWs was directly related to the size of mask opening, and their length became longer for smaller mask opening. In the present experiment, the obtained size of NW was 50 to 100 nm in diameter and 5 to 10 μm in length. For the electrical measurements, we fabricated NW-FETs. Firstly, InGaAs NWs were mechanically scraped and dispersed on p-type Si substrate with 1 μm thick SiNx layer. To fix NWs, 200 nm thick SiO2 was deposited on the substrate. The source and drain patterns were defined using electron-beam lithography at the both end of NWs. After the development of the electron-beam resist patterns, SiO2 and possible native oxide of InGaAs NW were removed by buffered fluorinated acid (BHF) and dilute hydrochloric acid (HCl : H2O = 1 : 10) treatment. Then the source and drain electrodes were formed by metal evaporation and a lift-off process. Thermal treatment was carried out at 320 °C for 3 min after the evaporation of Ti/Au (40/310 nm) for InGaAs NW. The Si substrate was used for a back gate. Current-voltage measurement on InGaAs NW FETs revealed linear and symmetric Ids-Vds characteristics at room temperature, indicating the formation of good ohmic contacts between InGaAs and Ti/Au. In addition, Ids increased by the application of positive back gate voltage Vgs, which indicates an n-type conduction in nominally undoped InGaAs NWs and FET action. The two-terminal resistivity and the slope of Vg-Ids characteristics differ from NW to NW, presumably because of the difference of the incorporation of residual impurities.
Q15.50
Intrinsic Characteristics of Semiconducting Oxide Nanobelt Field-Effect Transistors and Their Applications. Yi Cheng1, Peng Xiong1, Lenwood Fields2, Jim P. Zheng2, Rusen Yang3 and Zhong Lin Wang3; 1Physics Department, Florida State University, Tallahassee, Florida; 2Department of Electrical and Computer Engineering, FAMU/FSU College of Engineering, Tallahassee, Florida; 3School of Materials Science and Engineering, Georgia Institute of Technology, Atlanta, Georgia.
Single-crystalline semiconducting oxide (ZnO and SnO2) nanobelts with thickness of 10 -150 nm were synthesized by catalyst-free thermal evaporation of oxide powders. Field-effect transistors (FETs) based on individual SnO2 and ZnO nanobelts with multi-terminal electrical contacts have been fabricated and characterized. Simultaneous two-terminal and four-terminal measurements enable direct correlation of the FET characteristics with the nature of the contacts. Devices with high-resistance non-ohmic contacts exhibit Schottky barrier FET behavior dominated by contact modulation. In contrast, low-resistance ohmic contacts on the nanobelt lead to high-performance n-channel depletion mode FETs with well-defined linear and saturation regimes, large “on” current, and “on/off” ratio as high as 10^7. Electrical measurements also revealed high channel transconductance and intrinsic field-effect mobility for these nanobelt FETs. Distinguishing the Schottky-barrier and channel-limited FET behavior is not only critical to the understanding of the intrinsic properties of the semiconducting oxide nanobelts but also the clarification of sensing mechanisms for chemical and biological detection. Surface effects including gas modulation of the FET characteristics have been studied on SnO2 nanobelt based devices and the results were correlated with surface oxygen adsorption and desorption. The field-effect carrier mobility and carrier density were further estimated and found to be consistent with surface charge transfer. Importantly, the FET characteristics of the SnO2 devices showed significant modification by a 2% hydrogen gas flow at room temperature. The utilization of the channel-limited nanobelt FETs for protein detection will also be reported. *This work is supported by NSF NIRT grant ECS-0210332.
Q15.51
Ab initio Study of the Electronic Structures and Optical Properties of ZnO Single-wall Nanotubes. Shelly L. Elizondo and John W. Mintmire; Department of Physics, Oklahoma State University, Stillwater, Oklahoma.
In this work, we report simulations of the electronic structures and optical properties of ZnO single-wall nanotubes. Ultrathin nanowires and ZnO graphitic-like strips and sheets were calculated for comparison. The electronic structures were investigated within a first-principles, all-electron, self-consistent local density functional approach (LDF). Using the first-principles electronic structure results, the optical cross sections were calculated using an Ehrenreich-Cohen formalism. Due to its many favorable characteristics, including geometrical versatility and bio-friendliness, ZnO has received considerable attention lately, especially regarding applications involving optoelectronics and sensing. Although ZnO can form a variety of structures on the nanoscale, no single-wall nanotubular structures have been reported to date. Such nanotubular structures are typically grown from materials with a layered structure, such as graphite or graphitic-like structures. Nonetheless, theoretical reports have addressed the possibility of ZnO existing in flat graphitic-like sheets as well as a single-wall nanotubular structure analogous to carbon nanotubes. Herein, we were interested in changes in the electronic and optical properties of the ZnO quasi-one-dimensional structures with varying geometry. Because the LDF procedure implemented herein has been adapted for helical symmetry, both chiral and non-chiral ZnO single-wall nanotubes were studied with radii ranging from approximately 1.20 Å to 3.13 Å. Although the band-gaps of the single-wall nanotubes remained comparatively constant, the optical absorption spectra exhibited a fascinating trend. While the first optical absorption peak associated with a direct transition remained relatively stationary as nanotube radius varied, the second optical peak resulting from a direct transition displayed a blue-shift with decreasing nanotube radius. As a final point, the total energies of the single-wall nanotubes compared well to the energy of a 2D sheet, implying that if ZnO can form into a graphitic-like sheet, it should be possible to make ZnO into single-wall nanotubes. This work was supported by the DoD HPCMO CHSSI program through the Naval Research Laboratory.
Q15.52
Growth, Structure, and Doping of Gold-Catalyzed Si Nanowires: A First Principles Study. Soohwan Lee, Kyoung-Eun Kwon and Gyeong S Hwang; Chemical Engineering, The University of Texas at Austin, Austin, Texas.
The growth direction, diameter, and surface structure of semiconductor nanowires can be controlled by varying process conditions and metal catalysts. Such ability to manipulate their structural properties on the atomic scale makes semiconductor nanowires attractive for a variety of future applications in electronics, optoelectronics, and sensors. Earlier studies have identified the underlying mechanisms for metal catalyzed Si nanowire growth, involving Si diffusion into a metal catalyst, eutectic Si-catalyst alloy formation, and Si precipitation at the catalyst-nanowire interface. However, many fundamental aspects of Si nanowire growth and structure have not been clarified yet. In addition, de