1

Biomaterials Plenary Session Room: C-201 - Session BP-SuA

Biomaterials Plenary Session Moderator: K. Healy, University of California, Berkeley

3:00pm BP-SuA1 Visions of Science at BESSY SASE FEL in Berlin Aldershof , W. Eberhardt, BESSY GmbH, Germany INVITED BESSY is planning to build a SASE FEL facility covering the photon energy range from 20 eV to 1 keV at the site next to the existing BESSY II storage ring. This new facility will offer laser like photon beams with fully coherent, high power (mJ) pulses of ≤20 fs duration, enabling a whole set of novel experiments dedicated to understand dynamical processes in matter or for the investigations of very dilute systems. This SASE FEL covers the traditional BESSY II photon energy range, which is especially suited for electronic structure investigations of atoms, molecules, clusters, and solids. With the anticipated temporal resolution of ≤20 fs charge transfer processes and time resolved 'femtochemistry' studies as well as magnetization dynamics in magnetic materials establish some of the major areas of scientific interest in this new facility. Furthermore in microscopy on soft-matter and biological samples it is possible to acquire an image using a single laser pulse. Thus stroboscopic time resolved images of dynamical processes in living cells become possible. In general, the science planned at this facility is complementary to the science envisioned for the planned TESLA X-FEL facility at DESY and Linac Coherent Light Source (LCLS) at Stanford. Following the presentation of the parameters and the layout of the proposed facility, the Scientific Case will be presented as it was developed by the prospective user community in the course of several scientific workshops and in discussions.

3:40pm BP-SuA3 Large Scale Integration of Microfluidic Devices, and Applications to Biology, S. Quake, Caltech INVITED

4:20pm BP-SuA5 Biochips Designs, Challenges, and Bioanalytical Applications, L.J. Kricka, University of Pennsylvania Medical Center INVITED Biochips, in all of their guises are the most active area of research and development in the analytical sciences. These micro-miniature devices are produced using techniques originally developed in the microelectronics and in the printing industry. Although some microchip devices have been commercialized (e.g., capillary electrophoresis chips), many challenges and issues remain for the routine implementation of these micro-analytical devices. These include surface chemistry effects in the sub-microliter confines of a microchip chamber, the interface between the

2

Applied Surface Science Room: C-106 - Session AS-MoM

SIMS Moderator: F.A. Stevie, North Carolina State University

8:20am AS-MoM1 Maximizing and Automating Information Extraction in Time-of-Flight Secondary Ion Mass Spectrometry, S.J. Pachuta, 3M Company Numerical techniques have been found useful in extracting information from the large data sets produced by time-of-flight secondary ion mass spectrometry (TOF-SIMS). Simple univariate approaches to quantitation in static SIMS, including spectral normalization, smoothing, curve fitting, and generation of spectral overlays, will be described. The complexity of TOF-SIMS data makes multivariate approaches particularly useful, and multivariate statistical techniques such as principal components analysis (PCA) will be described in detail. PCA, in simplest terms, reduces a data set to its essential elements. A significant advantage of PCA over univariate methods is that it greatly compresses the data by combining variables. This is particularly beneficial in TOF-SIMS, where a data set may consist of a 256x256 pixel matrix in which each pixel contains a complete mass spectrum consisting of upwards of 100,000 mass channels. Applications of PCA which will be described here include automating complex tasks previously performed manually by the analyst, finding non-obvious information in data sets, and distinguishing relevant from non-relevant information. Situations in which PCA is particularly applicable are those in which differences are sought. Typical examples might be a comparison of spectra from a series of materials with different performance characteristics, or chemical imaging of a surface. Automation of some common imaging tasks, such as choosing which peaks to map or extracting spectra from regions of interest, is one of the key concepts to be presented.

8:40am AS-MoM2 G-SIMS - Molecular Structure, I.S. Gilmore, M.P. Seah, National Physical Laboratory, UK G-SIMS or gentle SIMS has already been demonstrated to provide a straightforward way to simplify static SIMS spectra.1,2 These spectra are composed of parent fragment ions amongst a large number of high intensity degradation products. The fragmentation is quantified in terms of the partition functions of the fragments emitted from a surface plasma with effective temperature, Tp. It is found that fragmentation is least for high mass incident ions at low energies. By extrapolation of the data to low Tp, a new spectroscopy, known as gentle-SIMS or G-SIMS is formed. The significant peaks in the G-SIMS spectra are those peaks which would be emitted as a result of an ion impact that generates a surface plasma of very low Tp and thus have little post-emission rearrangement or fragmentation. Those peaks are, thus, directly characteristic of the material without rearrangement and enable direct interpretation and identification. G-SIMS has been successfully tested on a range of polymers, molecules and complex organics.3 G-SIMS not only provides the parent molecule mass, and hence composition in terms of numbers of C,H,O,N etc atoms, but is also allows the molecular structure to be evaluated. Current work explores the re-building of parent molecules using the fragmentation pathways which are mapped out as Tp is varied. This new dimension to the technique has significant potential information not previously obtainable in static SIMS. G-SIMS-MS has some of the attributes of MS-MS. These effects and the use of G-SIMS with mixtures will be discussed. 1I S Gilmore and M P Seah, Appl. Surf. Sci. 161 (2000) 465. 2I S Gilmore and M P Seah, SIMS XIII proc, Appl. Surf. Sci. in the press. 3I S Gilmore and M P Seah, Appl. Surf. Sci. 187 (2002) 89.

9:00am AS-MoM3 Recent Advances in Time-of-Flight SIMS, E. Niehuis, ION-TOF GmbH, Germany INVITED When TOF-SIMS was introduced in the early 80's, it appeared to be the ideal instrument for surface analysis in static SIMS mode. It's most striking features were a very high transmission close to 100 %, a parallel detection of all masses and an unlimited mass range. At that time, the current density of the pulsed primary ion beam was so low that the lifetime of the uppermost monolayer exceeded thousands of seconds. Early applications of TOF-SIMS focused on the analysis of involatile molecules, prepared as thin layers on metal substrates. Improvements in mass resolution to a level above 10,000 and the development of an efficient charge compensation opened many new fields in surface analysis, like the detection of trace elements and surface analysis of bulk organic materials such as polymers. The combination with liquid metal ion guns towards the end of the 80's added powerful imaging capabilities with a lateral resolution well in the

sub-micron range. In the mid 90's, TOF-SIMS started to become a depth profiling technique using the so -called dual beam mode. A low energy sputter gun was applied for sample erosion with a reasonable speed while the center of the sputter crater was analyzed with the pulsed high energy beam. In the recent years, TOF-SIMS has become a very powerful depth profiling technique with applications ranging from the analysis of ultra-shallow implants to the profiling of thick films. The combin ation of a small spot analytical beam with a low energy sputter beam offers new capabilities in 3-dimensional micro analysis. In this paper recent developments in TOF-SIMS instrumentation will be discussed and analytical examples from a variety of different fields will be given to highlight the capabilities of the technique.

9:40am AS-MoM5 Applications of Time-of-Flight Secondary Ion Mass Spectrometry in Materials Research, B.W. Schueler, Physical Electronics INVITED TOF-SIMS has gained increasing acceptance as a surface analysis techniques in scientific research and industrial applications. The most important features of the technique are its ability to (parallel) detect and identify all secondary ion species (molecular and organic)over a virtually unlimited mass range with high sensitivity. These properties result in an extremely efficient utilization of information from the analytical area and high detection sensitivity for identifying organic molecular species as well as trace element contaminants. The use of micro-focused primary ion beams enables measurement of the lateral surface distribution of molecules and atomic ion mapping with ~100nm resolution. Ion bombardment of organic surfaces results in the emission of characteristic fragments of the molecule and often the complete ionized molecule, providing a "fingerprint" spectrum of the molecule. This fingerprint may serve as unambiguous identification of the molecular species and functional groups. Parallel detection and high sensitivity of TOF-SIMS are equally important in inorganic (surface) analysis. For example, trace silicon surface metal contamination levels in the 0.1-1ppm levels across the periodic table are readily achieved. TOF-SIMS surface metal detection limits in Si are typically 10-100 times lower than those achievable by standard TXRF (Total Reflection X-Ray Fluorescence). This paper is intended to illustrate some key analytical and problem solving capabilities of TOF-SIMS with a range of materials research applications. The emphasis will be on semiconductor-related applications such as surface metal contamination transfer in process equipment, organic/inorganic contamination and etch/rinse residues, and depth profiling. Applications from the disk drive, and polymer industry will also be included. The relative merits of TOF-SIMS and other surface analysis techniques (i.e., XPS, AES, TXRF, etc.) will be discussed.

10:20am AS-MoM7 Method to Quantify the Comparison of Predicted vs. Experimental Isotopic Clusters in Time of Flight Secondary Ion Mass Spectrometry for High Mass Peak Identification, R.W. Nowak, C.M. Mahoney, State University of New York at Buffalo , A. Hawkridge, University of Arizona at Tucson, J.A. Gardella , State University of New York at Buffalo Investigation of the high mass range of the ToF-SIMS spectrum (800-4000 Da.) will yield information on long range polymer interactions which affect ion formation and will provide information about the long range forces between polymer chains. Peak assignment in the high mass region is not as straight forward as in the low mass region. Isotopes of the atoms that make up the fragments combine to yield clusters of peaks, not just single peaks as seen in the low mass regions. We are able to assemble predicted isotopic clusters to compare with the collected experimental data for identification. In past reports visual comparisons of the predicted and experimental isotope cluster results were used to judge the goodness of fit between two data sets. We have developed an analytical method to construct correlation plots for comparison with these data sets and to quantify the goodness of fit between the predicted and experimental results. This paper will illustrate a method to create correlation plots of the predicted vs. experimental results. Clusters from atomic constituents in the low mass region are used as a simple model to demonstrate the application. High Mass clusters from several previously published ToF-SIMS of high mass polymers will be utilized to show the effectiveness of this method.

10:40am AS-MoM8 TOF-SIMS Characterization of Mixed Decanethiol - Octadecanethiol Self-Assembled Monolayers, D.J. Graham, D.G. Castner, University of Washington Previous work in our labs has demonstrated the utility of TOF-SIMS in determining detailed chemical and structural information from self-assembled monolayers (SAMs). Much of this work has been done with homogeneous SAM surfaces. This study presents an investigation of a more

Monday Morning, November 4, 2002

3

complex mixed monolayer system using methyl-terminated SAMs. Other studies using mixed SAMs have used thiols with unique head groups to exploit a surface property or chemical signature. Thus, using methyl terminated SAMs increases the difficulty of interpretation due to the fact that there are no unique atoms to exploit between the thiol molecules. For this study, mixed monolayers of decanethiol (C10) and octadecanethiol (C18) were prepared from 1mM solutions at various mixture ratios (v/v). Using TOF-SIMS multivariate and univariate analysis, we demonstrate that the molecular specific information generated by TOF-SIMS enables both a detailed characterization of the mixed monolayer films and insight into their structure and surface arrangement. XPS analysis showed only the presence of the expected elements with all sulfur bound to the gold surface. Surface concentrations estimated by TOF-SIMS and XPS showed a significant enrichment of the longer chain thiol. Monitoring the yield of specific molecular ions in the TOF-SIMS spectra gave insight into the arrangement of the thiols on the gold surface. In particular the yield of the cluster ion Au[MC10-H][MC18-H] suggested the samples were phase segregated at least at a nanometer scale. Multivariate analysis of the TOF-SIMS data captured the changing composition across the series, and also highlighted structurally specific information about each thiol.

11:00am AS-MoM9 Secondary Ion Emission from Thick Organic Films: Influence of Primary Ion Bombardment Conditions, E. Tallarek, Tascon GmbH, Germany , F. Kollmer, ION-TOF GmbH, Germany, B. Hagenhoff, R. Kersting , Tascon GmbH, Germany Our ongoing studies focus on the secondary ion emission from thick organic layers under different primary ion bombardment conditions. As model analyte system spin coated layers of the polymer additve Irganox 1010 on low density polyethylene (LDPE) was used because the collision cascade takes place completely in organic material (simulation of "real" polymer conditions), the analyte of interest is only present in the uppermost monolayer (no contribution from deeper layers) and shows characteristic secondary ions covering the complete mass range from 1-1000u whereas the aliphatic substrate polymer LDPE emits ions only in the low mass range. The secondary ion parameters yield Y, disappearance cross-section σ and secondary ion emission efficiency E (yield per damaged area) were determined for primary ion bombardment with several monoatomic as well as polyatomic primary ions including Ga+, Cs+, Au+ as well as SF5+, Au2+ and Au3+. Additionally, the primary ion energies were varied between 4 and 25 keV. The results can be summarized as follows: For monoatomic primary ion bombardment the efficiency increases with the primary ion mass. Maximum enhancement factors compared to Ga+ bombardment are about 5-10. Polyatomic primary ion bombardment leads to enhancement factors of at least 60. Optimum primary ion energies depend on the respective ion species. These results have consequences for the achievable lateral resolution in ion imaging as well as for the detection limits in surface spectrometry. Examples will be given.

11:20am AS-MoM10 A New Approach to Measuring the TRUE Boron Profile Near the Si Surface using SIMS, T.H. Büyüklimanli, J.W. Marino , C.W. Magee, Evans East To date, SIMS measurements of the B distribution in the near-surface region have been made using an oxidizing ambient formed by either O2 flooding or by using normal incidence O2 bombardment to avoid possible near surface artifacts. However, the shape of the B depth profile obtained under these conditions at or near the surface has always been questioned. Separate measurements by poly -Si-encapsulation SIMS, RBS, ERDA and TEM have shown an implant peak below the surface contrary to the peak at the surface as usually depicted by SIMS with oxidizing ambient. Our paper investigates whether or not the apparent problems of profiling ULE B in Si under fully oxidizing conditions can be obviated by sputtering with oxygen under conditions that are not fully oxidizing. Correct measurement of the B profile shape in the top 1-2 nm of the sample would be useful to the TCAD modeling community. Our tests show that the B+ relative sensitivity factors with respect to Si+ were identical for both Si and SiO2. This suggests that the matrix effect due to the presence of surface oxide can be avoided by simply point-by-point normalizing the B+ intensity to the interpolated Si+ intensity. However, proper characterization of the detector electronics is required to record intensities accurately. The profiles acquired using low energy O2 bombardment at 60° incidence and processed using interpolated normalization resulted in doses matching values measured by nuclear reaction analysis (NRA). The profiles also show that B implant peaks for the range of low energies measured are, indeed, below the surface as measured by ERDA, RBS, TEM and poly -encapsulation SIMS. Another benefit of depth profiling without full oxidation is the ability to detect the surface oxide and location of the B implant with respect to the oxide. Profiles acquired under oxidizing ambient require a separate measurement to determine the oxide thickness and to correct for sputter rate changes from the oxide into the Si.

11:40am AS-MoM11 Study of Electron Beam Excited Plasma SNMS for High Detection Sensitivity , T. Noguchi, The Graduate University for Advanced Studies, Japan, S. Kato , KEK & The Graduate University for Advanced Studies, Japan SNMS has been developed for years to obtain precise depth profiling and high quantificability conquering the problems of SIMS. SNMS based on electron beam excited plasma (so called SNART: Sputtered Neutral Analysis-Riken Type) has several advantages; the high detection sensitivity with a high post ionization efficiency, the high depth resolution with low energy sputtering less than 100eV keeping a high sputtering rate, the relatively simple structure of the apparatus, the small dispersion of relative elemental sensitivity factors and the capability of insulator analysis without a charge neutralizer. In this study we attempted to improve the apparatus so as to obtain a higher sensitivity with maintaining a high depth resolution of around 1 nm. For this purpose, we have adopted a ToF mass spectrometer and made the plasma source be able to work in UHV. Detection sensitivities for metal surfaces were estimated to be less than 1 at.ppm keeping the high depth resolution on the basis of our preliminary experimental results of a sputtering rate, a post-ionization efficiency, a transmission of the ion optics and a total gain of the detection system. We will report our hardware development and results of experimental data compared with the estimated numbers.

Biomaterials Room: C-201 - Session BI-MoM

Theoretical Studies of Biosurfaces/Biotribology and Biorheology Moderator: R.A. Latour, Clemson University

9:00am BI-MoM3 Molecular Simulation Studies of Orientation of Antibodies Adsorbed on Charged Surfaces, J. Zhou, J. Zheng, S. Jiang , University of Washington Antibodies have found many applications in biotechnology and clinical medicine, including diagnostic assays, environmental testing, and process monitoring. It is well-known that Fab fragment of an antibody can bind its antigen with a very high specificity. Therefore, it is desirable to control the antibody orientation for immunoassay applications. In this work, Monte Carlo simulations are performed to study and predict the adsorption and orientation of antibodies as a function of surface and solution properties using hierarchical models, a simplified Y-shape 12-bead model antibody, a united-residue model, and an all-atom model. For all these three models, simulation results show that higher surface charge density and lower solution ionic strength favor narrower orientation distribution of adsorbed antibodies. Simulation results further show preferred antibody orientation under controlled surface and solution conditions, which are verified by our SPR and ToF-SIMS experiments. For the 12-bead model, it allows us to quickly map out the general trends of th e orientation behavior of antibodies on surfaces. For the residue model we developed, more detailed residue-distribution information of antibody near surfaces can be achieved. For the all-atom model, the conformation change of an adsorbed antibody was obtained with a proposed hybrid method. The fundamental understanding of antibodies on surfaces of this work will facilitate the effort to develop better biosensors.

9:20am BI-MoM4 Mapping the Free Energy State of Water in Hydration Layers and Its Importance for Ligand-Receptor Binding , G.W. Grahek, R.A. Latour, S.J. Stuart, Clemson University The thermodynamic contributions of solvent molecules during ligand-receptor binding are generally believed to be very important, but relatively little is actually understood regarding how the entropy, enthalpy, and free energy of hydration layers change as a ligand approaches and docks with its receptor. We assume that both ligands and receptors perturb the thermodynamic state of their localized hydration layers, and that these effects must be superimposed on the intervening water layers as a ligand approaches its receptor. It is hypothesized that this effect may have a significant influence on the height of the activation barrier for ligand-receptor binding, and thus may serve as an important medium-range modulator of ligand-receptor binding. Based on this underlying hypothesis, we are investigating the development of statistical mechanics based molecular modeling methods to calculate the entropy, enthalpy, and free energy values of water as a function of position surrounding a designated solute molecule. Simulations have been conducted using both molecular dynamics (AMBER 6.0) and Metropolis Monte Carlo (BOSS 4.2; OPLS) methods using TIP3P and TIP5P water, respectively, and periodic boundary conditions surrounding a centralized solute molecule. Entropy, enthalpy and

4

free energy are then mapped on a 3-dimensional grid. The simulations indicate that distinct changes do occur in the calculated free energy state of water molecules in the first two hydration layers surrounding the solute compared to bulk water. Further studies are being planned to investigate the effect of solute-solute separation distance on the intervening water layers. Following final development, these methods will be applied to actual ligand-receptor systems for the purpose of predicting the influence of water structure on binding. It is believed that these simulations may provide new insights that will facilitate drug design for specific receptor targets.

9:40am BI-MoM5 Growth of a Polymer Brush from Solution: Adsorption, Desorption, Conformational Conversion and Charging , H.J. Kreuzer, Dalhousie University, Canada INVITED Going beyond mean field theory we develop a model of a polymer brush that allows for inhomogeneity, confinement and lateral interactions. The model is developed for freely rotating chains and a realistic Interacting Chain Model for poly(ethylene glycol). The parameters in the latter are obtained from a first principles theory based on (i) ab initio (density functional theory) calculations of the potential energy surfaces of the polymer conformers, and (2) the proper statistical mechanics for which we succeeded to formulate and solve a Green's function approach (transfer matrix method) in the presence of an external force field. We set up kinetic equations for the time evolution of the growth of a brush from solution. Fo r PEG a detailed analysis and discussion of recent data is made that identifies two time regimes of pancake adsorption and collision-induced conformational conversion to stretched moeities, respectively. Lastly we discuss the possibility that auto-ionization of water in contact with the brush may lead to preferential adsorption of hydroxide and hydronium ions depending on the pH. H.J. Kreuzer, R.L.C. Wang, and M. Grunze, New Journal of Physics 1, 21.1 (1999). R.L.C. Wang, H.J. Kreuzer, and M. Grunze, Phys. Chem. Chem. Phys. 2, 3613 (2000). L. Livadaru, H.J. Kreuzer, and R.R. Netz. Interacting Chain Model for Poly(ethylene glycol) from First Principles. Macromolecules (in press). Kreuzer, H.J.; Payne, S.H.; Livadaru, L. Biophysical Journal 2001, 80(6), 2505-2514. Kreuzer, H.J.; Grunze, M. Europhys. Lett. 2001, 55(5), 640-646. M. Himmelhaus, T. Bastuck, S. Tokuitsu, M. Grunze, L. Livadaru and H.J. Kreuzer. Growth of a polymer brush from solution, (preprint).

10:20am BI-MoM7 Puzzles of Fluid Flow in the Biomaterials Environment, S. Granick , University of Illinois, Urbana-Champaign INVITED In areas from blood flow to biosensor applications, it is essential to predict fluid flow. The standard model states that fluid velocity is zero at solid surfaces, but evidence is accumulating against this in many situations, especially regarding aqueous solutions and surfaces coated with polymer cilia. We have studied flow of aqueous solutions containing variable amounts of monovalent and divalent electrolyte past solid surfaces whose charge was varied and whose 'softness' was varied by polymer cilia. Deviations from the standard model are observed when the wall shear stress exceeds a critical level whose magnitude depends on the system studied. In some respects this is understood, in other respects it is not. The puzzles will be emphasized.

11:00am BI-MoM9 Hydration Forces on a Switchable Bioactive Surface, B.-I. Kim, M.A. Samara, D.L. Huber, J.E. Houston, B.C. Bunker, Sandia National Laboratories Poly(n-isopropyl acrylamide) (PNIPAM) monolayers can be thermally switched between hydrophobic and hydrophilic states at a phase transition temperature of 35°C. Protein adsorption studies indicate that the hydrophilic state represents an anti-fouling state, while biomolecules form adherent monolayers on the hydrophobic state. We have used a scanning probe system called the interfacial force microscope (IFM) to probe the mechanisms for protein adsorption on this switchable polymer surface. With the IFM, we have simultaneously measured both normal and friction forces between a silica tip and a surface functionalized with PNIPAM as a function of separation distance. The results show that at room temperature, there is a repulsive hydration force between the tip and the substrate. As the phase transition temperature is approached, the repulsive force collapses, allowing the tip and substrate to come into adhesive contact. The transition from repulsive to attractive adhesive forces is accompanied by a doubling in lateral friction forces. IFM results obtained at different tip speeds at different temperatures suggest that the repulsive hydration force observed at room temperature is associated with the presence of ordered water structures within the polymer that break down at higher temperatures. Experiments are in progress with chemically -functionalized tips to provide us with fundamental insights of the parameters controlling the stability of this ordered water and its role in protein adsorption.

11:20am BI-MoM10 Structural Properties of Nucleosomal DNA Characterized by Atomic Force Microscopy , M.E. Greene, M.A. Ratner, J. Widom, M.C. Hersam, Northwestern University One of the fundamental problems in contemporary molecular biology involves whether a sequence dependence exists in nucleosomal DNA which gives the molecule certain structural properties leading to the formation of nucleosome with histone octamer. A way to approach the solution is to look at the isolated DNA molecules to discern the native structural properties in the absence of histones. Interfacing biological molecules with inorganic substrates and probing them using atomic force microscopy (AFM) allows for such study. AFM has been used to image surfaces with adsorbed biological molecules for over a decade, and in particular DNA has been characterized to an extent that imaging artifacts interfering with proper analysis have been identified. Several technical difficulties have been resolved as well, including substrate selection and a reproducible surface binding protocol, opening the door for AFM to be used as a powerful tool to investigate problems of genuine biological importance. In this investigation, a 342-bp strand of synthetic dsDNA dubbed "601" shown by Lowary and Widom to have a high affinity for binding to histone octamer is examined. This sequence is thought to mimic the behavior of DNA sequences found in chromatin. Preliminary analysis of AFM data of a natual nucleosomal DNA sequence isolated from chicken erythrocyte suggests agreement with the worm-like chain (WLC) model. Attention is given to the quantities of end-to-end distance, contour length, and intrachain bend angles in order to assess the persistence length, bendedness, and bendability of the sequence. AFM data is currently being gathered using Si cantilevers tipped with multiwalled carbon nanotubes as well as high aspect ratio Si tips with a nominal radius of curvature of 2 nm to obtain better lateral resolution and detailed measurements of bends and curvature fluctuations in the chains. An automated analysis methodology to allow the handling of large data sets will be introduced as well.

Electrochemistry and Fluid-Solid Interfaces Room: C-104 - Session EC+SS-MoM

Fuel Cells and Surface Electrochemical Reactions Moderator: J.G. Chen, University of Delaware

8:20am EC+SS-MoM1 Imaging of Water Ionization at Platinum Surfaces in High Electric Fields, C. Rothfuss, V. Medvedev, E.M. Stuve, University of Washington The high electric field intrinsic to the electrode/electrolyte interface plays an important role in electrochemical surface chemistry. To study these fields, which are of the order of 1 V/?, we employ a field ionization system in which water and other electrolytic species are adsorbed and ionized on Pt field emitter tips. Ions produced by the applied field are imaged onto microchannel plates and mass resolved with time-of-flight or ExB (Wien filter) mass spectrometers. Water ionization produces hydrated protons with 1-10 water molecules per proton, that are ejected from the tip. Images of ramped field ionization experiments show dramatic differences in ionization of amorphous vs. crystalline water. Below 135 K, where water exists in amorphous form, ionization is random overall, increasing in intensity with increasing field. Above 135 K, where water is crystalline, ionization occurs in long-lived zones that, with increasing field, increase in intensity and number and redistribute themselves about the surface so as to be as far apart as possible. Temperature dependent studies over the range of 80-300 K follow the energetic details of water ionization. Below 170 K the field required for dissociative ionization decreases linearly with increasing temperature. In a ramped field desorption experiment, ionization produces hydrated proton clusters with 2-7 water molecules per cluster. Above 170 K protonated clusters desorb sequentially beginning with the 6-water cluster and followed by progressively smaller clusters as the field increases. The disappearance of an n-water ion cluster results from loss of a water molecule to form cluster n - 1. The respective energies for water removal from clusters of n = 5, 4, and 3 were found to be 0.55, 0.76, and 0.85 eV. These numbers are in excellent agreement with previous measurements of water attachment energies. This work is supported by the Office of Naval Research.

8:40am EC+SS-MoM2 A Specular He Scattering Study of Water Adsorption, Desorption, and Clustering on Pt(111), J.L. Daschbach, B.M. Peden, R.S. Smith, B.D. Kay, Pacific Northwest National Laboratory Specular He atom scattering is used to probe the adsorption, desorption, and clustering kinetics of sub-monolayer H2O on Pt(111) over the temperature range 22 K to 185 K. Water deposited on clean Pt at low temperatures is shown to be arranged random ly on the Pt substrate. Over a narrow temperature range, as the clean substrate temperature is raised, the

5

deposited water transforms to a 2-D condensed phase. Rearrangement of randomly adsorbed H2O is studied as a function of coverage and temperat ure. At low initial H2O coverage the specular He waveform is dominated by the clustering of the isolated H2O molecules. At higher H2O coverage and temperature a second feature is manifest, which we interpret as the Oswald ripening of the 2-D islands. Adsorption and desorption kinetics are examined isothermally. Water, when fully clustered in two dimensions, gives rise to a He specular intensity that decreases linearly with coverage. Over essentially the entire sub-monolayer coverage regi m e the sample coverage changes linearly in time during both adsorption and desorption of H2O. This requires that the desorption rate be independent of coverage and thus the desorption kinetics are zero-order. The zero-order kinetics are a consequen ce of the coexistence of a 2-D H2O gas with a 2-D condensed H2O phase on the Pt surface. At higher temperatures, depending on flux, non-zero order kinetics are observed which are indicitive of a transition to a single H2O 2-D phase. Details of the experimental techniques and results will be presented. Pacific Northwest National Laboratory is a multiprogram National Laboratory operated for the Department of Energy by Battelle under Contract DE-AC067-76RLO 1830.

9:00am EC+SS-MoM3 Surface Chemistry of Solid Oxide Fuel Cells, R.J. Gorte, University of Pennsylvania INVITED Fuel cells are an attractive method for electrical power generation because they offer the possibility of very high efficiencies compared to normal heat engines. One of the major hurdles preventing their implementation for a wide variety of applications is the fact that, until recently, only H2 could be used as the fuel. We have recently demonstrated that stable power generation, without either internal or external reforming, can be achieved through the direct oxidation of hydrocarbons, including liquids, using a solid-oxide fuel cell (SOFC).1 The anodes in these direct-oxidation SOFC were composites made of Cu, ceria, and yttria-stabilized zirconia (YSZ). In this talk, the methods for preparing these anodes will be described. It will be demonstrated that surface chemistry and structure are crucial for improved performance of these fuel cells. Attempts to control the surface chemistry and structure will then be discussed. 1 S. Park, J. M. Vohs, and R. J. Gorte, Nature, 404 (2000) 265.

9:40am EC+SS-MoM5 Strategies for the Study of Methanol and CO Electrocatalysis on Solid Electrodes and Nanometer-Scale Supported Catalysts, C. Korzeniewski, G. Vijayaraghavan, L. Gao , Texas Tech University INVITED The electrochemical oxidation of methanol and related small molecules has been of special interest in relation to fuel cell research. The development of fuel cells that operate below 100 °C on methanol, or H2 has stimulated interest in the reaction steps involved in methanol and carbon monoxide oxidation at metal electrodes. In addition to being a by-product of methanol oxidation, carbon monoxide can also be present as an impurity in H2. Adsorption of carbon monoxide on the anode catalyst generally degrades its performance. We have approached the study of methanol and carbon monoxide oxidation with the use of electrochemical techniques in combination with in situ infrared spectroscopy, atomic force microscopy (AFM) and wet-analytical methods. This presentation will focus on the surface electrochemistry of methanol and carbon monoxide at supported Pt and Pt-Ru catalysts. In situ infrared measurements are being performed with Vulcan carbon supported fuel cell catalysts. The carbon supported materials are adsorbed onto a smooth gold electrode to enable infrared sampling in a standard reflectance geometry. A thermostatted cell allows in situ infrared measurements between ambient and 80 °C. Similar to the bulk metals, thermal effects on methanol oxidation at nanometer-scale catalysts are stronger for Pt-Ru (atomic percent Ru = 50%) than Pt. The influence that metal particle size distribution and spatial arrangement on carbon supports has on methanol oxidation pathways is being investigated by depositing metal particles on highly ordered pyrolytic graphite. The surface electrocatalytic properties of the supported particles are investigated with cyclic voltammetry. AFM is used to determine the catalyst size distribution and spatial arrangement at different stages of preparation and electrochemical characterization. Properties of nanometer-scale metal particles in relation to methanol oxidation pathways will be discussed.

10:20am EC+SS-MoM7 Potential Application of Tungsten Carbides as Direct Methanol Fuel Cell (DMFC) Electrocatalysts, H.H. Hwu, J.G. Chen, University of Delaware The Pt/Ru anode in direct methanol fuel cells (DMFC), though effective, is disadvantageous in terms of its prohibitively high costs and limited supplies. In this work, we are evaluating the effectiveness of tungsten and molybdenum carbides as alternatives to Pt/Ru electrocatalysts by studying their reactivities towards methanol, water, and carbon monoxide. Using Temperature Programmed Desorption (TPD) and High-Resolution Electron Energy Loss Spectroscopy (HREELS) the reaction pathways of these

DMFC molecules on carbide-modified Mo(110), W(110), and W(111) can be understood. On both W(110) and W(111) carbide surfaces, methanol readily decomposes into gas-phase CO, methane, hydrogen, and surface carbon and oxygen. In addition, both tungsten carbide surfaces are active toward the dissociation of CO and water. Preliminary studies on the Mo(110) carbide surface also show strong decomposition activity toward methanol, but through a different pathway than either the W(110) or W(111) carbide surfaces. Results from parallel studies of DMFC molecules on thin film tungsten carbides will also be presented.

10:40am EC+SS-MoM8 Ru Nanoparticles Prepared by Decomposition of Ru3(CO)12 on Au (111): Structural Characterization and Chemical Properties, T. Cai, Z. Song, Z. Chang, G. Liu, J.A. Rodriguez, J. Hrbek, Brookhaven National Laboratory Supported ruthenium metal particles prepared from ruthenium carbonyl have been shown as a most active catalyst for ammonia synthesis. In the emerging field of nanoscience, a goal is to make nanostructures with interesting functional properties. We have started a research program using metal carbonyls as precursors in the synthesis of nanoparticles on well-defined templates. In this study, we prepared and characterized a Au-supported Ru model catalyst under UHV by depositing metallic Ru on a Au (111) surface using triruthenium dodecacarbonyl, Ru 3(CO)12, as a molecular precursor. We used the reconstructed Au (111) surface as an inert template for metallic cluster growth. Carbonyl adsorbs molecularly on the surface at 90 K and starts to dissociate at 280 K by CO elimination, as shown in TPD studies. The complete decomposition of the carbonyl occurs above 500 K, leaving metallic Ru on the surface with no significant C or O as detected by XPS. Such an atomically clean Ru deposit is also obtained on Au (111) by MOCVD of Ru 3(CO)12 at an elevated substrate temperature of 550 K. The morphology of the Ru nanoparticles investigated by STM and their chemical reactivity toward simple molecules (CO, N2, NH3, O2, NO2) studied by XPS and TPD will be discussed. The research was carried out at BNL under Contract No. DE-AC02-98CH10086 with the U.S. DOE (Division of Chemical Sciences).

11:00am EC+SS-MoM9 Development of a Microreactor System for Electrocatalytic Studies of Methanol Oxidation, N. Arvindan, E.M. Stuve, University of Washington We report on the development of a microreactor for studies of methanol electro-oxidation on platinum catalysts. One of the primary benefits of the microreactor is the ease of temperature control and low consumption of reactants. Temperature can be adjusted and controlled nearly instantaneously over the range of 20 to 100 C. Higher temperatures are possible depending on the pressure limitations of the fluidic connec-tions to the microreactor. The microreactor enables studies of methanol electro-oxidation at high temperatures to achieve accelerated kinetics and freedom from CO poisoning. Methanol oxidation is measured at constant potential following a step from a non-reacting potential. Accumulation of surface species like CO is subsequently measured by linear sweep voltammetry. These two measurements enable the overall oxidation rate of methanol to be compared with the CO oxidation rate. Initial results demonstrate clean voltammetry of polycrystalline platinum electrodes for all temperatures. Reaction studies over the range of 80 to 100 C show that methanol oxidation occurs at the same rate as CO oxidation, consistent with the series reaction path (methanol to CO to carbon dioxide) being the dominant mechanism. The results conclusively show that thermal desorption of CO is insignificant, even at temperatures as high as 95 C. At 95 C turnover rates vary from 0.1 to 1 per second for the respective potential range of 400 to 600 mV vs. RHE. These results show that unmodified polycrystalline platinum is an effect catalyst for methanol oxidation at 95 C and support the feasibility of high temperature direct methanol fuel cells. This work is supported by the National Science Foundation and the UW Center for Nanotechnology.

11:20am EC+SS-MoM10 Combined Atomic Force Microscope and Acoustic Wave Devices: Application to Electrodeposition, J.-M. Friedt, L. Francis, K.-H. Choi, A. Campitelli, IMEC, Belgium We here present the development of an instrument based on a new combination of techniques including scanning probe microscopy (atomic force microscopy, AFM, in our case) and acoustic wave devices (quartz crystal microbalance - QCM - and acoustic wave resonators). We display the ways these two measurement techniques interact and show that their performances are not degraded through interaction. Using finite element analysis, we explain observations compatible with the generation of longitudinal acoustic waves in the liquid, creating standing wave patterns between the QCM sensing electrode and the AFM cantilever holder leading to resonance frequency instabilities of the QCM. QCM electrode vibration in liquid is also shown not to degrade AFM lateral resolution. We then show measurement results from electrodeposition of copper and silver on

6

gold electrode obtained using this instrument, and demonstrate how the data from both techniques (QCM-D and AFM) are complementary. Sin ce QCM-D allows simultaneous measurement of the resonance frequency at several overtones of the quartz crystal resonator as well as the dissipation (quality factor) of each of these overtones, we show how the relative frequency shifts of the overtones informs on the kind of interactions between the oscillating acoustic wave device and the surrounding media (electrodeposited layer and solution used for electrochemistry). This combined measurement was performed on AT-cut quartz resonators (QCM), SH-SAW lithium tantalate and quartz acoustic wave devices and Love mode quartz acoustic wave devices. Finally, we show that after identifying the types of interactions we can efficiently use electrodeposition as a mean of calibrating the sensitivity of acoustic wave devices. Sensitivities close to the theoretical values and compatible with previous values given in the literature are presented.

11:40am EC+SS-MoM11 Study of Bismuth Thin Film Electrodeposition and Oxide Formation on Au(111) , C.A. Jeffrey, D.A. Harrington, University of Victoria, Canada, S. Morin, York University, Canada Bismuth and bismuth oxide films have been well studied due to their magnetoresistive and semiconducting properties. In this work, the formation of electrodeposited bismuth thin films is studied using in -situ scanning tunneling microscopy (STM). Their growth mode and morphology provide useful information for the production of well-defined bismuth thin films. Electrodeposition of bismuth is performed on Au(111) in acidic solution and the bismuth film transformation to bismuth oxide in alkaline solution is studied using in -situ atomic force microscopy (AFM). Our study of the underpotential deposition process indicate that the reconstruction of Au(111) is lifted by the adsorbed bismuth, resulting in the formation of gold islands at potentials negative of 0.170 VSCE. Scans taken during the overpotential deposition process at potentials negative of -0.070 VSCE reveal 'needle' growth starting at step edges. These needles propagate over the surface and eventually form relatively uniform films. Atomic resolution images of the needle structures show the nearly rectangular unit cell 3.9 Å x 4.3 Å that contains one bismuth atom. The shorter side of the unit cell lies in the direction of the growth axis of the needle. This reduced spacing results in preferential incorporation of surface diffusing atoms at the needle tip, as opposed to along the edge, and accounts for the anisotropic growth. Bismuth oxide was formed by first forming the bismuth layer in acidic solution followed by a gradual shifting of the solution pH to a value of 10. Under these conditions, the transformation to the oxide film is monitored as the potential is made more positive. Close to the potential where the formation of bismuth oxide is expected, the morphology changes abruptly; small isolated protrusions form on the needle structures and cover the entire surface. The surface oxide formed can be reduced back to bismuth and this results in a disordered Bi film.

Electronic Materials and Devices Room: C-107 - Session EL+SC+MI-MoM

Semiconductors Moderator: A. Rockett, University of Illinois

9:00am EL+SC+MI-MoM3 Low-temperature Epitaxial Growth of the Wide Bandgap Semiconductor SiCAlN, I.S.T. Tsong , Arizona State University INVITED Two compounds, SiC and AlN, normally insoluble in each other below 2000C, are synthesized as a single-phase solid solution thin film by molecular beam epitaxy (MBE) at 750C using a unimolecular precursor H3SiCN and Al atoms. The growth of epitaxial SiCAlN films with hexagonal structure takes place on 6H-SiC(0001) and Si(111) substrates. The surface morphology, microstructure, and composition of the films are analyzed by atomic force microscopy (AFM), cross-sectional transmission electron microscopy (XTEM), Rutherford backscattering spectrometry (RBS) and high-resolution electron energy loss spectroscopy (EELS). Two structural models for the hexagonal SiCAlN films are constructed based on first-principles total-energy density functional theory calculations, each showing agreement with experimental XTEM observations. The predicted fundamental bandgap is 3.2 eV for the stoichiometric SiCAlN, in good agreement with photoluminescence (PL) measurements. Bandgap engineering is a distinct possibility by varying the composition of the pseudo-binary (SiC)-(AlN) film.

9:40am EL+SC+MI-MoM5 Evolution of Structure and Optical Properties of GaAsN Films Grown by Reactive Molecular Beam Epitaxy, M.J. Reason, W. Ye, X. Weng, V. Rotberg, R.S. Goldman, University of Michigan Narrow gap nitride semiconductor alloys have shown significant promise for a wide range of electronic, optoelectronic and photovoltaic applications. At present, the ultimate limit of nitrogen solubility in GaAs, as well as the effects of growth conditions on stress relaxation and optical properties of narrow gap nitride films are not well understood. In this work, we have examined the evolution of nitrogen incorporation, strain relaxation, and optical properties of GaAsN films grown by solid -source molecular beam epitaxy using an N2-rf plasma source. The samples consisted of 500 nm buffer layers of GaAs grown at 580C and 20 nm layers of GaAs grown at 500C, both using a high arsenic flux; followed by 100-500 nm thick layers of GaAsN grown at 400C using a 10% N2/Ar gas mixture at a 0.15 sccm flow rate, with a variety of arsenic beam equivalent pressures (BEP). The structure and properties of the samples were investigated by reflection high energy electron diffraction (RHEED), multi-beam optical stress sensing (MOSS), high resolution x-ray rocking curves (XRC), nuclear reaction analysis (NRA), atomic force microscopy, and photoluminescence. For all of the arsenic fluxes studied, in-situ RHEED during the GaAsN layer growth reveals a pattern similar to that observed during the growth of the GaAs layers. As the arsenic BEP is increased, MOSS shows that the film stress decreases, indicating a lowering of the apparent nitrogen incorporation into GaAsN. Interestingly, variatio ns in the absolute nitrogen concentrations determined from NRA analysis and a Vegard's law interpretation of XRC suggest significant nitrogen incorporation into interstitial sites. We will discuss the effects of arsenic flux on the stress relaxation and optical properties of a variety of GaAsN and InGaAsN films and heterostructures. This work was supported in part by the DOE (Photovoltaics Beyond the Horizon Program), the Air Force Office of Scientific Research (MURI Program), and the TRW Foundation.

10:00am EL+SC+MI-MoM6 Effects of LED Processing Steps on the Surface of Doped GaN Epilayers, K.H.A. Bogart, D.D. Koleske, A.A. Allerman, A.J. Fischer, K.W. Fullmer, K.C. Cross, C.C. Mitchell, Sandia National Laboratories Gallium nitride (GaN)-based materials are critical for the creation of UV optoelectronic devices such as light-emitting diodes (LEDs). Ohmic contacts with low contact resistivities to p-type (

7

11:00am EL+SC+MI-MoM9 N-type Diamond Electronics With Nitrogen Doped Ultrananocrystalline Diamond, J.E. Gerbi, Argonne National Laboratory , B.W. Alphenaar, University of Louisville, O. Auciello, Argonne National Laboratory, J. Birrell, University of Illinois at Urbana-Champaign, J.A. Carlisle, D.M. Gruen , Argonne National Laboratory Thin diamond films have extremely attractive properties for electronic device applications: high thermal conductivity, carrier mobility, and breakdown fields. However, efforts to create diamond based electronic devices have been hampered by the difficulty in incorporating dopants. Attempts to dope diamond films have resulted in low p-type carrier concentrations or unstable p-type surface layers. N-type doping has been even less successful, and it has not yet been possible to synthesize n-type diamond films with sufficiently high room-temperature conductivities. Ultra-nanocrystalline diamond (UNCD) is a fine-grained (3-5nm), phase-pure diamond material with atomically abrupt grain boundaries. Synthesized by microwave CVD using Ar-rich Ar/CH4 plasmas, both the structure and electronic properties of UNCD can be tailored by doping with a controlled amount of N2 in the plasma. As the N2 content in the plasma increases to 20% , the grain size and grain boundary width of the UNCD films increase. This microstructural change correlates with a striking increase in room-temperature conductivity . Most importantly, nitrogen doped UNCD films are n-type with activation energies as low as 0.05 eV. This is striking, as traditional nitrogen substitutional doping of diamond produces a very deep state of 1.7eV, rendering the material useless for room-temperature applications. We use this material to demonstrate the first n-type diamond MESFET that can be operated at room temperature. We have characterized the films using Raman spectroscopy, NEXAFS, SIMS, Hall mobility measurements, and HRTEM, and measure device properties such as I-V curves and transconductance. The ohmic vs. Schottky behavior of various contacts to nitrogen doped UNCD as a function of growth chemistry has also been studied. A discussion of the microstructure-property relationship of nitrogen-doped UNCD films will be presented in the context of the UNCD-based MESFET performance.

11:20am EL+SC+MI-MoM10 Electronic Structure and Spin-Polarization of Mn-containing Dilute Magnetic III-V Semiconductors, L. Kronik, M. Jain, J.R. Chelikowsky, University of Minnesota INVITED The systematic use of electron spin, in addition to its charge, holds great promise for a new class of semiconductor devices with unprecedented functionality. Recently, Mn-containing, "dilute magnetic", III-V semiconductors have emerged as candidate materials for such a technology. They can potentially produce charge carriers with well-defined spin, yet are compatible with already existing semiconductor technologies. In order to assess the performance limits of such materials theoretically, we present first principles pseudopotential - density functional calculations for the electronic structure of the dilute magnetic semiconductors MnxGa1-xAs and and MnxGa1-xN, with an experimentally relevant realistic x=0.063, in their ordered ferromagnetic phase. We predict that both materials allow, in principle, for a theoretical limit of 100% spin-injection, and that spin -polarized transport can be attained in both materials in the context of a simple band picture. This is because in MnGaAs, hybridization of As 4p and Mn 3d orbitals splits the valence band, resulting in a ~0.5 eV energy range where holes have a well-defined spin and an effective mass comparable to that of GaAs. In MnGaN, the situation is even more favorable: hybridization of Mn 3d and N 2p orbitals results in the formation of a ~1.5 eV wide impurity band, which supports effective mass transport. We will discuss the technological impact of these findings and compare our results to pertinent experimental data.

Magnetic Interfaces and Nanostructures Room: C-205 - Session MI+EL-MoM

Spintronic Materials and Hybrid Devices Moderator: B.T. Jonker, Naval Research Laboratory

8:20am MI+EL-MoM1 Spin-transport in Ferromagnet/Semiconductor Structures, R. Jansen, University of Twente, The Netherlands INVITED Taking full advantage of electron spin in spin-electronics will eventually require an intimate integration of ferromagnetic and semiconductor materials. While device concepts are emerging, the understanding of spin transport in such hybrid ferromagnet/semiconductor structures is still at its infancy. We have focused on transport of non-equilibrium, hot-electron spins, for which spin currents can be controlled and manipulated via the electron energy and momentum. A particularly useful device for that purpose is the spin -valve transistor1, consisting of a metallic spin-valve base, sandwiched between a semiconductor emitter and collector. Using the spin-valve transistor, we address the relative importance of interface,

volume and thermal scattering of hot electron spins, and present new insight into the sources of spin -asymmetry in hot-electron transport.2,3 From an application point of view, enhancing the output current of the transistor is desired. We demonstrate several routes to enhance the transfer ratio, culminating in an overall improvement by two orders of magnitude while preserving the low-field magnetic response above 200% at room temperature.4 We also present transport in novel structures such as the magnetic tunnel transistor and the hot-electron spin-filter, and demonstrate that the latter allows room temperature injection of almost fully spin -polarized electrons into semiconductors. 1 R. Jansen et al., J. Appl. Phys. 89, 7431 (2001). 2 R. Jansen et al., Phys. Rev. Lett. 85, 3277 (2000). 3 R. Vlutters et al., Phys. Rev. Lett. 88, 027202 (2002). 4 O.M.J. van 't Erve et al., Appl. Phys. Lett. 80, to appear 20 may 2002.

9:00am MI+EL-MoM3 Spin Dependent Electron Transport in Hybrid Ferromagnet/GaAs Structures at Room Temperature, S.J. Steinmuller, W.S. Cho, A. Hirohata, C.M. Guertler, G. Wastlbauer, T. Taniyama, J.A.C. Bland, University of Cambridge, UK We report on the investigation of room temperature (RT) spin dependent electron transport in ferromagnet(FM)/GaAs hybrid Schottky barrier structures by photoexcitation. Spin accumulation in the GaAs was achieved by optical pumping with circularly polarised light. The photon helicity and the applied magnetic field were both introduced perpendicular to the plane of the film. Various FM materials were used (NiFe, Fe and Co) and investigated at different thicknesses (t=2.5nm, 5.0nm and 7.5nm). Furthermore an antiferromagnetic Cr sample was prepared as a reference. We measured the helicity -dependent photocurrent (PC), that is the difference in PC for illumination with right (i+) and left circularly polarised light (i-), for applied magnetic fields in the range from -2 T to 2 T as well as the spin polarisation P=(i+ - i-)/(i+ + i-) of PC. NiFe and Fe showed a rather strong effect (P in the range 0.2-2%) increasing with film thickness, whereas almost no effect was observed in the Co. The magnetic field dependence of the helicity -dependent PC was in good agreement with polar MOKE measurements, proving that magnetic effects in the GaAs are negligible at RT. No field dependent effect was seen for the Cr as expected. Moreover we carried out measurements at different doping densities of the GaAs substrate (n- and p-type), showing the importance of the Schottky barrier in our experiment, and different photon energies. We also discuss the results of similar measurements on NiFe/Cu/Co spin valve structures. We show that our combined data provides strong support for our model of electron spin filtering at RT based on tunnelling of spin polarised electrons across the Schottky barrier followed by ballistic transport in the FM.

9:20am MI+EL-MoM4 Tunneling Transport Across Reverse Biased Ag/Fe/Ag/GaAs Schottky Barriers, D.A. Hite, S.E. Russek, D. P. Pappas, National Institute of Standards and Technology Electrical transport characteristics for the epitaxial Ag/Fe/Ag/GaAs(100) system have been studied under various growth conditions. The surfaces and structure of the multilayer were characterized by low energy electron diffraction and angle-resolved Auger electron diffraction at all steps of the fabrication. We have been able to prepare clean, well-ordered, epitaxial multilayers. The ultra-thin Ag buffer layer (~7 atomic layers) was prepared in a manner to create an ultra-thin layer to mediate the growth morphology of the Fe layer, to prevent the undesired intermixing associated with the Fe/GaAs system, and to create a tunneling barrier in reverse bias. In-situ conductance spectroscopy measurements were performed in order to characterize the rate of electron injection into the semiconductor as a function of bias voltage. We find that these multilayer diodes exhibit a reverse bias tunneling effect above 0.6 V. This is significant because it shows that we have been able to overcome the conductivity mismatch problem between the Fe and GaAs using an ultra-smooth, ultra-thin Ag buffer layer. The possibility of using these structures for direct spin injection from the Fe across the Ag/GaAs Schottky barrier will be discussed.

9:40am MI+EL-MoM5 Modeling of Spin Injection into Disordered Semiconductors, E.Y. Tsymbal, University of Nebraska-Lincoln , V.M. Burlakov, University of Oxford, UK, I.I. Oleinik, University of South Florida INVITED Spin injection into semiconductors is a topic of growing interest within the field of spin electronics. Developing a realistic model for sp in injection is important both for the understanding of basic mechanisms that govern this phenomenon and for the application of spin injection in semiconductor devices. All the existing models so far either take into account a realistic band structure but neglect disorder within the semiconductor or consider phenomenologically defect scattering within a free-electron-type model. This talk will address the approach which combines an accurate description of the atomic structure, the electronic structure, and the conductance within

8

a unique microscopic model. Within this approach the atomic structure is simulated using Metropolis MonteCarlo technique, the electronic structure is modeled using a multiband tight-binding approximation, and the conductance is calculated using the Landauer-Buttiker formalism including inelastic scattering. We will demonstrate results of the application of this model to amorphous silicon - a representative semiconducting material suitable for spin injection. We will discus decisive factors that control the efficiency of spin injection into disordered semiconductors.

10:20am MI+EL-MoM7 Surface and Interface Properties of a Half-Metallic Alloy, S.J. Jenkins, D.A. King , University of Cambridge, UK INVITED The ferrimagnetic semi-Heusler alloys have received considerable experimental and theoretical attention since the prediction in 1983 that they may show half-metallic behaviour (i.e. complete spin polarization at the Fermi level). Bulk properties of these alloys are now well understood. Nevertheless, application of these materials in practical situations is likely to be contingent upon the influence of their surface and interface properties, which have thus far been less thoroughly studied. This presentation will focus on recent calculations of the electronic structure of the NiMnSb(001) surface, with and without an Sb overlayer, carried out within the framework of density functional theory. Particular emphasis will be placed on the role of surface- and interface- localised electronic states in modulating the Fermi-level polarization.

11:00am MI+EL-MoM9 Electrical Spin Injection from NiMnSb into GaAs, W. Van Roy, P. Van Dorpe, V.F. Motsnyi, G. Borghs, J. De Boeck, Imec, Belgium We demonstrate electrical spin injection from NiMnSb into a GaAs light-emitting diode (LED). We compare single crystalline films grown epitaxially on GaAs(111)B with and without an additional AlAs tunnel barrier, and polycrystalline films grown on top of an AlOx tunnel barrier on GaAs(001). The LEDs and NiMnSb films were grown by MBE in two chambers connected under vacuum. For the deposition of AlOx tunnel barriers the samples were transported through air to a sputter system for the deposition of Al and oxidation in a controlled O2 atmosphere. Spin injection was measured optically at T = 80 K. Electrons were injected with an in-plane spin-component. We used the oblique Hanle effect to transform this spin ensemble into an out-of-plane ensemble and used the circular polarization of the light emitted in the surface-normal direction as a measure of the electrical spin injection. The results were corrected for the out-of-plane tilting of the NiMnSb magnetization in the small oblique magnetic field, and for the MCD effect. We find electrical spin injection of up to 5% for polycrystalline NiMnSb films on top of an AlOx barrier. The spin injection drops with increasing bias voltage. The low values indicate a strongly reduced spin polarization for the polycrystalline NiMnSb films. Epitaxial NiMnSb films, especially on (111)B interfaces, are expected to show a much larger spin polarization for the conduction carriers. However, we did not yet observe spin injection from these films. This is attributed to the low interface resistance of this configuration in combination with a NiMnSb surface polarization that, although larger than for the polycrystalline films, is still short of 100%.

11:20am MI+EL-MoM10 Epitaxial Growth and Annealing Studies of Single Crystal, Ferromagnetic Co2MnGa of GaAs (100), D.M. Carr*, S. McKernan, F.M. Abdulle, J.W. Dong, C.J. Palmstrom, University of Minnesota Spintronic devices that use electron spin in semiconductor devices are promising candidates for the next generation of electronic devices. Ferromagnetic metals with high spin polarization may be required for successful implementation of these devices. The Heusler alloys are a promising family of metals because the material properties such as lattice parameter, saturation magnetization, Curie temperature, and spin polarization can be altered by changing the elemental composition. In addition, their lattice parameters span most of the lattice parameters of the compound semiconductors. Films of the Heusler alloy Co2MnGa have been epitaxially grown on GaAs (100) using molecular beam epitaxy. In situ reflection high-energy electron diffraction patterns and ex situ x-ray diffraction patterns of 300 Å thick films indicate single crystal growth with an out-of-plane lattice constant of 5.94 Å, which suggests tetragonally distorted growth since the bulk lattice parameter is 5.77 Å. Variable temperature vibrating sample magnetometry measurements show the Co2MnGa films to be ferromagnetic with in-plane magnetization and a Curie temperature close to the bulk value of approximately 690 K. Ex situ annealing at different temperatures from 300 to 450 degrees C reveals an increase in the saturation magnetization and reduced coercivity for anneal

* Falicov Student Award Finalist

times as short as 5 minutes at 425 degrees C. Annealed films exhibit a smaller out-of-plane lattice constant suggesting relaxation of the strained films. Cross sectional transmission electron microscopy studies will be used to characterize the level of interfacial reaction before and after annealing.

11:40am MI+EL-MoM11 Growth Temperature Controlled Magnetism in Molecular Beam Epitaxially Grown Ni2MnAl Heusler Alloys, X.Y. Dong , J.Q. Xie, J.W. Dong*, T.C. Shih, S. McKernan, C. Leighton, C.J. Palmstrom, University of Minnesota The Heusler alloy Ni2MnAl is thought to be either antiferromagnetic or ferromagnetic depending on its crystal structure, B2 (disordered Mn-Al sublattice) or L21.1 This suggests that a ferromagnet/antiferromagnet interface should be possible to be formed with Ni2MnAl by controlling its crystal structure. Single crystal Ni2MnAl thin films have been grown by MBE on GaAs (001) using Sc0.3Er0.7As interlayers. The effects of growth temperature on its structural and magnetic properties were studied. For all the films grown at different temperatures, streaky RHEED patterns were observed during the growth. The Ni2MnAl / Sc0.3Er0.7As / GaAs (001) films were single crystals with cube on cube epitaxial relationship. The Rutherford backscattering spectrometry channeling minimum yield, χmin, of ~ 5 %, confirms a relatively good quality crystal. XRD and TEM show that the Ni2MnAl films have a tetragonally distorted structure with its c axis oriented along the growth direction. Higher growth temperature tends to result in ferromagnetic films suggesting a more L21-like structure, while lower temperature growth gives rise to non-ferromagnetic behavior, suggesting a more B2 -like structure. For the ferromagnetic Ni2MnAl film, the Curie temperature was determined to be approximately 220K. The exchange bias effect was observed for Co (70Å) / Ni2MnAl (360Å, B2-like structure) bilayers, suggesting that the low temperature grown Ni2MnAl is antiferromagnetically ordered. Therefore the self exchange biased Ni2MnAl bilayers can be expected to be grown by varying the temperature during the growth. In this presentation, the effect of growth temperature on the structural and magnetic properties will be emphasized as well as the approaches of making self exchange biased structures will be reported. 1 F. Gejima, Y. Sutou, R. Kainuma, and K. Ishida, Metal. Mater. Trans. A 30A, 2721 (1999).

Manufacturing Science and Technology Room: C-109 - Session MS+SE-MoM

In-Situ Monitoring and Metrology for Coating Growth and Manufacturing Moderator: A. Diebold, International Sematech

8:20am MS+SE-MoM1 Product Development and Yield Enhancement through Failure Analysis of Integrated Circuits with Scanning Capacitance Microscopy , P. Tangyunyong , C.Y. Nakakura, Sandia National Laboratories Scanning capacitance microscopy (SCM) has become a widely used metrology tool in the microelectronics industry due to its ability to measure two-dimensional free carrier profiles with nanometer-scale resolution. To date, SCM has been used primarily to characterize source/drain formation by imaging cross-sectioned, metal-oxide-semiconductor field effect transistors (MOSFETs). We have extended the role of SCM in our Fab from an off-line research instrument to a routinely -used failure analysis tool, active in providing feedback in new product development, process validation, and yield enhancement. The SCM measurement can be performed on any two-dimensions of the sample, thus providin g unique information that cannot be obtained with other analysis techniques. This information has been instrumental in helping to identify several yield -limiting defects in our CMOS device product line. In addition, SCM measurements are performed in-house with quick turnaround, yielding a considerable advantage over off-site analysis techniques, such as secondary ion mass spectroscopy. The methodology for performing both top-down (parallel to the wafer surface) and cross-sectional SCM measurements will be presented. We will show, in detail, several examples of how SCM information has been used to identify the root causes of device failures and discuss some of the corrective actions taken to reduce defects and improve yield. This work was performed at and supported by Sandia National Laboratories under DOE contract DE-AC04-94AL85000. Sandia National Laboratories is a multi-program laboratory operated by Sandia Corporation for the United States Department of Energy.

9

8:40am MS+SE-MoM2 Integrated CD Metrology for Poly Si Etching , G.P. Kota , C. Lee, Lam Research Corporation, T. Dziura, A. Levy , KLA-Tencor Corporation Advanced process control (APC) is gaining widespread use because of the costs associated with 300mm wafer processing and because of the stringent control required for CD and profile due to the shrinkage of critical feature dimension. APC can be used in Feed forward, Feed back and Fault detection control modes. The KT metrology module called iSpectra is integrated on to the Lam 2300 Versys etch platform. This modular design allows for real time APC. Integrated metrology also enhances the overall equipment efficiency. A comparison of iSpectra, CD-SEM, and x-SEM results will be presented. iSpectra shows good correlation to the CDSEM measurements as well as x-SEM profiles. In addition, iSpectra repeatability is superior to conventional methods such as CD-SEM. It is common knowledge that 193nm PR shrinks during CDSEM measurements due to exposure of the PR to e-beam. This shrinkage has been measured to be up to 17nm after about 30 repeated measurements on the CDSEM. In comparison, the iSpectra measurement technique results in minimal CD shrinkage.

9:00am MS+SE-MoM3 Metrology for Manufacturing, U. Whitney, KLA-Tencor INVITED

9:40am MS+SE-MoM5 Measurements of Shallow Trench Isolation by Normal Incidence Optical Critical Dimension Technique, J. Hu, D. Shivaprasad, F. Yang, R. Korlahalli, Nanometrics, Inc. Shallow Trench Isolation (STI) has emerged as one of the primary techniques for device isolation in complementary metal-oxide semiconductor (CMOS) technologies. This device isolation technology has become extremely important to satisfy the high density requirements of modern integrated circuits, It is of paramount importance to measure the critical dimensions of the STI structure. Currently used CD-SEMs cannot identify the rounding typically present at the bottom or top of the profile, and it is difficult to differentiate between the top and bottom line-width values. X-SEMs which can give the profile information require the destruction of the wafer. In this paper we present the work done on STI using the Optical Critical Dimension (OCD) technique. This technique measures line or trench profiles using Normal Incidence Polarized Reflectometry with a sensitivity to sub-50nm grating lines. In the OCD technique, a broadband polarized light beam is focused onto the grating surface, and the reflected 0th order is measured as a function of wavelength. The data obtained by measuring the grating structure gives a signature of the profile structure which is analyzed in real time using Rigorous Coupled Wave Analysis (RCWA). Since the data is fitted in real time, there is no requirement for library generation, which makes the analysis simpler and easier to extend to other structures without the need of lengthy re-generation of a new library of profile data. Data from STI wafers before and after ashing (removal of the developed resist) will be presented. Sensitivity to the oxide notching and repeatability data will also be presented.

10:00am MS+SE-MoM6 The Evolution of Single Atomic Steps on vicinal Si(111) in NH4F, J. Fu, National Institute of Standards and Technology , H. Zhou, University of Maryland , J.A. Kramar, R. Silver, National Institute of Standards and Technology Determining the width of a feature or the scale in a pitch measurement with appropriate accuracy is fundamental for process control in state-of-the-art semiconductor manufacturing. To meet these needs as well as the future measurement and calibration needs of the emerging nanomanufacturing industry, the National Institute of Standards and Technology (NIST) has been pursuing research and development on techniques for the fabrication and measurement of atom-based dimensional standards. The key elements in the development of atom-based standards are the ability to prepare atomically ordered surfaces,and the ability to count the atoms making up the features of interest. One of the most difficult challenges in atom-based metrology has been the fabrication of an appropriate atomic template. Atomically ordered surfaces provide an intrinsic template which have both scale and orthogonality. Using Scanning Probe Microscopy(SPM), We have examined the surface produced by etching several different vicinal Si(111) sample in 40% NH4F. In agreement with others, we find that deoxygenation of the etchant generally reduces the number of triangular etch pits. The formation of single atomic steps is evolved from these etch pits. These etch pits undergo nucleation, growth, merging, and corner rounding which can lead to single atomic steps. We also find that for maximum uniformity and minimum root mean square roughness, a certain minimum miscut angle is required. This angle is related to the maximum clear terrace width, which in turn is related to the relative etching rate of the step-edge sites and the terrace sites. The time evolution of the surface-smoothing etching process was also examined.

10:20am MS+SE-MoM7 Real Time in situ Spectroellipsometry , J.A. Woollam, B. Johs, J. Hale, J. A. Woollam Co., Inc. INVITED This talk reviews applications of spectroscopic ellipsometry for in situ monitoring and control during deposition, thermal processing, and etching of surfaces and thin -films. In situ spectroscopic ellipsometry is valuable for calibrating film growth and etch rates, controlling the thickness of each layer in multi-layer structures, and investigating nucleation phenomenon. It is also useful for measuring surface and interfacial roughness, substrate and film optical constants (with and without surface oxides), alloy composition, and substrate temperature. There have been numerous challenges to implementing in situ spectroscopic ellipsometry, including how to deal with substrate wobble and the effects of windows, and how to accurately measure thickness and material properties during growth of large numbers of layers in multi-layer stacks. Solutions to these practical problems will be discussed, and example applications described.

11:00am MS+SE-MoM9 Real Time Process Control by Spectroellipsometry , D. Daineka, P. Bulkin, T. Novikova, B. Drévillon , CNRS, Ecole Polytechnique, France In situ ellipsometry is well known to be the most sensitive, non-invasive tool for monitoring and control of thin film growth. In the fabrication of optical coatings and thin films in general the refractive index of the material is usually assumed to remain constant within a single layer. With such assumption only optical thickness of the layer can be controlled. For modern complex structures, however, even insignificant variation in the refractive index can be very detrimental to the final performance of the coating. Simultaneous real-time determination of refractive index and growth rate is required in order to comply with strict specifications. If the index departs from the pre-calculated target value, one has to adjust process parameters. In PECVD such control variables are gas flows of the precursors. We report on the closed-loop control of the silicon oxynitrides deposition by in situ phase modulated kinetic spectroellipsometry using a direct numerical inversion algorithm for the real-time reconstruction of refractive index and layer thickness. This technique is tested on constant index layers as well as on graded refractive index profiles and shown to be efficient and reliable.

11:20am MS+SE-MoM10 In-Situ Studies of the Amorphous to Microcrystalline Transition of Hot-Wire CVD Si:H Films Using Real-Time Spectroscopic Ellipsometry , D.H. Levi, B.P. Nelson, J.D. Perkins, National Renewable Energy Laboratory In-situ real-time spectroscopic ellipsometry (RTSE) provides detailed information on the evolution of the structural and optical properties of Si:H films during growth.1 We have used in-situ RTSE to characterize the morphology and crystallinity of hot-wire CVD (HWCVD) Si:H films as a function of substrate temperature Ts, hydrogen dilution R=[H]/[H+SiH4], and film thickness db. Transitions from one mode of film growth to another are indicated by abrupt changes in the magnitude of the surface roughness during film growth. The degree of crystallinity of the film can be determined from the bulk dielectric function. We have studied the growth parameter space consisting of R from 0 to 14, T s from 250oC to 550oC, and db from 0 to 1 µm. For each set of R and T s values, the structural evolution of the film can be characterized by the shape of the surface roughness thickness ds versus bulk thickness db curve. In contrast to studies done by Collins et al on PECVD growth of Si:H films, our studies of HWCVD growth find no conditions where ds remains constant after coalescence of the initial nucleation centers. Most of the films grown within the range of parameters studied exhibit a secondary nucleation and coalescence signature. The transition between a-Si:H and uc-Si:H growth is near the R=3 to R=4 dividing line. Initial coalescence of purely uc-Si:H material does not occur until R>8. We have verified the RTSE crystallinity classification using ex-situ Raman scattering. 1 R.W. Collins, Joohyun Koh, H. Fujiwara, P.I. Rovira, A.S. Ferlauto, J.A. Zapien, C.R. Wronski, R. Messier, Appl. Surf. Sci., 154-155, 217-228 (2000).

11:40am MS+SE-MoM11 Post-Deposition Control of Resistivity and Anisotropy in ZnO Thin Films, J.S. Lewis, B. Stoner, C. Pace, MCNC A method for post-deposition control of the resistivity of ZnO thin films has been developed, and a method for providing anisotropic sheet resistance in the plane of the film has been demonstrated. Military needs for real-time image processing can be met using thin film analog image processor (TAIP) devices. TAIP chips provide compact and power-efficient analog processing, including high- or low-pass spatial frequency filtering. The analog spatial filters are based on the RC time constant of the circuit, and therefore require thin films with controlled, repeatable sheet resistance in the range of MΩ/sq. This range of sheet resistance can be difficult to achieve with good repeatability for inorganic films. ZnO thin films were sputtered from an undoped ZnO target by RF magnetron sputtering. The as-deposited sheet resistance of the films was in the range 5-50 kΩ/sq. Post

10

deposition processing yielded films with sheet resistance in the range from the as deposited value to > 100 MΩ/sq. Target values of sheet resistance were obtained routinely. Using an in-situ monitor of sheet resistance during processing resulted in much better repeatability than that possible for as-deposited films. For TAIP chips, anisotropic sheet resistance in the plane of the film can allow more sophisticated algorithms for image processing. Post processing techniques were used to fabricate ZnO thin films with sheet resistance anisotropy ratios in the range of 2:1 to 25:1, and larger anisotropies should be possible. This work was sponsored by DARPA (contract no. DAAD19-00-1-0002).

Nanometer Structures Room: C-207 - Session NS-MoM

Nanomechanics Moderator: N.A. Burnham, Worcester Polytechnic Institute

9:00am NS-MoM3 Structural Properties of Polymeric Nanostructures, J.J. de Pablo , University of Wisconsin-Madison INVITED

9:40am NS-MoM5 Dynamics and Mechanics of Nanoscale Adhesive Contacts, K.J. Wahl, U.S. Naval Research Laboratory , S.A.S. Asif, Hysitron, Inc. INVITED Recent advances in atomic force microscopy (AFM) and nanoindentation enable examination of surface mechanical properties of ultrathin films and compliant materials with far greater resolution and accuracy than ever before. In our laboratory, we have implemented dynamic mechanical analyses of nanoscale adhesive contacts using a 'hybrid' nanoindenter, coupling depth -sensing nanoindentation with AFM positioning capabilities. This combination allows surface sensitive, quantitative mechanical properties measurements of nanostructures and thin films, at a single point as well as while scanning. We illustrate these expanded capabilities with several examples: 1) a dynamic nanoscale Johnson-Kendall-Roberts (nano-JKR) adhesion test, and 2) scanning nanomechanics. The nano-JKR test allows study of processes that occur during the formation and breaking of adhesive contacts with diameters smaller than the optical limit, and can be used to measure dynamic visco-elastic properties including loss and storage moduli, adhesion energy, and strain energy release rate. Scanning nanomechanics provides a means of directly imaging mechanical response and properties with sub-micron spatial resolution. We will discuss how these new capabilities can be used to test the models and limits of continuum contact mechanics.

10:20am NS-MoM7 Creep Compliance and Stress Intensity in Small Viscoelastic Contacts, W.N. Unertl, University of Maine, M. Giri, Hewlett-Packard - Corvallis Adhesive contacts to viscoelastic materials with dimensions smaller than a few microns are difficult to analyze. This is due, in part, to the inability to measure the contact size directly. One consequence is the lack of a quantitative method to measure time-dependent mechanical properties. We demonstrate a method to overcome these difficulties. First, we extend a theory of viscoelastic contact1 to show how the contact radius, the stress intensity at the contact edge, and the creep compliance function can be extracted directly from load vs. deformation data. Then, we apply this analysis to load controlled indentation data for a paraboloidal diamond probe on a styrene-butyl acrylate substrate with 27 C glass transition temperature. The probe is brought into contact, the load is increased linearly to a predetermined maximum, and then decreased until the contact ruptures. Loads up to 3 mN result in deformations up to 2 mm in depth depending on the loading rate and contact time. Viscoelastic effects, indicated by the occurrence of maximum penetration after maximum load, were largest for contact times near 20 s. Calculated contact radii are up to 6 mm. The creep compliance for this material is described by a power law in time with exponent near 0.8. In contrast to predictions of simple fracture mechanics models, the stress intensity is not a unique function of the speed of the contact edge. This suggests either an interaction potential between the probe and polymer that is rate dependent or a polymer response that is non-linear under the conditions that occur at the contact periphery in these experiments. These results bring into question all previously reported nanoscale measurements of the mechanical properties of viscoelastic materials. 1C.Y Hui, J.M. Baney, and E.J. Kramer, Langmuir 14, 6570 (1998).

10:40am NS-MoM8 Mechanical and Electrical Properties of Mo3Se3-Nanowires and Nanowire-networks, A. Heidelberg, G. Staikov, J.W. Schultze, Heinrich-Heine-Universität Düsseldorf, Germany , J.J. Boland, University of North Carolina at Chapel Hill Nanowires and nanotubes have attracted enormous interest as potential buildin g blocks for nanotechnology.1 This interest can be traced to the novel structural, mechanical and electronic properties of these nanomaterials. Here we describe a study that measures these properties in the case of the (Mo3Se3)--nanowire system.2 The mechanical properties of single nanowires or bundles were studied using an SPM-nanomanipulator. This instrument allows us to controllably apply forces (µN-nN range) to supported Mo3Se3-nanowires to effect nanoscale manipulations. Using the lateral force data of the manipulations, the mechanical properties like Youngs modulus and tensile strength of the nanowires can be calculated. The electrical properties of bare LiMo3Se3-nanowires have previously been shown to have metallic behaviour.3 Exchanging the Li-counterion to alkylammonium, alkylpyridinium or alkylpiperazinium counterions produces network structures of the nanowires with a defined interwire spacing. Conductivity measurements at different temperature and oxidation times show that these networks act as percolation networks and have semiconducting behaviour. In addition the corrosion rate of the Mo3Se3-nanowires is slowed down in the wire networks, demonstrating that the organic coating forms partially insulating layers. 1 J. Hu, T. W. Odom, C. M. Lieber; Acc. Chem. Res. 32 (1999) 435 2 J. M. Tarascon, F. J. DiSalvo, Solid State Commun. 52 (1984) 227 3 J. H. Golden, F. J. DiSalvo, J. M. J. Fréchet, Chem. Mater. 7 (1995) 232.

11:00am NS-MoM9 Towards the Sensing of Atomic Interactions by Nanoindentation with Extremely Sharp Tips, J. Fraxedas, ICMAB-CSIC, Spain , S. Garcia-Manyes, CBEN and University of Barcelona, Spain , P. Gorostiza, University of California, Berkeley , F. Sanz, CBEN and University of Barcelona, Spain A force F applied to a surface acts directly on the surface atoms and is transmitted to the bulk atoms via the crystal lattice. The bonds play thus a crucial role in the mechanical response because of their strength and spatial distribution. For a point force only few atoms are involved. In this case the elastic deformation of the surface critically depends on in-plane interactions. In order to demonstrate the relevance of such in teractions we have done nanoindentation experiments on 2D materials and ionic single crystals with an AFM. The stiffness k of the crystal and an estimated radius ds of the elastically perturbed surface can be evaluated from expression F(δ)=kδ(1-ds/√(δ2+ds2)),1 where δ stands for the surface deformation. k is related to the Debye frequency ωD (kD=mωD2), where m represents the mean atomic mass. The calculated values of kD are very close to the experimentally derived values of k (i. e., k=84±13 Nm-1 and kD=86 Nm-1 for NaCl). Nanoindentation thus reveals the collective behavior of nanoscale volumes since many atoms are involved in the process (ca. 140 ion pairs for the alkali halides). We observe that k/d s=c11, where c11 represents the (1,1) component of the elastic tensor. Feynman developed a simple model relating the anion-cation interatomic interaction kac to elastic constants for small strains for NaCl-type crystals assuming central forces. 2 Within this approximation we obtain kac

11

Nanotubes: Science and Applications Topical Conference Room: C-209 - Session NT-MoM

Nanotubes: Growth and Characterization Moderator: S.B. Sinnott, University of Florida

8:20am NT-MoM1 Synthesis and Applications of Vertically Aligned Carbon Nanofibers, V.I. Merkulov, A.V. Melechko, M.A. Guillorn, D.K. Hensley, D.H. Lowndes, M.L. Simpson, Oak Ridge National Laboratory INVITED Vertically aligned carbon nanofibers (VACNFs) prepared by direct-current (dc) plasma enhanced chemical vapor deposition (PECVD) are important for various applications including electron field emitters, tips for scanning microscopy, and biological probes, among others. To date, the crucial advantage of using VACNFs is the ability to grow them deterministically, i.e. their location, height, tip and base diameters, and, to s