Session P1: Buckley Prize, Bouchet Award, and Apker Award Talks

Quantum spin dynamics and coherence in solids

Even while their existence as individuals is fundamentally quantum mechanical, derivable from atomic physics, for many purposes the spins in solids behave as classical objects. In particular, virtually the entire science of solid state magnetism, including phenomena as diverse as phase transitions, domain walls, and magnons can be understood in terms of classical rotors, subject to varying anisotropic potentials determined by external as well as internal (crystal) fields, coupled to each other via dipolar and exchange interactions. Even so, there has long been theoretical suspicion that when the interactions are antiferromagnetic rather than ferromagnetic, quantum mechanics might manifest itself on scales larger than the interatomic spacings associated with crystal fields and exchange interactions. Recent experiments on antiferromagnets, where the quantum fluctuations are enhanced by coupling to mobile electrons or by geometric factors such as system dimensionality, reveal this suspicion to be justified. Some of the materials, including even the simple two- dimensional Heisenberg antiferromagnet, display anomalous spin wave amplitudes and dispersion, while for others, classical order itself is replaced by mesoscopic quantum ‘order’ or coherence. Apart from their intrinsic interest as magnets with strong quantum fluctuations, systems of this type are also of more general importance as they include both the heavy fermion metals and high temperature superconductors, many of whose interesting properties are thought to be derived from the same fluctuations. Work is currently supported by a Wolfson Royal Society Fellowship and the Basic Technologies Progamme of the UK Research Councils.   

Spintronics: Semiconductors, Molecules, and Quantum Information Processing

There is a growing interest in exploiting electronic and nuclear spins in semiconductor nanostructures for the manipulation and storage of information in emergent technologies based upon spintronics and quantum logic. Such schemes offer qualitatively new scientific and technological opportunities by combining elements of standard electronics with spin-dependent interactions between electrons, nuclei, electric and magnetic fields. Here we provide an overview of recent developments in the field through a discussion of temporally- and spatially-resolved magneto-optical measurements, initially designed for probing local moment dynamics in magnetically doped semiconductor nanostructures. We demonstrate new electrical schemes for the local generation and manipulation of spins in conventional semiconductor heterostructures, thereby providing a compelling proof-of-concept that quantum spin information can be controlled within high-speed electrical circuits. Furthermore, we discuss a different experimental approach that enables the molecular wiring and assembly of colloidal semiconductor nanostructures to engineer hybrid systems for room temperature coherent spin transport. These experiments explore electronic, photonic, and magnetic control of spin in a variety of nanostructures, and show significant steps towards spin-based quantum information processing in the solid state.   

Quantum Tunneling of the Magnetization in Molecular Nanomagnets

Molecular nanomagnets, sometimes referred to as single molecule magnets, have attracted a great deal of recent attention for interesting behavior that is borderline between the classical and quantum mechanical regimes, and because of their potential usefulness for high-density data storage and quantum computation. Quantum mechanical processes are observed in these materials on a macroscopic scale in the form of steps in the magnetization curves.* Two particularly simple prototypes, Mn$_{12}$-acetate and Fe$_{8}$, have been studied in great detail. Typical behavior of the class will be examined by considering Mn$_{12}$-acetate: the structure of the molecule, the tetragonal (four-fold symmetric) crystal, the Hamiltonian that models the behavior, and the tunneling process that gives rise to the magnetization steps. * J. R. Friedman, M. P. Sarachik, J. Tejada, and R. Ziolo, Phys. Rev. Letters, \textbf{76}, 3830 (1996).   

Single-Electron Transport and Device Applications

In recent years, there has been a considerable amount of theoretical and experimental activity involving surface acoustic waves (SAWs). Interest in this field has centered around the evidence for single -electron transport in piezoelectric materials such as GaAs/AlGaAs heterostructures and its potential use as a current standard. Several device applications involving SAWs for transporting electrons (and holes) include quantum computing, where the spin is the qubit that is moved across a network of quantum gates by SAWs, and a single-photon source through the recombination of electrons and holes. The SAW quantum computer is a “dynamic” qubit type where the quantum information actually travels during the computation. This type of qubit has the advantage of delivering information quickly through the circuit when decoherence times are short. We consider specific architectures that will allow quantum entaglement and electron-hole recombination. We also propose a novel scheme of photon detection which uses SAWs for transporting photo-generated electrons and holes. Potential applications of the concept include imaging arrays in the visible and infrared regions and single photon detection. A preliminary feasibility study has indicated that the acoustoelectric photon detectors/imaging arrays can feature an extremely low dark current rate. The concept has advantages over Charge Coupled Devices (CCDs) because of its simplicity, speed of operation and high sensitivity (down to the single photon level). The device uses SAWs to transport charge along wave guides. We explore the application of the detector to the improvement of security.   

Asymmetry in RNA Pseudoknots

Single-stranded RNA can fold into a topological structure called a pseudoknot, composed of non-nested double-stranded stems and single-stranded loops. Our examination of the {\tt PseudoBase} database of pseudoknotted RNA structures reveals asymmetries in the stem and loop lengths and provocative composition differences between the loops. By taking into account the difference between the major and minor grooves of the RNA double helix, we explain much of the asymmetry with a simple polymer physics model.   

Session P2: New Techniques for Biological Structure Determination

Organic-inorganic templates in biomineralization of shells, bone, teeth, and bacterial biofilms

Recent experiments with the Spectromicroscope for PHotoelectron Imaging of Nanostructure with X-rays (SPHINX)[1] on the biofilm formed by Fe-oxidizing bacteria in fresh, ground water, demonstrated that microbially extruded polysaccharide filaments provide the precipitation site for amorphous FeOOH filaments [2]. Upon aging the mineralized filaments crystallize to ferrihydrite (2-line FeOOH), with one curved pseudo-single crystal of akaganeite $\beta$-FeOOH), at the core of each filament. The crystals are only 2 nm wide and up to 10 micron long (aspect ratio 1:1000:1), and their structure and morphology is unprecedented. Furthermore, akaganeite should not form in fresh water, therefore a templation mechanism was hypothesized, and supported by SPHINX analysis of carbon XANES. The results indicate that after formation of the crystal fiber, the polysaccharide structure is also altered, and C1s spectra suggest that the COO$^{-}$ group is involved in the templation mechanism. This was the first successful attempt to understand the organic-inorganic chemical interface in a biomineralized system. Many more templated biomineral systems can and will now be analyzed with this new approach. \begin{enumerate} \item Ultramicroscopy \textbf{99}, 87-94 (2004). \item Science \textbf{303}, 1656-1658 (2004). \end{enumerate}   

X-Ray Tomography Generates 3-D Reconstructions of the

Soft X-ray microscopy is an emerging new technique that can image whole, hydrated, biological specimens with a spatial resolution 5-10 times better than that obtained with light microscopy. X-ray imaging at photon energies below the K- absorption edge of oxygen exploits the strong natural contrast for organic material embedded in a mostly water matrix. With a transmission X-ray microscope using Fresnel zone plate optics, specimens up to 10 microns thick can be examined. The highest X-ray transmission in hydrated samples is obtained at a wavelength of 2.34 nm but, due to the low numerical aperture of zone plate lenses operated in first order diffraction mode (NA$\sim $0.1), the structures resolved are much larger than the X-ray wavelength. To date, soft X-ray microscopy has been used to resolve 30 nm structures in frozen hydrated specimens. Because of the low NA of X-ray lenses, combined with the effect of polychromatic illumination and a wavelength dependant focal length, the effective depth of field is large (6-10 microns). In this paper, we show tomographic reconstructions of rapidly frozen live cells in a 10 um diameter glass capillary (200 nm wall thickness). The image sequences span 180 degrees and consist of either 45 or 90 images spaced by 4 or 2 degrees, respectively. Computed tomographic reconstructions generate 3-D images of whole cells at better than 50 nm isotropic resolution. Using the x-ray linear absorption coefficient, quantitative information is obtained from the reconstructed data. Data sets containing 180 images, made possible by our new fully automated cryo-rotation stage, will generate images at resolution approaching 30 nm. This stage also enables collection of data in less than 3 minutes, making soft x-ray tomography the first high-throughput, high-resolution imaging technique for biological specimens.   

Biological x-ray microscopy: from biochemical mapping to lensless imaging

Cell structure has been very succesfully studied using light and electron microscopy. However, x rays ofer new insights, by imaging whole cells at 20-40 nm resolution using zone plate lenses, and in particular by combining this with spectroscopic sensitivity to organic functional groups. While spectra of single compounds can provide exquisite information on electronic states, a cell is much more complex. Pattern recognition algorithms provide a way to deal with this complexity and obtain insights into biochemical organization at a fine spatial scale, as illustrated in an ongoing study of the correlation of morphology with biochemical content in sperm. Another approach to biological imaging is to abandon the use of lenses and their resolution limits. The purest form of measurement is to collect x rays scattered by a cell with no optics-imposed losses. By using iterative phasing algorithms, this diffraction data can be phased to deliver a real-space image of a complex cell (at present, 30 nm resolution in studies of freeze-dried yeast) with a possible ultimate extension to atomic resolution imaging of proteins using x-ray free electron lasers.   

3D Diffraction Microscope Provides a First Deep View

When a coherent diffraction pattern is sampled at a spacing sufficiently finer than the Bragg peak frequency (i.e. the inverse of the sample size), the phase information is in principle encoded inside the diffraction pattern, and can be directly retrieved by using an iterative process. In combination of this oversampling phasing method with either coherent X-rays or electrons, a novel form of diffraction microscopy has recently been developed to image nanoscale materials and biological structures. In this talk, I will present the principle of the oversampling method, discuss the first experimental demonstration of this microscope, and illustrate some applications in nanoscience and biology.   

In Vivo Microtesla Magnetic Resonance Imaging

We have developed a magnetic resonance imaging (MRI) system which operates at magnetic fields of 132 microtesla, corresponding to proton Larmor frequencies of 5.6 kHz. The main advantages of performing MRI at low magnetic fields ($<$ 10 mT) are the reduced costs compared to conventional high- field MRI, and the reduction of nuclear magnetic resonance line broadening caused by inhomogeneous magnetic fields and susceptibility variations in the sample. Our technique combines prepolarization of the nuclear spins in a magnetic field up to 300 mT and signal detection at 132 microtesla using an untuned superconducting input circuit coupled to a superconducting quantum interference device (SQUID) to achieve a signal amplitude independent of the measurement field. We employ a standard spin-echo pulse sequence to acquire three-dimensional images in less than 6 minutes. Using encoding gradients of about 100 $\mu$T/m we obtain images of bell peppers and water phantoms with a resolution of 2 mm x 2 mm x 8 mm. Three- dimensional images of a human forearm were acquired at 132 microtesla with an average prepolarization field of 50 mT showing a signal-to-noise ratio (SNR) of 10 and an in-plane resolution of 3 mm x 3 mm. We have shown that for certain materials the longitudinal relaxation time (T$_1$) contrast is greatly enhanced at low magnetic fields. This enhancement is expected to lead to novel applications in specialized clinical imaging of human subjects, for example, low-cost tumor screening. To make such applications feasible further improvements of the SNR and resolution of the system are necessary. By employing a SQUID detector with a lower magnetic field noise and by raising the maximum polarizing field, an improvement of the SNR by an order of magnitude should be possible.   

Session P3: Physics and Sustainable Development

Doing Physics With Third World Collaborators

For more than a decade now we have been in collaborative projects with scientists from the Third World, specifically India, Turkey and Uzbekistan. Collaborations with Third World physicists bring their own unique problems and rewards. Third World scientists are no less talented than those in the West; they do, however, lack facilities, opportunities and funding. Collaboration of these Third World scientists and their Western counterparts can go a long way towards ameliorating these difficulties in a variety of ways. Some of the problems, unique to Third World collaborations, will be discussed. Furthermore, funding aimed specifically at supporting such collaborations is available from a variety of sources; some details of funding opportunities will also be presented. As a result of our collaborations, a variety of research projects involving theoretical studies of photoabsorption by free and confined atoms and ions are ongoing and have resulted in a significant quantity of publications; a representative sample of the research results over the years [1-6] will also be discussed. [1] E. W. B. Dias, H. S. Chakraborty, P. C. Deshmukh, S. T. Manson, O. Hemmers, P. Glans, D. L. Hansen, H. Wang, S. B. Whitfield, D. W. Lindle, R. Wehlitz, J. C. Levin, I. A. Sellin and R. C. C. Perera, Phys. Rev. Lett. \textbf{78}, 4553 (1997). [2] Z. Altun and S. T. Manson, Phys. Rev. A \textbf{59}, 3576 (1999). [3] V. K. Dolmatov and S. T. Manson, Phys. Rev. Lett. \textbf{83}, 939 (1999). [4] H. S. Chakraborty, A. Gray, J. T. Costello, P. C. Deshmukh, G. N. Haque, E. T. Kennedy, S. T. Manson and J.-P. Mosnier,~ Phys. Rev. Lett. \textbf{83}, 2151 (1999). [5] H. Wang, G. Snell, O. Hemmers. M. M. Sant'Anna, I. Sellin, N. Berrah, D. W. Lindle, P. C. Deshmukh, N. Haque and S. T. Manson, Phys. Rev. Lett. \textbf{87}, 123004 (2001). [6] A. Baltenkov, V. Dolmatov, S. Manson and A. Msezane, Chem. Phys. Letters \textbf{378}, 71 (2003).   

Energy and the Environment in the 21st Century

.   

Panel Discussion

As part of the celebration of the World Year of Physics, the World Conference on Physics and Sustainable Development will be held in Durban, South Africa, 31 October - 2 November 2005. Participants from developed and developing nations will join together to formulate action-oriented plans for the contributions that physics and physicists can make to society. This session will begin a discussion of the topics that will be further developed at the World Conference.   

Session P4: Structure in Solutions and Melts

Self-Assembled Liquid Crystalline Gels Designed from the Bottom Up

Block copolymers with long side-group liquid-crystalline (LC) midblocks and LC-phobic end-blocks form a physical network that swells readily in a small molecule LC to form model nematic gels with well-defined molecular weight between crosslinks. Ultralong ($>$800 kg/mol) SGLCP midblocks enable gelation at relatively low concentration ($\sim $5{\%} polymer), which preserves the fast dynamics of small molecule LCs. Similar to LC elastomers, an initially unaligned, polydomain gel aligns under applied strain. Further, the resulting monodomain is so well oriented that it generates clear conoscopic figures. Due to the coupling between nematic order and polymer elasticity, a novel stripe pattern forms when the gels are subjected to electric fields or when the order parameter of the LC solvent changes. Meyer and coworkers have described these patterns using a linear stability analysis that connects the band formation with the spontaneous anisotropy of the SGLCP backbone. Small-angle neutron scattering (SANS) on analogous SGLCP homopolymers confirms that greater chain anisotropy favors band formation and that the sense of anisotropy (prolate or oblate) dictates the initial band orientation. The physical junctions comprised of the LC-phobic endblocks perturb the director field on a nano-scale manifested in a reduction in the mean order parameter of the LC host characterized by both refractive indices (n$_{e}$, n$_{o})$ and NMR $^{2}$H quadrupole splitting. The physical principles demonstrated for nematic gels allow rational design of gels exhibiting higher-order LC phases, such as ferroelectric gels.   

Telechelic amphiphilic polymers: assembly in water and at the air/water interface

Associative polymers exhibit a rich rheology when dissolved or dispersed in water. The quintessential associative polymers are poly(ethylene oxides) (PEO) bearing hydrophobic groups at each end. Our research interests have focused on another water-soluble polymer, poly(N-isopropylacrylamide) (PNIPAM) and its hydrophobically-modified derivatives. Like PEO, PNIPAM is a non-ionic polymer soluble in water at room temperature. We succeeded recently in preparing telechelic PNIPAM samples of narrow molecular weight polydispersity that bear an octadecyl group at each chain end and carried out a systematic study of the assembly of this polymer as a function of concentration, from the dilute to the concentrate regimes and as a function of temperature (10 to 40 $^{o}$C) using static and dynamic light scattering measurements, microcalorimetry, and fluorescence spectroscopy for investigations of dilute solutions. In the concentrated regime, telechelic PNIPAM forms gels which were examined under oscillatory shear.   

Network Phases of ABC Triblock Copolymers

Fundamental exploration of the melt state phase behavior in linear ABC triblock copolymers has uncovered a fantastic array of over two dozen unique morphologies in just over a decade of limited scrutiny. These structures range from simple three-domain analogs of the classic diblock copolymer phases to exquisite ``decorated phases'' characterized by the presence of A/C interfaces not inherently required by the natural connectivity of the copolymer. In this presentation I will focus on our extensive research efforts targeting the discovery of multiply continuous network phases within the expansive ABC parameter space. Adopting a strategy involving block connectivities precluding A/C interface formation and compositions aimed at breaking symmetry between two and three domain lamellar regions, we synthesized a series of 43 poly(isoprene-$b$-styrene-$b$-ethylene oxide) (ISO) triblock copolymers (ranging from 15 to 25 kg/mol) to systematically explore network formation in ABC systems. Employing a battery of complementary analysis techniques including TEM, SAXS (static and under reciprocal shear), dynamic mechanical spectroscopy and static birefringence, coupled with mathematically generated level set models that bridge real and reciprocal space, we have identified a total of three independent network phases formed in this single triblock copolymer system. Two cubic network phases, Q$^{230}$ (core-shell double gyroid, \textit{Ia}$\bar {3}d)$ and Q$^{214}$ (alternating gyroid, $I$4$_{1}$32), and an unprecedented orthorhombic network phase, O$^{70}$ (\textit{Fddd}), were found to define a significant region of contiguous phase space, with order-order transitions (OOTs) found between network phases in some samples. Quite remarkably, the topology of each of these networks shares a common structure based on ordered arrays of connected 10-node loops, with each node trivalently joined to other nodes in the network. The universal presence of such networks in other ABC systems will be discussed.   

Conformations and Structure in Aqueous Poly(ethylene oxide) Solutions

Atomistic molecular dynamics (MD) simulations have resulted in important insights into the influence of water on the local conformations and chain dimensions of poly(ethylene oxide) (PEO) as well as the role of PEO-water polar interactions and PEO-water and water-water hydrogen bonding interactions on solution structure as a function of composition and temperature. Results of these simulations will be presented and discussed. In addition, results of recent simulation studies of PEO brushes and the interaction of PEO-modified nanoparticles in aqueous solution will be considered.   

Control of contents and release kinetics in block copolymer vesicles

Block copolymer vesicles have received considerable attention recently because of a wide range of potential applications. In our group, the thermodynamic aspects of vesicle formation, including curvature stabilization, as well as active loading and release from vesicles have been the focus of recent research. The vesicles are prepared from an amphiphilic diblock copolymer such as polystyrene-block-poly(acrylic acid) at a low pH (2.5) by adding water to a solution in a common solvent; then the extenal pH is raised to 6.5, and the compound, such as doxorubicin or another amine, is added. Since the compund inside the vesicle becomes ionized at the low pH, it can only escape at a rate very much slower than that of the loading process. The permeability of the wall can be controlled by the presence of plasticizers for the polystyrene wall; the plasticizers partition between the wall and the external aqueous solution with a known partition coefficient, and can be removed from the wall by dialysis. Release is then studied under perfect sink conditions and is diffusional. It is noteworthy that the rates of both loading and release can be varied by more than two orders of magnitude by controlling the plasticizer content. Also, between the loading and release processes, the vesicle wall can be hardened by removal of the plasticizer by dialysis. This degree of control makes block copolymer vesicles a promising delivery vehicle for a range of materials, including drugs.   

Session P5: Nitride-Based Microelectronics

Toward a Self-Consistent Model of Carbon States, Semi-Insulating Conductivity and Yellow Luminescence in GaN:C Grown by Molecular Beam Epitaxy

Carbon doping of GaN is of great interest to generate semi-insulating (SI) buffer layers used in AlGaN/GaN heterojunction field effect transistors grown by molecular beam epitaxy (MBE). However, the specific mechanism(s) responsible for SI behavior involving C-related bandgap states in GaN until recently had been un-verified experimentally due to difficulties in determining deep level properties within SI wide bandgap materials. Moreover, C-related states in GaN are under increasing scrutiny due to the observation of yellow luminescence (YL) in \textit{both} n-type and SI GaN:C since the prevalent model for YL in GaN requires requires gallium vacancy (V$_{Ga})$ defects, which are not expected to form in significant concentrations for SI GaN. However, the fact that YL is observed for SI GaN doped with C while not being observed for SI GaN doped with Fe, suggests additional roles for C-related states in the GaN bandgap that may also have implications for potential parasitic effects in GaN electronics. Hence a full understanding of C-related deep states in SI GaN:C is necessary. This presentation will focus on each of the aspects noted above. First, our recent development of a lighted capacitance-voltage defect profiling measurement that allows full quantification of deep level concentrations and energy levels within SI GaN throughout its bandgap will be described. By applying this method with deep level optical spectroscopy (DLOS) to a systematic MBE-grown GaN sample set with well-controlled carbon doping, carbon states responsible for the SI behavior are identified. From this knowledge and by comparing to photoluminescence studies of these same samples to monitor the YL dependence on both conductivity and carbon concentration, the controversy over the existence of YL in SI GaN:C is addressed. A model based on a coordinate configuration diagram will be presented, showing the first self-consistent picture of C-related bandgap states and how they influence both SI behavior and deep level YL.   

Dispersion in AlGaN/GaN HEMTs

Gallium nitride high electron mobility transistors (HEMTs) are living up to their potential as the high power high frequency transistor of the future. The record power density at 4 GHz now exceeds 30 W/mm [1], which is more than an order of magnitude greater than GaAs-based transistors. Achieving this record performance required controlling the phenomena known as ``dispersion'' in AlGaN/GaN HEMTs. Dispersion is the term used to indicate that the dynamic characteristics of a device are different from the static characteristics. The two sources of dispersion are trapped charge and self-heating [2]. Although electron trapping can occur in many parts of the device, the surface has been identified as one of the dominant sources of dispersion in AlGaN/GaN HEMTs. At frequencies below Ka-band, a combination of surface passivation (typically SiN) and a field plate structure can be used to control dispersion. As the frequency of operation is pushed above 30GHz, the penalty paid in capacitance introduced by field plates prevents field plate structures from being used. In addition, as the gate length is decreased in order to increase the operating frequency, the peak electric field in the channel increases causing dispersion to increase and reliability to decrease. Understanding and controlling dispersion is the key for obtaining good power performance and reliability in AlGaN/GaN HEMTs. This talk will explain dispersion from the virtual gate model of the surface. Two primary methods for measuring dispersion (pulsed I(V) and RF I(V)) will be shown. Finally, data showing how dispersion affects reliability of AlGaN/GaN HEMTs will be given. [1] Y.-F. Wu, et al., IEEE Electron Device Lett., vol. 25, no. 3, pp. 117-119, March 2004. [2] P. H. Ladbrooke et al., Electron. Lett., vol. 31, no. 21, pp. 1875-1876, Oct. 1995.   

III-V Nitride Materials and Electron Devices

The electronic properties of III-V nitride materials make them applicable to high power microwave transistors. They are grown by molecular beam epitaxy or metal organic vapor phase epitaxy, and are direct band gap materials, with alloy gaps ranging from $<$.7V to $>$6.2V. With no native substrates, SiC and sapphire are used. The dislocation properties are acceptor-like for larger band gaps, and donor-like for smaller band gaps. These Wurtzite Crystal materials, grown on the Ga face along the C-axis, have strong spontaneous and piezoelectric polarization. AlGaN/GaN heterojunctions have net polarization sheet charge, inducing an electron sheet charge. Growth without intentional impurity doping, yields $\sim $1X10$^{13}$/cm$^{2}$ electrons in a sheet of 25{\AA} FWHM thickness. Field-effect transistors are fabricated for microwave power amplification.. Their properties include CW normalized power levels $>$10W/mm at 10 GHz. Breakdown electric field strength in the GaN is 3 megavolts/cm, and the sheet current density is 1A/mm. The semiconductor surface is passivated, with silicon nitride, for charge stability. The experimental electron transit velocity of 1X10$^{7}$ cm/s is one third of that originally predicted by Monte Carlo simulations, since the build-up of longitudinal optical phonons was ignored in the simulations. Processing and properties of these transistors will be covered.   

Nitride semiconductor material growth by rf-MBE for electronic device applications

Gallium nitride and related materials are now beginning to realize their potential for electronic device applications, including high electron mobility transistors (HEMTs). Rf-plasma-assisted MBE is an attractive method of growing these materials due to its low background impurity incorporation, and recently there have been impressive results on MBE-grown electronic devices. However, significant growth issues remain, including the elimination or reduction of buffer conduction, threading dislocation densities, and trapping in or near the two-dimensional electron gas (2DEG) at the AlGaN/GaN interface. Recent work at the U.S. Naval Research Laboratory has addressed these issues, including the use of Be-doped GaN to reduce buffer conduction, the effect of the AlN nucleation layer on buffer conductivity and dislocation density, homoepitaxial growth of GaN on free-standing GaN substrates to reduce the threading dislocation density, and the investigation of trap states in AlGaN/GaN HEMT structures. For example, we have observed that buffer conduction can be reduced by several orders of magnitude by using Be-doped GaN layers and that the detailed growth conditions of the AlN nucleation layer on SiC substrates can significantly affect buffer conductivity and Hall mobilities in the 2DEG. Further, we have achieved room-temperature Hall mobilities of 1920 cm$^{2}$/V-s in AlGaN/GaN HEMT structures grown on free-standing GaN substrates.   

Session P6: Spectroscopy of Novel Magnetic Materials

Ultrafast Magneto-Optics in Ferromagnetic III-V Semiconductors

The carrier-density-dependent magnetic properties of Mn-doped III-V semiconductors allow us to explore novel ultrafast optical phenomena. Photo-generated, spin-coherent, and transient carriers can break the equilibrium among charges, spins, phonons, and ferromagnetic order, triggering an array of dynamical phenomena and providing ways to control magnetism optically. Here, we report results of ultrafast magneto- optical studies on ferromagnetic InMnAs and GaMnAs using non- degenerate time-resolved magneto-optical Kerr effect and transient reflectivity spectroscopies. We observe very short carrier lifetimes ($\sim$2 ps) and multi-level decay dynamics, due to low temperature MBE growth and heavily $p$-type magnetic doping. In an InMnAs/GaSb heterostructure, we observe a transient coercivity decrease (“softening”) during the free carrier lifetime. Furthermore, both in InMnAs and GaMnAs, we observe {\it ultrafast demagnetization}, similar to but much more drastic than what has been observed in itinerant ferromegnets. Above a pump fluence of $\sim$10 mJ/cm$^2$, we observe a complete quenching of ferromagnetic order, implying an ultrafast phase transition into a paramagnetic state.\\This work was carried out in collaboration with J. Wang, C. Sun, G.A. Khodaparast, A. Oiwa, H. Munekata, G.D. Sanders, C.J. Stanton, L. Cywinski, and L.J. Sham. We acknowledge support from DARPA (Grant No. MDA972-00-1-0034), ONR (Award No. N000140410657), and NSF (Grant Nos. DMR-0134058, DMR-0325474, and INT-0221704).   

Electrical spin injection from ferromagnetic metals and semiconductors

Electrical injection, transport, manipulation and detection of spin polarized carriers in a semiconductor are essential requirements for utilizing the spin degree of freedom in a future semiconductor spintronics technology. Electrical injection has been a particularly vexing issue. We describe results using both ferromagnetic semiconductor (FMS) and FM metal contacts, and review the on-going effort to understand the fundamental issues. Spin-LED structures serve as test platforms, and the quantum selection rules provide a quantitative measure of the electron spin polarization, Pspin, produced in an AlGaAs/GaAs QW. The large values of Pspin=85{\%} obtained using semimagnetic ZnMnSe contacts enable detailed analysis of spin scattering by interface defects. True FM materials are preferable as contacts, and FMSs are promising candidates -- their exchange split band edges offer both spin injection and spin-selective transport. An $n$-type FMS is especially attractive. We describe measurements of electrical spin injection from such an n-type FMS, CdCr$_{2}$Se$_{4}$, into AlGaAs/GaAs LEDs, and discuss interface structure and band offsets. FM metals offer many desirable attributes as spin injecting contacts -- high Curie temperatures, low coercive fields and fast switching times -- and the fundamental issue of interface conductivity mismatch can be circumvented by utilizing a tunnel barrier. We have successfully injected polarized electrons from a reverse-biased Fe Schottky contact into AlGaAs/GaAs, with Pspin $>$ 32{\%}. We demonstrate via the Rowell criteria that tunneling is indeed the dominant transport process, and confirm that majority spin electrons are responsible. We compare these data with spin injection using Fe/Al$_{2}$O$_{3}$ contacts into identical structures.   

Photoemission and magnetic circular dichroism studies of magnetic semiconductors

Recently, a series of novel ferromagnetic semiconductors have been synthesized using MBE and related techniques and have attracted much attention because of unknown mechanisms of carrier-induced ferromagnetism and potential applications as "spin electronics" devices. Some new materials show ferromagnetism even well above room temperature. Photoemission spectroscopy has been used to study the $d$ orbitals of the dilute transition-metal atoms, mostly Mn, and their hybridization with the host band states [1]. Soft x-ray absorption spectroscopy (XAS) and magnetic circular dichroism (MCD) at the transition-metal 2$p$-3$d$ absorption edges are useful techniques to study the valence and spin states of the transition-metal atoms. Furthermore, since MCD has different sensitivities to the ferromagnetic and paramagnetic components at different temperatures and magnetic fileds, if the sample is a mixture of ferromagnetic and non-ferromagnetic transition- metal atoms, it can be used to separate the two components and to study their electronic structures. In this talk, results are presented for the prototypical diluted ferromagnetic semiconductor Ga$_{1-x}$Mn$_x$As [2] and the room-temperature ferromagnets Zn$_{1-x}$Co$_x$O and Ti$_{1-x}$Co$_x$O$_2$.\\ I acknowledge collaboration with Y. Ishida, J.-I. Hwang, M. Kobayashi, Y. Takeda, Y. Saitoh, J. Okamoto, T. Okane, Y. Muramatsu, K. Mamiya, T. Koide, A. Tanaka, M. Tanaka, Hayashi, S. Ohya, T. Kondo, H. Munekata, H. Saeki, H. Tabata, T. Kawai, Y. Matsumoto, H. Koinuma, T. Fukumura and M. Kawasaki. This work was supported by a Grant-in-Aid for Scientific Research in Priority Area "Semiconductor nano-spintronics" (14076209) from MEXT, Japan.\\ 1. J. Okabayashi et al., Phys. Rev. B 64, 125304 (2001).\\ 2. A. Fujimori et al., J. Electron Spectrosc. Relat. Phenom., in press.\\   

Optical Properties of Magnetic Semiconductors

We have employed Infrared Sprectroscopy (IR) and Ellipsometry to explore the band structure of thin films and digitally doped superlattices of Ga$_{1-x}$Mn$_{x}$As, prepared in the group of D.D. Awschalom (UCSB). These measurements reveal the important role played by the Mn induced impurity band in the band structure and ferromagnetism of Ga$_{1-x}$Mn$_{x}$As. Our IR work on Digital Ferromagnetic Heterostructures reveals a unique ability to tune their optical properties as well as their intrinsic electronic structure without changing the doping/defect level. This work is in collaboration with E.J. Singley, D.N. Basov (University of California, San Diego) J. Stephens, R.K. Kawakami, and D.D. Awschalom(University of California, Santa Barbara).   

Ferromagnetic resonance studies of dilute magnetic semiconductors

We describe ferromagnetic resonance (FMR) measurements on ferromagnetic II$_{1-x}$Mn$_{x}$VI semiconductor alloys in thin film form. These include Ga$_{1-x}$Mn$_{x}$As layers grown by low-temperature molecular beam epitaxy on various buffers used to obtain different strain conditions. The analysis of the FMR provides values of cubic and uniaxial magnetic anisotropy fields -- i.e., those associated with the natural (undistorted) zinc-blende structure, and those arising from strain. Similar studies were also carried out on In$_{1-x}$Mn$_{x}$As, providing analogous information. Finally, we applied the FMR technique to Ga$_{1-x}$Mn$_{x}$As/Ga$_{1-y}$Al$_{y}$As heterostructures modulation-doped by Be. Here it was found that the increase in doping -- in addition to raising the Curie temperature of the Ga$_{1-x}$Mn$_{x}$As layers -- also leads to a significant increase of their uniaxial anisotropy field. The FMR data for modulation-doped heterostructures further show that the effective $g$-factor of Ga$_{1-x}$Mn$_{x}$As is strongly affected by the doping, thus providing a direct estimate of the free hole contribution to the magnetization of Ga$_{1-x}$Mn$_{x}$As.   

Session P7: Oh the Places You'll Go, Career Paths in Physics

A Physicist as President of the University

My wife, physicist Frances Hellman, is fond of referring to me as a ``restless soul,'' and I do not dispute her. In the 40 years since graduating from the University of Western Ontario with a bachelor's degree in mathematics and physics, I went on to earn master's and doctorate degrees in physics and an honorary doctor of science degree from McMaster University. In 22 years working at AT{\&}T Bell Laboratories, I held five positions, was department head in two departments, and director of one laboratory. At the University of California, San Diego, I was a Professor of Physics, chair of the Department of Physics, senior vice chancellor and then chancellor. Currently, in addition to being a professor of Physics, I am president of the University of California. The ``restless'' trajectory of my career from physics undergraduate to university president follows the nature of physics itself. In physics, you are constantly seeking challenges, experimenting, creating hypotheses, looking for and finding solutions. I recall having a structured view of the world as a boy, a sense that there was a guiding ``master plan'' to most things and that wise, educated, benevolent people were there to implement the plan. ``They'' would do the right thing. Along the way, I realized, ``there is no `they' there; there is only us.'' Acknowledging the laws of thermodynamics-- ``you can't win, you can't break even, and you can't get out of the game'' --I nonetheless believe that if you have a restless mind, an open heart, and intellectual honesty without giving into wishful thinking, physicists can do anything. .   

Becoming a Physicist at a National Laboratory

In this talk I will describe my experiences during 17 years of employment at Lawrence Livermore National Laboratory, a nuclear weapons laboratory. The stories will include how I came to work there, some of the projects on which I worked, and how my career evolved during the end of the Cold War. Recently, I have moved from Lawrence Livermore to Lawrence Berkeley Lab, another (but non-nuclear) Department of Energy Laboratory. I will reflect on the scientific challenges currently facing both laboratories and talk about the advantages of working in each environment and the advantages and disadvantages of the DOE laboratories vs. an academic environment.   

Between Industry and Academia: A Physicist's Experiences at The Aerospace Corporation

The Aerospace Corporation is a nonprofit company whose purposes are exclusively scientific: to provide research, development, and advisory services for space programs that serve the national interest, primarily the Air Force's Space and Missile Systems Center and the National Reconnaissance Office. The corporation's laboratory has a staff of about 150 scientists who conduct research in fields ranging from Space Sciences to Material Sciences and from Analytical Chemistry to Atomic Physics. As a consequence, Aerospace stands midway between an industrial research laboratory, focused on product development, and academic/national laboratories focused on basic science. Drawing from Dr. Camparo's personal experiences, the presentation will discuss advantages and disadvantages of a career at Aerospace, including the role of publishing in peer-reviewed journals and the impact of work on family life. Additionally, the presentation will consider the balance between basic physics, applied physics, and engineering in the work at Aerospace. Since joining Aerospace in 1981, Dr. Camparo has worked as an atomic physicist specializing in the area of atomic clocks, and has had the opportunity to experiment and publish on a broad range of research topics including: the stochastic-field/atom interaction, radiation effects on semiconductor materials, and stellar scintillation.   

Session P10: Focus Session: Magnetic Semiconductors: Oxides

Oxygen vacancies and ferromagnetism in Co-doped anatase

Cobalt-doped titanium dioxide, or CTO, has emerged in the past few years as a semiconducting, transparent, room-temperature ferromagnet. Very recently it has been shown that the magnetism in anatase-structure CTO often originates in surface nanoparticles or Co-rich regions that have a much-enhanced substitutional Co content up to 40{\%} of Ti sites, so that magnetic CTO is not a true dilute magnetic semiconductor (DMS), but rather a fairly high-density spin system. In this work we describe a computational study of Co-rich CTO using the Generalized Gradient Approximation (GGA) to density functional theory (DFT) within the supercell model. Our total energy calculations show a strong tendency for Co-atom clustering or segregation on Ti sites. There is also a strong tendency for the oxygen vacancies to form complexes with the Co atoms. In addition, we find that the oxygen stoichiometry plays an essential role in determining the system's magnetic order. The largest ordered moments require at least enough oxygen vacancies to put all of the Co atoms in the +2 charge state, as they indeed appear to be experimentally, so that the conventional DMS mechanism could only apply via n-type carriers. We find a small but not negligible spin density associated with Ti atoms near the vacancy sites, suggesting an F-center-mediated interaction between the much larger Co moments.   

Magnetic circular dichroism (MCD) measurements of cobalt-doped titanium dioxide films

Cobalt-doped TiO$_2$ has generated interest as a dilute magnetic oxide displaying room-temperature ferromagnetism with $T_c \ge 650\,$K for low-doped materials. However, controversy surrounding the mechanism for such a high $T_c$ and the observation of Co clusters cast doubts on this system as an intrinsic ferromagnetic oxide. A recent study\footnote{S. R. Shinde \textit{et al.}, Phys. Rev. B \textbf{67}, 115211 (2003).} reporting the importance of growth conditions on Co solubility confirms the existence of ferromagnetism in films showing no direct evidence of clustering. MCD offers promise as a technique to characterize the intrinsic nature of magnetism and probe the band structure. A sensitive heterodyne technique using a photoelastic modulator measures MCD in the visible frequency range at magnetic fields up to 1.5\,T. We report a comparison of MCD measurements on thin films of well-oxygenated anatase Ti$_{1-x}$Co$_x$O$_{2-\delta}$ ($x\le 0.07$) exhibiting no clustering with those on clustered films. Additionally, we compare MCD results to optical absorption measurements,\footnote{J. R. Simpson \textit{et al.}, Phys. Rev. B \textbf{69}, 193205 (2004).} which reveal a shift of the band edge upon cobalt doping and an absence of mid gap cobalt impurity states.   

Origin of ferromagnetism in transition metal doped TiO$_2$

Reports of robust room-temperature ferromagnetism in oxide-based dilute magnetic semiconductors, such as Co-doped TiO$_2$ and ZnO, make these materials promising candidates for device applications. However, the origin of this robust ferromagnetism and the factors that influence its strength have not yet been fully understood. One model, proposed by Venkatesan and Coey, suggests that the observed ferromagnetism in doped ZnO and SnO$_2$ is mediatied via an electron trapped in a bridging oxygen vacancy [1,2]. On the other hand there is also evidence, e.g. in the case of Co-doped TiO$_2$, that free carriers play a role in mediating the magnetic interaction (e.g.[3]). To test the validity of the different models for the TiO$_2$-based dilute magnetic semiconductors, and to modify them where necessary, we study transition metal doped TiO$_2$ in the anatase structure by means of ab-initio density-functional band-structure calculations. We analyse the interaction of the transition metal dopants with the electronic states of the host as a function of the distribution and concentration of the impurity atoms. Finally we determine the influence of point and extended defects on the magnetic interactions. \newline [1] D.M.C. Coey et al., J. Appl. Phys. Lett. {\bf 84}, 1332 (2004). [2] M. Venkatesan et al., cond-mat/0406719. [3] H. Toyosaki et al., Nature {\bf 3}, 221 (2004).   

Structure and magnetic properties of Co doped anatase TiO2 particles

Co$_{0.028}$Ti$_{0.972}$O$_{2-\delta }$was synthesized via sol-gel method. It's annealed at 600$^{\circ}$C for one hour in air. The X-Rays diffraction study confirmed that samples are of anatase structure and no detectable cobalt clusters or other impurities are observed. The magnetic properties are characterized by SQUID in a broad temperature range 5K $\sim $ 300K, and no ferromagnetic property was observed. The magnetization verse temperature curve fits well with the Curie-Weiss law and the extracted atomic effective moment suggests that the Co(III) is in high spin state. These paramagnetic particles can be turned into ferromagnetic phase with a Curie temperature above 300K after heat treatment in a mixture of H$_{2}$ and Ar gases at 600$^{\circ}$C for one hour. The origin of ferromagnetism has been studied by XPS, SQUID, XRD and TEM.   

Characterization of transition metal doped CVD-grown ZnO films and nanostructures

Diluted magnetic semiconductors (DMS) are intriguing materials that offer the possibility of studying magnetic phenomena in crystals with a simple band structure and excellent magneto-optical and transport properties. Theoretical and experimental studies indicate that ZnO is a promising DMS candidate for room temperature spintronics applications. We have characterized the chemical, compositional, and magnetic properties of TM-doped ZnO films grown by MOCVD and sputter deposition on a variety of substrates. Doping with Mn, and Fe by either diffusion, co-sputtering, or ion implantation has been investigated, and each doping method results in very different dopant depth profiles as revealed by Rutherford backscattering spectrometry. Soft x-ray absorption spectroscopy (SXAS) indicates that the TM dopant may be in either the 2+ or 3+ oxidation state and depends upon doping method and/or sample processing. Furthermore, the XAS results are consistent with the TM ions being substitutional for Zn. Squid magnetometry shows that some doping methods yield films exhibiting ferromagnetic behavior, with some Fe-doped films having the Curie temperatures above room temperature. Finally, we discuss the properties of MOCVD-grown ZnO nanotips that have been doped by TM ion implantation.   

Magnetism in transition doped ZnO

We present results of our detailed density functional investigation of ZnO doped with a series of \textit{3d} transition metal ions (Cr, Mn, Fe, Co, Ni and Cu). We have calculated the strength of the magnetic interactions when a single atom type is used as a dopant as well as the effects of simultaneous doping with two different transition metal ions. In addition, we have also done simulated \textit{p}-type doping in ZnO by substituting one of the Zn atoms by the monovalent ions , Li$^{+}$ and Cu$^{+}$ and determined the influence on the magnetism. We have also introduced defects in the form of O and Zn vacancies and its effect on magnetism. We find that our results are highly sensitive to the details of the calculations, including energy and k-point convergence, structural optimization and choice of exchange-correlation functional. However, our highly converged results suggest that above room-temperature ferromagnetism is not possible in transition metal doped ZnO without additional carriers and the experimental reports of high temperature ferromagnetism in this system could be due to the presence of secondary phases.   

Ferromagnetism in Ti-Doped ZnO Nanoclusters

Ferromagnetic behavior at room temperature is observed when a small percentage of non-magnetic titanium is combined with ZnO to form nanoclusters in the presence of oxygen atmospheres. The 5{\%} Ti- doped ZnO nanocluster film is prepared at room temperature using a technique that is a combination of high pressure sputtering and aggregation. A SQUID magnetometer measures the magnetic properties of this cluster film at various temperatures. The coercivity of the samples decreases exponentially with the increase of temperature. The maximum value of coercivity is 204.76 Oe obtained at 5K. The remanent magnetization increases at low temperatures up to 30K and decreases after wards. A distorted hysteresis curve is observed at 45K, 50K, 55K and 300K, where as at 5K, 30K and 90K the hysteresis curve showed normal ferromagnetic behavior. The field cooling (FC) and zero-field cooling (ZFC) measurements reveal a phase transition mechanism related to the spin ordering/disordering, which depends on the temperature.   

Carrier Mediated Ferromagnetism above 300 K in ZnO:Mn

We will present evidence that Zn$_{1-x}$Mn$_{x}$O thin films, grown by reactive magnetron sputtering, are ferromagnetic at temperatures significantly above 300 K. The onset of the ferromagnetic behavior is sensitive to the exact growth conditions - in addition to the Mn concentration, the magnetic properties strongly depended on the substrate type, film growth temperature and Oxygen partial pressure. Anomalous~Hall Effect shows that the charge carriers are spin-polarized electrons, participating in the observed ferromagnetic behavior. Specifically, Zn$_{1-x}$Mn$_{x}$O on Al$_{2}$O$_{3}$(0001) substrates are single-phase, as characterized by XRD and TEM and the magnetic moment for a Mn concentration of x=0.03 is 4.8$\mu _{B}$/Mn at 350 K, one of highest moments yet reported for any Mn doped magnetic semiconductor. Growth of Zn$_{1-x}$Mn$_{x}$O films on Si/SiO$_{2}$ substrates leads to the formation of secondary phases and no ferromagnetism is observed in these cases.   

Metastable defect ferromagnetic phases by low temperature interface reactions between transition metal oxides

Recently we showed [Nature Materials 3, 709 (2004)] that the room temperature ferromagnetism observed in low temperature (500 $^{\circ}$C) processed mixtures of 2 at{\%} MnO$_{2}$ with ZnO [Nature Materials 2, 673 (2003)] is caused by an interface phase suggested to be of the form Mn$_{2-x}$Zn$_{x}$O$_{3-\delta , }$wherein Zn is incorporated into Mn$_{2}$O$_{3}$. In order to establish the anticipated generic nature of the process, experiments were performed on low temperature sintering of 2 at{\%} MnO$_{2}$ with other transition metal oxides such as NiO, TiO$_{2}$ or CuO. Room temperature ferromagnetism was observed in all the three cases. Our results suggest that ferromagnetism in these new cases also resides at the interface. The same mixtures when sintered at 800 $^{\circ}$C resulted in compound phases (e.g. NiMn$_{2}$O$_{4}$ in the Ni-Mn-O case) that are nonmagnetic at room temperature. We suggest a new mechanism of ferromagnetism based on the valence and spin controlled defect state in the interface phases.   

Paramagnetic properties of Mn$^{2+}$ in Mn-added ZnO

Spin injected semiconductors, in which spins are introduced into the lattice, have been intensively studied due to their wide potential applications. Ferromagnetic ordering above room temperature in some of Mn-added ZnO was theoretically predicted by Dietl et al [1] and experimentally observed in Mn-doped ZnO [2]. In this work, we report the electron magnetic resonance (EMR) studies as well as physical properties of Zn$_{1-x}$Mn$_{x}$O ( x = 0.005 - 0.20 ) powders and a thermally Mn diffused ZnO crystal at 1000 $^{\circ}$C. The crystal structure of all samples showed a hexagonal wurtzite. However, even for the lowest Mn content (x=0.005) the samples turn out to contain a secondary phase [3], ZnMn$_{2}$O$_{4}$, from the XRD pattern. As the Mn content in the samples increases, so does the concentration of the secondary phase. In addition the electron magnetic resonance signal intensity of the paramagnetic Mn$^{2+}$, successfully incorporated into ZnO powder, decreases as the incorporated Mn content increases. This means that the Mn-rich secondary phase can more easily be formed than the Mn incorporated ZnO powder at 700 $^{\circ}$C. Paramagnetic Mn$^{2+}$ ions in a Mn-diffused ZnO crystal are turned out to sit on the Zn site from the EMR spectra. [1] T. Dietl et al, Science 287, 1019 (2000). [2] P. Sharma et al, Nature Materials 2, 673 (2003). [3] Y. M. Kim et al, Solid State Comm., 129, 175 (2004).   

Magnetic studies of ion-implanted p-GaN,\\Al$_{0.35}$Ga$_{0.65}$N, and ZnO with tran\-sition metals

Examination of the viability of ion-implantation for creating dilute magnetic semiconductors with ferromagnetic properties persisting to room temperature has been undertaken. Samples of $p$-GaN, Al$_{0.35}$Ga$_{0.65}$N and ZnO (film and nanotips) have been implanted with Fe, Mn and Cr at doses of 5$\times $10$^{16}$ cm$^{-2}$ and Ni at 3$\times $10$^{16}$ cm$^{-2}$. The samples were annealed at temperatures ranging from 600 to 800$^{\circ}$ C. The GaN and AlGaN samples were annealed in flowing N$_{2}$ for 5 min, and the ZnO samples in flowing O$_{2}$ for 10 min to determine the effect of annealing temperature. Using a superconducting quantum interference device (SQUID) magnetometer, we quantify ferromagnetism by the magnitude of coercive fields and show that an optimum annealing temperature is reached and passed within the range tested for a majority of the material/dopant combinations. Finally, we measure field-cooled and zero-field-cooled magnetization versus temperature.   

Preparation and Magnetic Properties of Transition-metal-doped SnO$_2$

There is strong interest in the ferromagnetism (FM) at room temperature in doped oxides, e.g., Co or Fe doped SnO$_{2 }$[1, 2]. We report the preparation and magnetic properties of transition-metal-doped SnO$_{2}$ (Sn$_{1-x}$TM$_{x}$O$_{2}$, TM=V and Mn, x=0--0.05). Bulk and thin-film samples were prepared by solid-state reaction and pulsed-laser deposition, respectively, and characterized by X-ray diffraction, electron microscopy and SQUID magnetometry. The magnetic properties strongly depend on the sample-processing temperature. Room-temperature FM has been observed in the bulk samples sintered at a low temperature of 500 $^{\circ}$C, but not in those sintered or annealed at higher temperatures (650 $^{\circ}$C and 900 $^{\circ}$C for TM=V and Mn, respectively). Additional Sb-doping has no strong effect on the FM, while vacuum annealing enhances the FM. Effects of sample-processing conditions and additional Sb-doping on the magnetic properties will be discussed. This research is supported by ONR, NSF-MRSEC, W. M. Keck Foundation and CMRA. [1]. S. B. Ogale, et al., Phys. Rev. Lett. 91, 077205 (2003). [2]. J. M. D. Coey, et al., Appl. Phys. Lett. 84, 1332 (2004).   

Variable range hopping behavior in the magnetic semiconductor SnO$_2$:Co

To explore the origin of ferromagnetism in doped oxide systems we have grown SnO$_2$: Co thin films on R-plane Al$_2$O$_3$ substrates by laser ablation. We present detailed structural (x-ray diffraction and TEM), transport and magnetic measurements on films grown at different temperatures, growth rates and oxygen pressures. TEM studies show no evidence of cobalt clustering in any of the SnO$_2$:Co films investigated. Variations in growth conditions have been found to affect the physical properties of the material. For example, films with high crystallinity, grown at low growth rates, show transport consistent with Mott variable range hopping, with $\log(\rho)\propto T^{-1/4}$. This is in contrast to the $\log(\rho)\propto T^{-1/2}$ dependence found previously in TiO$_2$:Co containing metallic cobalt clusters$[1] $. We will discuss implications of these findings on the ferromagnetism observed in SnO$_2$:Co.\\ This work was supported by the DARPA SPINS program. The TEM work was supported by NSF DMR-0084173.\\ $[1]$ R.J. Kennedy, et al, Appl. Phys. Lett. 84 2832 (2004).   

Intrinsic Defect Ferromagnetism: The case of Hafnium Oxide

In view of the recent experimental reports of intrinsic ferromagnetism in Hafnium Oxide (HfO$_{2}$) thin film systems \footnote{M. Venkatesan, C. B. Fitzgerald, J. M. D. Coey Nature {\bf 430}, 630 (2004) Brief Communications}, we carried out first principles investigations to look for magnetic structure in HfO$_{2}$ possibly brought about by the presence of small concentrations of intrinsic point defects. {\it Ab initio} electronic structure calculations using Density Functional Theory (DFT) show that isolated \textbf{cation} vacancy sites in HfO$_{2}$ lead to the formation of high spin defect states which couple ferromagnetically to each other. Interestingly, these high spin states are observed in the low symmetry monoclinic and tetragonal phases while the highly symmetric cubic flourite phase exhibits a non-magnetic ground state. Detailed studies of the electronic structure of cation vacancies in the three crystalline phases of Hafnia show that symmetry leading to orbitally degenerate defect levels is not a pre-requsite for ferromagnetism and that the interplay between Kinetic, Coulomb and Exchange energy together with favourable coupling to the Crystalline environment can lead to high spin ferromagnetic ground states even in extreme low symmetry systems like monoclinic HfO$_{2}$. These findings open up a much wider class of systems to the possibility of intrinsic defect ferromagnetism.   

Electric Field Modulation of Ferromagnetism in Diluted Magnetic Insulating Co:TiO$_2$

In this work we report the first successful implementation of an external electric field modulation of ferromagnetism in an oxide-based diluted magnetic system. An anatase TiO$_{2 }$layer with 7{\%} Co doping and a ferroelectric PbZr$_{0.2}$Ti$_{0.8}$O$_{3 }$(PZT) layer were epitaxially grown on a conducting SrRuO$_{3 }$buffered LaAlO$_{3}$ substrate by pulsed laser deposition. The Co:TiO$_{2}$ channel grown in this case at a high temperature of 875 $^{0}$C is insulating in nature. The magnetic hysteresis loops of the Co:TiO$_{2}$ were measured by SQUID after positive or negative electric poling on PZT. A 15{\%} difference in the room temperature saturated magnetic moment as well as the coercive filed of Co:TiO$_{2}$ is observed according to the two polarization states of PZT, which can be modulated over several cycles. This first demonstration of electric field effect in an oxide based diluted ferromagnetic insulator system provides evidence of its intrinsic nature. Possible mechanisms for insulating ferromagnetism and its modulation by an electric field are discussed. This work was supported by DARPA SpinS program (through US-ONR) and the NSF-MRSEC (DMR 00-80008) at Maryland and by a grant from Brookhaven National Laboratory.   

Session P12: Superconducting Materials

Structure and Physical Properties of Hydrated Sodium Cobalt

Using an electrochemical de-intercalation technique, we have produced single crystal samples of hydrated sodium cobalt oxide (Na$_{0.3}$CoO$_{2}\cdot$1.3H$_{2}$O). Thermodynamic and transport measurements reveal the low temperature properties of these samples to be consistent with those of a Fermi liquid with a strong mass enhancement. We have used neutron scattering and x-ray scattering to study the structure and excitations of this compound. We find that there are multiple stable structures which differ in the stacking arrangement of the planes. We discuss the bulk physical properties of the various compositions measured.   

Revised superconducting phase diagram of hole doped Na$_{x}$CoO$_{2}\cdot y$H$_{2}$O

We have studied the superconducting phase diagram of Na$_{x}$CoO$_{2}\cdot y$H$_{2}$O\space as a function of electronic doping, characterizing our samples both in terms of Na content $x$ and the Co valence state.[1] Our findings are consistent with a recent report that intercalation of H$_{3}$O$^{+}$\space ions into Na$_{x}$CoO$_{2}$, together with water, act as an additional dopant indicating that Na sub-stochiometry alone does not control the electronic doping of these materials. We find a superconducting phase diagram where optimal T$_{c}$\space is achieved through a Co valence range of 3.24 - 3.35, while T$_{c}$\space decreases for materials with a higher Co valence. The critical role of dimensionality in achieving superconductivity is highlighted by similarly doped non-superconducting anhydrous samples, differing from the superconducting hydrate only in inter-layer spacing. The increase of the interlayer separation between CoO$_{2}$ sheets as Co valence is varied into the optimal T$_{c}$\space region is further evidence for this criticality.[1] C.J. Milne $et al.$, $Phys. Rev. Lett.$, in press (2004). Also cond-mat/0401273.   

Evidence of two-gap superconductivity in Na$_{0.35}$CoO$_2\cdot$1.3H$_2$O

The recent discovery of superconductivity in the layered cobalt oxyhydrate Na$_{0.35}$CoO$_2\cdot$1.3H$_2$O [1] has attracted considerable attention in the scientific community because of its structural similarity to high-T$_c$ cuprates. Although intensive studies have been performed to understand the nature of superconductivity in this compound, no consensus has been reached on many important issues and the symmetry of order parameter still remains open. The low temperature behavior of the magnetic penetration depth $\lambda(T)$ provides a useful probe of the low-lying excitations in superconductors and hence of the symmetry of the superconducting energy gap. In this contribution, we present a high-precision measurement of $\lambda(T)$ on singe crystalline Na$_{0.35}$CoO$_2\cdot$1.3H$_2$O down to 90 mK, using a tunnel-diode based, self-inductive technique at 21 MHz. It is found that $\lambda(T)$ can be fit by a quadratic power-law above $T\simeq 1$ K. However, $\lambda(T)$ changes to an exponential decay at the lowest temperature ($T < 0.8$ K), indicating that the material is fully gapped. Detailed analysis shows that $\lambda(T)$ can be nicely fitted with a two-band model, resembling the case of MgB$_2$. These findings are consistent with the recent report of specific heat results [2] and suggest s-wave superconductivity in Na$_{0.35}$CoO$_2\cdot$1.3H$_2$O. [1] K. Takada et al, Nature \bf{53} (2003). [2] R. Jin et al, cond-mat 0410517.   

High Pressure Structure of Na$_{0.75}$CoO$_2$

High pressure x-ray diffraction experiments were performed on Na$_{0.75}$CoO$_{2}$, using synchrotron x-rays and dimond anvil cell up to 25 GPa, at ambient and cryogenic temperatures down to 10 K. The hexagonal structure of this compound is found to be stable at both conditions, and no structural changes were found around the unconventional magnetic state reported at 22 K. The bulk modulus obtained at ambient temperature, by fitting the pressure volume (PV) data shows, the compound is less compressible than its hydrated analogues. A rapid decrease observed in the c/a ratio under pressure, at ambient and low temperatures, indicates the presence of strong lattice anisotropy in this system similar to the high T$_{c}$ cuprates.   

Pulsed Laser Deposition of Na$_x$CoO$_2$ Thin Films

Na$_{x}$CoO$_{2}$ has been discovered to have very large thermoelectric power, which shows that it may be used in potential integrated heating spreading solution. Recently it was also found to be superconducting at certain sodium concentration after intercalated with water. It has a layered structure similar to the cuprates and considered to be helpful to the understanding of the mechanism of the high temperature superconductor. We have successfully grown c-axis oriented thin films of Na$_{x}$CoO$_{2}$ on substrates of polycrystalline sapphire and (0001) sapphire by pulsed laser deposition. The in-plane transport and magnetic measurements has been performed in the Na$_{x}$CoO$_{2}$ films and show similar behaviors as in the single crystal samples. Their structure properties as well as physical properties will be discussed.   

Effects of Disorder on the Normal State of Pr$_{2-x}$Ce$_{x}$CuO$_{4-\delta}$

We present a study on epitaxial thin films of the electron-doped superconducting cuprate Pr$_{2-x}$Ce$_{x}$CuO$_{4-\delta }$ (PCCO) as a function of disorder and oxygen content. Overdoped samples (x = 0.17, 0.19) were made using a pulsed laser deposition technique. Oxygen content was adjusted during the annealing process and disorder was induced by proton irradiation. The evolution of the resistivity and the Hall coefficient is examined, with a focus placed on the normal state properties in the regime T $<$ T$_{c}$ and H $>$ H$_{c2}$. Results are compared between optimally prepared, oxygenated, and irradiated samples.   

Superstructural lattice distortion within electron doped superconductor Pr$_{0.88}$LaCe$_{0.12}$CuO$_{4}$

It is well-known that electron-doped Nd$_{0.85}$Ce$_{0.15}$CuO$_{4}$ can be reversibly converted between the superconductor and non-superconductor by appropriate high-temperature treatments in reducing or oxidizing environments, respectively. Samples that exhibit superconductivity also present a crystalline Nd$_{2}$O$_{3}$ (bixbyite structure) impurity phase as well as diffuse (0, 0, L) rods of scattering at superlattice positions in the (H, K, 0) plane that coincide with the reciprocal lattice of the larger 2$\surd $2 $\times $ 2$\surd $2 impurity cell. We present a diffuse scattering analysis of the rod scattering in related compound Pr$_{0.88}$LaCe$_{0.12}$CuO$_{4}$ (PLCCO) and demonstrate that they are evidence of a superstructural distortion of the CuO$_{2}$ sheets, rather than an impurity effect. This in-plane superstructure may prove to be a necessary condition for superconductivity in the electron-doped cuprates.   

The effect of annealing to the transport and magnetic properties of electron-doped Pr$_{0.88}$LaCe$_{0.12}$CuO$_{4}$

The single crystals of electron doped cuprate Pr$_{0.88}$LaCe$_{0.12}$CuO$_{4+\delta}$ have been grown by floating-zone method. As-grown, the samples exhibit long-range antiferromagnetic order without superconductivity. Superconductivity only appears after annealing the sample in pure Ar or vacuum. We present a detailed investigation on how annealing conditions affect the in-plane and $c$-axis resistivity and hall coefficients. We will also discuss the chemical compositions of these samples before and after annealing.   

Thin Film Growth via Pulsed Laser Deposition and Characterization of the Electron-Doped Superconductor Sm$_{2-x}$Ce$_x$CuO$_{4-y}$

We report the growth and characterization of the electron-doped superconductor Sm$_{2-x}$Ce$_x$CuO$_{4-y}$ (SCCO). The growth conditions and in-situ annealing procedures for the thin films grown on yttria-stabilized zirconia (YSZ) substrates have been established. These are reported along with x-ray diffraction measurements, magnetic susceptibility, and magnetoresistance R(H, T) data. Scaling analysis of the transport measurements yields a diagram of the vortex-glass melting line. The films have a superconducting transition temperature as high as T$_c$ = 17.8 K and a transition width $\Delta$T$_c$ = 0.93 K. A comparison between our results on thin films and those on single crystals is also made. The results from additional underdoped (x $<$ 0.15) and overdoped (x $>$ 0.15) concentrations are also presented. High magnetic field and low temperature data were taken at the National High Magnetic Field Laboratory (NHMFL) in Tallahassee, Florida. This research was supported by the US Department of Energy under Grant No. FG03-86ER-45230 and the CULAR program no. 9985-001.   

Origin of superconducting carriers in ``non-doped'' T$^{\prime}$-(La, RE)$_{2}$CuO$_{4}$ (RE = Sm, Eu, Gd, Tb, Lu, and Y)

We have reported the isovalently-substituted new superconductors T'-La$_{2-x}$\textit{RE}$_{x}$CuO$_{4}$ ($T_{c}\sim $20-25K) prepared by MBE with no effective dopant. As regards the origin of the carriers in these nominally non-doped superconductors, there seems to be two possible scenarios: (i) oxygen deficiencies at the regular oxygen sites serve as a source of effective electron carriers, and (ii) they are not Mott insulators and have intrinsic carriers. Since precise information on the site- specific occupancy of oxygen is very difficult to obtain, alternatively, we investigated the in-plane lattice constant $a_{0}$ with changing \textit{RE} concentration $x$, with a view to examining possible $a_{0}$ expansion due to electron doping. In each \textit{RE} substitution, the $a_{0 }$of the T'-La$_{2-x}$\textit{RE}$_{x} $CuO$_{4 }$linearly decreases with increasing $x$, whose extrapolation to $x$=2 agrees well with the reported value for bulk T'-\textit{RE}$_{2}$CuO$_{4}$. This$_{ } $variation can simply be understood based on the difference in the ionic radius of \textit{RE}$^{3+ }$vs La$^{3+}$, suggesting that these superconductors$_{ }$are not electron-doped, at least substantially, and that the second scenario is the more plausible. This conclusion is also supported by the results of transport and photoemission experiments.   

Electric-field-induced modulation of the magnetic penetration depth of superconducting La$_{2-x}$Sr$_{x}$CuO$_{4}$ ultrathin films

A study of the electric-field-induced change of the in-plane magnetic penetration depth $\lambda _{ab}$ of an underdoped La$_{2-x}$Sr$_{x}$CuO$_{4}$ (LSCO) ultrathin superconducting (S) film is reported for the first time. Using MBE, a two unit-cell (UC) thick (x$\approx $0.1) LSCO S-film was grown epitaxially on a 12 UC thick normal (x=0.4) LSCO buffer layer deposited on a monocrystalline. SrLaAlO$_{4}$ substrate. A capacitor structure was then patterned after growing on top of the S-film a 15 nm thick HfO$_{2}$ insulating layer with a dielectric constant $\varepsilon \approx $15 and a Pt gate electrode. The inverse kinetic inductance 1/L$_{k}\propto $ 1/$\lambda _{ab}^{2}$ of the LSCO film was measured using an inductive two-coil technique. Both the temperature (T) and magnetic-field dependences of 1/L$_{k}$ were investigated by applying gate voltages corresponding to electric fields E = $\pm $ (2 x 10$^{8})$ V/m. For the largest E-field modulation ($\Delta $E $\equiv $ 4 x 10$^{8}$ V/m) a relative change $\Delta $L$_{k}^{-1}$/L$_{k}^{-1}\approx $ 18{\%} was observed at low temperature in good agreement with an elementary theoretical estimate. The nonmonotonic T-dependence of $\Delta $L$_{k}^{-1}$/L$_{k}^{-1}$ (a maximum is observed where L$_{k}^{-1}$(T) has the largest slope) can be accurately described by a simple model assuming that L$_{k}^{-1}$(0) $\propto $ T$_{c}$.   

Experimental observation of magic doping fractions and two-dimensional charge ordering in La$_{2-x}$Sr$_x$CuO$_4$

Competing order is currently an issue of controversy in the study of high temperature superconductivity. In LSCO, neutron scattering experiments have found one-dimensional spin stripes; however, it is unclear whether LSCO has some sort of charge-ordered state. If there is a charge-ordered state in LSCO, the charge ordering tendency is expected to be pronounced near certain doping levels where the charge modulation is commensurate with the underlying lattice; therefore, to examine the nature of possible charge ordering in LSCO, we have carefully measured the hole-doping dependence of the in-plane resistivity using a series of high-quality single crystals. Our detailed measurements find a tendency toward charge ordering at particular rational hole doping fractions of 1/16, 3/32, 1/8, and 3/16. This observation is most consistent with a recent theoretical prediction of the checkerboard-type ordering of the Cooper pairs at rational doping fractions $x=(2m+1)/2^n$, with integers $m$ and $n$.   

Synthesis and Charateriztion of BSCCO System Doped with Ru, Nb and Ho

Samples of Bi$_{2}$Sr$_{2}$CaCu$_{2-x}$ A$_{x}$O$_{8}$ with A=Ru, Nb, Ho and X=0.0, 0.10, 0.15, 0.2 were synthesized using ammonium nitrate melt method. Appropriate amounts of Bi$_{2}$O$_{3}$, SrCO$_{3}$, and the oxides of the doping elements, etc were mixed together. A small amount of ammonium nitrate was added to the mixture and reground thoroughly using an agate mortar and a pestle. The resulted precursor then heated between 160-170C under hood. Further heating at 400-500 C resulted in the formation of dark colored powder. The powder was then pressed into small pellets and annealed at 950 C in flowing oxygen for 24 hours and then furnace cooled to room temperature. The samples were characterized by measuring their magnetic as well as superconducting properties which will be presented.   

Growth and characterization of large HgBa$_{2}$CuO$_{4+\delta}$ single crystals

Using flux techniques, we have been able to grow unprecedentedly large HgBa$_{2}$CuO$_{4+\delta}$ (Hg1201) single crystal, exceeding 20 mm$^{3}$ in volume. Hg1201 is a model high-temperature superconductor, with the highest T$_c$ ($\sim 97$ K at optimal doping) among all single-layer cuprates and a simple tetragonal crystal structure. X-Ray and neutron scattering measurements demonstrate the single-grain nature of our crystals. We report results for the uniform susceptibility and the resistivity. Measurements of the $c$-axis resistivity and magnetoresistance were used to determine the pseudogap temperature at several hole densities.   

Growth and superconductivity of single crystals La$_{1.875}$Ba$_{0.125}$CuO$_4$

The origin of high temperature superconductivity in cuprate materials is one of the biggest puzzles in material science. Since the discovery of the significant anomalous suppression of superconductivity in high temperature superconducting oxide La$_{2-x}$Ba$_{x}$CuO$_{4 }$(x=1/8) $^{[1]}$, the so-called 1/8 anomaly has been a subject of considerable research attention. Many attempts to grow the single crystals have been made, but no single crystal La$_{2-x}$Ba$_{x}$CuO$_{4 }$(x=1/8) has been successfully grown. In this work, the effects of the growth condition and the compositions of a feed rod on the crystal growth of La$_{2-x}$Ba$_{x}$CuO$_{4 }$has been studied by an infrared image floating zone method. The experimental result shows that a planar solid-liquid growing interface tends to break down into a cellular interface when the growth velocity is more than 1 mm/h. When the planar solid-liquid growing interface break down into a cellular interface, the single crystal size decreases abruptly and the as-grown rod is not single phase. The large single crystals of La$_{2-x}$Ba$_{x}$CuO$_{4 }$(x=1/8) has been successfully grown. The single crystals of La$_{2-x}$Ba$_{x}$CuO$_{4 }$(x=1/8) up to 8 mm diameter and 55 mm length have been cut from the as-grown bars. The superconductivity transition temperature T$_{c}$ of as-grown single crystals is 2.5 K. The static stripe order in the large single crystals has been studied by neutron scattering method$^{[2]}$. [1] A. R. Moodenbaugh \textit{et al}, Phys. Rev. B, 38(1988)4596. [2] J. M. Tranquada \textit{et al}, Nature, 429(2004)534.   

Session P13: Superconducting Thin Films and Transport

Electrical properties of epitaxial junctions between Nb:SrTiO$_{3}$ and optimally doped, underdoped and Zn-doped YBa$_{2}$Cu$_{3}$O$_{7-\delta }$

Epitaxial thin films of optimally doped, underdoped and Zn-doped YBa$_{2}$Cu$_{3}$O$_{7-\delta }$ were grown on single crystal (001) Nb:SrTiO$_{3}$ substrates by pulsed laser deposition (PLD) and the electrical properties of the corresponding interface junctions were examined. The growth conditions were optimized in each case to get the appropriate crystalline quality of the films as well as the desired normal state and superconducting properties. The films/heterointerfaces were characterized by x-ray diffraction, Rutherford backscattering channeling (RBSC) spectrometry in normal and oxygen resonance modes, magnetic susceptibility, four probe in-plane resistivity, and the temperature dependent junction current-voltage (I-V) characteristics. Non-linear I-V curves (forward and reverse) were obtained in all the cases, revealing some characteristic differences and interesting temperature evolution. These data were analyzed within the framework of a standard description of the transport across the metal-semiconductor (Schottky) interface and the curve fitting parameters were extracted in each case. An attempt is made to relate the observed parametric differences to the existing knowledge about the normal state and superconducting properties of the films.   

Ce doping in T-La2CuO4 films: Broken electron-hole symmetry in high-Tc superconductivity

We attempted Ce doping in La$_{2}$CuO$_{4}$ with the K$_{2} $NiF$_{4}$ ($T)$ structure by molecular beam epitaxy. With low growth temperature and appropriate substrate choice, we found that Ce can be incorporated into the K$_{2}$NiF$_{4}$ lattice up to $x \quad \sim $ 0.06, which has not yet been realized in bulk synthesis. The doping of Ce made $T$-La$_{2-x}$Ce$_{x}$CuO$_{4}$ more insulating, which is in sharp contrast to Ce doping in La$_{2}$CuO$_{4}$ with the Nd$_{2}$CuO$_{4}$ structure, which made the compounds superconducting. The observed smooth increase in resistivity from hole-doped side ($T$-La$_{2-x}$Sr$_{x}$CuO$_ {4})$ to electron-doped side ($T$-La$_{2-x}$Ce$_{x}$CuO$_{4})$ indicates that electron-hole symmetry is broken in the $T$-phase materials. We propose that the nature of the insulating state in $T$-La$_{2-x}$Ce$_{x} $CuO$_{4}$ is of a Kondo insulator instead of a Mott insulator. The insulating mechanism based on Kondo interaction between Cu3d spins and O2p holes explains the global evolution of the resistivity and also the pseudo gap phenomenon from hole-doping to electron doping.   

Magnetic impurity doping effect of Bi2201 superconducting thin films

We have studied the effect of doping Mn into Cu-O planes of Bi$_{2}$Sr$_{2}$CuO$_{6}$ thin films grown by atomic layer-by-layer Molecular Beam Epitaxy in an ozone environment of 8E-6 Torr. These films were grown on SrTiO$_{3 }$substrates at 680$^{o}$C. The in-situ Reflection-High-Energy-Electron-Diffraction showed two-dimensional atomically flat surfaces during the whole growth. As Mn concentration on Cu site is varied, we see significant changes in transport. Bi2201 films with no Mn are superconducting with Tc of about 12.5K. By substituting 3.5{\%} of the Cu sites with Mn, we observed insulating behavior characterized by a variable range hopping mechanism with dimensionality between 2 and 3.   

Complex Conductance Measurements of Ultra-thin MoGe films near the Superconductor-Insulator Transition

The application of a magnetic field destroys the superconducting state and gives rise to unusual conducting or insulating states in two-dimensional samples. [1,2] This field-tuned transition has been extensively studied using conventional electrical transport measurements and analyzed within the context of critical behavior near a quantum phase transition. We report on a new approach to study the magnetic field-tuned transition using a two-coil mutual inductance technique, which has been integrated into a top-loading dilution refrigerator. Using this experimental setup, we have measured the complex conductance of Mo$_{43}$Ge$_{57}$ thin films as function of temperature and magnetic field. These measurements are used to determine the behavior of the superconducting electron density in the vicinity of the field-tuned transition. $^{1}$ A Yazdani and A. Kapitulnik \textit{Phys. Rev. Lett.} \textbf{74}, 3037(1995). $^{2}$ N. Mason and A. Kapitulnik.,\textit{ Phys Rev.} B \textbf{64}, 60504-1 (2001).   

Crossover from 2D-XY to 3D-XY superconducting fluctuations with hole-doping in dynamical conductivity of La$_{2-x}$Sr$_x$CuO$_4$ thin films by broadnband technique

We report the systematic study of dynamical complex ac conductivity, $\sigma(\omega)=\sigma_1(\omega)+i\sigma_2(\omega)$, of high-quality La$_{2-x}$Sr$_x$CuO$_4$ (LSCO) ($x$=0.07 to 0.24) thin films ate temperatures ($T$s) just above the superconducting transition temperature, $T_c$, by using a broadband microwave technique, where both components of the complex conductivity can be obtained as a detailed function of frequency by sweeping the microwave frequency continuously (0.1 to 12~GHz). For all superconducting films, we observed a definite contribution of the superconducting fluctuation to $\sigma(\omega, T)$, which could not be described by the conventional Aslamazov- Larkin term. Detailed analyses using a dynamic scaling theory clearly indicated that the fluctuation can be well described by the 2D-$XY$ critical behavior for the underdoped LSCO samples, whereas it changes into the 3D-XY critical behavior in the optimally doped samples. We are now fixing the fluctuation behavior in the overdoped samples, which is very important to judge whether or not the quantum-criticality picture proposed in some theories is appropriate for the description of the electronic state of the high-$T_c$ cuprate supercondutors.   

Vortices and the superconductor-insulator transition

We present results from a study of the temperature ($T$) and magnetic field ($B$) dependence of disordered, superconducting, amorphous indium-oxide thin-films. Application of a perpendicular $B$ weakens superconductivity until, at a well- defined critical $B$, the system is driven into an insulating state. We find that our samples follow a simple power-law dependence on $B$ that holds over a wide range of $T$ and resistance. Surprisingly, this power-law dependence continues, uninterrupted, into the $B$-driven insulating state. These results indicate that vortices play a central role in determining the transport properties of our films.   

Fractional-exponent behavior of magnetization near $T_c$ in ${\rm Bi_2Sr_2CaCu_2O_8}$

Using high-resolution torque magnetometry, we have investigated in detail how long-range phase coherence develops as the critical temperature $T_c$ (88.7 K) is approached in optimally-doped $\rm Bi_2Sr_2CaCuO_{8+\delta}$ with field $\bf H||c$. Three distinct regimes are observed. Above $\sim$92 K, $|M|$ increases rapidly as $T\rightarrow T_c$ in step with the vortex Nernst signal. $M$ is strictly linear in $H$ in weak $H$, but shows strong curvature at large $H$ (5-14 T). The curvature provides a determination of the correlation length $\xi_{sc}$ which grows as a power law, viz. $\xi_{sc}\sim 1/t^\nu$. In the second regime, $86 < T < 92$ K, $M$ becomes nonlinear in $H$, viz. $M\sim H^{\alpha(T)}$, where the exponent $\alpha(T)$ decreases from 1 to 0. This interesting fractional-exponent behavior is highly unusual and fits poorly with conventional pictures of `fluctuating diamagnetism.' As previously known, $M$ is virtually $H$ independent below 2 Tesla at the ``crossing temperature'' $T_{cr} $ = 86 K. Below $T_{cr}$, $M$ is a function of $\log H$. We compare this behavior with predictions of the 3DXY and Kosterlitz-Thouless theory. Supported by funds from the U.S. National Science Foundation under grant DMR 0213706.   

Normal-Superconducting Phase Transition Obscured by Current Noise

There is a large volume of experimental research on the normal- superconducting phase transition, both in zero field and the so- called ``vortex-glass'' transition in a field. For these phase transitions, resistive behavior at low currents is expected for $T > T_c$, and non-linear current versus voltage isotherms are expected below $T_c$. We show theoretically and experimentally that the addition of current noise to nonlinear voltage versus current curves will create ohmic behavior. Thus, current noise will create ohmic behavior at low currents even for temperatures below $T_c$, and isotherms that are actually \textit{below} $T_c$ will appear to be \textit{above} $T_c$. This obscures the phase transition and leads to incorrect values for $T_c$ and the critical exponents $\nu$ and $z$. Most reports in the literature do not explicitly mention filtering, yet the transition temperature and the critical exponents extracted from the conventional analysis are inaccurate if current noise is not filtered out. Thus, current noise is a possible explanation for the wide range of critical exponents found in the literature.   

Phase Transition from a Spin-Glass Metal to Superconductor in La2-xSrxCuO4 (x<0.05) Single Crystalline Films by Epitaxial Strain

Epitaxial growth of films is an effective way to induce high anisotropic pressure on films. In this work we have grown La$_{2-x}$Sr$_{x}$CuO$_{4}$ (LSCO) (0.03$<$x$<$0.05) films, having metallic behavior in bulk, by liquid phase epitaxy technique along the $<$100$>$ direction. For the desired kind of strain we have chosen overdoped LSCO ($\sim $x=0.19) as the substrate. We observed that the resistance versus temperature measurement, in a configuration where the film and substrate are in series, shows a two-step superconducting transition. We suggest that the other transition, which does not coincide with the T$_{c}$ of substrate, is due to superconductivity in film. The T$_{c}$ of the films with x=0.034, 0.040 and 0.045 are found to be 25K, 30K and 34K, respectively. Where as the x=0.0 film is not superconducting. In our work we have shown possible transition of a spin-glass metallic phase to a superconducting phase in LSCO (0.03$<$x$<$0.05) films induced by epitaxial strain.   

Critical currents of $ex-situ$ YBCO thin films on ``RABiTS'' substrates: thickness, field and temperature dependencies

The critical current density $J_c$ flowing in thin $YBa_2Cu_3O_ {7-\delta}$ (YBCO) films of various thicknesses $d$ has been studied magnetometrically, both as a function of applied field $H$ and temperature $T$. The films, grown by a BaF$_2$ $ex- situ$ process and deposited on buffered ``RABiTS'' substrates of Ni-5$\%$W, have thicknesses $d$ ranging from 28 nm to 1.5 $\mu$m. Isothermal magnetization loops $M(H;T)$ and remanent magnetization $M_{rem}(T)$ in $H=0$ were measured with $H$ $\|$ c-axis (i.e., normal to film plane). The $J_c(d)$ values, which were obtained from a modified critical state model, increase with thickness $d$, peak near $d \sim$ 150 nm, and thereafter decrease as the films get thicker. For a range of temperatures and intermediate fields, we find $J_c \propto H^{-\alpha}$ with $\alpha \sim (0.56 - 0.69)$ for all materials. This feature can be attributed to pinning by large random defects. At higher fields approaching the irreversibility line, $J_c(H)$ decreases faster. The $J_c$ at self field varies as $J_c(T,sf) \sim (1-T/T_c)^n$ with $n \sim$ 1.2 - 1.4. This points to ``$\delta T_c$ pinning'' (pinning that suppresses $T_c$ locally) in these YBCO materials. Work at UTK was supported by AFOSR Grant F49620-02-1-0182. ORNL is managed by UT-Battelle, LLC for the USDOE.   

Current Percolation Characteristics of IBAD Coated Conductors

We have investigated IBAD coated conductors by variable temperature scanning laser microscopy (VTSLM) to map how current flows in superconducting transition and superconducting region. We examined various factors that make non-uniform current distribution and found that the current flow was affected by the grain boundary network, the physical bending of samples, and the surface contaminations. On the basis of data analysis, we deduced 65$\mu $m $\sim $ 150$\mu $m as percolation cluster feature sizes. Below the critical temperature, we have investigated J$_{c}$ to establish the relationship of the T$_{c}$ and $\delta $V and J$_{c}$, and we conclude that lower T$_{c}$ and higher $\delta $V area has lower J$_{c}$. We also found several circular shapes of about 100$\mu $m $\times $ 130$\mu $m diameter from a sample. Those dots had definitely higher T$_{c}$ than the surrounding area and created non-uniform current flow. In this presentation, we will report more about VTSLM images around the dots and the characteristics of current bottleneck features which are related with the lower J$_{c}$.   

Reversible and irreversible magnetostriction of untwinned YBa$_2$Cu$_3$O$_{6.999}$ single crystals

We present anisotropic magnetostriction measurements of untwinned $YBa_2 Cu_3 O_{6.999} $ single crystals between 40 K and 150 K and in fields up to 10 T applied along the c-axis. The isothermal magnetostriction, which probes the pressure dependence of the magnetization, is nearly reversible above 55 K, which allows a thermodynamic analysis of the data. These data are used to derive the uniaxial pressure dependences of $T_c $, the thermodynamical critical field $H_c^o $and the electronic specific heat coefficient$\gamma _{electronic} $. The data also provide a unambiguous measure of the broadened$H_{c2} $, which shows excellent 3D-XY scaling. Below 55 K and above 6-8 T, the magnetostriction becomes irreversible due to increased flux pinning at the Bose-glass to vortex-glass transition of the vortex matter. In contrast to the magnetization, which shows the typical monotonic peak effect, the irreversible magnetostriction exhibits reproducible fine structure within the transition region. This may be due to nucleation of large vortex domains in the crystal and points to a first-order phase transition.   

Superconductors in a Strong Low-Frequency ac Electric Field

The electric-field induced ball formation has been observed for high temperature superconducting particles, MgB$_2$ powder, and low temperature superconducting particles in a low frequency ac electric field. Different from the situation with a static electric field, the superconducting particles in an ac field first form chains along the field direction if the electric field is below a critical value $E_{c1} $. As soon as the field exceeds $E_{c1}$, the chains are broken and the particles aggregate into balls. The experiment has found that $E_{c1}$ is a function of frequency $\omega$. To understand the experimental results, we consider a bulk superconductor in an ac field. The electric field is along the x direction and the bulk superconductor has its surface at $x=0$, perpendicular to the field and is located at $x\ge 0$. The electric field penetrates into the superconductor: for $x>0$, ${\vec E}(x)={\vec e_x}E \exp(-x/l_s)\cos(\omega t)$, where $l_s$ is the electric-field's penetration depth and $E$ is the electric field at the surface of $x=0$. With this model, we have found that if the electric field is strong enough, Cooper pairs near the surface are depleted and a positive surface energy is produced. This induced surface energy is responsible for the formation of superconducting balls. The critical electric field to produce the positive surface energy $E_{c1}$ is found to be related to the binding energy of a Cooper pair$\Delta(T)=2\epsilon_f-\epsilon$ and the frequency $\omega$. As $\omega$ increases, $E_{c1}$ goes up, too. A comparison between the theory and experimental results will also be made.   

Tuning the Insulator-Superconductor Transition in Ultrathin Films By Use of the Electric-Field Effect

A 10 {\AA} thick film of amorphous bismuth has been prepared in an electric-field effect device geometry. Its low temperature electrical properties have been continuously tuned from weakly insulating to fully superconducting by increasing the gate voltage. The systematics of this insulator-superconductor transition will be discussed in the context of a quantum phase transition. This work is supported in part by the National Science Foundation under grant NSF/DMR-0138209.   

Session P14: Focus Session: Hydrogen Storage I: Media

A Variety of Metal-N-H System for Hydrogen Storage

Metal nitrides and imides exhibit strong affinity towards hydrogen molecules. Such strong interactions enable these substances to be used as potential materials for hydrogen storage. In the previous investigations, maximum of 11.5wt{\%} and 7.0 wt{\%} of hydrogen storage capacities have been determined in lithium nitride and lithium imide, respectively. However, relatively high operating temperatures place a serious restriction onto the application of those substances. It is clear that to lower down the operation temperatures the composition and structure of the subject material have to be altered in order to sit within a suitable thermodynamic range. Successful attempts have been made by introducing Mg or Ca into the Li2NH binary system, in which considerable reduction in hydrogen absorption and desorption temperatures and uprising H2 desorption plateau pressures have been achieved. Ternary imide Li2MgN2H2, as an example, could reversibly store 5.8wt{\%} of hydrogen at 180ºC with desorption plateau pressure higher than 10 bars. Previous investigations reveal that the hydrogen-rich phase of Li-N-H and Li-Mg-N-H systems comprise of metal amides and hydrides. It is probably the great potential for the union of H$^{\delta +}$ in amide and H$^{\delta -}$ in hydride to H2 that drives the two chemicals to react and give out hydrogen. According to this hypothesis, a serial of new Metal-N-H systems can be developed by reacting various amides with hydrides. Interaction between amide of Li or Mg with LiAlH4, MgH2, NaH and CaH2, respectively, has been investigated by an in situ planetary ball mill, TPD, volumetric Release-Soak techniques and FTIR etc. Novel Metal-N-H systems have been developed accordingly, among which three systems can release substantial amount of hydrogen (more than 5.0wt{\%}) near ambient temperature..   

Glassy materials as a hydrogen storage medium

The adsorption of molecular hydrogen on a glassy material and its relatives is studied with a use of pseudopotential density functional method. The binding energy and distance of adsorbed hydrogen is particularly calculated. It is found that the desorption temperature of hydrogen in layered boron oxide is significantly higher than that in carbon nanotubes as much as twice, which is attributed to heteropolar bonding in boron oxide. The effect of water addition to boron oxide on hydrogen adsorption is also investigated. Our results indicate that water may reduce the surface area of boron oxide but does little affect the hydrogen adsorption energy. We also calculated an optimum pore size for hydrogen diffusion into boron oxide. Current study demonstrates a pathway to the finding of a new class of materials for hydrogen storage media that can hold hydrogen at ambient conditions through physisorption.   

Enhanced hydrogen affinity in single-walled carbon nanotubes relative to activated carbons

The present body of work represents a meticulous and thorough investigation of single-walled carbon nanotube properties and processing in relation to hydrogen storage. Nanotube samples were characterized by Raman spectroscopy, light scattering, microscopy, TGA, ICP, differential pressure adsorption (DPAU), and BET surface area. A reproducible carbon nanotube cutting method was developed and characterized. A number of nanotube variables, such as average length, were then evaluated for their effects on hydrogen capacity. Thorough characterization reveals the strongly variable nature of carbon nanotube materials. Diameter, length, purity, structural integrity, as well as secondary and tertiary morphology must be determined in order to establish any structure-property relationships. Measurements of hydrogen capacity indicate that commercially available single-walled carbon nanotubes in their pristine, non-functionalized form are not a viable hydrogen storage option. However, it is also clear that single-walled carbon nanotubes have a higher affinity for hydrogen than do activated carbons. At a given BET surface area, activated carbons have a fraction of the hydrogen capacity of single-walled carbon nanotubes. This information, in addition to Air Product's experimental modeling results, leads to a promising path forward. In particular, experiments with smaller diameter nanotubes and charge transfer intercalation are planned.   

New Reversible Complex Metal Hydrides

Novel, reversible complex metal hydrides with the general formula $A_{3-x}A'_x$AlH$_6$ ($A$, $A' =$ Na, Li, K, Mg, Ca) were synthesized and structurally characterized using synchrotron x-ray diffraction. The hydrogen absorption/desorption characteristics and thermodynamic properties were studied using pressure-composition isotherms. These results demonstrate that the partial substitution of the alkali metal can change the equilibrium pressures substantially. As an example, the substitution of one Li for Na in Na$_3$AlH$_6$ cryolite to form Na$_2$LiAlH$_6$ elpasolite increases the dissociation enthalpy by $6.5 \pm1.6$ kJ/mol H$_2$. This thermodynamic change lowers the plateau pressure by 30 bar at 518 K. Similar trends were observed in the potassium cryolite and elpasolite phases. This form of thermodynamic tuning may be applied to other, high capacity alanates (e.~g. LiAlH$_4$ and Mg(AlH$_4$)$_2$), which are currently hindered by reaction enthalpies that are largely unfavorable for PEM fuel cell applications.   

Hydrogen storage in BN cage

Recently hydrogen is being considered as a potential candidate to meet the increasing energy need of both the developing and developed world. To this end it is important to find efficient means for storing hydrogen. Carbon nanotubes, especially single-wall tubes, were initially considered to be better candidates for hydrogen storage than other materials. However, the early experiments have met with some controversy and very different results for the hydrogen storing capacity of carbon nano-tubes have been reported. To search for other non-carbon system, we studied (H$_{2})_{n}$@B$_{36}$N$_{36}$ cage by using first-principles method. It has been found that H$_{2}$ can go into the cage through the hexagonal face. The maximum number of H$_{2}$ that can be inserted into the cage without breaking the cage is 18, resulting in a weight percentage of 4{\%}. However, the storage of H$_{2}$ in BN cage needs external energy, which increases with n$^{2}$ (n is the number of hydrogen molecules) while the HOMO-LUMO gap decreases as n$^{3}$. The energy cost is associated with the increase in the B-N bond length and decrease in the H-H bond length as n increases.   

Effect of Mg Doping in NaAlH$_4$

First principles calculations based on density functional theory have been carried out to search for suitable catalysts that can lower the hydrogen desorption temperature from sodium-alanate (NaAlH$_{4})$. We focus here on the possibilities of doping Mg in bulk sodium alanate. The result shows that Mg prefers to occupy the Na site and weakens the covalent bond between Al and H. The energy needed to remove a hydrogen atom from (Mg,Na)AlH$_{4}$ is found to be significantly lower than that from NaAlH$_{4}$. The effect is similar to Ti doping which is supposedly the best catalyst found to date for hydrogen desorption from sodium alanate. Furthermore, the Mg doping is shown to promote the formation of Na vacancy which in turn plays an important role in the hydrogen desorption process.   

Cohesion of Mg$_2$FeH$_6$ and Related Complex Hydrides

The stability and bonding of complex K$_2$PtCl$_6$ structure hydrides is analyzed using results of density functional calculations. The cohesion is dominated by ionic contributions. The 18-electron rule generally followed in these compounds results from their ionic character combined with crystal field effects. Density functional results for the formation energies are presented and implications for hydrogen storage are discussed.   

New Quaternary Hydride Li$_{3}$BN$_{2}$H$_{8}$ with $>$10 wt{\%} Hydrogen: I. Material Synthesis and Structural Characterization

We report a new quaternary hydride Li$_{3}$BN$_{2}$H$_{8}$ synthesized from mixed LiNH$_{2}$ and LiBH$_{4}$ powders in a 2:1 molar ratio by ball milling. X-ray diffraction (XRD) results show that as milling time increases, the LiNH$_{2}$ and LiBH$_{4}$ diffraction peaks weaken and a new set of peaks emerges. At 40 min, the sample is substantially converted to the new phase, with only a small remnant of LiNH$_{2}$ in the XRD pattern. After 300 min the conversion is complete, and continued milling up to 960 min produces no further change. The final XRD pattern appears to be single phase, except for a small amount of Li$_{2}$O impurity, and has a background intensity that is essentially unchanged with milling time, implying that ball milling does not produce an amorphous phase. All of the observed XRD peaks can be indexed as a single BCC quaternary phase with a =10.76 {\AA}. Our \textit{in-situ} XRD data show that Li$_{3}$BN$_{2}$H$_{8}$ forms when mixed LiNH$_{2}$ and LiBH$_{4}$ powders are heated to above $\sim $95$^{\circ}$C without ball milling, then melts at $\sim $190$^{\circ}$C, and finally forms a mixture of solid Li$_{3}$BN$_{2}$ polymorphs upon H$_{2}$ gas release above $\sim $250$^{\circ}$C.   

New Quaternary Hydride Li$_{3}$BN$_{2}$H$_{8}$ with $>$10 wt{\%} Hydrogen: II. Hydrogen Desorption Measurements

We report thermogravimetric, volumetric, and calorimetric measurements of hydrogen desorption from the new quaternary hydride Li$_{3}$BN$_{2}$H$_{8}$ (11.9 wt{\%} theoretical hydrogen capacity). Li$_{3}$BN$_{2}$H$_{8}$ releases $\ge $10 wt{\%} hydrogen at temperatures above $\sim $250$^{\circ}$C. Simultaneous mass spectrometry residual gas analysis shows that a small amount of ammonia (2-3 mole{\%} of the generated gas) is released concurrently. Independent volumetric and gravimetric measurements are in excellent agreement regarding the quantities of hydrogen and ammonia released. Differential scanning calorimetery and in-situ x-ray diffraction show that Li$_{3}$BN$_{2}$H$_{8}$ melts at $\sim $190$^{\circ}$C, thus hydrogen evolution occurs from the molten state. It dehydrides to the solid product Li$_{3}$BN$_{2}$, and the evolved gas satisfactorily accounts for all of the available hydrogen content. Preliminary calorimetric measurements suggest that hydrogen release is exothermic, and, hence, not easily reversible; to date, rehydriding has not been achieved.   

Towards 9 weight percent, reversible, room temperature hydrogen adsorbents: Hydrogen saturated organometallic bucky balls

A new concept for high-capacity hydrogen absorbents is introduced by first-principles calculations. Transition metal (TM) atoms bound to fullerenes are proposed as a medium for high density, room temperature, ambient pressure storage of hydrogen. TMs bind to C60 or C48B12 by charge transfer interactions to produce stable organometallic bucky balls (OBBs) and bind to multiple dihydrogen molecules through the so-called Kubas interaction [1]. A particular scandium OBB can bind as many as eleven hydrogen atoms per TM, ten of which are bound in the form of dihydrogen molecular ligands that can be adsorbed and desorbed reversibly. In this case, the calculated binding energy is around 0.3 eV/H2, which is ideal for use on-board vehicles. The theoretical maximum retrievable H2 storage density is about 9 weight percent. This work was supported by the U.S. DOE EERE, BES/MS, and BES/CS under contract No. DEAC36-99GO10337. [1] G.J. Kubas, J. Organometallic Chem. 635, 37 (2001).   

Session P15: Focus Session: Relaxation and Phonons in Nanostructures

Properties of the LO-Phonon in GaN Nanocrystallites

Resonant Raman scattering in wurtzite structured GaN nanocrystallites of various morphologies were studied. The LO polar mode exhibited Fr\"{o}hlich-type resonant Raman scattering whose characteristics were found to depend weakly on the morphology of the crystallites. In contrast, the UV-laser heating and heat retention in the porous media of a crystallite ensemble were discovered to drastically modify the Raman properties. An ensemble temperature on the order of 550 K was inferred from the electron-phonon interaction model, a result that was verified via Raman scattering experiments at the elevated temperature regime. Complementary photoluminescence investigations concur with the Raman findings. The LO behavior of the GaN nanocrystallites, at temperature range 77 K- 900 K was investigated as well; the behavior is discussed in terms of the anharmonic decay mechanisms and the phonon dispersion curve of GaN of the wurtzite structure   

Intraband carrier relaxation in semiconductor quantum rods: Competition between phonon-assisted cooling and Auger heating

Quantization of electronic and phonon energies and large surface-to-volume ratios significantly modify energy relaxation mechanisms in nanoscale semiconductors compared to bulk materials. In the case of ultrasmall, semiconductor nanocrystals (NCs), strong quantum confinement leads to greatly enhanced carrier-carrier interactions that open new NC-specific energy relaxation and recombination channels. Here, we analyze the effect of Auger heating on the energy relaxation dynamics in elongated CdSe nanocrystals [quantum rods (QRs)]. At low carrier densities, less than 2-3 photoexcited electron-hole (e-h) pairs per QR on average, we observe bulk-like, fast, phonon-assisted carrier cooling with a time constant of around 0.5 ps. At high pump-intensities (more than 2-3 e-h pairs per QR), we detect a dramatic, orders-of-magnitude reduction in the energy relaxation rate resulting from efficient Auger heating. In this regime, energy relaxation directly correlates with recombination dynamics, which is an effect that has never been observed either in bulk or low-dimensional materials. Furthermore, we find that Auger heating differs in short and long QRs that can be explained by the difference in the scaling of Auger rates with respect to the carrier density in zero-dimensional (0D) and 1D semiconductors.   

Confined optical phonons in PbSe quantum dots

We characterized the confined optical phonons of PbSe quantum dots (QDs) by Raman scattering and far-infrared (FIR) absorption spectroscopy. The size dependence of the Raman spectrum is consistent with theoretical calculations within a dielectric continuum model, considering the phonon dispersion in the bulk material. The electron-phonon coupling strengths inferred from overtones in the Raman spectra is three orders of magnitude larger than expected from the calculated electron wave functions. The FIR absorption spectra exhibit an asymmetric broad peak near the bulk LO phonon frequency. The measured spectra are inconsistent with the ?hard-boundary? condition that has been used successfully in prior work on lead-salt QDs. We will discuss the size-dependence of the confined optical phonon and the possible mechanism to explain the FIR absorption spectra of PbSe QDs.   

Effects of Band Structure on the Electronic and Optical Properties of Semiconductor Nanocrystals: Lead Selenide vs. Cadmium Selenide

PbSe and CdSe nanocrystals (NCs) can be synthesized with high monodispersity, have a size-tunable band gap, and can exhibit near unity photoluminescence quantum yields. These two materials have significant differences with respect to bulk band gap, crystal structure, Bohr radius, and carrier effective masses that result in very distinct energy structures. Here we perform a side-by-side comparison of the optical and electronic properties of these two materials in both the single and multiexciton regimes for NCs of comparable sizes. Femtosecond transient absorption (TA) spectroscopy is used to study inter- and intraband relaxation of photo-generated carriers including various types of Auger-relaxation processes. We have discovered that for PbSe NCs, photo-generated single excitons with sufficient energy in excess of the band gap are able to relax by producing multiple excitons (carrier multiplication). In carrier multiplication, intraband excess energy is transferred to a valence band electron that is excited into the conduction band, resulting in the formation of two or more excitons per initially photo-excited exciton [Phys. Rev. Lett. 2004, v.92, 186601/1-4]. We have found that this effect of multiexciton generation, which has never been found to occur with significant efficiency in bulk semiconductors, can occur with up to 100{\%} efficiency in PbSe NCs depending upon the absorbed photon energy and occurs at wavelengths that are relevant to solar energy conversion. This process, which is an enabler of Generation III solar cells, has the potential to considerably increase the power conversion efficiency of NC-based photovoltaics. Pump-power dependent TA studies performed with a probe pulse tuned to near the photoluminescence maximum have revealed that both PbSe NCs and CdSe NCs can exhibit optical gain and, when incorporated into high optical quality sol-gel waveguides, are capable of producing size-tunable amplified spontaneous emission [J. Phys. Chem. B 2003, v.107, 13765-8]. Similar TA studies performed with a probe pulse tuned to the band-edge absorption feature reveal that differences in crystal structure cause the two materials to have significantly different exciton degeneracy, which directly results in different thresholds for gain.   

Investigation of the Phonon Frequency Shifts in ZnO Quantum Dots

Nanostructures made of ZnO have recently attracted attention due to their proposed applications in low-voltage and short-wavelength electro-optical devices. However, the origin of the observed phonon frequency shifts in such nanostructures is not always understood. We carried out both resonant and non-resonant Raman measurements for 20 nm-diameter ZnO quantum dots (QDs) and bulk ZnO reference samples [1]. A comparison with a recently developed theory [2], allowed us to clarify the origin of the phonon frequency shifts in ZnO QDs. It was found that the phonon confinement results in phonon frequency shifts of only few cm$^{-1}$. At the same time, the UV laser heating of the QD ensemble was found to induce a large red shift of phonon frequencies for up to 14 cm$^{-1}$. The authors acknowledge the support of MARCO and its Functional Engineered Nano Architectonics (FENA) Focus Center. [1] K.A. Alim, V.A. Fonoberov, and A.A. Balandin, Appl. Phys. Lett., in review (2004). [2] V.A. Fonoberov and A.A. Balandin, Phys. Stat. Solidi C \textbf{1}, 2650 (2004); cond-mat/0405681; cond-mat/0411742.   

Ultrafast flipping of exciton spin orientation in colloidal CdSe quantum dots

A nonlinear optical spectroscopy is demonstrated that retrieves exciton spin orientation dynamics using linearly polarized excitation. Rotationally averaged optical selection rules for quantum dots dictate that the sign of the signal is reversed when the spin state flips. Results are reported for CdSe nanocrystal samples with mean diameters from 3.1 nm to 5.0 nm. Ultrafast exciton spin flip times correspondingly range from 236 fs to 1.2 ps. Implications for quantum computation and spintronics applications are that exciton transitions can be used to induce long-lived spin polarization, but memory of exciton spin orientation decays on times less than 1 ps.   

Ultrafast carrier dynamics and carrier localization in thermally annealed Si nanoclusters

We present results of time-resolved terahertz pulse spectroscopy experiments on thermally deposited and annealed Si nanoclusters embedded in a SiO$_{2}$ matrix. These clusters range in size but are on the order of 5 nm in diameter, which is comparable to the Bohr radius in Si where carriers are expected to be strongly localized. The frequency-resolved conductivity of these samples after excitation by a 400 nm, 100 fs pump pulse is non-Drude like with a real component that increases with frequency and a negative imaginary component indicative of carrier localization. A series of samples is investigated as a function of anneal temperature, showing transient absorption decays ranging from a few picoseconds to several hundred picoseconds as the anneal temperature increases. Several models to explain the observed response are discussed. The authors acknowledge financial support from NSERC, CIPI and iCORE.   

A Model for the Powerlaw Behavior of Blinking Dynamics of Single Quantum Dots

Quantum dots (QDs) are a new generation of fluorescent markers for biological labeling with exceptional optical properties. However, single QD fluorescence blink due to a photo-induced electron transfer (ET) process, where the QD core loses an electron to a surface based trap site generating a non-fluorescent core-shell charge-separated state. This QD blinking behavior complicates their imaging applications. We observed that CdSe/ZnS QD blinking is largely suppressed in cysteine and histidine solutions and provided a mechanism for the blinking suppression. Cysteine and histidine act as small molecule ligands and bind to the Zn$^{II}$ based surface trap sites. The electronic interaction between the ligand and Zn$^{II}$-based trap sites raises the energy of the non-fluorescent core-shell charge-separated state and thus shuts down the photo-induced ET process, leading to blinking suppression. Based on ligand binding mechanism, we developed a model that accounts for the observed power law behavior of QD blinking kinetics.   

Session P16: Focus Session: Optical Resonances and Techniques in Nano-Optics

Cavity cooling of a microlever

Few years ago P. F. Cohadon, A Heidmann, and M. Pinard demonstrated that laser-cooling of the vibrational modes of a mirror can be achieved by subjecting it to a radiation pressure, actively controlled through a servo loop adjusted to oppose its brownian thermal motion. Atoms in an optical trap on the other hand can be laser-cooled passively without such active feedback, because their random motion is intrinsically damped through their interaction with radiation. In very close analogy with laser-cooling of atoms we have developed a simple and direct experimental method for passive optical cooling of a micromechanical resonator. We established that exploiting cavity-light induced forces allow to quench efficiently the brownian vibrational fluctuations of a gold-coated silicon microlever from room temperature down to an effective temperature of 18 K. Extending this method to optical-cavity-induced radiation pressure might enable the quantum limit of the lever vibrational ground state to be reached.   

Optical forces at Morphology Dependent Resonance

A strong optical force can be induced on a pair of transparent dielectric microspheres by exciting the morphology dependent resonance (MDR). We investigate such a resonant optical force through rigorous calculations using multiple scattering theory for the electromagnetic field and the Maxwell stress tensor for the electromagnetic force. The bonding and anti-bonding modes of the electric field of the MDR's correspond to strong attractions and repulsions respectively. At resonance, the force can be enhanced by orders of magnitude as compare to the off-resonance case. With a modest incident light intensity, it is showed that the MDR-force can be stronger than thermal fluctuations and the van der Waals forces when the separation between the spheres is more than a few tens of nano-meter, thus achieving the goal of manipulation. It is showed that stable binding of the spheres is possible. The dependence of the force on separation between the spheres, and the role of absorption by the material, and the robustness against sphere size dispersions are also discussed.   

Single Metal Nanoparticle Optical Interference

Optical interference of plasmon light scattering from a single gold nanoparticle is experimentally observed by placing a plane mirror nearby. The unique interference patterns in both spatial and spectral domians are reproduced by simulations based on the Huygens-Fresnel diffraction theory. The large spectral resonance enables us to determine the distance to the mirror with a 10 nm resolution without scanning the mirror [1]. The image dipole from a spherical mirror's reflection interferes with the real dipole of a single gold nanoparticle attached to an optical fiber tip [2], resulting in enhancement and inhibition of the resonant scattering rate by the modulation in the scattered light intensity collected outside the interference solid angle. [1] S.-K. Eah, H.M. Jaeger, N.F. Scherer, G.P. Wiederrecht, {\&} X.-M. Lin, submitted to Phys. Rev. Lett. (Nov. 2004). [2] S.-K. Eah, H.M. Jaeger, N.F. Scherer, G.P. Wiederrecht, {\&} X.- M. Lin, Appl. Phys. Lett. (in press).   

A Simple Route to Tunable Ordered Arrays of Quantum Dots

This work presents an innovative method to grow structurally defined systems of CdSe@ZnS quantum dots (QDs). By firstly utilizing nanosphere lithography (NSL) to generate hexagonally patterned metallic islands, self-assembly of QDs was achieved with the aid of appropriate linker molecules. The ordered arrays were analyzed using AFM imaging and confocal microscopy techniques. The method for building this type of arrays is quite simple and permits the scale-up and scale-down of the size of the arrays. The next step was the construction of a 3D structure by organizing a bi-functionalized chemical linker on the QDs surface, building a structure made of several layers of QDs. Optical properties were studied here. Such arrays may have a large range of applications, as CdSe@ZnS QDs are well-known as the photonic crystals used in waveguides or information storage. Jessica Pacifico, Daniel Gomez and Paul Mulvaney, Advanced Materials, in press, 2004.   

Optical Studies of a Nanoporous Array in Silicon

We have studied the effects of nanometer-scale texturing on the optical properties of silicon. Surfaces are textured using a non- lithographic template method combined with plasma etching. The process results in an array of cylindrical nanopores 60 nm in diameter and varying in depth up to 1 micron. We observe a significant reduction in near-normal specular optical reflectance from silicon textured this way. This reduction is across a broadband range of photon energies from 2.0 to 6.0 eV, and varies systematically with increasing pore depth. We develop a two-dimension Maxwell-Garnett effective medium approximation to model the specular reflectance. Agreement between the model and data is good provided we take into account the roughness of the nanoporous surface. Micro-Raman scatter is found to depend strongly on the texturing, exhibiting an increase in intensity with pore depth. These results support the notion that both insertion and extraction are enhanced by the nanopore surface treatment. We will summarize the nanotexturing approach, the optical properties, and the effective medium model.   

Numerical Aperture Increasing Lens Assisted Microscopy and Spectroscopy

A Numerical Aperture Increasing Lens (NAIL) is used for high resolution far-field microscopy and spectroscopy of semiconductor nanostructures. Incorporating NAIL into our low temperature confocal microscope, we have been able to perform high collection efficiency spectroscopy of single, self-assembled InGaAs/GaAs quantum dots. We plan to exploit the measured six-fold collection efficiency increase in our system to enhance the signal-to-noise ratio in a Hanbury-Brown Twiss (HBT) interferometer. In an attempt to quantify the far-field optical resolution of our NAIL assisted thermal microscope, we are using a pulsed UV laser to generate a thermal radiation source in Si with a spatial extent less than .5 micrometers. Previously, we experimentally demonstrated an optical resolution of 1.4 micrometers when imaging semiconductor integrated circuits.   

Nano-patterning Using an Embedded Particle Monolayer as an Etch Mask

A new nano-fabrication technique using a self-assembled nano-particle monolayer as an etch mask is developed, which forms a homogeneous particle monolayer over a large area. A periodic nano-cone pattern, which acts as an antireflective structure, was fabricated onto the SiO2 substrate by transferring the nano-particles. A trapping layer of thermoplastic resin was formed on the substrate, and then nano-particles were spread to form a multilayer of particles. As it was heated, the particles of the bottom layer were spontaneously embedded in the trapping layer. Excess particles were washed away and the bottom layer remained. The substrate below was etched using the particle monolayer as an etch mask by CF4 reactive ion etching (RIE). As a result, conical nano-cone patterns were transferred over the entire surface of the substrate. Using this method, an antireflective structure with a two-dimensional grating known as a moth-eye surface was fabricated on the SiO2 substrate. Compared with the flat surface, the reflectivity was reduced more than 40{\%}. This can be applied to large flat display.   

Higher external efficiency of LEDs having nano-structured surface fabricated by self-assembled block-copolymer

Recently, the internal quantum efficiency of LED has improved, but the external efficiency remains as low as a few {\%} due to the high refractive index of semiconductors (n=3-3.5). In order to extract more light, a nano-pillar structure on the LED surface was fabricated. This structure has two functions; as an antireflective layer completely transmitting light below a critical angle to the LED surface, and as gratings that diffract the light larger than the critical angle to extract -1st order light. To fabricate this structure on the semiconductor (GaP) surface by dry etching, we employed a periodic dots pattern with self-assembled diblock-copolymer process. The diblock copolymer of polystyrene (PS) - polymethyl methacrylate (PMMA) was used in this study since the PMMA is etched much faster than the PS. The GaP was dry-etched by Cl-based inductively coupled plasma using the remaining PS dots as a mask, and the nano-pillars with a diameter of 100 nm, a period of 150-200 nm, and a height of 400 nm were fabricated. As a result, we improved the external extraction efficiency of the fabricated surface more than 100{\%} compared with the flat one.   

Ultra-small Ni/NiOx/Ni tunnel junctions for terrahertz and ir applications

We have studied current-voltage, conductance, second derivative and temperature dependence of Ni/NiOx/Ni tunnel junctions having areas less than 0.1 um2. These overlap junctions have very small capacitance and are being studied for use as room temperature, high frequency detectors. The junctions are fabricated using a photolithography derived shadow mask and angular depositions of the Ni. The oxide barrier is formed using a plasma oxidation process. Electron micrographs of the overlap junctions are used to estimate junction area. For RT operation a nonlinear i-v is required and is generally observed. The temperature dependence of the iv characteristics from 100 C to 4 K appears to be that of tunneling although at RT other parallel conduction paths may exist. The nonlinearity of the iv curves at 4 K is much greater than at RT the ratio being about 5 to 1 at 0.1 V applied. Tunneling spectroscopy studies at 4 K reveal scattering mechanisms that can be related to Ni-O excitations. Detailed simulation of both the temperature dependence and the i-v curves shows that the barrier height is about 0.2 eV.   

An alternative mechanism for surface enhanced spectroscopy: second harmonic surface plasmon resonance

Strong ultraviolet luminescence having intensity comparable with device quality GaN epifilms has been observed in Au nanoparticles. It is identified that the luminescence involves radiative recombination of electrons in band 6 (sp conduction band) with holes in band 4 (secondary top d band), near the L symmetry point. We show that the strong emission is a consequence of the second harmonic surface plasmon resonance (SHSPR), which is an inherent nature of metallic nanoparticles with high density of surface plasmons. However, for SHSPR we only need a dim pumping source, which is used to trigger the second harmonic absorption of very dense quantum particles, such as surface plasmons in Au nanoclusters. Unlike in the conventional SHG process, the high density quantum particles responsible for the nonlinearity in the SHSPA process preexist in the studied material, which are not due to external perturbation. In addition to Au nanoparticles, we demonstrate that SHSPR provides a very efficient way to enhance the luminescence of a material incorporated with metal nanoparticles. As an example, incorporation of Au nanoparticles into SiO$_{2}$ nanoparticles can enhance the luminescence intensity of the SiO$_{2 }$nanoparticles by a factor as large as 2 orders of magnitude. We thus point out that SHSPR can serve as one of the underlying mechanisms responsible for surface-enhanced nonlinear optical phenomena.   

Session P17: Focus Session: Materials and Device Physics for Quantum Computing II

Controlling Spin and Charge in Quantum Dots and Nanotubes

This talk summarizes recent experimental progress toward realizing exchange-coupled spin qubits in both few-electron GaAs quantum dots, and gate-defined carbon nanotubes. In the GaAs quantum dots, measurement of the singlet-triplet spin relaxation for separated spins has been realized. These measurements are made at low magnetic fields, and can be studied as a function of magnetic field. We find T2-like relaxation times of order 100 microseconds at 100 mT, with enhanced relaxation at zero field. The experimental method uses pulsed confining gates of a double dot with integrated quantum point contact charge sensors. The relaxation time appears to be limited by hyperfine coupling to nuclear spins. In the carbon nanotube system, we have realized gate-confined double dots and are currently measuring nonlinear double-dot transport, aiming to observe Pauli blockade effects. Up-to-the-minute results will be reported.   

Transport and charge detection spectroscopy of few electron quantum dot molecular states.

. Few electron lateral quantum dot molecules have been fabricated and studied using both transport and charge detection techniques. Such devices are ultimately planned for quantum information applications such as charge and spin based qubits. Measurements have been performed for several specific configurations (n,m) relevant for quantum information where n and m refer to the measured number of electrons in each dot e.g. (5,5) the lowest occupancy configuration with a filling factor 2 state. Measurements were made as a function of magnetic field and the tunnel coupling between the dots.   

Fermionic Bell state analyzer for spin qubits

In a seminal proposal Knill, Laflamme, and Milburn showed that quantum computation with photons is possible using only linear optics [1]. Partial measurements of a quantum state are sufficient and, most remarkably, coupling qubits with each other via gates is no longer required. It is an important task, therefore, to extend this concept to other qubit systems as well and, in particular, to search for physical realization of such partial measurements for Fermionic qubits. In an important step in this direction, Beenakker et al. recently proposed to combine partial Bell measurements of charge qubits with single qubit operations [2], however, no concrete read-out scheme was discussed. To address and solve this fundamental problem, we propose here to consider spin instead of charge, which is considered a promising candidate for a scalable qubit system [3]. We consider partial Bell state measurements on two spin qubits and argue that it can be performed with available techniques, based on spin-to-charge conversion and charge detection. This opens up the possibility to implement quantum computing without the need of two-qubit gates. \\ \noindent [1] E. Knill, R. Laflamme, and G.J. Milburn, Nature { \bf 409}, 46 (2001). \\ \noindent [2] C.W.J. Beenakker, D.P. DiVincenzo, C. Emary, and M. Kindermann, Phys. Rev. Lett. { \bf 93}, 020501 (2004). \\ \noindent [3] D. Loss and D.P. DiVincenzo, Phys. Rev. A { \bf 57}, 120 (1998).   

Pulsed-gate measurements of the singlet-triplet relaxation time in a two-electron double quantum dot

We use a pulsed-gate technique to measure the singlet-triplet relaxation time in a two-electron double quantum dot when the singlet and triplet states are nearly degenerate. Transitions from the (1,1) to (0,2) charge state involve spin selection rules. Measurements of this transition probability as a function of pulse time and perpendicular magnetic field are used to determine the (1,1) singlet-triplet relaxation time and the (0,2) singlet-triplet splitting. We find a singlet-triplet relaxation time $\geq$70 $\mu$s for our double dot. Experiments aimed at measuring the spin T$_2$ time will be described.   

Enhancement Single Electron Transistor for Quantum Computing

We propose a novel scheme to build single spin quantum dots as the building block for quantum computing. In contrast to the depletion mode single electron transistors (SETs) commonly used for creating quantum dot qubits, our approach is based on an enhancement mode SET using InAs/GaSb composite quantum wells through bandgap engineering. The enhancement mode SETs host no electrons at zero applied voltage, compared to thousands of electrons in depletion dots to start with. When a voltage is applied to a single top metal gate, two symmetric tunneling barriers are created between GaSb and InAs quantum wells. These tunneling barriers define an InAs quantum dot and a single electron can tunnel there. This novel approach has a number of advantages for scalable quantum computing. In this talk, we will discuss the structure design, quantum dot simulation, and device fabrication. We will also present experimental results that provide proof-of-principle demonstrations.   

Anisotropic exchange interactions between electron spins in coupled semiconductor quantum dots

Electron spins in semiconductor quantum dots (QDs) are of considerable interest for qu-bits in quantum computing. Quantum gates (QGs) can be obtained by pulsing the exchange interaction between two spins by external fields. The anisotropic parts of the exchange determine the pulse shapes and also provide a dephasing mechanism affecting the gate fidelity. We have derived an anisotropic term caused by the electron-electron interaction, by treating the Coulomb interaction and the \textbf{k}$\cdot $\textbf{p} band mixing on an equal footing. This term arises from the coupling of spins in addition to single spin effects such as spin-orbit coupling, asymmetry of the confining potential, and inversion asymmetry of the bulk material. The anisotropic contribution obtained here can represent $\sim $ 10 $^{-2 }$ of the isotropic exchange, which has a significant effect on gate shapes and fidelity. We use a general model of elliptically shaped dots in arbitrarily oriented magnetic fields. We give results for vertically coupled InAs/GaAs quantum dots and laterally coupled GaAs electrostatic dots, and we describe the operation of the XOR gates.   

CCD-like Architecture for Quantum Computing on Liquid Helium

Electrons floating above liquid helium form a 2-dimentional gas with high mobilities at low densities due to the absence of a lattice and associated defects and impurities. There are almost no spin-orbit interactions or other magnetic coupling with the surroundings which in turn leads to a long electron spin coherence time. This system therefore lends itself to a quantum computing architecture similar to those proposed for electron spins in semiconductors but with mobile qubits. Scalability requires the ability to shift individual electrons around the surface to either store them in remote locations that can be thought of as memory or to bring pairs close to one another to interact and realize the two-qubit operations. We present a CCD-like architecture for transporting spins above the surface of liquid helium and the ongoing work pertaining to it.   

Imaging Coherent Electron Flow in a Two-Dimensional Electron Gas with a Perpendicular Magnetic Field

Scanning probe microscopy (SPM) at liquid He temperatures can be used to image electron flow from a quantum point contact (QPC) in a GaAs/AlGaAs two-dimensional electron gas (2DEG) in a perpendicular magnetic field. A charged SPM tip creates a perturbation directly beneath it, which scatters electrons incident from the QPC (1). Recording the changes in the QPC conductance as we vary tip position creates an image of the flow, including interference fringes that provide information about the accumulated phase of the electrons. Our new SPM allows us to reach He-3 temperatures and to apply a magnetic field. At zero magnetic field with the QPC biased to the first conductance plateau, the image shows a well-defined branch of electron flow. Introducing a magnetic field of about 50 mT breaks the time-reversal symmetry, by enclosing magnetic flux inside the football-shaped roundtrip path, and the amplitude of the conductance decays in the image at longer distances until the branch is no longer visible. (1) Topinka, M.A., Westervelt, R.M., Heller, E.J. ``Imaging Electron Flow.'' Physics Today 56 (12) 47-52 DEC 2003   

Imaging Electrons in Few-Electron Quantum Dots

Single-electron quantum dots are important candidates for quantum information processing. We have developed a new technique to image electrons inside a single-electron quantum dot in the Coulomb blockade regime, using a scanning probe microscope (SPM) at liquid He temperatures (1). A single-electron quantum dot was formed in a two-dimensional electron gas (2DEG) inside a GaAs/AlGaAs heterostructure by surface gates. Spatial images of an electron inside the dot were obtained by fixing the tip voltage and recording the dot conductance while scanning the SPM tip above the quantum dot. The images show a ring of increased conductance about the center of the dot, where the dot conductance is on the Coulomb blockade conductance peak between 0 and 1 electrons. Simulations show that this technique can be used to extract the wavefunction of electrons inside the dot if the tip perturbation is narrower than the wave function (2). A charged SPM tip promised to be a useful tool for manipulating electrons in quantum dot circuits. 1) P. Fallahi, A.C. Bleszynski, \textit{et al} submitted to Nanoletters. 2) P. Fallahi,\textit{ et al }Proc. 27 Int. Conf. on Physics and Semiconductors (ICPS27), Flagstaff, July 26-30, 2004, in press. \newline *This work was supported at Harvard University by DARPA DAAD19-01-1-0659 and by the NSEC, NSF PHY-01-17795, and at UCSB by QUEST NSF Science and Technology Center.   

Local Gating of Nanostructures with a Low-Temperature Scanned Probe Microscope

We have recently constructed a low-temperature scanned probe microscope designed to operate down to dilution refrigerator temperatures. There has been increasing interest in gaining spatial information about transport through nanostructures by using scanned probe microscopy$^{1-3}$. Our microscope incorporates a commercial stick-slip positioner for coarse approach and alignment to nanostructures. Here we present our solutions to some of the challenges faced in designing such an instrument, as well as a first set of low-temperature scanned probe images obtained from several nanostructures fabricated on different high-mobility GaAs/AlGaAs 2DEG heterostructures. \newline\newline $^{1}$ M.A. Topinka et al., Science \textbf{289}, 2323 (2000). \newline $^{2}$ R. Crook et al., Phys. Rev. Lett.,\textbf{ 91}, 246803 (2003). \newline $^{3}$ A. Pioda et al., Phys. Rev. Lett., \textbf{93}, 216801 (2004).   

Spin Qubit Quantum Computing with RKKY interaction

Motivated by recent theoretical work that demonstrated the importance of the nonlocal coupling between two localized spins via RKKY interaction, we investigate the feasibility of GaAs quantum dot spin-qubit quantum computing scheme utilizing RKKY interactions as a means of spin exchange operation. We estimate the strength and decoherence of RKKY spin exchange with comparison to direct nearest neighbor spin exchange. We also estimate errors associated with RKKY spin interaction and develop schemes of the corresponding error corrections. This work is supported by LPS, ARO, and ARDA.   

Spin Exchange Interaction between Localized and Itinerant Carriers

The exchange interaction between localized and itinerant carriersis a key requirement in indirect spin coupling mechanisms. These interactions have been investigated intensively due to their potential applications in spintronics and in implementations of gates for quantum computation. A quantitative evaluation of spin coupling effects requires an accurate description of this exchange interaction, which has not been available to date. Here we present a microscopic formulation of the spin exchange interaction between localized and itinerant carriers. The effects of correlation and hybridization of continuum and localized states are included by performing a set of nested canonical transformations on an Anderson-type Hamiltonian, bringing it to an s-d spin exchange form. These results are extended to address the problem of spin exchange interaction between a localized electron and an itinerant exciton. This formulation also facilitates the understanding of magnetic properties of Kondo systems. In particular we address the interplay between RKKY and Kondo interactions.   

Session P18: Surfaces and Point Defects in Semiconductor

Improved Simulation of Charged Defect Profiles in Scanning Tunneling Microscopy Images

Scanning tunneling microscopy is often used to study charged defects at semiconductor surfaces. Charge-induced band bending affects the local tunnel current by changing the electronic state density between tip and sample Fermi levels. Imaging is usually carried out under constant current feedback conditions with adjustable tip-sample spacing and a constant tunnel gap voltage. In order to understand the defects, surface profiles through the defect screening regions are often simulated using either standard tunneling theory or a scattering theory approach. These produce qualitative agreement with observations but could be improved by using a more accurate form of the electrostatic potential and by self-consistently calculating the band bending, screening length, and resulting tip response under constant current conditions at each point along the profile. We have done this calculation for the tunneling theory approach. We will describe the details of the calculation and then show the results of applying it to charged features near GaAs(110) surfaces.   

Dimer and NEB studies on Si $I_{4}$ chain and the ground state $I_{4}$

The dimer method explores the energy landscape of two important Si four-interstitials: 1) the ground state and 2) chain extending in the [110] direction. DFT-GGA calculations find the ground state is an extremely stable with 3.3-8.4 eV barrier to five dimer-identified neighboring minima. Those minima are 0.9-1.8 eV above the ground state. In contrast, the chain's most accessible local minimum has 1 eV barrier to the three-interstitial ground state plus a nearby single interstitial with a tiny 0.1 eV barrier to form chain. The tight-binding, nudged-elastic-band pathway from the chain to the four-interstitial ground state has two structurally close local minima with a separation of 4.4 \AA. The chain-to-ground-state barrier is 1.4 eV, suggesting that the tri-interstitial could grab an interstitial to form a four-interstitial chain that could grow into a longer chain without falling into the four interstitial ground state.   

Energy Scaling and Surface Patterning of Halogen-Terminated Si(001) Surfaces

We show that the steric repulsion energies between halogen dimers on a passivated Si(001) surface scale with square of the principle quantum number of the halogen, and arise mostly from bonding with Si substrate. We exemplify the scaling from previously calculated steric interactions of F, Cl, and Br, predict the interactions for I and At, and verify the prediction by density-functional calculations. From the energetics, we explain the patterning of the halogen-terminated Si(001), providing a better understanding of the halogen-roughening process, and predict a crossover to a new vacancy-line defect for large halogens.   

Diffusion Monte Carlo Formation Energies of Silicon Self-Interstitial Defects

Silicon self-interstitial defects can hinder the fabrication of semiconductor devices. Several stable single-, di-, and tri-interstitial clusters found with \emph{ab initio} and tight-binding simulations are believed to form in silicon\footnote{D. A. Richie et al. \emph{Phys. Rev. Lett}. \textbf{92}, 45501 (2004).}. Since experimental detection of self-interstitials remains a challenge, accurate theoretical methods are needed to study their properties. The first Diffusion Monte Carlo (DMC) calculations found single-interstitial defect formation energies to be about 1 eV higher than predicted by density functional theory (DFT)\footnote{W.-K. Leung et al. \emph{Phys. Rev. Lett.} \textbf{83}, 2351 (1999).}. This indicates that DFT may be insufficient for the study of silicon self-intersitials. We confirm the discrepancy between DMC and DFT formation energies for three single-interstitial structures (X, H, and T) and extend the comparison to several di- and tri-interstitial clusters.   

Interactions between native point defects in ZnGeP$_2$.

First-principles calculations of the native point defects $V_{Zn}$, $V_{Ge}$, $Zn_{Ge}$ and $Ge_{Zn}$ show that under Zn-poor conditions, the dominant defects are the $Ge_{Zn}$ and $V_{Zn}$. Since these are respectively a donor and an acceptor, one may expect them to attract each other. The formation of complexes of the type $V_{Zn}-Ge_{Zn}-V_{Zn}$ was studied and found to be favorable.A simple molecular model is proposed for the electronic structure of this complex. Optical excitation of electron paramagnetic resonance (EPR) studies by Gehlhoff et al. [1] were used by these authors to extract energy levels in the gap associated with these defects. The model proposed by these authors assumes that the $Ge_{Zn}$ EPR centrum in irradiated samples becomes activated by a two step process in which an electron from a $V_{Zn}^{2-}$ is optically excited to the conduction band and subsequently trapped at a $Ge_{Zn}^{2+}$ site converting the two defects in EPR active sites $V_{Zn}^-$ and $Ge_{Zn}^+$. We instead propose a direct transition between the two defect states without the intervening conduction band and show that our calculated occupation energy levels agree with such a model. The $V_{Ge}$ on the other hand is found to have a high energy formation and to be unstable towards the formation of a $V_{Zn}$ and a $Zn_{Ge}$ antisite. [1] W. Gehlhoff, R. N. Pereira, D. Azamat, A. Hoffmann, and N. Dietz, Physica B {\bf 308-310}, 1015 (2001).   

Hartree-Fock Cluster Investigation of Locations for Erbium in Silicon

Using the Hartree-Fock Cluster Procedure, we have investigated three locations hexagonal and tetrahedral interstitial (H$_{i}$ and T$_{i})$ and substitutional (S) for Er$^{3+}$ in Silicon including relaxation effects of the Si neighbors of Er$^{3+}$. Our results for the binding energies show that S is the most stable site for Er$^{3+}$, in contrast with the results from the most recent channeling measurements,\footnote{M. B. Huang et al., Appl. Phys. Lett. 81, 2734 (2002)} which can best be explained assuming that Er$^{3+}$ is at T$_{i}$ site. Possible reasons for the difference will be suggested. Magnetic hyperfine fields obtained for the Er nucleus at various sites will be discussed.   

A LSDA+U Study on Bulk CuO: Electronic Structures and Native Defects

We have performed a first-principles study on the electronic structure and the formation of native defects in the monoclinic CuO by using the LSDA+U method with the PAW potentials. The optimized structural parameters of the crystal are in good agreement with the experimental data. The band structure of the crystal is calculated. An indirect band-gap of ~1.0 eV is obtained, which agree with the experimental semiconducting property of the material. This is qualitatively different from a LDA or LSDA calculation, which predicts a metal with the Fermi level below the top of valence band. The formation energies of various native defects as well as their charged states in CuO are carefully studied. The influence of Fermi level and the stoichiometry to the defects formation will be discussed.   

Does the zinc vacancy in ZnGeP$_2$ exhibit a Jahn-Teller distortion?

The Zn-vacancy is one of the dominant defects in ZnGeP$_2$. Its single negative charge state is EPR active. The hyperfine splitting shows that the unpaired electron is primarily localized on a pair of P atoms. In contrast, first-principles 64 atom supercell calculations using both the FP-LMTO and the VASP method of the $V_{Zn}^{-}$ state show that the defect maintains $S_4$ symmetry with the wave function spread equally over 4 P atoms. Here a group-theoretical analysis is presented. When including only the nearest neighbors, the system has $D_{2d}$ symmetry. While the one electron state of the unpaired electron is non-degenerate, a doubly degenerate $e$-state lies only about 10 meV below it. We show that a P-pairing distortion mode splits this $e$-state in two states which are even with respect to one of the mirrorplanes and odd with respect to the other and thus can only contain two of the P-dangling bonds. Calculations in which a pairing of P atoms is enforced while relaxing the remaining atoms confirm this model. Remaining puzzling aspects of this defect will be discussed.   

Study of the stability of Se passivation layers on Si(001) surfaces using time-of-flight positron annihilation induced Auger electron spectroscopy

The stability of Se passivation layers on a Si(001) surface was investigated using a non-destructive surface sensitive technique: Time-of-Flight Positron annihilation induced Auger Electron Spectroscopy (TOF-PAES). After 10 days of exposure in the air, the Se passivation layer was observed to incorporate some oxygen but to remain largely intact. Part of the adsorbed oxygen was removed during annealing at 400\r{ }C in the UHV-environment, however, some oxygen remained on the surface until high temperature annealing at 1030\r{ }C. We posit that that the oxygen that remained after the low temperature anneal was chemisorbed on the Si at defects in the Se passivation layer. The Se passivation layer was stable up to an annealing temperature of $\sim $ 800\r{ }C. The stability of the Se-passivated Si(001) surface is attributed to the saturation of the Si dangling bonds at the surface and to the strong Se-Si bonds.   

Uniaxial stress study of the ro-vibrational transitions of HD in Si

The vibrational spectroscopy of interstitial H$_{2}$ in Si gave rise to a number of perplexing puzzles that concerned the rotational motion of the defect [1]. Most experiments were interpreted in terms of a static defect whereas theory suggested that there should be a very small barrier to rotation. The position and intensity of the HD vibrational line were also anomalous. The key to the solution of these puzzles was the discovery of a new vibrational line for HD and the recognition that certain ro- vibrational transitions are possible for HD that are not possible for the H$_{2}$ or D$_{2}$ homonuclear molecules in Si. H$_{2}$ in Si is a nearly free rotator after all. New experiments have been performed for HD in Si in which IR spectroscopy combined with uniaxial stress has been used to confirm the assignments of the ro-vibrational transitions of HD that underpin our understanding of H$_{2}$ in Si. This work is supported by NSF Grant DMR 0403641. 1. M. Stavola, E E. Chen, W.B. Fowler, G.A. Shi, Physica B \textbf{340-342}, 58 (2003), and references contained therein.   

B couples formation and dissolution in ion implanted Si.

The off-lattice displacement of electrically active, substitutional B in presence of Si interstitials generated by light ion irradiation has been studied by channeling along the $<$100$>$ and $<$110$>$ axes. The channeling yield $\chi $ of B increases with the ion fluence until it saturates at $\chi \approx $ 0.5 suggesting a non-random B displacement. At the saturation B is not electrically active and accurate angular scans indicates the formation of B-B couples aligned along the $<$100$>$ direction in agreement with first principle calculations. The same kind of defect is formed upon B implantation at room temperature as demonstrated also by angular scans with $\chi _{B}\approx $0.5. A peculiar behavior is observed upon annealing: at 800 \r{ }C a significant increase of randomly located B occurs and $\chi _{B}\approx $1, at higher temperatures B recovers progressively into substitional site. The $\chi _{B}$ reaches 0.1 at 950 \r{ }C and the carrier concentration coincides with the amount of substitutional B. The increase of $\chi _{B}$ at 800\r{ }C can be due to the dissolution of B couples and to an intermediate off lattice location of B before to occupy a substitutional site.   

Scanning Probe Study of Donor Layer Charging in a Gallium Arsenide Heterostructure

We use a cryogenic scanning probe technique to study the charging behavior of silicon dopants in a GaAs/AlGaAs heterostructure sample. The sample contains a delta doped layer which is 60 nm below the exposed surface and 20 nm above an underlying two-dimensional electron system. We locally induce charge to enter the donor layer by applying an ac excitation voltage to a sharp metal tip situated a few nanometers above the surface. The resulting image charge appearing on the tip provides a local measure of magnitude of charge entering the donor layer. Here we report measurements as a function of dc voltage and magnetic field.   

Si as an acceptor in (110) GaAs for high mobility p-type heterostructures

We implement metallic layers of Si-doped (110) GaAs as modulation doping in high mobility p-type heterostructures, changing to p-growth conditions for the doping layer alone. The strongly auto-compensated doping is first characterized in bulk samples, identifying the metal-insulator transition density and confirming classic hopping conduction in the insulating regime. To overcome the poor morphology inherent to Si p-type (110) growth, heterostructures are fabricated with only the modulation doping layer grown under p-type conditions. Such heterostructures show a hole mobility of $\mu = 1.75 \times10^5~ \textnormal{cm}^2/\textnormal{V}\textnormal{s}$ at density $p=2.4\times10^{11}~ \textnormal{cm}^{-2}$. We identify the zero field spin-splitting characteristic of p- type heterostructures, but observe a remarkably isotropic mobility and a persistent photoconductivity unusual for p- heterojunctions grown using other doping techniques. This new modulated growth technique is particularly relevant for p-type cleaved-edge overgrowth and for III-V growth chambers where Si is the only dopant.   

Session P19: Spectroscopy of Semiconductor Nanostructures II

Dynamics of the Inter-Landau-Level Magnetoplasmon Coherence in a Quantum Hall system

Collective many-electron behavior of the cold 2DEG in a magnetic field leads to novel Quantum Hall (QH) effects and electronic excitations like the inter-Landau-level magnetoplasmon (MP). Using 3-pulse four-wave mixing (FWM) spectroscopy, we study coherent MP dynamics in a QH system. By delaying the arrival of one pulse relative to the others, the FWM signal shows striking beats for short time delays ($\sim $300fs), followed by a rise ($\sim $10ps) and then a decay ($\sim $200ps). We analyze the experiment in the coherent regime (short time delays) by extending the Dynamics Controlled Truncation Scheme to the case of a strongly correlated ground state. We infer that the beats are due to quantum interference of the MP and magnetoexciton coherences. The decay of the beats gives the decay of the MP coherence. We perform a comprehensive study of these effects as a function of excitation frequency, magnetic field, excitation intensity and temperature.   

Infrared spectroscopy of 2D electron gas in high magnetic field: a case study of graphite

We present the first systematic investigation of the optical constants of HPOG graphite in magnetic fields up to 17T. The ab plane magneto-reflectance in the frequency range 15-3000 cm$^{-1}$ was measured with the field in c-axis. The optical conductivity was obtained from Kramers-Kronig analysis augmented with ellipsometry data. These experiments have allowed us to monitor the field-induced transfer of the electronic spectral weight from the Drude mode to cyclotron resonance (CR) modes. In applied fields, the conductivity in the limit of $\omega \to $0 is depleted by several orders of magnitude in accord with notoriously large positive magneto-resistance of graphite. A close examination of the lineshape of CR modes is indicative of the coexistence of carriers with 3D and 2D character. The latter mode reveals a $\sqrt H $ dependence of the cyclotron frequency long anticipated for Dirac quasiparticles with linear dispersion. In this fashion, our magneto-optics experiments have allowed us to explore novel aspects of charge dynamics in this prototypal quasi-2D material.   

Observation of Optical Gain in InAs Nanocrystals

We developed processes that enable the inclusion of InAs nanocrystals, emitting at 1.55 microns, into a transparent polymer matrix while preserving their optical properties. This provides a flexible platform for integrating the functionality of the nanocrystals into the current photonic circuit technologies. Using three-beam, time-resolved pump-probe measurements, we observed strong evidence of optical gain in a polymer film containing InAs nanocrystals for the first time. We measure the gain dynamics in that system, extracting the gain lifetime and the gain recovery time. These processes and measurements are the first step towards incorporating semiconductor nanocrystals into active devices, such as lasers and amplifiers, on an integrated photonic circuit.   

Cathodoluminescence study of thermal activation of carriers in InAs/GaAs(001) self-assembled quantum dots

We have examined state-filling and thermal activation of carriers in buried InAs/GaAs(001) self-assembled quantum dots (SAQDs) with excitation-dependent cathodoluminescence (CL) imaging and spectroscopy. The dependence of the CL intensity of the ground and various excited state transitions on excitation density was studied. The measurements reveal that carriers escape and are recaptured as excitons or correlated electron-hole pairs. At sufficiently high excitations, state filling and spatial smearing effects are observed together with a sublinear dependence of the CL intensity on excitation. Thermal quenching of the CL intensity of the QD ground and first excited state transitions at low excitations in $\sim $230 to 300 K temperature range is attributed to dissociation of excitons from the QD states into the InAs wetting layer. At high excitations, significantly lower activation energies of the ground and excited states are obtained, suggesting thermal reemission of single holes from QD states into the GaAs matrix is responsible for the observed temperature dependence of the QD luminescence in $\sim $230 to 300 K temperature range.   

Exciton Determination of Strain Parameters in InSb/AlInSb Quantum Wells

Excitons in semiconductors can be used as a tool to probe various material and structural properties. We studied strain-related material parameters in the strained and nonparabolic InSb /AlInSb quantum well system. The strain present in the InSb wells alters the spectrum from that for unstrained systems by introducing a shift in both the heavy and light hole band gaps. With the change of Al concentration in the barrier layers, the strain in the quantum well system can be tuned continuously. Using FTIR spectroscopy, we were able to trace the strain-induced shifts of the exciton energies. The different strain dependence of the light-and heavy-hole band edges allows us to determine deformation potentials a and b simultaneously. Our samples were nominally undoped InSb/AlInSb quantum wells with Al concentration ranging from $5$ to$15\%$ grown by molecular beam epitaxy on GaAs substrates. This work is supported by the National Science Foundation under Grants No. DMR-0080054 and DMR-0209371   

Carrier Relaxation in Self-Assembled ZnTe/ZnSe Quantum Dots

Carrier relaxation in the type II self-assembled ZnTe/ZnSe quantum dots have been studied using ultrafast photoluminescence upconversion technique. We found that PL exhibits fast decay for ZnTe/ZnSe QDs grown in Volmer-Weber mode than that in Stranski-Krastanow mode. The dependence of PL decay time on energy was found only for QDs grown in Stranski-Krastanow mode. We attribute the different behaviors of PL decay for these two modes to the wetting layers.   

Carrier Relaxation in Multi-Stacked InAs/GaAs Quantum Dots

We report the ultrafast time-resolved photoluminescence study of multi-stacked InAs/GaAs quantum dots (MSQD) using the photoluminescence upconversion technique. MSQD with thicknesses of GaAs spacer of 30, 15, and 10 nm were studied to elucidate the dynamics of carrier coupling in both growth and lateral directions. The PL decay time decreases with the thickness of the GaAs spacer. The PL exhibits fast decay as the energy increases. We attribute the energy dependence of PL decay time to carrier tunneling in growth direction. We also found that the carrier tunneling is less effective for GaAs spacer of thickness 30 nm.   

Green's Function Monte Carlo Study of Two-Dimensional Quantum Dots

Two-dimensional quantum dots of interacting electrons in zero magnetic field have been studied with various Monte Carlo methods for some time. To our knowledge, however, Green's function Monte Carlo (GFMC) has not been used. We show that GFMC is a useful method of calculating the ground state of these quantum dot systems. We calculate ground state energies for two and four electrons using a weighted average of a trial wave function's local energy, $(\mbox{\^{H}}\Psi_{T})/\Psi_T$. Our trial wave function $\Psi_T$ uses a two-body (electron-electron) Jastrow function to incorporate correlations between electrons and is variance-minimized. The GFMC method used is an exact-cancellation method which locates the nodes of the true ground state wave function. This method is in principle ``exact,'' outside of a statistical error common to all Monte Carlo methods. The energies obtained are slightly lower than those of earlier diffusion and variational Monte Carlo work.   

Study of the Trion Correlations in Quantum Dots with the Tight Binding Method

The existence of a negatively or positively charged exciton bound state, also known as trion, has been proposed by Lampert in 1958. The binding energy of this complex is only of the order of a meV or less, making the task of observing such state and studying its properties particularly challenging. The importance of studying the properties of trions resides in the fact that this quasi-particle can be manipulated both electrically and optically. Many electro-optical devices utilizing the properties of trions have been proposed, such as a mobile light emitter, a memory device, and even a quantum computer. We study the properties of the trion system by means of the empirical tight-binding method, where the trion Hamiltonian is explicitly described at an atomistic level and the correlation among the constituents is included in the configuration interaction. This approach has been successfully applied to study the binding energy of excitons in quantum dots (QDs). We will extend this previous approach to describe the binding energy of trions in QDs.   

Exciton Confinement in Traps Formed by a Laterally Modulated Gate Voltage

In semiconductor materials, cold gases of bosons can be realized in the system of indirect excitons in coupled quantum well structures. Boson confinement in potential traps improves the critical temperature for Bose-Einstein condensation and allows manipulation of the bosons by varying the trap potential [1,2]. Here, we present in-plane potential traps for indirect excitons, where the traps are formed by a laterally modulated gate voltage. The calculated trap design allows effective exciton confinement as well as in situ manipulation of excitons by the gate electrodes that control the confining potential. The design also ensures that the electric fields caused by the confining potential are well below the threshold for the exciton dissociation. Experiments with indirect excitons in the traps are presented as well. 1. E.A. Cornell, C.E. Wieman, Rev. Mod. Phys. 74, 875 (2002). 2. W. Ketterle, Rev. Mod. Phys. 74, 1131 (2002).   

Condensation of microcavity polaritons with a disorder induced distribution of oscillator strengths

Partly motivated by recent experiments $[1,2]$, a model for condensation of disordered excitons coupled to cavity photons is investigated. The inhomogeneous broadening of the excitonic energies is combined with a distribution of oscillator strengths, allowing for correlations between the exciton energies and their coupling to light. Results are discussed in terms of the mean-field phase diagram, the spectrum of excitations and the polariton photoluminescence. \\[1em] \noindent $[1]$ P.~G.~Lagoudakis {\it et al.}, J.Appl.Phys. {\bf 95}, 2487 (2004)\\ \noindent $[2]$ M. Richard {\it et al.}, J. Phys. C {\bf 16}, S3683 (2004)   

Polariton Condensation with Localised Excitons and Propagating Photons

We calculate the condensation temperature for a model of microcavity polaritons, constructed from localised excitons and propagating photons. This condensation may be described in two different ways. At low densities, it may be considered as B.E.C. of weakly interacting bosons; at high densities a mean-field theory of self-consistent polarisation and optical fields is more appropriate. Considering fluctuations on top of the mean-field theory causes a crossover of the phase boundary from a B.E.C-like $T_c \propto \rho$ at low densities to the mean field theory when $T_c$ reaches the Rabi splitting. Due to the photon component of polaritons this regime occurs at densities much lower than those at which excitons overlap and becomes relevant for current experiments aimed at polariton B.E.C. From the excitation spectrum of fluctuations, one can also predict a number of experimentally accessible signatures which could indicate the presence of a condensate. In particular, we discuss the excitation spectrum, and momentum distribution of emitted photons.   

Condensation of interacting excitons in a microcavity

We consider the ground state of excitons interacting both via their Coulomb forces and via photons in a microcavity. We propose a mean field ansatz for the wavefunction, generalised from [NOZ1], that encapsulates the physics of both high and low densities. We discuss the phase diagram in regimes where the photon- or Coulomb-mediated coupling is dominant, and the excitations have a character that is either excitonic, or polaritonic, or an electron-hole plasma. \newline \newline [NOZ1] Comte, C. and Nozi\`{e}res, P., J. Physique \textbf{43}, 1069--1081 (1982)   

Characterization of Macroscopic Ordering in Exciton Rings

Recently observed complex PL patterns in 2D QW structures exhibit the inner [1,3] and the outer [1-4] exciton rings, localized bright spots [1,3], and the macroscopically ordered exciton state (MOES) [1,3]. The latter appears at the outer ring via its fragmentation into a periodic array of aggregates. While the gross features have been explained within classical framework, attributing the inner rings to nonradiative exciton transport and cooling [1], and the outermost rings and the bright spots to macroscopic charge separation [3,4], the origin of the MOES remains unidentified [5]. Here, for the first time, we report experiments demonstrating the exciton energy modulation over the MOES as well as the phase diagram of MOES in exciton density and temperature coordinates. The experiments shed new light on the dynamical origin of MOES. Besides, we present the studies of dynamical processes within MOES including the observation of aggregate instabilities and bifurcations that point to the spontaneous character of the instability.[1] L.V. Butov, A.C. Gossard, D.S. Chemla, Nature 418, 751 (2002). [2] D. Snoke, S. Denev, Y. Liu, L. Pfeiffer, K. West, Nature 418, 754 (2002). [3] L.V. Butov, L.S. Levitov, A.V. Mintsev, B.D. Simons, A.C. Gossard, D.S. Chemla PRL 92, 117404 (2004). [4] R. Rapaport, G. Chen, D. Snoke, S.H. Simon, L. Pfeiffer, K. West, Y. Liu, S. Denev PRL 92, 117405 (2004). [5] L.S. Levitov, B.D. Simons, L.V. Butov, cond-mat/0403377.   

Spin relaxation of stress-split orthoexcitons in cuprous oxide*

By applying Hertzian strain fields of $\sigma $ = 0 -- 1.1 kbar along (100) direction, we break the cubic crystal symmetry of the cuprite, causing the triply-degenerate orthoexciton to split into a singlet and doublet. The interconversion rate between the orthoexciton singlet and lower-lying doublet is measured at temperatures from 2 to 15 K by using sub-nanosecond time-resolved luminescence. Based on the experimentally observed stress and temperature dependence, we propose that this transition process occurs via TA-phonon scattering associated with a shear tensor field, in contrast to the vector field scattering mechanism for the ortho-para conversion. At high excitation levels, both singlet and doublet transients are well explained by a density-dependent Auger recombination process. *Supported by DOE Grant DEFG02-96ER45439   

Theoretical study of electronic properties of Cu2O and Cu1.5Mn0.5O

Cu$_{2}$O has excellent optical properties and is believed to be a good host to achieve diluted magnetic semiconductors because doped Cu$_{2}$O is a p-type and direct wide bandgap semiconductor. We present electronic structure calculations of Cu$_{2}$O and Cu$_{1.5}$Mn$_{0.5}$O using density functional theory (DFT) within the generalized gradient approximation (GGA). In ground state of Cu$_{2}$O, the calculated lattice constant $a$ = 4.295A and bulk modulus $B_{o}$ = 109GPa are in excellent agreement with experimental values of $a$ = 4.267A and $B_{o}$ = 102GPa. The density of states and band structure show that Cu$_{2}$O has \textit{0.5eV} direct band gap at \textit{$\Gamma $} point. Very interestingly, Cu$_{1.5}$Mn$_{0.5}$O shows half-metal behaviors with \textit{0.6eV} band gap for spin up band at \textit{$\Gamma $} point. The magnetic moment for each manganese atom is about \textit{4$\mu $}$_{B}$, obtained from spin polarized calculation.   

Session P20: Surface Localized States

Chemical Tuning of Metal-Semiconductor Interfaces

We report a study of the Schottky barrier for Pb films grown on Si surfaces terminated by various metals (Ag, In, Au, and Pb) to explore the atomic-scale physics of the interface barrier and a means to control the barrier height. Electronic confinement by the Schottky barrier results in quantum well states in the Pb films, which are measured by angle-resolved photoemission. The barrier height is determined from the atomic-layer-resolved energy levels and the line widths. A calculation based on the known interface chemistry and the electronegativity yields predicted barrier heights in good agreement with the experiment.   

Band structure calculations of the surface linear optical response of the clean and hydrogenated Si(100) surface

We calculate the reflectance anisotropy (RA) and the reflectance difference (RD) spectra for a clean Si(100) surface and two H covered Si(100) surfaces. The clean surface we consider is a $2\times 1$ surface reconstruction characterized by a tilted dimer formed between the two top-most Si atoms. One of the H covered surfaces is a monohydride surface in which the two dangling bonds of the dimer are H saturated (rendering a flat dimer), and the other is a dihydride surface in which the H saturates each of the two dangling bonds leading to a bulk ideally terminated surface. This dihydride surface is thought to be created if enough H is added to the surface. The optical response is calculated both with pseudopotential and all-electron LAPW band structures including a scissors shift, and compared. In the pseudopotential we neglect the non-local contribution to the momentum matrix elements. We contrast the two methods and trace their differences to the physics involved in each one. Finally, we compare our results with available experimental measurements on this surface.\footnote{Yves Borensztein, private communication}   

Effects of one-dimensional multiband electronic structure on indirect interaction between Pb atoms adsorbed on the In(4x1)-Si(111) surface

As predicted theoretically indirect interactions between adatoms on a metal surface can be mediated by the exchange of electrons trough the conduction band of the substrate. The long-range oscillating character of these interactions is attributed to the Friedel oscillations of the conducting electron density which screen adatoms and lattice distortions caused by adatoms. The period of the oscillations is determined by the Fermi wave-length. We have studied the growth of Pb submonolayer films on the quasi-one-dimensional In(4x1)-Si(111) which as known from spectroscopic experiments has a triple-band metallic electronic structure. A novel ordered phase of Pb is discovered with a non-primitive unit cell. Islands of this phase form well below its stoichiometric concentration ($\theta $=6/32ML). The period of the structure is 8 times the substrate lattice constant $a$ and its unit cell includes three nonequivalent Pb dimers. The average distance between the dimers is 2.66$a$.The origin of this structure is the long-range indirect interaction between Pb atoms and the observed periodicity is explained by the contribution of the three surface electronic bands with different Fermi wave-lengths.   

Unusual Band Dispersion in Pb Films on Si(111)

Uncommon effects are observed in thin films, in part because the influence of the substrate is measurable. Atomically uniform Pb films are prepared on Si(111), and the quantum well states corresponding to confined valence electrons in the film are probed by angle-resolved photoemission. The subband structure shows a free-electron-like dispersion near the zone center, but the dispersion turns sharply toward higher binding energies at larger in-plane momenta. The effective mass at the zone center shows large variations of up to a factor of seven, and in one instance, the sign of the effective mass is reversed. These anomalous results are explained in terms of the bulk band structures of Pb and Si and an anticrossing coupling.   

Bimodal electronic structure of isolated Co atoms on Pt(111)

Co atoms deposited onto Pt(111) have been found to have a giant magnetic anisotropy energy ($\sim $9meV), and therefore constitute a promising system for nanomagnetism and quantum information based applications [1]. We have used low temperature (5K) scanning tunneling microscopy and spectroscopy to probe the local electronic structure of Co adatoms on the Pt(111) surface. We observe two varieties of Co adatoms (with equal probability) that differ in their dI/dV spectra at energies around 80meV below the Fermi energy. We find that this contrast in spectral density depends on the binding site of the Co adatoms. Atoms at different surface lattice sites (i.e., fcc versus hcp sites) exhibit different local density of states (LDOS). Manipulation of a Co atom from one kind of lattice site to the other results in the expected change in the LDOS behavior. dI/dV spectra measured in this study were normalized using a new method that compensates for differences in tunneling current that occur at different surface sites. Such normalized spectra are quite useful in predicting the spatial contrast of dI/dV maps. [1] P. Gambardella \textit{et al.}, \textit{Science} \textbf{300}, 1130 (2003).   

Breakup of Quasiparticles in Thin-Film Quantum Wells

Quantum well states in thin films are commonly described in terms of a quasiparticle confined in a quantum box, but this single-particle picture can fail dramatically near a substrate band edge, as shown by this angle-resolved photoemission study. Atomically uniform Ag films are prepared on Ge(111) to facilitate accurate line shape and dispersion relation measurements. A quantum well peak is observed to split into two peaks near the Ge valence band edge. The unusual line shapes are shown to be due to many-body interactions and are quantitatively explained by a Green's function calculation.   

Correlation between Electron Reflectivity and Quantum Bound States Observed by Scanning Tunneling Spectroscopy on Ag Thin Film

The transmission coefficient of a metal film for free electrons at low energy often reveals peaks that are associated with quantum well (QW) resonances above the vacuum level. We have observed that QW resonance peaks can also be observed on Ag films grown Si(111)7$\times $7 with scanning tunneling spectroscopy (STS). It indicates the transmission density of states probed by STS is equivalent to the transmission coefficient in free space. This equivalence is no longer valid for the reflectivity of the free electron. The distinct quantum bound states (QBS), which do not appear in the reflectivity spectra of free electron, are observed by STS. The total density of states of each QBS is equal to a total electron reflectivity in the energy distribution of the QBS, manifesting a correlation between electron reflectivity and QBS.   

Unoccupied Metallic Quantum Well States and CO Adsorption on Ni/Cu(100)

When ultrathin metal films are grown on metal surfaces, reflection at the interface and at the film surface give rise to so-called metallic quantum well (MQW) states in the valence levels. The Cu/Ni/Cu(100) system is anomalous in that occupied MQW states in the Cu overlayer disperse upwards with increasing overlayer thickness, while the Cu-induced features seen in inverse photoemission (IPE) disperse downward. To better understand the origin of this phenomenon, we performed an IPE study of the Ni/Cu(100) and CO/Ni/Cu(100) systems as a function of Ni thickness. For thin Ni layers, the film grows pseudomorphically, while for thicker layers the Ni relaxes to its bulk lattice parameter. IPE spectra from Ni films up to 10 ML exhibit two features that increase in energy with increasing Ni thickness. CO adsorption strongly modifies the spectrum indicating that one feature is a Ni surface resonance while the other is a state confined to the Ni/Cu interface. In the range from 10 ML to 20 ML, the spectra exhibit a single strong feature, similar to Ni(100) surface but at a higher energy, which also appears dominated by the surface resonance. Above 20 ML the energy of the feature returns to the value for Ni(100) and the spectrum is only weakly affected by CO. Our results suggest that the Ni surface resonance and the Cu/Ni interface state play key roles in the anomalous dispersion of the unoccupied states in the Cu/Ni/Cu(100) system.   

Electronic Structure and Dynamics of Quantum-Well States in thin Yb-Metal Films

By low-temperature scanning tunneling spectroscopy, we have studied quantum-well states above the Fermi energy in thin Yb(111)- metal films deposited on a W(110) single crystal. These states are laterally highly localized and give rise to sharp peaks in the tunneling spectra. Due to the high lateral resolution of STS, the quantum-well states and their film-thickness dependence can be observed on rather rough films with variations of the local thickness over a range of several monolayers. A quantitative analysis of the spectra yields the bulk-band dispersion in $\Gamma - \mbox{L}$ direction as well as quasi-particle lifetimes. The quadratic energy dependence of the lifetimes is in quantitative agreement with Fermi-liquid theory. cond-mat/0411580.   

Local potential profile and confined states on the Si(111)- $\sqrt 3$ x $\sqrt 3$-Ag surface by scanning tunneling spectroscopy

It has been known that on the Ag-covered Si surface forms two-dimensional metallic states, like the cases of (111) surface of the noble metals, in the band gap of the semiconductor substrate. Confined potential around defects and steps on the surface might, therefore, create 0D and 1D electronic states, respectively. We investigated in two-dimensional tunneling spectroscopic measurements using low-temperature (6K) STM how the electrical potential changes and what kinds of electronic states are formed around these sites. Our STS measurements reveal that the steps and Ag-adatom induced defects lower the potential around them. It is found that the potential variation around step edges spreads in a range of a few nm and can be fitted well with a 2D Thomas-Fermi screened potential. These can be explained with a model of positive charging at the step edges. Electronic states related with the varied potential were also observed and their details will be discussed in the presentation.   

Positron Trapping at Quantum-Dot Like Particles on Metal and Semiconductor Surfaces

The results of studies of sputtered surfaces of the Fe-Cu alloy with quantum-dot like Cu nano-particles embedded in Fe and submonolayer films of Au and Pd deposited on Cu(100) and Si(100) using Positron-Annihilation-Induced Auger-Electron Spectroscopy are analyzed by performing quantum mechanical calculations of positron surface states and annihilation characteristics. Estimates of the positron binding energy, work function and annihilation characteristics performed for studied surfaces reveal their strong sensitivity to nano-particle coverage. Trapping of positrons at nano-particles on studied surfaces is determined from calculated positron surface state wave functions and comparison of theoretical core annihilation probabilities with experimental ones estimated from the measured Auger peak intensities.   

Effect of terrace width on site selectivity of C and S adsorption on stepped Pd surfaces

We have performed \textit{ab initio} density functional theory calculations of C and S adsorption on stepped Pd surfaces to extract the role of terrace width on adsorption characteristics. Previous calculations$^{1}$ show that both C and S prefer to sit on hollow site between step edge and corner atom on Pd(211), although C atoms penetrate deeper into the surface and form much stronger bond with the neighboring Pd atoms. We present results here for C and S on Pd(533) which consist of 4-atoms wide terraces instead of 3 atoms on Pd(211). The trend in adsorption energies for C and S and the nature of the local bonding with the Pd atoms will be presented in detail along with the comparison between Pd(533) and Pd(211). Implications will be drawn for the relevance of the results to those for the Pd nanoparticles. This work is supported in part by NSF Grant No. CHE0205064, and by DOE under Grant No. DE-FGO3-03ER15445. $^{1}$S. Stolbov, F. Mehmood, T. S. Rahman, M. Alatalo, I. Makkonen, and P. Salo, Phys. Rev. B \textbf{70}, 155410 (2004).   

On mechanism of influence of alkali adsorbates on the interaction of co-adsorbed molecules with metal surface

It is well known that adsorption of alkali atoms on metal surfaces dramatically changes their properties. Alkali adsorbates promote various chemical reactions on catalyst surfaces, form quantum wells at the Cu(111) surface, and substantially reduce vibration frequencies of co-adsorbed molecules. To understand the nature of these effects, we study the electronic structure of Cu and Pd surfaces adsorbed with alkalis and reveal groove-like or plateau-like features of surface potential formed upon the adsorption. These features are found to produce high density low-energy electronic excitations strongly delocalized toward vacuum with respect to the ground states. We find this phenomenon to be responsible for strong enhancement of the electronic polarizability in the vicinity of the surface that leads to dramatic softening of the vibration frequencies of co-adsorbed molecules. We also discuss the connection of this phenomenon to the promotion of surface reactivity and unusual optical properties of quantum wells formed at Cu(111) upon alkali adsorption.   

Contribution of Vibrational Dynamics to Adatom Diffusion on Metal Surfaces

We have calculated the vibrational dynamics and thermodynamics for a series of scenarios where an adatom is adsorbed of flat and stepped surfaces for both cases of hollow site and saddle-point adsorption. We have used the embedded atom method for the interatomic potential for Cu. The local vibrational densities of states were calculated using real space Green's function formalism and the thermodynamical functions were evaluated using the harmonic approximation. Activation energies for the static systems show a strong anisotropy for Cu(110) and near a step. For Cu(110), for example, we find activation energies of 0.230 and 1.146 eV for diffusion along and perpendicular to the open channel respectively. The change in the vibrational free energy for these two cases was found to be about 50 meV which represents about 20{\%} for the first case and less than 5{\%} for the second. Trends extracted from our detailed study on the contribution of the vibrational dynamics to the diffusion prefactors will be presented and comparisons with experimental data will be made when available.   

Computational studies on small silicon clusters deposited on a graphite substrate

The structural and electronic properties of small silicon clusters adsorbed on a graphite (0001) substrate are studied by Density Functional Theory (DFT) adopting periodic boundary conditions. Monolayer coverage of single Si atoms on the graphite surface is considered as well as the adsorption of Si$_ {n}$ clusters with n = 2, 3. By virtue of covalent bond formation between Si and C, the adsorption of Si$_{n}$ clusters deposited on graphite turns out to be distinctly more stable than that of alkali metal clusters $^{2}$ used as a standard for comparison in the present work. The structure and electronic properties of the Si$_{3}$ cluster are strongly affected by the interaction between adsorbate and substrate. As shown by Density of States (DOS) analysis, the energy gap of the free Si$_{3}$ cluster shrinks due to the influence of the graphite surface. The resulting energy gap of about 0.4 eV is within the experimental range $^{3}$. \\[4pt]$^{2}$K. Rytk\"onen, J. Akola, and M Manninen, Phys. Rev. B {\bf 69}, 205404 (2004); $^{3}$B. Marsen, M. Lonfat, P.Scheier, and K Sattler, Phys. Rev. B {\bf 62}, 6892 (2000).   

Session P21: Nanotubes, Nanowire and Nanoparticles in Biology

Electrical Detection of ssDNA by Single Wall Carbon Nanotubes

We report conductance measurements of single-walled carbon nanotubes (SWNT) in the presense of single-stranded DNA (ssDNA). The characteristic I-V curves of our metallic SWNT samples changed from linear (ohmic) to non-ohmic in the presence of ssDNA dissolved in DI water, and remained so when the sample was dried. The results imply possible applications of SWNT in the electronic detection of ssDNA, detection of hybridization of ssDNA, and sequencing of DNA.   

Sturdier DNA nanotubes via ligation

Self-assembly of DNA nanotubes from double crossover tiles results in cylindrical lattices of tiles joined by overlapping sticky ends. We show that nicks formed at overlapping sticky ends can be successfully ligated in the tube geometry, resulting in increased thermal and mechanical stability of the nanotubes. We compare the melting temperature and persistence length of ligated and unligated nanotubes.   

Interaction between proteins and carbon nanotubes field effect transistors.

The interaction between proteins and carbon nanotube network field effect transistors in a biological buffer environment has been investigated. In general, chemical or biological species can affect conduction through the network either by charge transfer or by introduction of a scattering potential on the nanotubes. A method employing the real-time analysis of transistor transfer characteristics allows us to distinguish between these two effects. Based on the available experimental data, we argue that the mechanism by which proteins influence carbon nanotubes is the charge transfer from --NH$_{2}$ groups.   

Self-Assembled Gold Nanowires from Nanoparticles

We present a new technique for fabricating gold nanowires using carbon nanotubes as the template. By applying an ac voltage to an electrically contacted single walled carbon nanotube, we generate highly non-uniform ac electric fields in the vicinity of the nanotubes. These ac electric fields serve to polarize 2 nm gold nanoparticles dispersed in solution. The induced dipole moment in the nanoparticles is attracted to the high-intensity field regions at the surface of the nanotube, thus causing a gold nanowire to grow on the surface of the nanotube[1]. Interestingly, we find gold nanowires grow even on nanotubes that are not electrically contacted but in close proximity to the electrodes. This process is also visualized using fluorescently labeled nanoparticles. Future applications of this work include DNA sensors based on functionalized Au nanoparticles. Additional work on the electronic trapping of proteins and DNA[2] using AC electric fields will also be presented. [1] L. Zheng, S. Li, J. Brody, and P. J. Burke, ``Manipulating nanoparticles in solution with electrically contacted nanotubes using dielectrophoresis'', Langmuir, 20, 8612-8619, 2004; [2] Zheng, L.; Brody, J. P.; Burke, P. J. "Electronic Manipulation of DNA, Proteins, and Nanoparticles for Potential Circuit Assembly", Biosens. Bioelectron., 20, 606-619 (2004).   

Accurate and efficient modeling for carbon nanotubes in biological applications

Carbon nanotubes (CNTs) hold great promise for applications in biomedicine and biotechnology, in particular, as biosensors. For such applications, it is essential to understand the interaction of CNTs and water and/or other biomolecules in the aqueous environment. In this regard, the short-ranged van der Waals interaction together with the Coulomb interaction arising from atomic partial charges and dielectrically induced charges on the CNT play an important role. We have developed an accurate, yet computationally efficient, empirical method to model the electrostatics of finite-length single-walled armchair CNTs. Atomic partial charges are fitted to electrostatic potentials computed at a B3LYP/6-31G* level of density functional theory. The dielectric properties are calculated self-consistently from a third-nearest-neighbor tight-binding Hamiltonian, and are found to be in good agreement with density functional theory results. We demonstrate our description for water transport through a finite-length CNT channel. The atomic partial charges on the edges are found to greatly contribute to the total interaction energy and may influence water entering the CNT, while the polarizability of the CNT significantly lowers the electrostatic energy in the tube center.   

Importance of specific magnetic moment and size monodispersity of magnetic nanoparticles for biomedical applications

Magnetic nanoparticles with suitable biocompatible coatings are becoming increasingly important recentlt in biomedical applications. In most cases people just use nanoparticles but don't pay much attention to their magnetic properties and size effects, which could improve greatly the applications. There are very few publications dealing with underlying physics and discussing how the magnetic properties and size distribution of nanoparticles influence the applications. Most magnetic particles or beads used currently are based on ferromagnetic iron oxides with low magnetic moment and large size distribution. In this paper we will discuss the important role of high magnetic moment and monodispersity of magnetic nanoparticles from physics point review. Physics of hyperthermia treatments of nanoparticles, and biomolecule detection using Brownian rotation dynamics will be discussed in detail. As an example we will show how we produce monodispersive core-shell iron nanoparticles with ultrahigh magnetic moment and the significant results in biomedical applications.   

Incorporation and release of materials into/from nanohorns used as drug delivery systems

We have established methods of incorporating organic materials into nanohorns in the liquid phase at room temperature in a quality-controlled manner. The methods are classified into two types: quasi-equilibrium or non-equilibrium. We can choose a suitable type depending on the affinity between guest molecules and nanohorns. When nanohorns are used as drug carriers in drug delivery systems, the incorporated materials must be released, ideally over controllable period of time. We found that the release process included fast components that release 50--70{\%} of the materials quickly as a result of the weak binding among the molecules in the central region inside the hollow spaces of the nanohorns. We also found methods of slowing down the quick release, which should improve the applicability of nanohorns as drug carriers in drug delivery systems.   

Drug-loaded single-wall carbon nanohorns: adsorption and release of dexamethasone in vitro

Oxidized single-wall carbon nanohorns (SWNHox) are spherical aggregates (80-100 nm) of elongated graphitic tubes and have holes of $<$2 nm on the surface that enable incorporation of small guest molecules into them. Here we evaluate the \textit{in vitro} capacity of SWNHox to adsorb and release an anti-inflammatory drug, dexamethasone (DEX). The total amount of DEX adsorbed to SWNHox was determined to be 200 mg/g using [$^{3}$H]-labeled DEX. Kinetic analysis showed DEX slowly released from the complexes into PBS: about a half of DEX adsorbed was released in two weeks. Experiments with mammalian cells indicated that functional DEX was released from the complex. The results obtained here showed that SWNHox is an attractive candidate for a controlled drug carrier.   

Cell Structure Assembly with Magnetic Nanowires

The ability to precisely position and organize mammalian cells plays a significant role in numerous biological applications, including the assembly of multi-dimensional cell structures. Ferromagnetic nanowires, used in conjunction with patterned micromagnets, are shown to provide a highly effective tool for cell manipulation. The nanowires are fabricated by electrodeposition, allowing for precise control of their dimensions and magnetic properties. Their high aspect ratio and large remanent magnetization allow suspensions of cells bound to nanowires to be controlled with low magnetic fields through the alignment of the wires' moments. We have shown that these characteristics enable self-assembly and patterning of 3T3 mouse fibroblast cells with Ni nanowires in one and two dimensions in the presence of an applied horizontal field. Vertical orientation of the applied field promotes the creation of vertical cell columns which can be organized to form more complex structures. The geometry of these structures is controlled through selection of different micromagnet patterns, and the resulting three-dimensional structures can be made permanent by stabilization in hydrogel. These methods are potentially useful in the synthesis of engineered tissues.   

Imaging of DNA/Nanosphere Condensates

DNA forms condensates in a variety of environments. In chromatin, DNA is condensed around 10-nm-diameter, positively-charged histone complexes. To model chromatin formation in cells, lambda-phage (16 microns long) and herring sperm (0.03 to1 micron) DNAs were mixed with polystyrene nanospheres of diameter 40nm and 930nm containing 1.8x10$^{4}$ and 2.6x10$^{8}$ positive surface charges, respectively, to form condensates. Sphere concentrations were 1-2 times the isoelectric concentration. Condensation vs time was imaged at various concentrations, pH's, viscosities, and ionic strengths. Bright-field and fluorescence (YOYO-1 dye bound to DNA) images were recorded. In general HS DNA aggregate size increased with time. Except in 0.5-0.8 M KCl, herring sperm DNA formed one huge aggregate (100's of microns) and depleted other areas, both in 10{\%} and 20{\%} glycerol. Phage DNA samples rapidly formed longer, fiber-like aggregates. Within 2 hours it formed ordered structures and in most samples, empty, apparently depleted regions were found in the viewing area. Shapes of the phage-DNA aggregates in 20{\%} glycerol, in contrast, formed small clumps like HS DNA.   

Theory of Excluded Volume Effects in Tethered Particle Experiments

Tethered particle motion is a class of single molecule experimental techniques that has been used to explore numerous macromolecular properties including the motion of kinesin and RNA polymerase; proteins synthesis of DNA and protein mediated loop formation in DNA. In this experimental technique an imagining bead is attached to the macromolecule of interest. Despite the diversity of experimental studies theoretical work regarding the relationship between the motion of the bead and that of the molecule is lacking. In this work we present a theoretical analysis of tethered particle motion. Our theoretical analysis reveals that the nature of the experimental protocol gives rise to a volume exclusion effect resulting in an effective force acting on the molecule. This effective force causes the molecule to swell changing its statistical properties and in some cases it biological functionality. Statistical properties of the tethered bead (experimentally measurable) are then related to that of the molecule (not observed). We then apply this theory to the analysis of dynamical macromolecular interactions. In particular, we demonstrate how the rate of loop formation of dsDNA generated by the lac-repressor protein is decreased due to the volume exclusion effect. Finally, we apply this theory to the analysis of protein digestion of DNA, revealing a simple manner in which tethered particle motion can be used to the study of such interactions.   

Session P23: Focus Session: Biological Hydrodynamics II

Self-organization of hydrodynamically entrained sperm cells into an array of vortices

The emergence of spatiotemporal patterns is of great interest in many scientific disciplines. Here we report a new dynamically self-organized pattern formed by hydrodynamically entrained sperm cells at planar surfaces. The sperm cells form vortices resembling quantized rotating waves. These vortices form an array with local hexagonal order. Using a novel order parameter, we show that the array is only formed above a critical sperm density. Supported by numerical simulation we suggest a mechanism for the appearance of the array and we estimate the strength of the hydrodynamic coupling between the cells. The vortex array represents a new chiral active gel and may serve as an experimentally accessible model for the metachronal wave of ciliated epithelia and other non-equilibrium phenomena in general. Finally we discuss the biological implications of our work.   

Individual and collective dynamics of gyrotactic algae

When a swimming microorganism has an anistropic distribution of mass, rotational drag due to shear can strongly affect the orientation of its swimming trajectories. {\it Gyrotaxis} is motility guided by this combination of torques. An important example is provided by upward swimming of slightly negatively buoyant organisms, which results in unstable density stratification in the fluid, then descending curtains that become plumes and bioconvection patterns. The details of these self-concentrative dynamics are determined by gyrotaxis and the associated flows. We report novel experiments on the development of these instabilities, their eventual settling into steady states or, occasionally, secondary dynamical instabilities. Details of the collectively generated flow fields were obtained using PIV, Particle Imaging Velocimetry. The implications of self-generated flow fields with large Peclet numbers are discussed.   

Tracking the bacterial dynamics in three dimensions

We present experimental results on the tracking of swimming {\it Escherichia coli} cells using a novel 3D particle tracking method. {\it E.coli} is a single cell organism, it is about 1-3 $\mu$m in size. Under favorable condition, it can run at a speed of ~30 times of its body length in one second. Swimming {\it E.coli} cells provide us an unique opportunity to probe the transport properties of a nonequilibrium system, where it consists of self propelled objects. In our experiments, a wildtype {\it E.coli} (RP437) is used for its known motile behavior. The cells are made fluorescent by the insertion of a plasmid that expresses GFP (Green Fluorescent Proteins) constitutively. As a result, the cells emit fluorescent light all the time. The cells are placed in a 5mm diameter and 1.5 mm depth well before filming. The 3D trajectories of multiple swimming {\it E.coli} cells are obtained for the first time using a novel defocused particle tracking technique. Various types of locomotion of bacteria are observed, running, tumbling, and wobbling. Using the track data, we evaluated the diffusion coefficient of the swimming cells. It demonstrated a ballistic behavior at the early time, and gradually develop into a random walk in later time. The Diffusion coefficient is about $10^{3}$ orders of magnitude larger than a system with nonmotile microorganisms of similar size.   

A Computational Model of Deformable Cell Rolling in Shear Flow

Selectin-mediated rolling of polymorphonuclear leukocytes (PMNs) on activated endothelium is critical to their recruitment to sites of inflammation. The cell rolling velocity is influenced by bond interactions on the molecular scale that oppose hydrodynamic forces at the mesoscale. Recent studies have shown that PMN rolling velocity on selectin-coated surfaces in shear flow is significantly slower compared to that of microspheres bearing a similar density of selectin ligands. To investigate whether cell deformability is responsible for these differences, we developed a 3-D computational model which simulates rolling of a deformable cell on a selectin-coated surface under shear flow with a stochastic description of receptor-ligand bond interaction. We observed that rolling velocity increases with increasing membrane stiffness and this effect is larger at high shear rates. The average bond lifetime, number of receptor-ligand bonds and the cell-substrate contact area decreased with increasing membrane stiffness. This study shows that cellular properties along with the kinetics of selectin-ligand interactions affect leukocyte rolling on selectin-coated surfaces.   

Pressure-driven polymer dynamics in nanofluidic channels

The pressure-driven transport of different lengths of DNA molecules in flat, rectangular nanofluidic channels was studied using fluorescence microscopy. The molecular speeds were observed to fall between the maximum fluid velocity in the parabolic flow profile and the average fluid velocity. The dependence of polymer speed on channel height was characterized by two distinct transport regimes: In channels larger than $\sim $1 $\mu $m, 21 $\mu $m-long molecules traveled faster than 3.8 mm-long molecules. In channels smaller than $\sim $1 $\mu $m, the observed speeds coincided. This behavior reflects the dynamical properties of polymer coils in solution, whose statistical size is characterized by the radius of gyration, R$_{g}$. In large channels, DNA coils can explore all regions of the channel cross section up to a distance $\sim $R$_{g}$ from the walls. The center of mass of large molecules is therefore confined to regions of higher fluid velocity than small molecules, and travel faster in a pressure-driven flow. In thin channels, molecular conformations are confined in height, leading to cross-sectional profiles that are independent of length, and that travel at the same speed.   

Polymer dynamics and fluid flow in actin-based cell motility

In living cells, nonequilibrium protein polymerization reactions are frequently used to convert chemical energy into mechanical energy and thereby generate useful force for cellular movements. We have examined the polymer and fluid dynamics in two biological cases where the assembly of branched actin filament networks generates force: the intracellular movement of the bacterial pathogen \textit{Listeria monocytogenes}, and the extension of the leading edge of skin epithelial cells during wound-healing. In both cases, net actin filament assembly occurs at the front of the network structure and net disassembly occurs at the rear. Actin protein subunits and other network components must be recycled through the fluid phase to the front of the polymerizing network in order for forward movement to continue at steady state. For actin-based movement of \textit{Listeria monocytogenes}, we have found that actin recycling is not rate-limiting; instead, the speed of movement is governed by the cooperative dissociation of groups of noncovalent protein-protein bonds attaching the filamentous network to the bacterial surface. In contrast, rapid actin-based extension at the leading edge of moving epithelial cells is associated with unusual perturbations in intracellular fluid flow.   

Formation of 2D ligand-receptor bonds under shear

Formation and dissociation of specific molecular links between opposing surfaces are omnipresent events of the living world. They very often take place in highly dynamic conditions like in blood stream where wall shear rates are believed to vary from 150 to 1600 s$^{-1}$. Thus, a better understanding of shear effects on 2D ligand-receptor bonds formation is a key step towards improving our conception of biological molecular recognition. Using streptavidin-biotin as a model receptor-ligand pair, we have probed the establishment of specific bonds between the surface of a B-lymphocytes and grafted micrometric particles under controlled shear stress. The results showed that shear stress had a determining effect on cell-particle interactions, introducing a ligand surface density condition for the binding and very likely cell mechanical compliance leading to increased binding at high shear. This view will be discussed in the light of other results that we have obtained using a purely colloidal experimental model, made of particles grafted either with streptavidin or biotin and brought into contact under controlled shear stress.   

Swimming, Stirring, and Hydrodynamic Scaling in the {\it Volvocales}

The {\it Volvocales} constitute a family of colonial algae ranging in size from tens to hundreds of microns. The surface of these nearly spherical algae is packed with cells, each having two flagella. Their incessant, periodic flailing moves the water in which the organisms live. The {\it Volvocales} and especially their largest member {\it Volvox} constitute a model system to study the coordinated action of multiple flagella in self-propulsion and the fluid mixing they produce. Using flow visualization and micromanipulation we have measured the swimming speed, fluid velocities, and propulsive forces for {\it Volvocales} over several orders of magnitude in organism size. The associated Peclet number varies from $\sim 0.01$ for the smallest species to $\sim 100$ for the largest, spanning the range from diffusion-dominated to advection-dominated transport of vital molecules dissolved in the suspending medium. Over this same range we quantify scaling relations for swimming speed and propulsive efficiency. These results are interpreted in terms of metabolic and physical tradeoffs.   

The effect of solution conditions on the conformation of clathrin triskelions

The major component in the protein coat of certain endocytic vesicles is clathrin, a three-legged heteropolymer (known as a ``triskelion'') that assembles into polyhedral cages composed primarily of pentagonal and hexagonal facets. In vitro, this assembly depends on the pH, with cages forming more readily at low pH and less readily at high pH. We have developed novel techniques to make physical measurements of the clathrin triskelion under conditions where assembly occurs. By sedimentation velocity and laser light scattering, we measure changes in Stokes radius, r$_{H}$ and radius of gyration, r$_{g}$ of the clathrin triskelion as the pH is lowered. Calculations, with the program HYDRO, on a rigid molecular bead model of clathrin show that measured changes may be accounted for by a pH dependent puckering of the arms at the vertex. This is consistent with the idea that a change in clathrin conformation may play a role in clathrin cage assembly.   

Lizard locomotion on weak sand

Terrestrial animal locomotion in the natural world can involve complex foot-ground interaction; for example, running on sand probes the solid and fluid behaviors of the medium. We study locomotion of desert-dwelling lizard {\it Callisaurus draconoides} (length 16 cm, mass=20 g) during rapid running on sand. To explore the role of foot-ground interaction on locomotion, we study the impact of flat disks ($\approx$ 2 cm diameter, 10 grams) into a deep (800 particle diameters) bed of $250~\mu m$ glass spheres of fixed volume fraction $\phi \approx 0.59$, and use a vertical flow of air (a fluidized bed) to change the material properties of the medium. A constant flow $Q$ below the onset of bed fluidization weakens the solid: at fixed $\phi$ the penetration depth and time of a disk increases with increasing $Q$. We measure the average speed, foot impact depth, and foot contact time as a function of material strength. The animal maintains constant penetration time (30 msec) and high speed (1.4 m/sec) even when foot penetration depth varies as we manipulate material strength. The animals compensate for decreasing propulsion by increasing stride frequency.   

Choice of High-Efficacy Strains for the Annual Influenza Vaccine

We introduce a model of protein evolution to explain limitations in the immune system response to vaccination and disease [1]. The phenomenon of original antigenic sin, wherein vaccination creates memory sequences that can increase susceptibility to future exposures to the same disease, is explained as stemming from localization of the immune system response in antibody sequence space. This localization is a result of the roughness in sequence space of the evolved antibody affinity constant for antigen and is observed for diseases with high year-to-year mutation rates, such as influenza. We show that the order parameter within this theory correlates well with efficacies of the H3N2 influenza A component of the annual vaccine between 1971 and 2004 [2,3]. This new measure of antigenic distance predicts vaccine efficacy significantly more accurately than do current state-of-the-art phylogenetic sequence analyses or ferret antisera inhibition assays. We discuss how this new measure of antigenic distance may be used in the context of annual influenza vaccine design and monitoring of vaccine efficacy. 1) M. W. Deem and H. Y. Lee, Phys. Rev. Lett. 91 (2003) 068101. 2) E. T. Munoz and M. W. Deem,q-bio.BM/0408016. 3) V. Gupta, D. J. Earl, and M. W. Deem, ``Choice of High-Efficacy Strains for the Annual Influenza Vaccine,'' submitted.   

Hydrodynamic Forcing of Spontaneous Helical Growth

Inspired by the biological system of actin ``comet tails'' formed during bacterial motility, we have modelled the natural evolution of a cylindrical gel growing from a fixed object. We have found that a stable solution, is a kind of helical growth in which the helix pitch and diameter increase, and the pitch angle varies, but the axis remains constant. The growth can be defined by a rotation vector $\vec k$. The instantaneous magnitude of this vector is related to the intantaneous pitch and diameter, and is a simple inverse linear function of the contour length, $s$, along the helix: $\left| \vec k \left( s \right) \right| \approx c s$ for some constant c. The instantaneous pitch angle $\chi(s)$, is a function of the components of $\vec k$. Helical growth can spontaneously emerge from random initial conditions. However, by starting with "seed" helices of various shapes we found that a critical value of the pitch angle, $\chi_{crit} \approx \tan^{-1}\frac{4}{\pi}$, governs the growth of the helices. For example, when $\chi (s) = \chi_{crit}$ then $\frac{d\chi}{ds} = 0$. However, this critical value is an unstable fixed point, so growth does not converge to $\chi_{crit}$.   

A Newtonian fluid meets an elastic solid: Simulating fluid flow past compliant walls

We present a novel algorithm that couples the dynamics of a fluid with the mechanical behavior of the confining walls. The fluid is simulated with the lattice Boltzmann model, an efficient solver of the Navier-Stokes equations. The solid walls are modeled by the lattice spring model, which simulates the dynamics of a continuum elastic material. We implemented solid-fluid interactions that give stick boundary conditions for the fluid and allow for a dynamic interaction between the elastic walls and the confined fluid. Here, the fluid and the solid are coupled through pressure and shear forces that are excerted across the interface. We applied the model to a study of the impact of a microcapsule (a fluid-filled elastic shell) with a regular smooth wall as well as an adhesive surface for varying fluid and shell properties. The results show that we can accurately and efficiently simulate the interaction between microcapsules and a variaty of solid surfaces. This is crucial in the study of many bio-mechanical applications.   

Session P24: Focus Session: Jamming: Rheology and Failure

Avalanche behavior in yield stress fluids

We show that above a critical stress, typical yield stress fluids (gels, clay suspensions) and soft glassy materials (the colloidal glass of Laponite) start flowing abruptly and subsequently accelerate, leading to avalanches that are remarkably similar to those of granular materials. Rheometrical tests reveal that this is associated to a bifurcation in rheological behavior: for small stresses, the viscosity increases in time: the material ``ages,'' and eventually stops flowing. For slightly larger stresses the viscosity decreases continuously in time: the flow accelerates and we observe a 1 ``rejuvenation'' of the material by the flow. We show that for the Laponite system, both the aging and the shear rejuvenation can be observed directly using Diffusive Wave Spectroscopy. We propose a simple physical model capable of reproducing the rheological observations. These results may have some implication in geophysics: they shed some light on certain landslides of clayey soils, and the way quicksand works.   

Three Dimensional Observation of Force Chain Formation in Emulsions

The spatially heterogeneous distribution of forces may be a common feature of jammed granular, foam and emulsion systems. This is thought to be due to the presence of force chains, structures of particles, droplets or bubbles which bear the bulk of a force exerted on the system. In emulsions, adjacent droplets exert forces on one another, which results in the deformation of droplet interfaces. We use fast laser-scanning confocal microscopy to observe the local deformations of each droplet in a sample in three dimensions, in order to visualize the force chains. Furthermore by placing jammed emulsions in a sheer-cell we observe the dynamics of force chain formation in real-time.   

Collision times and stress in gravity driven granular flow

We investigate, using simulations, collision times and stress distributions in two and three-dimensional steady-state granular matter in jammed versus diffuse flows. Surprisingly we find little dependence on the dimensionality of space. Grains remain separated with similar power-law distributions of times between collisions and the origin of stress transfer follows a similar mechanism in both cases. We compare our simulations to experimental results.   

Microrheology and the jamming transition in colloidal suspensions

We study concentrated colloidal suspensions, a model system which has a jamming transition (the colloidal glass transition). We use an optical confocal microscope to view the motion of these colloidal particles in three dimensions. These are suspensions of small solid particles in a liquid, and exhibit glassy behavior when the particle concentration is high; the particles are roughly analogous to individual molecules in a traditional glass. This allows us to directly study the microscopic behavior responsible for the macroscopic viscosity divergence of glasses. In particular we use small magnetic particles to locally ``poke'' on the colloidal samples, a form of active microrheology. We find a yield force (below which there is no motion), which grows as the glass transition is approached. Above this force, the magnetic particle moves, disturbing the surrounding colloidal particles. The results are interpreted in the framework of microrheology.   

Stressed out colloids: The effects of gravity on sedimented suspensions

Using laser scanning confocal microscopy we investigate dense hard-sphere colloidal sediments. Confocal microscopy allows for both high temporal resolution and real-space three dimensional imaging. By density mismatching the particles and solvent, and introducing slight polydispersity, we produce sediments without crystalline order that are dynamically arrested. Within these sediments, distinct clusters of particles with reduced local volume are observed. We will present our analysis of these structures, and postulate their equivalence to characteristic features of analogous jammed systems.   

Yield stresses and stress relaxation in latex dispersions

Static and dynamic yield stress as functions of temperature and time are used to study dynamics of jammed systems. Similar to hard spheres, inter-particle bonds which are function of volume fraction, crowding of clusters and aggregates can be used to and modulus is a powerful tool. Particle to particle interaction dominate in concentrated dispersions resulting in a complex microstructure due to ordering of particles. Extent of ordered and disordered regions in a microstructure depends on the magnitude of shear rate ( Peclet number ). Flow fields influence particle to particle collisions which are subject to Brownian, hydrodynamic, electrostatic and van der Waal forces moving a dispersion from having a dominant ordered structure to a disordered structure and then eventually into a dominant ordered region in the dispersion. Polydispersed particles of acrylic latexes were treated to both steady and oscillatory shear and their shear response was analyzed to determine shear rate dependence, dynamic and steady yield stress. The aggregate particles stemming from primary particle form fractal-like structures, which coupled with volume fraction of particles, relate to correlation length of the network. By considering chains from a bundle of primary particles in a network undergoing creation, evolution and annihilation, stress and relaxation modulus can be calculated with relaxation time constant as dependent on fractal size and volume concentration. The shear modulus and relaxation time constant from dynamic experiments are calculated and their dependence on aggregate size is determined and compared to network models.   

Universal Breakdown of Elasticity at the Onset of Material Failure

We show that, in the athermal quasi-static deformation of amorphous materials, the onset of failure is accompanied by universal scalings associated with a \emph{divergence} of elastic constants. A normal mode analysis of the non-affine elastic displacement field allows us to clarify its relation to the zero-frequency mode at the onset of failure and to the crack-like pattern which results from the subsequent relaxation of energy.   

Sum rules for the visco-elastic response of disordered solids at zero temperature

We study exact results concerning the non-affine displacement fields observed by Tanguy et al [Europhys. Lett. {\bf 57}, 423 (2002), Phys. Rev. B {\bf 66}, 174205 (2002)] and their contributions to elasticity. A normal mode analysis permits us to estimate the dominant contributions to the non-affine corrections to elasticity, and relate these corrections to the correlator of a fluctuating force field. We extend this analysis to the visco-elastic response of the system.   

Jamming and dynamics in Confined Quasi-One Dimensional Systems

Geometrically confining particles to a quasi-one dimensional arrangement so that they can only interact with their nearest neighbours simplifies the way the particles can pack to the extent that we can calculate the distribution of jammed packings exactly, making them ideal systems for exploring the connection between jamming and dynamics. We study the mean squared displacement (MSD) of a system of two dimensional hard discs subject to inertial motion and confined to a single file by two hard lines. At low densities the MSD of the discs increases linearly with time, consistent with the Einstein relation for normal diffusion. However, at high densities the system exhibits anomalous diffusion, where the MSD is proportional to $t^{1/2}$. We show how this dynamic transition is related to the nature and distribution of jammed structures. We also use this simple system to examine the role of dynamic heterogeneity in the motion of dense confined fluids.   

Signatures of dynamical heterogeneities in dense granular flows

Recent interest in understanding the dynamical arrest in both thermal and athermal systems has led to questions about the nature of these jamming transitions (PRL {\bf 86}, 111 (2001), Nature {\bf 411}, 772 (2001)), as well as the role extended structures may play in determining the dynamics of the system (Science {\bf 287}, 627 (2000)). Simulations of steady-state gravity-driven flows of inelastically colliding hard disks show the formation of large-scale linear chains of particles with a high collision frequency even at flow velocities well above the jamming transition (EPL {\bf 66}, 277 (2004)). These chains can be shown to carry much of the collisional stress in the system due to a dynamical correlation that develops between the momentum transfer and time between collisions in these "frequently-colliding" particles. Several striking features develop which may be connected to the presence of the chains, including a strong anisotropy in the distribution of collision angles and an increase in collision frequency as the granular temperature is decreased. The granular temperature displays a different dependence on flow velocity than is predicted by current kinetic theory (PRE {\bf 65}, 011303 (2001)); the velocity fluctuations are observed to die out more rapidly than expected, as observed in experiments (Science {\bf 275}, 1920 (1997)). Understanding the effects that these long-lived dynamical stress chains have on dense, flowing granular materials can lead to further insight into the nature of these systems.   

Is a ``homogeneous'' description of dynamic heterogeneities possible?

We study the simplest model of dynamic heterogeneities in glass forming liquids: the one-spin facilitated kinetic Ising model introduced by Fredrickson and Andersen [G.H. Fredrickson and H.C. Andersen, Phys. Rev. Lett. \textbf{53}, 1244 (1984); J. Chem. Phys. \textbf{83},5822 (1985)]. We previously showed that the low-temperature, long-time behavior of the excitation density autocorrelation function predicted by a scaling approach can be obtained from a self-consistent mode-coupling-like approximation. Here we use a similar approach to investigate diffusion of a test particle and the incoherent intermediate scattering function.   

Heterogeneous Dynamics in Thin Films of Glassy Polymers

In this talk, we present an analysis of the heterogeneous dynamics in ultrathin polymeric films near the glass transition. Specifically, behavior of polymer films supported by an absorbing structured surface is studied using molecular dynamics simulation. A coarse-grained, bead-spring model is used for the polymer chains. We define a specific criterion to characterize the polymer bead mobility and use this to determine the mobile and immobile beads in the system. The immobile beads are found to occur throughout the film, but their distribution is inhomogeneous, with the probability of their occurrence decreasing with distance from the substrate. Still, enough immobile beads are located near the free surface to cause them to percolate in the direction normal to the substrate surface, at a temperature near the glass transition temperature. The immobile beads block or “jam” the overall molecular motion in the film and hence cause the type of dynamic arrest, typically associated with glass transition. This result is in agreement with a recent theoretical model of glass transition [D. Long, F. Lequeux, Eur. Phys. J. E 4, 371 (2001)].   

Origins of jamming in the zero-temperature dynamics of the Sherrington-Kirkpatrick model

We consider zero-temperature dynamics of the Sherrington-Kirkpatrick spin-glass model. Such dynamics consistently converges to an energy above that of the ground state. We argue that this jamming cannot be explained soley by the presence of large numbers of metastable states. We elucidate the origins of the jamming by modelling the dynamics as a Markov process in the single-site energies. We discuss the features of this process which cause the jamming, and present an approximate derivation of the dynamics that captures these features.   

Session P25: Focus Session: Novel and Complex Oxides: Multiferroic and Other

Magnetoelectric Coupling in Multiferroic Materials

There has been increasing recent interest in magnetoelectric multiferroics, which are materials that show spontaneous magnetic order and ferroelectricity in the same phase. In addition to the fascinating physics resulting from the independent existence of two or more ferroic order parameters in one material, the coupling between magnetic and electric degrees of freedom gives rise to additional phenomena. \par In this talk we will discuss possible coupling scenarios between the ferroelectric polarization and the magnetization in magnetoelectric multiferroics. As an example we present results for the magnetoelectric multiferroic bismuth ferrite. Using first-principles calculations in the framework of density functional theory we analyze the nature of the electric polarization, the magnetization, and the coupling between these two quantities. We show that weak ferromagnetism occurs in this material, and that the resulting magnetization is strongly coupled to the structural distortions. We explore the possibility of electric-field-induced magnetization reversal and show that, although it is unlikely to be realized in bismuth ferrite, it is not in general impossible. Finally we outline the conditions that must be fulfilled to achieve such a switching of the magnetization using an electric field.   

Orbitally-driven Peierls state in correlated oxides

In studying superstructures in transition metal oxides such as charge or orbital ordering, one usually considers site-centered superstructures. Hovewer there exist another possibility: bond-centered superstructures, such as e.g. the Peierls state in low-dimensional systems. In this talk we will consider the possibility of existence of site-centered and bond-centered structures on a few examples. We will show that in spinels and in some other frustrated systems a site-centered orbital ordering (ODW-Orbital Density Wave) may lead to the formation of bond-centered singlet Peierls-like states [1]. This picture gives a simple explanation of extremely strange superstructures observed recently below metal-insulator transitions in MgTi$_2$O$_4$ [2] and CuIr$_2$S$_4$ [3], and may be relevant for several other materials, such as NaTiO$_2$, La$_4$Ru$_2$O$_{10}$ [4] and some others. We will also give a general discussion in which cases bond-centered structures may be favourable, and discuss the role of orbital degrees of freedom for the insulator-metal transitions. [1] D.Khomskii and T.Mizokawa, cond-mat/0407458 [2] M.Schmidt et al., Phys.Rev.Lett. \textbf{92}, 056402 (2004) [3] P.G.Radaelli et al., Nature \textbf{416}, 155 (2002) [4] P.Khalifah et al., Science \textbf{297}, 2237 (2002)   

Andreev Edge State on Semi-Infinite Triangular Lattice: Detecting the Pairing Symmetry in Na$_{0.35}$CoO$_{2}$.yH$_{2O}$

We study the Andreev edge state on the semi-infinite triangular lattice with different pairing symmetries and boundary topologies. We find a rich phase diagram of zero energy Andreev edge states that is a unique fingerprint of each of the possible pairing symmetries. We propose to pin down the pairing symmetry in recently discovered Na$_{x}$CoO$_{2}$ material by the Fourier- transformed scanning tunneling spectroscopy for the edge state. A surprisingly rich phase diagram is found and explained by a general gauge argument and mapping to 1D tight-binding model. Extensions of this work are discussed at the end. ref: cond-mat/0407187   

Finite Size Effects in Nanocrystals of Magnetoelectric Multiferroic Oxides.

The coexistence of coupled ground states of ferroelectricity and magnetism is an intriguing phenomenon, observed in some multiferroic materials. Among the many interests in such materials is the possibility of using an electric field as a means to control magnetic spin. The ability to maintain a small grain size is an important parameter in memory devices, and therefore it is important to study the effect of reduced crystal size on properties. We report studies of a select group of ABO3-type oxide materials with controlled particle size in the 20-100 nm size range, fabricated using solution techniques (samples $\sim $ 20nm, 33nm, 56nm and 80nm particle size). We have studied particle-size induced changes in the lattice structure, and in the electric and magnetic properties. We also find a change in the electronic band gap and the phonon spectrum. We compare these results with bulk single crystals and conclude that our size effects mostly arise from a size induced change in lattice symmetry.   

Far-infrared phonon behavior in the undersconstrained negative thermal expansion system Zr(WO$_4$)$_2$

Zr(WO$_4$)$_2$ is an unusual material in that it contracts isotropically as it is heated over a very broad temperature range from about 10 to 1000 K. Temperature dependence of the lattice volume and specific heat indicate that the important energy range for the mechanism of this negative thermal expansion (NTE) phenomenon is about 2 to 12 meV. Using infrared spectroscopy, we have studied[1] the optic phonons as a function of temperature. Comparison of energy levels to the measured specific heat and neutron density-of-states indicates that anomalous features arise in precisely the spectral region where NTE phonons are believed to exist. In addition, lattice-dynamical modeling suggests that most of the low-energy modes involve a combination of twisting and translational motion of WO$_4$ tetrahedra. We will discuss the relationship of these data to the origins of negative thermal expansion and the possible role of geometrical frustration in sustaining the unusual environment that supports NTE in this under-constrained open-structured system. [1] J. N. Hancock et al, Phys. Rev. Lett. 93, 225501 (2004), cond-mat/ 0409533.   

Eleectric Field Gradients and Born Effective Charges of PST

Relaxor behavior in some complex ferroelectrics oxides is thought to be related to the local chemical environment. High magnetic field MAS NMR measurements have recently shown great promise as a microscopic probe of local structure of relaxors \footnote{G. Hoatson et al., {\it Phys. Rev. B} {\bf 66} 224103, (2002).} by their ability to resolve electric field gradient (EFG) splittings. It is thus of considerable interest to be able to calculate EFGs in these materials. Here we present local density functional EFG calculations of Pb(Sc$\frac{1}{2}$Ta$\frac{1}{2}$) (PST) using the linear augmented planewave (LAPW) method. Our calculations focus on PST unit cells with different chemical ordering and ferroelectric phases. Trends of EFGs of PST and the correlation between EFGs and Born effective charges are discussed to better understand how local environments induce changes in EFGs. \newline \newline * Supported by ONR   

High-yield growth of semiconducting tungsten oxide nanowires

We characterized the growth, composition, and electrical properties of crystalline WO$_3$ nanowires grown using a catalyst-free chemical vapor deposition method. We showed that growing the wires in a mixture of methane and hydrogen dramatically increases the yield compared to growing the wires in argon carrier gas. The high yield makes simple nanowire `harvesting' schemes feasible. Additionally, we demonstrated that field-effect transistors can be produced using single WO$_3$ nanowires as the transistor channel. Devices made by using Ni as source and drain contacts are $n$-type and have good ON-currents and reasonable ON/OFF ratios. Scanning tunnelling spectroscopy gives a bandgap of about 2.2~eV.   

Synthesis and characterization of MgO and ZnO nanoparticle

We will discuss recent synthetic, thermodynamic and electron microscopicy investigations that probe the topological, adsorption and chemical properties of MgO and ZnO nanometer sized particles. Using a novel, patented process we find that pure and doped particles of metal oxides can be produced in large quantities with well-defined crystal habitat. Our thermodynamic investigations indicate that at low temperatures several layering transitions of methane and hydrogen are easily observable. Our electron microscopy studies indicate that numerous shapes including cubes, rods, plates and tetrapods can be selectively produced. We will discuss the effectiveness of these materials as catalysts when small (20nm) sized metal clusters of Au and Pd are deposited. If time permits the optical properties of the ZnO materials will also be discussed.   

Non-integral spin moment and electron correlation effects in magnetite

In order to directly probe the electronic ground state of magnetic electrons, we have carried out temperature dependent magnetic Compton scattering experiments on an oriented single crystal of magnetite (Fe$_3$O$_4$). First principles band theory computations using the conventional local density approximation (LDA) as well as computations treating correlation effects beyond the LDA are used to gain insight into these measurements. The magnetic moment associated with unpaired spins in magnetite is found to be insensitive to temperature over the range of 10-300K with a value of about 3.6 $\mu_B$/formula unit, including the region of the Verwey transition. The non-integral value of the spin moment implies that some majority spin states must be present at the Fermi energy ($E_F$) at all temperatures so that the polarization of electrons at the $E_F$ cannot be 100\%. Our analysis emphasizes the role of Fe$^{2+}$ ions on the octahedral sites in producing a correlated ground state of magnetite and gives insight into the nature of the order parameter for Verwey transition and the lack of quenching of the orbital magnetic moment in the system. Work supported by the USDOE.   

Theoretical studies on strongly correlated systems: bulk and surfaces of magnetite Fe$_3$O$_4$

Transition metal oxides (TMOs) are important materials because of their wide range of properties, the underlying physics and the tremendous implications for tomorrow's technology. Magnetite, Fe$_3$O$_4$ is one technologically important TMO that undergoes a first-order metal-insulator transition at $T_V$=120 K as reported first by E. Verwey. We use density functional theory adding a Hubbard-U parameter (DFT+U) to account for intra-atomic interactions for the strongly correlated Fe:3$d$ electrons of Fe$_3$O$_4$. Applying plane-wave DFT within the generalized gradient approximation and appropriate parameters, we examine the electron-phonon effects that cause a small structural distortion and lead to the insulating state with low symmetry. The electronic structure for this phase presents a sub-band of partially-delocalised minority spin electrons below the Fermi level. Here we investigate the competing roles of the screened coulomb repulsion, Fe-O hydridization and Fe-Fe overlap. We also study the Fe$_3$O$_4$ (001) surface: $(\sqrt{2}\times\sqrt{2})R45^{\circ}$ slab is used and we consider both tetrahedral and octahedral terminations.The surface reconstruction, electronic structure, magnetic properties and surface energies are computed. The calculated superficial density of states of our optimal slab is compared with scanning tunneling microscope data. Finally we analyze the effect of the electron-phonon interaction in the electronic structure of the surface and the possible existence of charge ordering.   

Effect of the electron-phonon coupling on the magnetism in the Nickelatematerials Li$_{x}$Na$_{1-x}$NiO$_{2}$

The absence of magnetic and orbital ordering in the nickelate LiNiO$_{2}$ has long been a subject of speculation, especially in light of the fact that its sister compound NaNiO$_{2}$ exhibits both magnetic and orbital structure. Although this issue has attracted much attention in recent years from both the theoretical and experimental fronts, the unusual spin-glass state of lithium nickelate remains a mystery. We are able to account for the observed type A magnetic structure of NaNiO$_{2}$ by computing the intra- and inter-layer exchange couplings using a model Hamiltonian which includes electronic hopping, on- site energy, and Coulomb interaction. The electronic structure parameters are obtained via \emph{ab initio} density functional theory calculations using the linear muffin-tin orbitals method. The dynamical electron-phonon coupling is then introduced by quantization of the motion of the Na/Li ion. We compute the ground-state of the full Hamiltonian by exact diagonalization as well as using a Lang-Firsov unitary transformation. We find that the coupling of the electronic degrees of freedom to the motion of the metallic ion decreases the exchange coupling.   

Hole dynamics in spin and orbital ordered vanadium perovskites

We present a theory of the doped perovskite vanadates with spin and orbital orders [1]. Two kinds of spin-orbital orders are found in the ground state: the G-type (three-dimensional (3D) staggered) spin order (SO) with the C-type (rod type) orbital order (OO) (the alternative $d_{xy}^1d_{yz}^1/d_{xy}^1d_{zx}^1$ configuration) termed (SG/OC) in YVO$_3$, and the C-type SO with the G-type OO termed (SC/OG) in LaVO$_3$. Mobile holes are strongly renormalized by spin excitations (magnons) in the spin G-type and orbital C-type (SG/OC) order, and orbital excitations (orbitons) in the spin C-type and orbital G-type (SC/OG) one. It is found that hole dynamics in a staggered $t_{2g}$ orbital array is distinct from that in a antiferromagnetic order as well as the $e_g$ orbital one. The anomalously fragile character of the (SG/OC) order observed in Y$_{1-x}$Ca$_x$VO$_3$ is attributed to the orbiton softening induced by a reduction of the spin order parameter. [1] S. Ishihara, cond-mat/0408395.   

P-O and Al-O Bonding in Alumina-Calcia-Monazite Melts Studied by Raman Scattering and Ultra High-temperature NMR

Raman scattering and NMR of $^{27}$Al have been used to investigate the structure of molten samples of ceramics in the Al$_{2}$O$_{3}$-CaO-LaPO$_{4}$ system. Raman spectra of quenched samples indicate the presence of PO$_{4}$ structures similar to those of metaphosphate glasses, involving Q$_{2}$ tetrahedral$_{ }$units. NMR of $^{27}$Al in melts shows strong 4-fold coordination, but also 5- and 6-fold Al-O bonding and diffusivities far more rapid than those expected for networked AlO$_{4}$ tetrahedra. Evidence that Al cross links tetrahedral chains, in addition to forming these tetrahedra, will be discussed.   

Session P26: Computational Nanoscience IV

Water at a hydrophilic solid surface probed by ab-initio molecular dynamics: inhomogeneous thin layers of dense fluid.

We present a microscopic model of the interface between liquid water and a hydrophilic, solid surface, as obtained from \textit{ab-initio} molecular dynamics simulations. In particular, we focused on the (100) surface of cubic SiC, a leading candidate semiconductor for bio-compatible devices. Our results show that, in the liquid in contact with the clean substrate, molecular dissociation occurs in a manner unexpectedly similar to that observed in the gas phase. After full hydroxylation takes place, the formation of a thin ($\sim $ 3 {\AA}) interfacial layer is observed, which has higher density than bulk water and forms stable hydrogen bonds with the substrate. The liquid does not uniformly `wet' the surface, rather molecules preferably bind along directions parallel to the Si dimer rows. Our calculations also predict that at $\sim $ 1 nm, the structural and electronic properties of liquid water are weakly affected by one-dimensional confinement between hydrophilic, solid substrates. This work was performed under the auspices of the U.S. Department of Energy by University of California Lawrence Livermore National Laboratory under contract No. W-7405-Eng-48.   

One dimensional growth of styrene on H-Si(001)-(3×1): a Density Functional Theory study.

Recent experimental work on the addition of styrene on hydrogenated Si (001) and (111) surfaces has provided evidence for a surface chain reaction mechanism initiated at isolated H vacancies. In contrast with the island-type growth on the H-terminated Si(111) surface, styrene is found to form one dimensional lines on the hydrogenated Si(001) surfaces. Using periodic Density Functional Theory (DFT) calculations, together with a recently developed method to find reaction pathways, we have studied the initial steps of the radical chain mechanism on the H-Si(001)-(3 $\times $1) surface. Our results suggest a preference for a one dimensional growth in the direction perpendicular to the Si dimer rows, in agreement with experiment. This preference is partly due to a smaller activation energy for hydrogen abstraction from the nearest dihydride site compared with hydrogen abstraction from a neighboring dimer, as well as to a strong repulsion between hydrogen atoms of the styrene molecules and monomer sites, when the growth is parallel to the dimer rows.   

Surface Reaction of Alkynes and Alkenes with H-Si(111) : A DFT study

There is currently a strong interest in the organic functionalization of semiconductor surfaces. One of the most promising approaches for preparing such functionalized surfaces is via a radical-initiated surface chain reaction of terminally unsaturated molecules with hydrogen-terminated surfaces. The adsorbing organic molecule reacts with the Si dangling bond, and forms an intermediate metastable state in which a carbon centered radical is present. Abstraction of a hydrogen atom from a neighboring H-Si surface unit results in a stable adsorbed species and a new hydrogen vacancy. We have studied the initial stages of this chain reaction on the H-Si(111) surface using the first principles string molecular dynamics approach [1] which couples the Car-Parrinello scheme with an efficient method to determine reaction pathways. We find that the relative values of the energy barrier for hydrogen abstraction and for the desorption of the adsorbed carbon centered radical at the metastable state is the crucial factor in determining the viability of the surface chain reaction. The roles of the molecule and surface electronic structures in the reaction are discussed. Results for the adsorption of C$_2$H$_2$, C$_2$H$_4$, Phenylacetylene, and Styrene on the H-Si(111) surface are presented. \newline \newline [1] Y. Kanai, A. Tilocca, A. Selloni, and R. Car, J. Chem. Phys. 121, 3359 (2004).   

Structure, Bonding, and Dynamics of Alkanethylhiolates on Copper and Gold Clusters and Surfaces

The interaction of alkanethiolates with small coinage metal clusters and (111) surfaces of copper and gold was studied based on density functional theory with a focus on the metal-thiolate junction. Calculation of fragmentation energies indicate that for Cu cluster-thiolate (n=1,3,5,7, and 9) there is a progressive lowering in energy for the fragmentation of the S-C bond in the thiolate from a value of 2.9 eV for n=1 to 1.4 eV for n=9. The detailed electronic origins of this specific weakening are attributed to a polarization of electron density in the S-C bond as induced by bonding with the Cu cluster. For the gold analogues this effect is not observed and fragmentation at the S-C bond experiences only a slight 10\% destabilization as n increases from 3 to 9 On the Cu(111) surface the metal to thiolate charge transfer which leads to a non-direction partially ionic bonding with a concurrent flat adsorption energy landscape, As a result, occupation of fcc-hollow, hcp-hollow and fcc-bridge sites is observed during the coarse of a short finite temperature ab-initio molecular dynamics simulation as opposed to a static model where only the hollow sites are stable minima. Comparison of our results with the available experimental evidence and consequences of the electrostatic profile of the metal-molecule interface are presented. The difference between Cu and Au are discussed in the context of relativistic effects.   

Strain in Layered Nanocrystals

Layered nanocrystals consist of a core of one material surrounded by a shell of a second material. We present computation of the atomistic strain energy density in a layered nanocrystal, using an idealized model with a simple cubic lattice and harmonic interatomic potentials. These computations show that there is a critical size r$_{\ast }$ for the shell thickness r$_{s}$ at which the energy density has a maximum. This critical size is roughly independent of the geometry and material parameters of the system. Moreover it agrees with the shell thickness at which the quantum yield has a maximum, as observed in several systems.   

Molecular Dynamics Simulation of Nanostructured Thin Film Growth

Properties of nanostructured films are sensitive to atomic defects. Molecular dynamics (MD) simulations of growth can reveal defect formation mechanisms that are difficult to explore using other approaches. There are mainly two challenges. First, nanostructures often utilize different materials with metallic, ionic and covalent bondings. The MD must hence use an interatomic potential transferable to different local bonding environments encountered during growth. Secondly, growth is simulated by randomly adding atoms on the surface. The interatomic potential must hence accurately predict surface properties under various surface configurations. The newest MD approaches have begun to enable the growth simulation for a wide range of materials. The embedded atom method (EAM) potential was successfully used to simulate the growth of giant magnetoresistance metal multilayers. Our integrated EAM and charge transfer ionic potential is transferable between metallic and ionic materials and has been successfully used to simulate the growth of spin tunnel junction multilayers. Stillinger-Weber, Tersoff, and our analytical bond order potentials are compared for simulating covalent semiconductor growth.   

Self-Teaching Kinetic Monte-Carlo Scheme For Small Cluster Diffusion on Cu(111)

We have developed a new version of Self-Teaching Kinetic Monte Carlo technique (ST-KMC) for simulations of processes relevant to growth on fcc(111). In this method, adsorbed atoms may occupy fcc or hcp sites for 2D cases. Additionally, adatoms may occupy ``top'' sites in the case of 3D simulations. The environment of any diffusing adatom is mapped using a 211-site template mimicking the layer stacking on fcc surfaces. We applied this new approach to the diffusion of small copper adatom clusters with size from 1 to 7 atoms on a Cu(111) surface. Activation energies for all mechanisms were calculated using the drag method, for saddle points search, and Embedded Atom Method (EAM) for interaction potentials. With this new approach, it was possible to incorporate multi-atom concerted motion involving the occupancy of fcc and hcp sites, which has been suggested by experimental findings and MD simulations of small clusters diffusion on fcc(111). Diffusion coefficients and their scaling with size and temperature will be presented and contrasted with MD and KMC results.   

Preferential Growth of Pt Particles on Rutile TiO$_2$

Pt/TiO$_2$ is the prototype system exhibiting strong-metal- support-interaction phenomenon. The characterization of a real Pt/TiO$_2$ catalyst system through a combination of atomic resolution Z-contrast images and electron energy loss spectroscopy in the scanning transmission electron microscope has revelaed an unexpected result: Pt particles have a strong tendency to nucleate on the rutile phase of TiO$_2$ rather than anatase. In order to address the selective growth of Pt on rutile, Pt atom binding energies on stoichiometric and reduced TiO$_2$ surfaces and surface oxygen vacancy formation energies have been calculated using first principles density functional theory calculations for both rutile and anatase phases.   

Mechanical properties of diamond/a-C nanocomposite films

Nanostructured amorphous carbon (na-C) is a hybrid form of carbon in which nanocrystallites are embedded in an a-C matrix. It has attracted considerable attention, because it offers the possibility to tailor the mechanical and electronic properties of a-C. We present here our studies of a particular form of na-C, containing diamond nanocrystals, using tight-binding molecular dynamics and empirical-potential Monte Carlo simulations. The calculations allow us to shed light into several properties of this material. We examine its structure, stability as a function of temperature and size of nanocrystals, stress state, and hardness. We find that the nanocrystals are stable only in dense, highly coordinated a-C matrices. The nanocomposite films possess negligible intrinsic stresses. The elastic moduli and yield stresses under tensile or shear load are consistently and considerably higher than those of pure a-C, making the nanocomposite diamond/a-C films suitable for ultra-hard mechanical coatings.   

Theoretical Investigation of the Vibrational and Electronic Properties of Titanium Carbide Nanocrystals

Stable titanium carbide nanoclusters with \verb+~+1:1 stoichiometry were first discovered in molecular-beam experiments in the early 1990's. These clusters are all indexed to perfect or nearly perfect $N_1 \times N_2 \times N_3$ fragments of bulk TiC in the rocksalt structure and are thus termed ``nanocrystals''. The most abundant member of this family is the $3\times 3 \times 3$ nanocrystal Ti$_{14}$C$_{13}$, indicating special stability for this species. Using Density Functional Theory, we have carried out a detailed theoretical analysis of the structural, electronic, and vibrational states of Ti$_{14}$C$_{13}$ and its $3 \times 3 \times 3$ sibling, Ti$_{13}$C$_{14}$, which is not observed in the experiments. In this talk, we will present our theoretical results and show how our analysis sheds light on several previously unresolved experimental findings.   

Evolution of Electronic and Vibrational Polarity of NaF Nanocrystals

Density functional theory is used to study vibrations, electrical dipole moments, and polarizabilities of NaF clusters. We use the NRLMOL code with GGA exchange and correlation, and a large basis set of Gaussian orbitals. Because of prior experimental and theoretical studies, this is a good model system for tracking the evolution of the properties from diatomic molecule to bulk crystal. The predicted ratio of vibrational to electronic contributions to the polarizability increases dramatically with size $N$ in the closed shell clusters Na$_N$F$_N$. The open shell system Na$_{14}$F$_{13}$ has a greatly enhanced electronic polarizability. Contrary to previous studies on this system which treated only the outer electron by quantum mechanics, we find the O$_h$ cubic structure to be stable relative to the polar distorted structures such as C$_{3v}$. The size of the permanent dipole is predicted to be 2.01 and 5.12 in units $ea_B$ for the C$_4v$ systems Na$_9$F$_9$ and Na$_{18}$F$_{18}$ respectively.   

Simulations of amorphous nanoparticles: the effect of shape on surface structure and subsequent interactions with the surroundings

Molecular dynamics simulation is employed to study the effect of varying nanoparticle shape on the structure of boron oxide nanoparticles, and their subsequent influence on surrounding polyethylene oxide. While previous studies have focused on crystalline nanoparticles, this study is unique because the nanoparticles are amorphous. Shape has been shown to affect the electrical and optical properties of nanoparticles, in addition to the structure and dynamics of polymers surrounding nanoparticles. In this study, two nano-shapes of boron oxide are compared: a 16 angstrom diameter sphere, and a 16 x 16 x 16 angstrom cube. The networks are described by a short-range structure consisting of BO3 units, while the intermediate-range structure is described by six-membered planar boroxol rings. Both the fraction of boroxol rings and their locations differ between the two nano-shapes. All planar boroxol rings within the spherical simulation are located on the interior, while planar rings within the cubic simulation aggregate to the cube walls. Structural differences also appear at longer ranges, including the formation of ``layers'' aligned parallel to the walls of the cube. We also investigate how varying the nanoparticle shape influences the structure and dynamics of the surrounding polymer.   

Cage structures based on Polyhedral Oligomeric Silsesquioxanes (POSS) with atomic and ionic impurities

Endohedral and exohedral polyhedral cage molecules of the form (HAO$_{3/2}$)$_{8}$ (A = C, Si, Ge) with double four-membered ring D4R units complexed with the atomic or ionic species: Li$^ {+}$, Na$^{+}$, K$^{+}$, F$^{-}$, Cl$^{-}$, Br$^{-}$, He, Ne, Ar have been investigated using Density Functional Theory (DFT). Geometric, electronic and energetic properties were obtained. The symmetry of the endohedral complexes when X is a cation turned out to depend critically on the relative cation and cage sizes. The binding energies of the endohedral and exohedral complexes document a clear preference for the latter, except for halides, where the endohedral complexes are more stable. The formation of the endohedral complexes is discussed in terms of transition states that connect the exohedral and endohedral minima, as well as the activation barriers for insertion of the guest into the cage. Our studies predict that a fluoride anion can penetrate into the (HAO$_ {3/2}$)$_{8}$ cage without destroying it. For X = Cl$^{-}$, in contrast, the cage ruptures upon insertion of the impurity.   

Structure of nanocrystals embedded in amorphous carbon

Amorphous carbon (a-C) has been often found to contain crystalline regions with diameters in the nanometer scale. The so-called nanostructured amorphous carbon has attracted considerable attention, because of possible applications in MEMS/NEMS and optoelectronic devices. In this work, we study embedded nanocrystals in various a-C matrices using tight-binding molecular dynamics and empirical-potential Monte Carlo simulations. We are especially interested in faceted nanocrystallites, that deviate significantly from a spherical shape. We start by calculating the interface energy between various faces of the crystal and a-C. We then use the so obtained interface energies to predict the shape of the nanocrystal by means of a Wulff construction. Finally, we construct the atomistic configuration of the embedded nanocrystallite and examine its stability as a function of temperature and the nanocrystallite size.   

Session P27: Focus Session: Carbon Nanotubes: Functionalization II

Electrochemical Sensing with Individual Single-Wall Carbon Nanotubes

The transport properties of molecular electronic devices can be strongly modulated by immersion in a liquid electrolyte. For example, early investigations with single-wall carbon nanotubes (SWNTs) used the electrolyte as a "liquid gate." That is, the conductance of SWNTs in a field-effect-transistor configuration was tuned via an electrochemical potential applied to the electrolyte. This concept was also extended to sensor applications in which molecules impinging upon the SWNT surface cause a rearrangement of screening ions and a corresponding change of the device conductance. In these approaches, the coupling between the device and the electrolyte is solely electrostatic, and no charge is transferred across the liquid-device interface. Here we demonstrate that individual SWNTs can also be used as electrodes for electron transfer reactions, that is, electrochemical reactions in which electrons are exchanged between a SWNT and redox-active molecules in solution. The rate of electron transfer to SWNTs is observed to be very fast. It can nonetheless be resolved in dc transport measurements due to the high diffusive flux of redox molecules resulting from the nanometer diameter of SWNTs. Interestingly, metallic and semiconducting SWNTs yield similar current-voltage characteristics; we show that this behaviour is consistent with theories of electron transfer in which the electronic structure of the SWNTs is explicitly taken into account. Finally, we demonstrate that noble metals can be controllably and selectively electrodeposited from aqueous solution unto individual single-wall carbon nanotubes, opening new routes for the functionalization of SWNT devices. Work done with K. Besteman, H. A. Heering, I. Heller, J. Kong, J.-O Lee, B. M. Quinn, F.G.M. Wiertz, K. A. Williams and C. Dekker.   

Quantitative Characterization of Defect Densities in Single-Walled Carbon Nanotubes

Carbon nanotubes are often imagined to be pristine, defect-free objects, but different types of synthesis and processing are known to result in materials of different qualities. We have developed a method for the quantitative characterization of nanotube defect densities which can readily be used to compare nanotubes from different batches or processes. The method relies on the enhanced chemical reactivity of defect sites as compared to the graphene lattice. By tailoring the potentials used in electrochemical deposition, we selectively seed the growth of metal particles at reactive defect sites without decorating the bulk of a carbon nanotube. Because the metal particles can subsequently be grown 50 nm in diameter or larger, they are easily counted by low magnification SEM imaging, allowing for good statistics and wafer-scale characterization. We will demonstrate the results of this method on a batch of CVD-grown nanotubes and compare the measured defect density against nanotubes from other sources. This work has been supported by NSF-DMR.   

Probing the Phonon-Assisted Relaxation Process in DNA-wrapped Carbon Nanotubes Using Polarization Dependent Optical Spectroscopy

In this study, polarization-dependent spectroscopy is carried out on DNA-wrapped single walled nanotube hybrids, deposited onto a Sapphire substrate. By using a nanotube sample highly enriched in one specific (n,m) species and an intense light source, the various phonon-assisted excitonic relaxation processes, in addition to the commonly observed electronic interband transitions, can be separately identified and studied in detail. The phonon-assisted relaxation processes involving different phonon branches is emphasized in this study. The MIT authors acknowledge supports under the Dupont-MIT Alliance, NSF Grants DMR 04-05538.   

Electrical transport characteristics of DNA-wrapped carbon nanotubes contacted to palladium and palladium oxide electrodes

DNA-wrapped carbon nanotubes (DNA-CNT) have generated attention due the ability to disperse cleanly into solution, and by the possibility of sorting nanotubes according to size and conductivity. In order to learn more about the effects of DNA on the electrical transport characteristics of single wall carbon nanotubes, we fabricate and test a series of devices consisting of DNA-wrapped CNTs placed across gold, palladium, and palladium oxide electrodes. In addition, we look at how DNA functionalized CNTs react to presence of hydrogen, which has previously been shown to affect the conductivity of CNTs when in contact with palladium.   

Detection of Pathogens Using AFM and SPR

A priori detection of pathogens in food and water has become a subject of paramount importance. Several recent incidents have resulted in the government passing stringent regulations for tolerable amounts of contamination of food products. Identification and/or monitoring of bacterial contamination in food are critical. The conventional methods of pathogen detection require time-consuming steps to arrive disembark at meaningful measurement in a timely manner as the detection time exceeds the time in which perishable food recycles through the food chain distribution. The aim of this presentation is to outline surface plasmon resonance (SPR) and atomic force microscopy (AFM) as two methods for fast detect6ion of pathogens. Theoretical basis of SPR and experimental results of SPR and AFM on E. coli O157:H7 and prion are presented.   

Anomalously soft dynamics of water in carbon nanotubes

Quasi-one-dimensional water encapsulated inside single-wall carbon nanotubes, here referred to as nanotube-water, was studied by neutron diffraction, inelastic and quasielastic neutron scattering. The study reveals that nanotube-water is a new state of water manifested by weak hydrogen bonds, extended intermolecular water-water distances and very large amplitude of water molecules vibrations. Molecular dynamics simulations well describe the observed spectra and give a possible nanotube- water structure in a form of a square-ice sheet wrapped into a cylinder next to the inner carbon nanotube wall and a water chain in the interior. The soft dynamics of nanotube-water arises mainly from the drastic change in hydrogen-bond connectivity of the central water-chain.   

Observations of Simple and Complex Liquid Transport in Carbon Nanotubes

Taking advantage of the optical transparency of template grown, carbon nanotubes (CNTs) with 15 nm thick walls, we studied experimentally and theoretically the capillary filling, condensation, and evaporation of glycerin, ethylene glycol, and DI water inside CNTs under room conditions [1]. All the liquids readily filled the CNTs by the action of capillary forces. The capillary filling was also used to study the filling of the tube with nanoparticles. Liquid, laden with 40nm-diameter fluorescent beads, was brought into contact with a 300nm diameter CNT. The liquid and the particles' transport were observed, respectively, with optical and fluorescence microscopy. Finally, a nanotube-based device that enables us to conduct controlled experiments of liquid and macromolecule transport with an electron microscope is described. [1] B. M. Kim, S. Sinha, and H. H. Bau, Nano Letters, 4, 2203 (2004).   

Microscopic and spectroscopic studies of the surfactant-assisted dispersion process of CoMoCAT carbon nanotubes in water

Recent Raman studies carried out on the precipitant and supernatant fractions after the centrifugation step of the surfactant-assisted dispersion of CoMoCAT carbon nanotubes in water have indicated an apparent selectivity of the dispersion process for some specific tubes. To get a better understanding of this phenomenon, we have combined SEM, Raman spectroscopy and photoluminescence techniques to investigate in details the suspendability properties of CoMoCAT samples in aqueous solutions of anionic surfactants under different experimental conditions.   

Noncovalent functionalization of single-walled carbon nanotubes with water-soluble porphyrins

We have employed water-soluble porphyrin molecules [meso-(tetrakis-4-sulfonatophenyl) porphine dihydrochloride] to solubilize individual single-walled carbon nanotubes (SWNTs), resulting in aqueous solutions that are stable for several weeks. The porphyrin-nanotube complexes have been characterized with absorption and fluorescence spectroscopy, and with AFM. We find that the porphyrin/SWNT interaction is specific to the free base form, and that this interaction increases the effective pKa value for the protonation of the free base. Under mildly acidic conditions (pH less than 5) nanotube-mediated J-aggregates form which are unstable in solution and result in precipitation of the nanotubes over the course of a few days. Porphyrin-coated SWNTs can be precisely aligned on hydrophilic poly(dimethylsiloxane) (PDMS) surfaces by combing SWNT solution along a desired direction and then transferred to silicon substrates by stamping. Parallel SWNT networks and SWNT crossbars have been fabricated in this manner.   

A Generic Method towards Periodically Functionalized Carbon Nanotubes

As a consequence of their extraordinary physical properties and large application potential, carbon nanotubes (CNTs) have attracted the interest of scientists and engineers since their discovery in 1991. However, in order to effectively explore the remarkable properties and manipulate CNT, one essential step involves their functionalization. Periodical pattering CNT is an extremely challenging yet attractive field of research, and up to date, very few reported CNT functionalization methods have dedicated on how to achieve the periodical pattern on CNT. Our study of functionalization of CNTs via controlled polymer crystallization method has resulted in ``nano hybrid shish-kebab'' (NHSK), which is CNT periodically decorated with polymer lamellar crystals. The morphology and periodicity of NHSK can be controlled by tuning experimental parameters such as concentration of polymers and CNT, crystallization temperature and time. Preliminary results show that the periodicity varies from 20-70nm. Similar results were successfully obtained from both Nylon 6, 6 and PE with different kinds of CNTs, which can prove this is a generic method to periodically functionalize CNTs.   

Single-walled carbon nanotubes in superacid: partly ordered H$_{2}$SO$_{4}$, nanotube-templated crystallization and evidence for protonation

Liquid anhydrous sulfuric acid forms a partly ordered structure in the presence of single-walled carbon nanotubes (SWNTs). Room temperature x-ray scattering from well-aligned nanotube fibers immersed in acid shows that H$_{2}$SO$_{4}$ molecules align along the nanotube axis and form cylindrical ``shells'' wrapped around nanotubes and ropes. Differential scanning calorimetry of SWNT-acid suspensions exhibits concentration-dependent supercooling/melting behavior, confirming that the partly ordered molecules are a new phase. Temperature-dependent x-ray scattering further shows that crystallization of the bulk-like acid surrounding the structured acid shells is templated by the aligned SWNTs, while the structured shells remain partly ordered, at least for temperatures from 100K to 500K. The (2 0 0) or (-2 0 2) planes of the templated H$_{2}$SO$_{4}$ crystallites are parallel to the nanotube axes. This provides solid evidence for the direct protonation of SWNT since these planes display exposed hydrogen bonds.   

Structural, thermodynamic and transport properties of functionalized carbon nanotubes from first-principles

We have studied the structural, thermodynamic and transport properties of carbon nanotubes functionalized via cycloaddition of nitrenes and carbenes, using our recently-developed band-structure and quantum conductance method based on maximally-localized Wannier functions. This approach allows us to study the effects of chemical functionalization on the transport properties of nanostructures containing thousands of atoms, while maintaining full ab-initio accuracy. We find that the stability of the two distinct sites of attachment on the sidewalls of achiral CNTs is strongly affected by diameter, due to the competition between elastic and chemical forces. A subtle interplay of electronic effects between $sp^2$ and $sp^3$ hybridization results in very significant differences for the transport properties of metallic nanotubes.   

Session P29: Transport and Electronic Structure of Organic Electronic Materials

Theory of Quantum Hopping In Metallic Polymers and Applications in Electronics

The low frequency electromagnetic response of highly conducting polymers (e.g., polyaniline and polypyrrole) in a metallic state$^{1}$, when analyzed within the standard theory of metals, is provided by an extremely small fraction of the total number of available electrons $\sim $ 0.1 {\%} (in contrast to $\sim $ 100 {\%} for common metals such as Cu, Ag, or Ni) but with anomalous long scattering time $>$ 10$^{-13}$ s ($\sim $ 100 times longer than for common metals). We show that a chain-linked network of metallic grains (the polymer's crystalline domains) connected by resonance quantum tunneling through strongly localized states in surrounding disordered medium produces this behavior. The small fraction of electrons is assigned to the low density of resonance states and the long scattering time is related to the narrow width of energy levels in resonance. Recently a new interesting phenomenon, an electric field effect, was reported for the doped highly conducting polymers. Upon applying the gate voltage of a few volts the conductivity of the polymer film drops by a several orders of magnitude$^{2}$. This observation is in conflict with the fact that the electric field cannot penetrate into a conductor further that the `lattice constant', and therefore its effect on the polymer film of $\sim $ 100 nm thickness should be negligible. We suggest that the field effect in doped conducting polymers is an electric field induced conductor-nonconductor transition described by the chain-linked granular model in the presence of mobile ions. The ion motion under the gate voltage is breaking the interdot percolation network by removing critical hoping sites and as a result producing the conductor-nonconductor transition. The experimental evidences for the present mechanism of field effect in conducting polymers are presented. \begin{enumerate} \item R.S. Kohlman \textit{et al}.,\textit{ Phys. Rev. Lett}. \textbf{78}, 3915 (1997). \item A.J. Epstein \textit{et al}., \textit{Curr. Appl. Phys.}\textbf{ 2}, 339 (2002). \end{enumerate}   

Charge injection, transport and trapping in nanoparticle based memory devices

Blends of metallic nanoparticles in a semiconducting organic host show bistable electrical resistance at low ($\sim $1 V) reading voltage. When a layer of the blend, of order 100 nm thick, is sandwiched between electrodes, a low resistance (on) state is set by applying a voltage pulse of order 2 V to 3 V, and is switched to the high resistance (off) state by a pulse of about 6 V to 8 V. This behavior is observed for several different metals (e.g. Au, Ag, Al, Mg) and for both polymeric and small-molecule semiconductors. We interpret the switching and bistability in terms of charge trapping and storage on the nanoparticles. When trapped charge density is high, the resulting space-charge field inhibits charge injection, yielding the off-state, and vice versa. A simple model based on Fowler-Nordheim tunneling shows that particles of order 2 - 5 nm in diameter exhibit sharp discharge thresholds in the range of a few volts, as observed in the experiments.   

Local EFM measurements of organic conducting materials at various temperatures

We have investigated the charge injection processes in regioregular poly-3-hexylthiophene (P3HT) and in a molecularly-doped polymer, triarylamine (TPD) dispersed in polystyrene (PS), by electric force microscopy under high vacuum and at various temperatures. This microscopic study unambiguously reveals space-charge limited conduction within TPD-PS, in contrast to bulk current-voltage measurements. The temperature dependence of charge injection will also be shown to compare with current theories on charge injection into organic semiconductors.   

On the electronic transport in doped polyaniline/polyethylene oxide nanofibers prepared via electrospinning

We fabricate and electrically characterize electrospun nanofibers of doped polyaniline/polyethylene oxide. Scanning conductance microscopy shows that fibers with diameter below 15 nm are electrically insulating. Single fiber I-V characteristics show that thin fibers conduct more poorly than thick ones and fibers with large asymmetry along their length between the electrical contacts show rectifying behavior [1]. A theoretical analysis of the conductance in the polymeric nanofibers is presented. The analysis is based on the model of a granulated metal [2], so the polymeric material is treated as a network of small metallic-like domains made out of densely packed polymeric chains embedded in an amorphous array of disordered chains. Assuming the electronic transport to be provided by the electron tunneling between the metallic grains through the intermediate resonance states, the theory of conductance in molecular wires [3] is employed to calculate the tunneling current. It is shown that nonlinear features in the I-V curves could appear when the coupling of the grains to the intermediate ``bridge'' is weak enough. Obtained results are in agreement with the experiments. [1]. Y. Zhou, M. Freitag, J. Hone, C. Staii, A.T. Johnson, N.J. Pinto and A.G. MacDiarmid, Appl.Phys.Lett., v.83, 3800 (2003). [2]. V.N. Prigodin and A.J. Epstein, Physica B, v.338, 310 (2003). [3]. S. Datta, Nanotechnology, v.15, S433 (2004).   

Dispersion and Current-Voltage Characteristics of Helical Polyacetylene Single Fibers

To study the transport properties of individual helical polyacetylene (PA) fibers, we developed a method to extract a single fiber from tightly entangled ropes of helical PA bulk film. After a few minutes of sonication of a piece of helical PA bulk film in an organic solution containing surfactant, a droplet of solution is deposited on the pre-patterned electrode under argon atmosphere. AFM images show that extracted helical PA fibers are typically 10 $\mu $m in length and 100--200 nm in diameter. We found that the helicity of bulk materials is conserved. We present the temperature dependencies of current-voltage characteristics of individual helical PA fibers doped with iodine.   

Characterization of the Porphyrin Molecule as an Electronic Component

Porphyrins have recently generated great interest as a potential ``electronic material'' for use in functions such as sensing (i.e.-carbon monoxide), biomolecular and medical physics applications (i.e. antiviral agents) and for computer memory capability. Fundamental to the latter of these three and to the realization of molecular based electronic devices is the phenomenon of self-assembly. In this work, we investigate the electronic characteristics of porphyrins, using directed self-assembly monolayer of n-alkanethiols. SAM of alkanethiols on gold surfaces has been shown to form stable surface structures. It is expected that the exchange of molecules is most active at SAM defective sites, substrate step edges and substrate vacancy islands. We measure the I-V characteristics of porphyrin molecules by directed self-assembly in the SAM defect sites. It is observed in ambient conditions using conductive probe atomic force microscopy (CPAFM). The conductivity of porphyrin molecules in alkanethiol SAM is discussed here. This experimental result is further enhanced by our group through ``AFM study of current transport through porphyrin based molecules.''   

CP-AFM Study of Current Transport Through Porphyrin – Based Molecules

Conductive Probe Atomic Force Microscopy (CP-AFM) is used to study current transport through dithiolated porphyrin based molecules. Porphyrin molecules are inserted at defect sites into an alkanethiolate SAM on Au (111), and the exposed top terminal end of the porphyrin with thiol is attached to a gold nanoparticle. These gold nanoparticles (d$_{CORE}$ = 1 nm to 5 nm)$_{ }$ stabilized by phosphine ligands are introduced into solution where ligands are displaced by thiol groups of the porphyrin bound to Au surface. I(V) measurements are done with nanoparticles of varying sizes to determine the effect on transport properties. Measurements are done using CP-AFM, and contamination is reduced by immersing the sample in toluene. Complementary work from our group will also be presented as ``Characterization of Porphyrin as an Electronic Component'' at this meeting.   

The injection barrier at a metal/organic interface

The landscape for the thermionic injection of electrons from a metal into a molecularly doped polymer is energetically disordered as a result of inhomogeneous electric fields coming from the non-uniform charging of dopant molecules in the vicinity of the metal/organic interface. This comprises an electric dipole (double - ) layer. We have determined the equilibrium composition of the dipole layer by simulated annealing, in order to study its influence on injection. We find that an electron must surmount an energetic barrier in order to escape into the organic, even when the LUMO level of the dopants is aligned with the Fermi level of the metal. For typical dopant densities, this barrier is on the order of 0.5 eV.   

Charge Injection into Cathode-Doped Amorphous Organic Semiconductors

We analyze electron injection at a wide variety of metal-organic semiconductor interfaces, and find remarkably universal characteristics at low temperature. The current voltage characteristics at T = 10 K follow power-law behavior, $J \quad \sim \quad V^{m}$, where $m $=~(20+/-1) for over 15 combinations of the metals Li, Mg, Al, Ag and Au and the organic semiconductors Alq$_{3}$, BCP, CBP, TAZ, and CuPC. The material independence of injection characteristics at low temperature is attributed to the effect of interface roughness on the image potential. We develop an analytic model to explain charge injection into organic semiconductors at disordered interfaces.   

Spectroscopy and Imaging of Metal-Organic Interfaces using BEEM

Charge injection from metal electrodes to organics is a subject of intense scientific investigation for organic electronics. Ballistic electron emission microscopy enables spectroscopy and imaging of buried interfaces with nanometer resolution. Spatial non-uniformity of carrier injection is observed for an Ag-PPP and an Ag-MEHPPV interfaces. Possible reasons are discussed. BEEM current images are found to correlate only marginally with the surface topography of the Ag film. We also determine the transmission function of the Ag-PPP interface. The transmission function is inferred from the derivative of the measured BEEM spectrum, and compared with theory, and with the transmission of metal inorganic semiconductor (MIS) interfaces. For Ag-PPP, we find a curvature opposite to that of MIS interfaces. This agrees well with the theoretical calculations on metal-phenyl ring interfaces. We demonstrate that patches of low Schottky barrier can nucleate current filaments and are likely responsible for the switching behavior observed in metal-organics.   

Session P30: Polymers - Inorganic Composites III

Novel Route to Mesoporous silica with perpendicular nanochannels from polymer/inorganic nanocomposite films

Mesoporous metal oxide films have generated intense interest due to their potential scientific and technological importance. For applications such as sensors, separations and detection devices, structures having cylindrical channels oriented normal to the surface are highly desirable, but have remained elusive. Recently we reported a new approach to mesoporous materials that involves the infusion and selective condensation of metal oxide precursors within one phase domain of a highly ordered, preformed block copolymer template dilated with supercritical carbon dioxide to yield a polymer/inorganic nanocomposite film. The organic component of the nanocomposite is then removed to produce the mesoporous oxide. To date ordered spherical and randomly oriented cylindrical morphologies have been replicated to yield silica/organosilicate mesostructures in films over micron thick while maintaining all the structural details of the sacrificial copolymer template. The preparation of phase-segregated block copolymer films with cylindrical domains oriented normal to the surface using controlled solvent evaporation has recently been realized. Here we report the replication of these templates to yield the corresponding silicate mesostructures containing arrays of perpendicular channels.   

A Fracture Resisting Molecular Interaction in Trabecular Bone: Sacrificial Bonds and Hidden Length Dissipate Energy as Mineralized Fibrils Separate

A molecular energy dissipation mechanism in the form of sacrificial bonds and hidden length was previously found in bone constituent molecules of which the efficiency increased with the presence of Ca$^{2}$+ ions in the experimental solution. Here we present evidence for how this sacrificial bond-hidden length mechanism contributes to the mechanical properties of the bone composite. From investigations into the nanoscale arrangement of the bone constituents in combination with pico-Newton adhesion force measurements between mineralized collagen fibrils, based on single molecule force spectroscopy, we find evidence that bone consists of mineralized collagen fibrils and a non fibrillar organic matrix which acts as a ``glue'' that holds the mineralized fibrils together. We propose that this ``glue'' resists the separation of mineralized collagen fibrils. Like in the case of the sacrificial bonds in single molecules, the effectiveness of this ``glue'' increases with the presence of Ca$^{2+}$ ions. We further investigate how this molecular scale strengthening mechanism increases the fracture toughness of the macroscopic material.   

Self-assembled anisotropic polymer particles by polycondensation in lyotropic surfactant mesophases

We report the formation of crosslinked polysiloxane particles whose morphologies are directed by the nonionic surfactant phase in which they self-assemble. Under conditions that allow slow condensation of silanol monomers, the polymer particles formed have a geometry similar to the parent mesophase: rod-like particles form in a hexagonal mesophase and sheet-like in a lamellar phase. The characteristic diffraction pattern obtained from the liquid crystalline surfactant assembly is preserved during polycondensation. Interestingly, while the geometry of the particle is similar to that of the mesophase, the particles are microns in size, three orders of magnitude larger than the characteristic size of the surfactant mesophase. Our observed morpholgies differ from those predicted by theories of ordering of colloidal particles in liquid crystalline phases. In our system too, ordering of polymer colloid particles is a result of minimization of elastic free energy due to distortion of the nematic matrix. However, the polymer particles in our experiments are not formed at once, but grow slowly as polycondensation proceeds. We speculate that slow condensation and cross-linking kinetics, gradual build-up of molecular weight and the non-linear architecture of the polysiloxane molecules might allow the dynamic organization of the particles by the liquid crystalline mesophase, leading to the formation of the observed particle geometries.   

Electron spin resonance on carbon nanotubes-polymer composites

Electron spin resonance (ESR) is used to assess the quality of carbon nanotubes. However, few studies were done on carbon nanotubes dispersed in polymeric matrices. Not annealed carbon nanotubes exhibit three electron spin resonance (ESR) line; a wide resonance line located at g values larger than g=2.0023 assigned to catalyst residues and two lines located close to g=2.0023 assigned to paramagnetic impurities and electrons delocalized over the conducting domains of carbon nanotubes. The annealing process reduces dramatically the intensity of the wide line. Qualitatively, same resonance lines were observed in carbon nanotubes-polymer composites. Samples of polystyrene loaded with various amounts of carbon nanotubes ranging from 0.1{\%} to 10.0{\%} (wt) were prepared. The effect of nanotube dispersion on the parameters of the resonance line is presented. The effect of temperature on the resonance line parameters was investigated. A matrix effect was observed within the glass transition range. This has been assigned to the adhesion of nanotubes to the polymeric matrix.   

Flow Based Control of Conductivity in Nanotube Composites

Nanotube composites are finding applications due to their ability to enhance the electrical conductivity of polymeric materials. They exhibit a percolation threshold in both rheological and electrical properties at mass fractions less than 0.01. We study the interrelationship between these two coupled transport properties by simultaneous dielectric spectroscopy and rheology. We find that the frequency dependent electrical conductivity is quite sensitive to shear flow near the percolation threshold; it can reversibly vary by six orders of magnitude and can become highly anisotropic. Interestingly, the shear dependence of the viscosity and the conductivity show distinct behaviors, indicating that different aspects of the nanotube network are probed by these two transport coefficients.   

Effect of Carbon Nanotube Alignment in Polymer Nanocomposites on the Electrical Conductivity

We studied the effects of nanotube alignment on the electrical conductivity of the nanotube / polymer nanocomposites. Nanocomposites with single-walled carbon nanotubes (SWNTs) and poly(methyl methacrylate) (PMMA) were prepared via our coagulation method. The nanotubes were subsequently aligned to various extents by melt fiber spinning and the degree of alignment was quantified by x-ray scattering. Electrical percolation in the nanocomposites with isotropic nanotubes occurs at $\sim $0.39wt{\%} SWNT. At all nanotube loadings investigated (0.5 to 3 wt{\%}), the electrical conductivity exhibits percolation behavior with decreasing nanotube alignment. At nanotube loadings just above the concentration threshold for percolation, a maximum electrical conductivity is observed at intermediate level of alignment. We attribute the existence of this optimal nanotube alignment to the competition between the number of tube-tube contacts and the distance between these contacts. A two-dimensional Monte Carlo simulation of sticks in a unit square was developed to mimic these nanotube/polymer nanocomposites and shows similar trends. In sum, we observed percolation behavior with respect to nanotube alignment and obtained the maximum electrical conductivity by partially aligning the nanotubes.   

Thermal Conductivity of Single-Walled Carbon Nanotube / Polyethylene Nanocomposites

The thermal conductivity of nanocomposites with single walled carbon nanotubes (SWNTs) and polyethylene are being investigated with attention to the effect of the degree of PE crystallinity and the alignment of both the PE and SWNT. The nanocomposites were prepared via the hot-coagulation method, resulting in a good dispersion of the SWNTs in the polymer matrix. Characterization methods include the comparative and modulated thermo-reflectance method to measure thermal conductivity, x-ray scattering to quantify SWNT and PE alignment, and SEM and AFM to determine SWNT dispersion. SWNTs act as nucleation sites for the PE and this crystalline interface might enhance heat transfer between the SWNT and matrix. At 30 wt{\%} SWNT, isotropic composites made with low density and high density PE have thermal conductivities of 1.8 and 3.5 W/m-K, respectively, clearly demonstrating the importance of degree of PE crystallinity. Alignment of both the SWNT and PE was produced by melt fiber spinning and results show that the thermal conductivity increase with orientation.   

Controlling the Dispersion and Properties of Single-Walled Carbon Nanotube-Polymer Nanocomposite

Carbon nanotubes possess extraordinary electrical and mechanical properties. Dispersing nanotubes in a polymer matrix provides an effective way to exploit these extraordinary properties, however this has been difficult to achieve due to strong inter tube interaction. Previous work in our lab has shown that optimized hydrogen bonding between a copolymer and an anisotropic filler enhances miscibility of the mixture. Controlling the extent of hydrogen bonding between a copolymer and carbon nanotube gives a well-dispersed nanocomposite for both single and multi-wall carbon nanontubes as indicated by Raman spectroscopy, dynamic mechanical analysis, electrical conductivity, optical microscopy and SEM. The amount of hydrogen bond interactions in the nanocomposite is controlled by varying the copolymer composition. The results are critical in understanding interfacial phenomenon in polymer and nanocomposites and provide a mechanism to design materials with tunable properties.   

From Carbon Nanotube Dispersion to Composite Nanofibers

Composite polymer nanofibers containing single-walled carbon nanotubes (SWCNT) are fabricated by electrospinning. We describe the path from dispersing individual SWCNTs or thin bundles in water using amphiphilic polymers, through a structural characterization of the polymer conformation in the SWCNT/polymer hybrid to the characteristics of the electrospun composite nanofibers. An alternating copolymer of styrene and sodium maleate (PSSty) and gum arabic (GA)-a highly branched natural polysaccharide were successfully used to produce stable aqueous dispersions. Measurements of small angle neutron scattering (SANS) show that both polymers form a thick corona of adsorbed coils on the nanotubes. The large coils introduce a significant steric barrier stabilizing the dispersions, in addition to electrostatic repulsion by charged groups. The composite nanofibers showed good distribution and alignment of the SWCNTs in the poly(ethylene oxide) (PEO) nanfubers, as revealed by transmission electron microscopy. X-ray diffraction demonstrated a high degree of orientation of the PEO crystals in the electrospun nanofibers. Enhanced tensile properties were achieved due to the high degree of alignment of both nanotubes and polymer crystals, and a strong interface, especially with PSSty. The morphology and possible applications of these composite nanofibers will be discussed.   

Enhanced alignment of Multi-Walled Carbon Nanotubes in Electrospun PS/PMMA Polymer Blends

Electrospinning has been recognized as an efficient technique to obtain ultrafine polymeric nanofibers. A variety of polymers have been successfully electospun into continuous polymeric fibers having micron to submicron diameters. In this work, multiwalled carbon nanotubes are incorporated into electrospun nanofibers which consist of PS/PMMA polymer blends during the electrospinning process. It was observed that multiwalled carbon nanotubes(MWNT) are linearly oriented along the fiber axis in which internally co-continuous phase morphology of the PMMA is formed in the PS matrix This highly oriented MWNT structures in the fiber are characterized by transmission electron microscopy. The internal morphology of polymer blends are determined by selectively etching the PMMA with a good solvent using scanning electron microscopy, and staining the PS portion with osmium tetra teroxide.   

Processing Phase Diagram of Polymer Carbon-Nanotube Composites

The `phase diagram' of a model polymer/carbon-nanotube melt composite is measured as a function of concentration, shear stress and geometrical confinement. We observe a hierarchy of flow-induced structure, including dispersed (para) nematics, a variety of aggregates, and `jammed' fractal networks. By applying simple scaling arguments from polymer physics to rigid-rod gels, our data suggest that the portion of the network responsible for the arrest of flow is more diffuse than the full elastic network, akin to `force chains' in granular media.   

Measurements of particle orientation in simple shear and channel flows of polypropylene/clay nanocomposites

We report studies of flow-induced orientation in dispersions of organically modified montmorillonite clay in polypropylene. The nanocomposite samples were prepared using two methods. Melt blending in a twin-screw extruder led to intercalated samples in which the layered structure of the clay remains intact. An additional step of solid-state shear pulverization leads to samples with a higher degree of exfoliation of individual clay sheets. In situ x-ray scattering was used to probe particle orientation in steady shear using an annular cone and plate shear cell which provides information about particle orientation in the flow-gradient plane. The more highly exfoliated pulverized sample shows significantly lower orientation than the intercalated melt-blended sample. Both samples were also studied in extrusion-fed channel flows. In slit-channel geometries, the dominant shear rate direction is parallel to the x-ray beam, allowing information about orientation in the flow- vorticity plane to be obtained. In fact, little scattering was observed in these configurations, confirming the tendency of clay particles to `lie down' in the shear flow. Superposition of extension via contractions or expansions in slit-channel flows did not reorient particles sufficiently to bring them `into view' in these geometries.   

Rheology of Non-dilute Polystyrene/Cloisite/Toluene Solutions

We have previously described a simple model of spin casting for polymer/clay nanocomposite films in which the viscosity of the polymer solution at low solvent concentration is a critical parameter. We have therefore examined the shear dependent viscosity of polystyrene/Cloisite-6A/toluene solutions over a wide range of weight fractions, and for various molecular weights. The data is well described by the Carreau model $\eta -\eta _\infty =(\eta -\eta _0 )(1+\lambda ^2\dot {\gamma }^2)^{-N}$, where the parameters in the model show a clear dependence on the clay/PS ratio. We will discuss the trends observed in the viscosity data, and their impact on the uniformity of spin cast films.   

Melt rheology studies of polymer chain dynamics in the presence of nanofillers

We will present results from recent and ongoing melt rheology runs on a series of nanocomposites with polystyrene as the matrix and silica nanoparticles (10-15 nm in diameter) as the fillers. A number of combinations of the matrix molecular weight and nanofiller concentrations are being studied to elucidate the trends in the system dynamics for a number of different concentrations of the nanofillers in a given matrix molecular weight and for a given nanofiller concentration over a range of matrix molecular weights. Melt rheology runs are being performed also on the pure polymers for comparison purposes. This elucidates the long term relaxation dynamics of a matrix polymer in the presence of nanofillers; when compared with the corresponding pure polymer in the terminal region, the nanocomposite shows a lower G$'$ slope indicating a slowing down of the relaxation dynamics in the presence of the nanoparticles.   

Polymer nanocomposites: permeability, chain dynamics, mechanical properties

Polymer nanocomposites based on dispersion of surfactant treated expandable smectite clays such as montmorillonite layered silicates (MLS) have shown promise as organic-inorganic hybrids with the potential to improve barrier properties. Separately, flexible displays based on plastic substrates have reduced lifetimes tied to the low barrier properties. While there has been a general attribution of improved barrier properties to the tortuous path, this does not consider the influence the introduction of a secondary filler has on the morphology of the host polymer. Here we examine the influence of MLS nanoplatelets on the barrier properties and chain dynamics of polymers. We investigate the potential for host polymer modification by comparing two crystallizable polymers nylon and PET and resulting well dispersed nanocomposites. We study mechanical, cyclic fatigue and permeability of films. Permeability of the biaxially stretched film and when the film undergoes fatigue of 50 and 10000 cycles are also measured. Chain dynamics were modeled based on the Burger model fit to creep-recovery data. A systematic approach to predict the permeability considering amorphous, crystalline and MLS content and comparison with experimental values were done. We also conducted water absorption measurements to highlight the water absorption differences in the two polymers. Dimensional stability of PET was studied by measuring coefficient of thermal expansion of thin film on Si substrate by ellipsometry method.   

Session P31: Biopolymers: Molecules, Solutions and Networks I

Nonlinear elasticity of semiflexible polymer networks

Networks of filamentous proteins play a crucial role in cell mechanics. These cytoskeletal networks, together with various crosslinking and other associated proteins largely determine the (visco)elastic response of cells. Such biopolymers have also provided new insights into basic aspects of polymer physics. In contrast with conventional polymer materials, the response of these networks is highly non-linear, and their rheological properties can be tuned with small changes in density and local network connectivity. We discuss recent theoretical and experimental efforts to understand these essential materials of the cell.   

RNA gels with negative Poisson ratio

We present a simple model for the elastic properties of very large single-stranded RNA molecules linked by partial complementary pairing, such as a viral RNA genome in solution. It shown that the sign of Poisson's Ratio is determined by the convexity of the force-extension curve of single-stranded RNA. The implications of negative Poisson Ratio's for viral genome encapsidation will be discussed.   

DNA intercalation by ethidium bromide: A quantitative binding study using DNA stretching and force-induced melting

The interactions between single DNA molecules and the non- covalent binding agent ethidium bromide are investigated using an optical tweezers instrument and the effects of this intercalator on the structure and mechanical stability of DNA molecules are quantitatively analyzed using our model of force- induced melting. The DNA force-extension cycles in the presence and absence of drug are recorded. It is found that the drug binds preferentially to double-stranded DNA and stabilizes the double helix. There is clear evidence of the force induced melting transition at low concentrations of drug, while at higher concentrations the drug is able to prevent the melting transition. The DNA contour length is obtained as a function of ligand concentration directly from the stretching curves. From this data we obtain the complete ethidium bromide dsDNA binding isotherm, which is used to find the binding constant and the binding site size of the intercalator. Out data also allows us to quantify directly the effect of ethidium bromide on the free energy of the helix-coil transition in dsDNA. This single molecule study brings new insights into the molecular mechanisms which drive drug-DNA complex formation.   

Stretching DNA by a Constant Field

We consider the problem of stretching DNA by a constant field, such as an electric field or a hydrodynamic flow field. We obtain analytical expressions for the elongation of DNA under both weak and strong applied fields, in two and three dimensions. In the weak field limit we consider the effect of self-avoidance, which leads to a 9-fold enhancement of the average end-to-end distance over the result obtained when self-avoidance is ignored in two dimensions, and 3-fold increase in three dimensions. In the strong stretching regime we obtain the exact force-extension relation by mapping the problem to the Schr\"{o}dinger equation for a simple harmonic oscillator in a time dependent potential. We use our theoretical results to comment on the experiment of Maier {\em et al.} \footnote{B. Maier, U. Seifert, and J. O. R\"{a}dler, {\it Europhys. Lett.}, {\bf 60}, 622 (2002).} on DNA adsorbed on a lipid bilayer in the presence of an in-plane electric field. In particular, we find that their estimate for the effective charge density of the DNA molecule, made on the basis of an approximate theory, requires significant corrections in light of our calculations. This work was supported by the NSF through grants DMR-9984471 and DMR-0403997. JK is a Cottrell Scholar of Research Corporation.   

Beyond Wormlike Chain: An effective theory of mesoscale DNA mechanics

The wormlike chain model has come to dominate physical discussions of DNA conformation, due in part to its spectacular success in modeling the force-extension of single molecules. This model rests upon an assumption of linear bending elasticity in a rod-like polymer chain. But we show that force-extension relations are actually rather insensitive to the details of the stress-strain relation, and in particular do not test the linear-elasticity hypothesis. A renormalization-group flow toward the linear-elastic model hides deviations from linear elasticity on length scales larger than a few helical turns. Our results can be seen as validating the wormlike chain model for the long-scale (small curvature) regime of DNA mechanics, but many important biological processes such as DNA looping operate on shorter scales. We will show how recent experiments on DNA cyclization, and DNA contour analysis by scanning force microscopy, imply a stress-strain relation on intermediate length scales that is quite different from the simple linear form [cond-mat/0409003, Phys Rev E in press]. This revision of DNA mechanics has implications for the structure and dynamics of DNA loops essential in gene regulation.   

Semiflexible Chain Networks Formed via Self-Assembly of Beta-Hairpin Molecules

We present experimental results from a de novo designed oligopeptide that intermolecularly self-assembles into rigid hydrogel networks after an intramolecular folding event. The effect of ionic strength and beta hairpin peptide strand length on beta-sheet formation, self-assembly and resultant rheological properties were studied. The peptide molecules are locally amphiphilic with two linear strands of alternating hydrophobic valine and hydrophilic lysine amino acids flanking a central turn sequence. The beta-sheet formation of 24, 20, 16 and 12 amino acid long beta-hairpin molecules were studied by CD spectroscopy. The network properties and the nanostructure of the hydrogels were studied by rheology, TEM and SANS. The hydrogel network is composed of semiflexible fibrillar assemblies with viscoelastic behavior that follows the theoretical prediction for heavily crosslinked,semi-flexible polymer networks. SANS results show that the cross-sectional diameter of the fibrils, and thus, the bending modulus of the chains can be varied by changing the number of amino acids of strands of the molecules. Rheological measurements reveal that rigidity, creep and relaxation behavior of the hydrogels vary with the magnitude of stimulus and with the cross-section diameter of the chains.   

The Rheological Properties of the Biopolymers in Synovial Fluid

The polyelectrolyte hyaluronic acid (HA, hyaluronan), its interactions with anti-inflammatory drugs and other biopolymers, and its role in synovial fluid are being studied. We are investigating the rheological properties of sodium hyaluronate (NaHA) solutions and an experimental model of synovial fluid (comprised of NaHA, and the plasma proteins albumin and $\gamma $-globulins). Steady shear measurements on bovine synovial fluid and the synovial fluid model indicate that the fluids are highly viscoeleastic and rheopectic (stress increases with time under steady shear). In addition, the influence of anti-inflammatory agents on these solutions is being explored. Initial results indicate that D-penicillamine and hydroxychloroquine affect the rheology of the synovial fluid model and its components. The potential implications of these results will be discussed.   

The nonlinear elasticity of alpha helical polypeptides: Analytical and Monte Carlo studies

We study a minimal extension of the worm-like chain model to describe polypeptides having alpha-helical secondary structure. In this model presence/absence of secondary structure enters as a scalar variable that controls the local chain bending modulus. Using this model we analytically compute the extensional compliance of an alpha-helix under tensile stress, the bending compliance of the molecule under externally imposed torques, and the nonlinear interaction of such torques and forces on the molecule. We find that, due to coupling of the ``internal'' secondary structure variables to the conformational degrees of freedom of the polymer, the molecule has a highly nonlinear response to applied stress and bending torques. In particular we demonstrate a sharp lengthening transition under applied force and a buckling transition under applied torque. We use perturbative calculations and a mean field analysis to obtain these results. We also carry out Monte Carlo simulations of this model. The numerical results agree well with the mean-field and perturbative calculations where they are expected to do so. The Monte Carlo simulations allow us to examine the response of the chain to large forces and torques where the perturbative approaches fail. In addition we extend our mean-field analysis by studying the fluctuation dominated regime at the force-induced denaturation transition.   

Persistency of single-stranded DNA: the interplay between base sequences and base stacking

The chain persistency of single-stranded (ss) DNA at a high-salt limit mainly arises from the so called base-stacking interaction between consecutive bases along the strand; stacking is appreciable only for purine-purine (e.g., $A$-$A$) and purine- pyrimidine stacks (e.g., $A$-$T$), stiffening the strand, but is negligible for pyrimidine stacks ({\i.e.}, $T$-$T$, $T$-$C$, and $C$-$C$). We develop an exactly-solvable model for describing the stacking-induced persistency of heterogeneous ssDNA. Using this, we study the interplay between the heterogeneity of sequences and base stacking in determining the persistency. Our results demonstrate how the sequence information can influence the conformational properties of ssDNA.   

Unzipping DNA from the condensed globule state--effects of unraveling

We study theoretically the unzipping of a double stranded DNA from a condensed globule state by an external force. At constant force, we find that the double stranded DNA unzips an at critical force F$_{c}$ and the number of unzipped monomers M goes as M$\sim $(F$_{c}$-F)$^{-3}$, for both the homogeneous and heterogeneous double stranded DNA sequence. This is different from the case of unzipping from an extended coil state in which the number of unzipped monomers M goes as M$\sim $( F$_{c}$-F)$^{\chi }$, where the exponent $\chi $ is either 1 or 2 depending on whether the double stranded DNA sequence is homogeneous or heterogeneous respectively. In the case of unzipping at constant extension, we find that for a double stranded DNA with a very large number N of base pairs, the force remains almost constant as a function of the extension, before the unraveling transition, at which the force drops abruptly to zero. Right at the unraveling transition, the number of base pairs remaining in the condensed globule state is still very large and goes as N$^{3/4}$, in agreement with theoretical predictions of the unraveling transition of polymers stretched by an external force.   

Distance measurement along DNA molecules using fluorecent quantum dots

To create and design better micro- and nanofluidic devices, we need to understand how macromolecules behave when squeezed by lateral barriers to create pseudo-two-dimensional confinement. We present experiments in which we visualize DNA molecules of varying sizes (2 kbp - 50 kbp) trapped in 10 micrometer wide slits, the slit height varying from the radius of gyration of the unconfined molecule (micrometer) down to 25 nm (half the persistence length of DNA). We present data on the diffusion coefficient and electrophoretic mobility (no electroosmotic flow) of SYBR-gold labeled DNA molecules as a function of slit height. Simultaneously, we have assessed the DNA conformation by examining molecules that are end-labeled with differently colored fluorescent quantum dots. By determining the distance between labels, we measure directly the end-to-end distance - a conformational measure much discussed but rarely measured. Using the same approach but turning the problem around, we determined if contour length can be estimated from visualization experiments. The answer to this question becomes important when the distance between specific binding sites on the DNA backbone must be measured. One such application, for example, is the determination of haplotypes (genetic variability due to blocks of single nucleotide polymorphisms (SNP)) in diploid individuals.   

AFM Imaging of F-actin Network Formation on a photopolymer surface

We investigated the network formation of cytoskeltal filamentous (F-) actin in the presence of divalent cations by atomic force microscopy using a novel protein immobilization technique. The F-actin network was immobilized on the surface of a unique nonionic photopolymer containing azo-dyes (azopolymer), which upon photo-irradiation deforms along the contour of the proteins thus physically immobilizes them. Two-dimensional F-actin networks were formed and immobilized by spotting F-actin solutions on the azopolymer surface, which was then irradiated using an array of blue light emitting diodes. The structure of the F-actin network, which consists of multiple X-, Y-, T-shaped junctions, was influenced by the concentration of the divalent cations in the spotting solution. We observed that the angle between two crossing F-actins at a junction decreases with increasing concentration of divalent cations. Above a certain ionic concentration, the cross-linked networks of F-actin transform into close-packed parallel rafts and bundles. The results show promise in the fabrication of two-dimensional aligned F-actin sheets.   

Microtubule Bundling and Shape Transitions

Microtubules (MTs) are hollow cylindrical polymers composed of heterodimers of the protein tubulin that align end-to-end in the MT wall, forming linear protofilaments that interact laterally. Placing MTs under osmotic pressure causes them to reversibly buckle to a noncircular shape and pack into rectangular bundles at a critical osmotic pressure; further increases in pressure continue to distort MTs elastically. At higher osmotic pressures stressing polymers may be forced into the MT lumen causing the MTs to revert to a circle cross-section and pack into hexagonal bundles. This SAXRD-osmotic stress study provides a probe of the inter-protofilament bond strength and gives insight into the mechanisms by which microtubule associated proteins and the cancer chemotherapeutic drug Taxol stabilize MTs. We present further measurements of the mechanical properties of MT walls, MT-MT interactions, and the entry of polymers into the microtubule lumen. Supported by NSF DMR- 0203755, NIH GM-59288 and NS-13560, and CTS-0103516. SSRL is supported by the U.S. DOE.   

Higher Order Assembly of Microtubules by Counter-ions

Cellular factors tightly regulate the architecture of bundles of filamentous cytoskeletal proteins, giving rise to assemblies with distinct morphologies and physical properties in vivo, but it is unclear how the microscopic interactions between filaments result in the observed structures. We study a model system consisting of microtubules (MTs) and multivalent cations, and demonstrate the formation of distinct bundle phases. We have characterized the structure of these self-assemblies of MTs from the nanoscale to the mesoscale using synchrotron x-ray scattering and diffraction, video enhanced DIC and fluorescence microscopy, and electron microscopy. Tightly packed hexagonal bundles with controllable diameters are observed for large tri-, tetra-, and pentavalent counterions. Unexpectedly, in the presence of small divalent cations, we have discovered a living necklace bundle phase, comprised of dynamical assemblies of MT nematic membranes with linear, branched, and loop topologies. Supported by NSF DMR- 0203755, NIH GM-59288 and NS-13560, and CTS-0103516. SSRL is supported by the U.S. DOE.   

Session P32: Computational Methods Classical and Quantum Monte Carlo

Quantum Monte Carlo Study on NiO crystal

Quantum Monte Carlo (QMC) calculations using the variational (VMC) and diffusion (DMC) methods is performed on NiO crystal with gaussian basis and pseudo potentials. We report calculations of energy-volume plots obtained by VMC and DMC with 2*2*2 (16 ions) and 4*4*4 (128 ions) simulation cells. In order to obtain smooth dependences careful optimizations of the basis set at each lattice constant turned out to be indispensable, as well as the larger (4*4*4) simulation cell. The cohesive energy obtained by VMC (without Jastrow function) is around 6.6 eV, while DMC gives around 9.3 eV (near to the experimental value as 9.5 eV, though the present QMC is assuming ferromagnetic ordering).   

Interpretation of Hund's multiplicity rule for the atomic systems

We have studied Hund's multiplicity rule for the carbon atom using quantum Monte Carlo methods[1]. Our calculations give a high-level description of electron correlation and satisfy the virial theorem to high accuracy. This allows us to obtain accurate and reliable values for each of the energy terms and therefore to give a convincing explanation of the mechanism by which Hund's rule operates in carbon. We obtain the following results: (1) the energy gain in the triplet with respect to the singlet state is due to the greater electron-nucleus attraction in the higher spin state, and (2) the electron-electron repulsion in the triplet is greater than that in the singlet, in accordance with Hartree-Fock results and studies including correlation. Although our main topic is the carbon atom, we would also like to show our current results of the nitrogen atom.[1]K. Hongo, \textit{et al}., J. Chem. Phys. \textbf{121}, 7144 (2004).   

Pfaffian wavefunctions with pairing orbitals for electronic structure quantum Monte Carlo

The trial wavefunction nodal structure determines the accuracy of electronic structure calculations by the fixed-node quantum Monte Carlo (QMC). Wavefunctions based on Hatree-Fock (HF) or multi-reference HF provide about 95\% of correlation energy in real systems such as molecules and solids. On the other hand, antisymmetrized product of pairing orbitals (geminals) is a variationally richer form which can describe effects absent in HF such as electron singlet and triplet pairing. One of such forms is the BCS pairing wavefunction with singlet pairing which can be expressed as a determinant. An extension of BCS includes also triplet pairing and therefore requires a pfaffian. We explore the variational freedom of pfaffian wavefunctions and evaluate improvements of fixed-node QMC energies for cases of atomic, molecular and solid systems.   

Nodes of fermionic wavefunctions: coordinate transformations and topologies

We study fermion nodes for both spin-polarized and spin-unpolarized states of a few-electron ions and molecules with $s,p,d$ one-particle orbitals. We find exact nodes for some cases of two electron atomic and molecular states and also the first exact node for the three-electron atomic system in $^4S(p^3)$ state using appropriate coordinate maps and wavefunction symmetries. We analyze the cases of nodes for larger number of electrons in the Hartree-Fock approximation and for some cases we find transformations for projecting the high-dimensional node manifolds into 3D space. The node topologies and other properties are studied using these projections. We also propose a general coordinate transformation as an extension of Feynman-Cohen backflow coordinates to both simplify the nodal description and as a new variational freedom for quantum Monte Carlo trial wavefunctions.   

Dielectric Response of Periodic Systems from Quantum Monte Carlo

We introduce a novel approach to study the response of periodic systems to finite homogeneous electric fields using the diffusion Quantum Monte Carlo method. The interaction with the electric field is expressed through a generalized many-body electric-enthalpy functional; a Hermitian local potential is then constructed that determines the evolution towards the ground state. This local potential depends self-consistently on the Berry-phase polarization, and is evolved ``on-the-fly'' in the course of the simulation, with the polarization operator evaluated using forward-walking. To validate this approach we calculated the dielectric susceptibility of simple molecular chains, greatly over-estimated by standard density-functional approaches, and found good agreement with the results obtained with correlated quantum-chemistry calculations.   

Coupled Electron-Ion Monte Carlo Study of Hydrogen

We present details of the Coupled Electron-Ion Monte Carlo method (CEIMC) [1,2] applied to the problem of the equation of state of pure hydrogen. The aim is to develop a method that can predict state information outside the range of temperatures and pressures that are accessible with other existing methods, such as PIMC. \\ The CEIMC method centers on exploring the configuration space of the hydrogen nuclei (classical or quantum path integrals) using a modified Metropolis algorithm, with configurational energy differences computed from Born-Oppenheimer energies. Energy differences are computed with VMC or Reptation quantum Monte Carlo, both of which supply unbiased estimates of energy differences, the latter within a projector framework. New developments include a fast band-structure calculation for the trial function which should improve the localization of molecule-atom phase transitions. \\ (1) D. Ceperley, M. Dewing and C. Pierleoni, in Bridging Time Scales: Molecular Simulations for the Next Decade, eds. P. Nielaba, M. Mareschal and G. Ciccotti, Springer-Verlag, pgs. 473-500 (2002). \\ (2) C. Pierleoni, D. M. Ceperley and M. Holzmann, Phys. Rev. Lett. 93, 146402 (2004)   

Auxiliary Field Quantum Monte Carlo in Continuum Systems

The auxiliary field Quantum Monte Carlo method allows Monte Carlo to be performed in any basis. This is accomplished by using the Hubbard-Stratonavich transformation to transform two body interactions into an integral over one body interactions. In practice this method has been difficult to use because while exact, it suffered from a phase problem more severe than the sign problem encountered in Diffusion Monte Carlo. Recent work has suggested a phase free approximation that allows this phase problem to be overcome while sacrificing the exact nature of the method$^1$. We have implemented the auxiliary field Quantum Monte Carlo algorithm with the phase free approximation in a plane wave basis. The results of this code for the total energy of jellium and silicon are compared to previous work to assess the accuracy of the method and its approximation. We also discuss results obtained for the energy of jellium with a gap caused by modifying the kinetic energy operator. Results from this study may prove useful in developing more accurate functionals for density functional calculations$^2$. We conclude with a brief discussion of the strengths and weaknesses of the auxiliary field method within the phase free approximation. \begin{itemize} \item[{[1]}] S. Zhang, and H. Krakauer. Phys. Rev. Lett. {\bf 90}, 136401 (2003) \item[{[2]}] C. Gutle, et. al., Int. J. Quant. Chem. {\bf 75}, 885 (1999) \end{itemize}   

Quantum Monte Carlo Study of Composite-Fermions in Quantum Dots

Composite-fermion wave functions, projected onto the lowest Landau level, provide accurate wave functions for quantum dots in the limit of strong magnetic fields. We show that the range of validity of these wave functions can be greatly extended to smaller magnetic fields by incorporating Landau level mixing effects by multiplying them with a Jastrow factor, optimized using the variance minimization method. The energy and other expectation values can be further improved by projecting the wave functions onto the ground state using diffusion Monte Carlo within the fixed-phase approximation. Energies for 6-electron system are compared to energies obtained by exact diagonalization within 3 Landau levels. Excellent agreement between the two methods is obtained. We then apply our method to a 15-electron system, far beyond the capabilities of the exact diagonalization method, to study ground state properties as the magnetic field is varied.   

Quantum Monte Carlo examines accuracy of density functional approximations for defects and phase transformations in silicon

Silicon displays a variety of interstitial defects limiting device fabrication and performance and shows at least twelve crystallographic phases under pressure. While DFT-determined structures are reliable, defect energies and phase transformation pressures are sensitive to the specific exchange-correlation functional. Diffusion Monte Carlo calculations for silicon defects and phases test the accuracy of the current density-functional approximations LDA, PW91, PBE, and TPSS. Diffusion Monte Carlo predicts the correct cohesive energy of the diamond structure, however, the pressure for the transition to beta-tin is larger than in experiments. The transformation is sensitive to anisotropic stresses; an anisotropy of 2-3 GPa lowers the prediction to agree with experiment. Diffusion Monte Carlo for high-pressure Si phases and interstitial defect clusters shows that relative to diamond Si the energies of phases and defects are underestimated by DFT.   

Auxiliary Field Quantum Monte Carlo Study of Ground State Properties of Atoms and Molecules

We apply a recently developed quantum Monte Carlo (QMC) method \footnote{Shiwei Zhang, Henry Krakauer, Phys. Rev. Lett. {\bf 90}. 136401 (2003).} to calculate the ground state properties of several atoms and molecules. The QMC method projects the many-body ground state from a trial state by random walks in the space of Slater determants. The Hubbard-Stratonovich transformation is employed to decouple the Coulomb interaction between electrons. A trial wave function $|\Psi_T\rangle$ is used in the approximation to control the phase problem in QMC. We also carry out Hartree-Fock (HF) and density functional theory (with the local density approximation (LDA)) calculations. The generated single Slater determinant wave functions are used as $|\Psi_T\rangle$ in QMC. The dissociation and ionization energies are calculated for Aluminum, Silicon, Phosphorous, Sulfur, Chlorine and Arsenic atoms and molecules. The results are in good agreement with experimental values.   

Study of TiO and MnO using auxiliary field quantum Monte Carlo

We study the transition metal oxide molecules TiO and MnO using the recently developed auxiliary field quantum Monte Carlo approach [1]. This method maps the interacting many-body problem into a linear combination of non-interacting problems using a complex Hubbard-Stratonovich transformation, and controls the phase/sign problem using a trial wave function. It employs a random walk approach in Slater determinant space to project the ground state of the system, and uses much of the same machinery as density functional theory such as single particle basis and non-local pseudopotentials. In our calculations, we used a single Slater determinant trial wave function obtained from a density functional calculation, with no further optimization. The calculated dissociation energies are in good agreement with experiments. These together with previous results show the robustness of the method for studying sp- as well as d-bonded atoms, and molecules. Calculations of other observables and correlation functions will also be discussed. [1] S. Zhang and H. Krakauer, Phys. Rev. Lett. 90, 126401 (2003).   

Monte Carlo Calculations of Finite Temperature Transition Rates in Nickel

Experiments reveal that self diffusion for many fcc and bcc metals is enhanced at high temperatures, thereby deviating from expected Arrhenius behavior. For nickel it is expected the deviation is caused by the influence of the di-vacancy mechanism in addition to the dominant single vacancy mechanism, but zero temperature \emph{ab initio} calculations suggest this is not the case. We investigate finite temperature effects due to anharmonicity in the potential for the single vacancy mechanism in nickel using a classical EAM potential. We modify the recently developed Wang-Landau Monte Carlo method to extract the vibrational density of states along a continuous spatial reaction coordinate. This allows the calculation of the transition rate within full transition state theory at finite temperature. We find our results agree with the harmonic approximation at low temperature and we compare our high temperature results with experiment.   

Numerical Interpolation of Orbitals in Periodic Systems for Diffusion Monte Carlo Calculations

Diffusion Monte Carlo methods provide accurate energies for complex materials, however, the algorithms are computationally intensive. Representing the orbitals of the Slater determinant numerically with splines reduces the time scaling from O($N^3$) to O($N^2$) \footnote{A. J. Williamson, R. Q. Hood, and J. C. Grossman. PRL \textbf{87}, 246406 (2001).}. We compare memory and time requirements and the accuracy dependence on the number of grid points for cubic spline and Lagrange interpolation schemes in periodic systems. Both interpolation schemes have a small prefactor, providing speedup even for small systems. For example, in bulk silicon with 256 electrons, Lagrange interpolation reduces the computation time by a factor of 70. We are currently working on the implementation of different splines routines.   

A rejection-free Monte Carlo method for the hard-disk system

We construct a rejection-free Monte Carlo method for the hard-disk system. Rejection-free Monte Carlo methods preserve the time-evolution behavior of the standard Monte Carlo method, and it is confirmed for our method by observing nonequilibrium relaxations of a bond-orientational order parameter. The rejection-free method obtains much better performance than the standard method at high densities with new optimization methods to calculate a rejection probability and to update the system. This method should allow an efficient study of the dynamics of two-dimensional solids at high density.   

Optimal Ensemble Monte Carlo Simulations: Application to dense Lennard-Jones fluids

Broad-histogram Monte Carlo simulations directly calculate the density of states of a (quantum) system by sampling broad energy ranges and thereby give access to thermodynamic properties. While flat-histogram methods suffer from a critical slowing down, we have shown that the simulation of an optimized ensemble substantially speeds up equilibration and can efficiently overcome the entropic barriers which cause the slowdown [1]. In this talk, we present recent applications of the optimal ensemble method to dense Lennard-Jones fluids and particle-solvent models. Based on measurements of the local diffusivity an optimal ensemble that maximizes round-trip rates in radial coordinates can be simulated and the potential of mean force can be determined to high precision. The optimized histogram of the radial random walk reveals clear signatures of the intermediate transitions between shells of the dense fluid. [1] S. Trebst, D. A. Huse, M. Troyer, Phys. Rev. E {\bf 70}, 046701 (2004)   

Session P33: Quantum Mechanical Properties

Teleportation of electronic many-qubit states via single photons

We propose a teleportation scheme that relies only on single- photon measurements and Faraday rotation, for teleportation of many-qubit entangled states stored in the electron spins of a quantum dot system. The interaction between a photon and the two electron spins, via Faraday rotation in microcavities, establishes Greenberger-Horne-Zeilinger entanglement in the spin-photon-spin system. The appropriate single-qubit measurements, and the communication of two classical bits, produce teleportation. This scheme provides the essential link between spintronic and photonic quantum information devices by permitting quantum information to be exchanged between them.   

Coherence-decoherence crossover with subohmic bath

We study the spin-boson model with a subohmic bosonic bath using variational method. The coherence-decoherence crossover for a finite two-level splitting is analyzed. The results for the crossover as a function of temperature and strength of the bath will be discussed.   

A quantum measurement of the double barrier junction (DBJ) qubit

A quantum measurement on a double barrier junction (DBJ) qubit performed by a coupled by an SIS junction is studied. The DBJ qubit state $\left| s\right\rangle$ is monitored by entangling of it with the SIS ``meter'' state $\left| m\right\rangle$. The the coupling strength $J$ between the ABS qubit and the meter is controlled by a Josephson flux transistor. The efficiency of the measurement versus the coupling strength between DBJ and SIS is computed. It is found that the information gained in the quantum measurement depends on $J$ and on the ratio $\mu =\varepsilon_m /\varepsilon_q$ (where $\varepsilon_q$ and $\varepsilon_m$ are the qubit and the meter level splitting correspondingly. Using a linear circuit approach we also compute the spectral density $J_{eff}^s (\omega )$ of the qubit noise induced during the measurement. An evaluation of noise and decoherence made shows that they depend on the characteristic times of the equivalent circuit of the qubit system, and thus may be optimized to improve the overal qubit performance.   

Quantum measurement of a Cooper-Pair Box using a nanomechanical resonator

It has been shown by Irish and Schwab that the resonant frequency of a nanomechanical resonator depends on the quantum state of the Cooper-Pair box (CPB) coupled to it. In addition, the energy splitting of the CPB is shifted as a result of the interaction with the resonator. We explore the possibility of using the nanomechanical resonator to detect the state of a CPB at degeneracy. We also discuss the possibility of making quantum nondemolition measurements of Fock states of the nanomechanical resonator by probing the energy levels of the CPB. We will describe our experimental progress where we are currently working with a 10 MHz resonator which is capacitively coupled to the CPB. The resonator is read out using our newly developed impedance matched, capacitive technique.   

Noise and Back-action in Nanomechanical Resonators

We have recently demonstrated that a superconducting Single Electron Transistors (SSET) capacitively coupled to a nanomechanical resonator, can be used as a nearly-quantum limited position detector. The ultimate sensitivity of this scheme is limited both by forward-coupled charge fluctuations on the SET island and back-acting electrostatic potential fluctuations which drive the resonator. In this talk, we will present our recent measurements of SET back-action on nano-resonators. In particular, we will discuss the back-coupled power, damping, and frequency of the resonance as a function of the coupling between the SET and the resonator and compare these results to theoretical predictions.   

Non-Markovian qubit dynamics in a thermal field bath: Relaxation, decoherence and entanglement

We have studied the non-Markovian dynamics of a qubit interacting with an electromagnetic field bath initially at finite temperature in the Jaynes-Cummings model. Unlike studies in which the bath is assumed to be fixed, we have included the dynamics of the bath, thus allowing for the coherent evolution of the combined qubit-bath system. In this way we can see the development of quantum correlations and entanglement between the system and its environment. The non-Markovian effects are illustrated in various quantities including the decoherence, relaxation, fidelity and von~Neumann entropy, which are compared to the Markovian results.   

Quantum Adiabatic Evolution Algorithm via combinatorial landscapes

We analyze the performance of the Quantum Adiabatic Evolution algorithm (QAEA) on a variant of Satisfiability problem for an ensemble of random graphs parametrized by the ratio of clauses to variables, $g=M/N$. We introduce a set of macroscopic parameters (landscapes) and put forward an ansatz of universality for random bit flips. We then formulate the problem of finding the smallest eigenvalue and the excitation gap as a statistical mechanics problem. We use the so-called annealing approximation with a refinement that a finite set of macroscopic variables (versus only energy) is used, and are able to show the existence of a dynamic threshold $g=g_d$, beyond which QAEA should take an exponentially long time to find a solution. We compare the results for extended and simplified sets of landscapes and provide numerical evidence in support of our universality ansatz. We have been able to map the ensemble of random graphs onto another ensemble with fluctuations significantly reduced. This enabled us to obtain tight upper bounds on satisfiability transition and to recompute the dynamical transition using the extended set of landscapes.   

Effects of finite bandwidth and delay on Bayesian quantum feedback of a qubit

We analyze the effect of various imperfections on the performance of the Bayesian quantum feedback loop designed to maintain quantum coherent oscillations in a solid-state ``charge'' qubit for an arbitrary long time. For the feedback operation the qubit state is continuously monitored using the information provided by the noisy output of a weakly coupled detector (quantum point contact or single-electron transistor); this information is taken into account using the quantum Bayesian equations. Finite signal bandwidth reduces the monitoring accuracy and affects performance of the feedback; we study this effect by Monte Carlo simulation of the quantum measurement process. We also analyze the reduction of feedback fidelity due to additional time delay in the control loop. The simulations also take into account finite quantum efficiency of the detector and possible energy asymmetry of the qubit.   

Experimental Evidence for Violation of Bohr's Principle of Complementarity

We have implemented a novel double-slit which-way experiment which raises interesting questions of interpretation. Coherent laser light passes through a dual pinhole, and the resulting interference pattern is passed through a converging lens which produces well-resolved images of the two pinholes, providing full which-way information. A series of thin wires are then placed at the minima of the interference pattern upstream of the lens. No reduction in the total flux or resolution of the images is found, providing evidence for coexistence of perfect interference and which-way information in the same experiment, contrary to the common readings of Bohr's principle of complementarity. Implications of the experiment for the measurement theory are also briefly discussed.   

Interference and Trajectories

A trajectory representation of interference is presented. The trajectory representation manifests destructive interference and reinforcement as a dwell-time phenomenon of the trajectory. Evoking a probability density is unnecessary.   

Inference of Schr\"odinger's Equation from Classical Wave Mechanics[1]

A localized oscillatory point charge $q$ generates in a one-dimensional box electromagnetic waves which may be generally described by monochromatic plane waves $\{\varphi_i=C_K e^{i(KX- \Omega T+ \alpha_i)}\}$ of angular frequency $\Omega$, wavevector $K=\Omega/c$, velocity (of light) $c$, and initial phases $\{\alpha_i\}$. $q$ and $\{\varphi_i\}$ as a whole is here taken as a particle, which total energy ${\sf E}$ and mass $M$ are given by the basic equations ${\sf E}=\hbar \Omega=M c^2 $, $2\pi\hbar$ being Planck constant. (For example, $q=-e$ and $M=511$ keV give an electron.) $\{\varphi_i\}$ as incident and reflected and those from the charge as reflected in the box superimpose into a total wave $\psi=\sum \varphi_i$ that, as with $\varphi_i$, obeys the classical wave equation (CWE): $c^2 \frac{d^2\psi}{d X^2}= \frac{d^2\psi}{d T^2}$. If now the particle is traveling at velocity $v$, in a potential field $V=0 $ here (see Ref. 2004b for $V\ne 0$), then $\{\varphi_i'\}$ are Doppler effected and form a total wave $\psi'={\mit\Phi} {\mit\Psi}$, with ${\mit\Psi}= C \sin(K_d X)e^{i \Omega_d T}$ being the envelope about a beat wave and identifiable as de Broglie wave of angular frequency $\Omega_d= \Omega (v/c)^2$, and ${\mit \Phi}$ an undisplaced monochromatic wave. Using $\psi'$ in CWE gives upon decomposition a separate equation describing the particle dynamics, $-\frac{\hbar^2}{2M} \frac{\partial^2 {\mit\Psi}(X,T)}{\partial X^2}=i\hbar\frac {\partial {\mit\Psi}(X,T)}{\partial T}$, which is equivalent to Schr\"odinger's equation. \\[0cm] [1] J. X. Zheng-Johansson and P-I. Johansson, arXiv:Physics/0411134 (2004a); "Unification of Classical, Quantum and Relativistic Mechanics and the Four Forces" (in printing, 2004b).   

Session P34: Dynamics in Condensed Phase III

Langmuir Prize Talk: Pathways to forming glass: bubbles in space-time

This lecture describes a perspective of glass-forming materials that I have helped Juan P. Garrahan develop. It based upon the structure of trajectory space. Illustrations of this perspective are most often drawn from so-called ``facilitated'' or ``kinetically constrained'' lattice models. With these models, glassy dynamics emerges from the metric or geometrical constraints for molecular motion. More detailed atomistic models, and presumably natural glass formers as well, behave similarly. In the d+1 dimensions of trajectory space, one finds order-disorder phenomena that can be organized according to scaling and universality classes. Various predictions from this viewpoint, some yet to be verified experimentally, will be discussed.   

Oxidative Damage to DNA

The initial stage of the reaction of a guanine radical cation with water is investigated using ab-initio methods coupled with molecular mechanics. We explore the influence of the sugar- phosphate backbode, the complementary cytosine base, the adjacent base pairs in the B-DNA double strand, the state of hydration, and finally the distribution of counter-ions on the barrier and reaction path leading to 8-oxo-G.   

Glassy Dynamics in the Immune System Prevents Auto-Immune Disorders

A model of protein evolution is introduced. Hierarchical structures of the protein sequences or modularities play an important role in the dynamics. Computer simulations of the dynamics show that different evolving mechanisms(DNA swapping + point mutation v.s. point mutation ) lead to different stable(metastable) states. From the immunological point of view, point mutation corresponding to the metastable state has the advantage of preventing auto-immune disorders. The energy of the equilibrium states is determined only by the dynamics and independent of the initial states. Differences in initial states leads to different times of reaching equilibrium, and the binding energy is linear to the difference. Analytical arguments will be given as well to explain these features.   

Novel Method for Selective Probing of Ground-State Rotational Dynamics of Solutes in Solvents

We present an optical pump/probe method that allows selective measurement of ground-state rotational dynamics of solutes in liquids. It is known that because of different solute-solvent interactions, a solute in different electronic states could have markedly different rotational dynamics in the same solvent. However, this state-specific rotational dynamics has not yet been fully explored. It is particularly difficult to measure that of the ground state since the probe often cannot distinguish responses from various molecular species (ground-state solute, excited-state solute, and solvent) present in the solution. We employ two successive pump pulses that are adjusted to create an optical linear dichroism arising from the orientational distribution of only the ground-state solute molecules, hence allowing direct measurement of the ground-state rotational dynamics of the solute. Application of the technique to a dye-solvent system shows a large difference between rotational diffusion rates of the ground state and the excited state of the dye molecules. This work was supported by National Science Foundation.   

Equilibrium structure and phase separation in lipid mixture from dissipative particle dynamics simulations

Amphiphilic lipid molecules self assemble in aqueous solutions to form various structures interesting to both biological and engineering applications. We use dissipative particle dynamics simulations to investigate the equilibruium structures of lipid assemblies and map out their phase diagram. In addition, we study the effect of phase separation in two-lipid mixtures and report the resulting shape deformation and transition between micelles, miceller disks, miceller cylinders, bilayers and vesicle.   

Experimental Test of Hatano and Sasa's Nonequilibrium Steady-state Equality

This abstract has not been submitted yet.   

An ab initio study of crystalline and molecular biotin

The protein cofactor, biotin, has been studied due to its importance in human metabolism and its ability to selectively bind to proteins such as avidin. Understanding the selectivity of biotin from an analysis of its structural stability will help in developing new bio-sensor engineering applications. Experimental analysis of the structure of biotin is typically carried out using x-ray diffraction from crystalline samples. We analyze the differences in energetic, structural, and dynamical properties of biotin in its experimentally determined crystalline form and in its proposed molecular form. Using first principles density functional theory calculations we determine the cohesive energy of the crystalline phase. These calculations explore the limitations of density functional theory, under the generalized gradient approximation, in describing hydrogen-bonding and long-range order in molecular crystals. We analyze the structural stability of both crystalline and molecular phases by calculating the phonon spectrum. Particularly soft modes in the molecule are related to its change in conformation in transforming from the molecular to the crystalline phase.   

The Transition State in a Noisy Environment

Classical transition state theory is the cornerstone of reaction rate theory. It postulates a partition of phase space into reactant and product regions, which are separated by a dividing surface that reactive trajectories must cross. In order not to overestimate the reaction rate, the dividing surface must be chosen so that no reactive trajectory crosses back into the product region. Whereas most chemical reactions take place in a randomly fluctuating environment, as, e.g., a liquid, conventional transition state theory is not well equipped to handle this case because the no-recrossing condition is difficult to enforce in the presence of noise. To generalize the formalism of transition state theory to reactive systems driven by noise, we introduce a time-dependent dividing surface that is randomly moving in phase space so that it is crossed once and only once by each reactive trajectory.   

Session P35: Molecular Electronics I

Signal and thermal noise in [2]catenenane molecular electronic switches

Due to the flexible nature of molecules, an important consideration in molecular electronics would be the influence of molecular vibrations on the device charge transport characteristics. Employing the first-principles matrix Green's function method combined with classical force-field molecular dynamics, we examine the effect of thermal vibrations on the switching in [2]catenane molecular electronic devices. Previously, we have identified frontier molecular orbitals that systematically shift within the two co-conforamations of bistable [2]catenane molecules, which results in the switching of the device when it is probed via a small bias. Here, we compute the charge transport characteristics of the molecule in the switch-on and switch-off configurations at different molecular dynamics snapshots, and find that the fluctuation of resonant transmission peaks in each conformation is smaller than the shift of the peaks between the two conformations. Thus, we confirm that the thermal noise does not mask the switching signal in the [2]catenane molecular electronic devices, which is in accordance with the experimental observation of the switching at the ambient condition.   

Electronic Conduction and Switching in Metal / Molecule / Metal Structures

We report both physical and electrical characterization of several metal / organic monolayer / metal device structures which display electrical switching behavior. Devices comprised a planar lower metal electrode of aluminum (Al) or platinum (Pt), a Langmuir-Blodgett or self-assembled organic alkane monolayer, and an evaporated metal upper electrode of titanium (Ti) or platinum. Single crosspoint devices of area 1600 nm$^{2}$--100 um$^{2 }$incorporated 10$^{3}$-10$^{7}$ molecules in parallel. Electrode surfaces, monolayer structure, and electrode-monolayer interactions were very sensitive to sample preparation. X-ray photoelectron spectroscopy (XPS) indicated that the thickness and stoichiometry of PtOx and TiOx species at both metal-organic interfaces were strongly affected by process conditions including deposition pressures and plasma treatments. Infra-red spectroscopy (RAIR) using ultra-flat template-stripped metal substrates showed that the physical structure of the monolayer was similarly sensitive to nanometer-scale electrode roughness. Electrical conductance hysteresis was observed in Al/monolayer/Ti and Pt/monolayer/Ti devices. Local-pressure modulated atomic force microscopy (AFM) suggested that the electrical hysteresis was dominated by one or two nano-conduction channels $<$30 nm in diameter. The asymmetric, reversible conductance switching observed remains inconsistent with a simple dielectric breakdown process. Instead, for each electrode system we suggest either an interface electrochemical process or a reversible nanoparticle growth {\&} dissolution as primarily responsible for the observed electrical switching. Technology proof-of-principle demonstrations of ultra-dense nanoscale memory and logic integrated crossbar circuits, including latch circuits showing signal restoration, have successfully utilized these organic monolayer structures.   

Transport through thiol SAMs: Effect of monolayer order, dynamics and temperature

We discuss the self assembly and charge transport of organic thiol molecules and discuss the influence of structure, order and dynamics in the monolayer [1] on the transport and also the effect of temperature by scanning tunneling microscopy/spectroscopy (STM/S). Conjugated thiol molecular wires and organometals such as terpyridine metal complexes provide a new platform for molecular electronic devices. Molecular resolution STM imaging in vacuum reveals that the molecular wires adopt an incommensurate, almost vertical SAM structure with a rectangular unit cell, while terpyridine metal thiol complexes tend to lie flat on the Au(111) substrate. STS of the molecular wires show that inherent asymmetry in the molecular structure and asymmetric coupling to contacts results in asymmetric, weakly rectifying I-Vs. STS on alkanethiols do not show a marked temperature dependence down to 150K. We also show that packing and order greatly influence the transport measurements and that the presence of molecular order in the monolayer is very important for reproducible I-Vs. Thus a good control of the molecule-substrate interface needs to be ensured for device reliability. We also point out that molecular electronic devices need to be made tolerant to fluctuations as these cannot be totally eliminated in low dimensional soft systems. [1] Geetha R. Dholakia et. al, PHYSICAL REVIEW B 69, 153402 (2004).   

Molecular Engineering of Single Molecular Switches and Molecular Assemblies

We use molecular design, tailored syntheses, intermolecular interactions and selective chemistry to direct molecules into desired positions to create nanostructures, to connect functional molecules to the outside world, and to serve as test structures for measurements of single or bundled molecules. Interactions within and between molecules can be measured, understood and exploited at unprecedented scales. We look at how these interactions influence the chemistry, dynamics, structure, electronic function and other properties. Such interactions can be used to advantage to form precise molecular assemblies, nanostructures, and patterns. These nanostructures can be taken all the way down to atomic-scale precision or can be used at larger scales. We select and tailor molecules to choose the intermolecular interaction strength and the structures formed within the film. We selectively test hypothesized mechanisms for electronic switching by varying molecular design, chemical environment, and measurement conditions to enable or to disable functions and control of these molecules with predictive and testable means.   

Ab initio study of vibrational spectrum and electron-molecular vibration interaction in molecular electronic systems

For the last few years it was realized that the electron-molecular vibration (e-mv) interaction has to be taken into account to predict transport properties of molecular electronic devices. For realistic calculations, however, the eigenmodes of molecular vibrations and the coupling constants in an e-mv interaction Hamiltonian have to be found from the first principles. We developed a technique to calculate eigenfrequencies of the molecular vibrations (a full spectrum) at both equilibrium and non-equilibrium conditions using the DFT within Keldysh Green's function formalism implemented into our McDCAL code. Using the obtained vibrational spectrum and the self-consistent Kohn-Sham Hamiltonian of the electrons, we can build the e-mv interaction Hamiltonian. By averaging this Hamiltonian over the scattering states of the total Kohn-Sham Hamiltonian, we obtain a dimensionless e-mv interaction strength (both elastic and inelastic ones), for each vibrational mode of the spectrum. These numbers tell us which molecular modes build strong inelastic channels in an electron tunneling through a molecule. As an example, we consider a dithiol-benzene molecule in a good covalent contact with two identical Al electrodes. In the vibrational spectrum of the dithiol-benzene in this device, we show which modes are important at different applied bias voltages.   

Switching of the Fe Oxida\-tion State in Ferro\-cene-Capped Alkanethiols

Molecular electronics has been a rapidly-growing area, due to the simplicity of building molecular devices by self-assembly and the promise of extremely low power consumption as a result of pushing the size down to a few molecules per device. A self-assembled monolayer (SAM) of ferrocene-capped alkanethiols is produced in two stable oxidation states of Fe (Fe$^{2+}$ and Fe$^{3+})$. The oxidation states of Fe are probed with sub-monolayer sensitivity by Near Edge X-ray Absorption Fine Structure (NEXAFS) spectroscopy at the iron L$_{2, 3}$ edges $^{[1]}$. NEXAFS provides a direct method to distinguish between the oxidation states of submonolayer by comparing with the bulk sample spectrum. The native Fe$^{2+}$ layer is converted chemically to Fe$^{3+}$, and the Fe$^{3+}$ layer can be switched back to Fe$^{2+}$ or possibly Fe$^{0}$ by irradiation with soft x-rays. The results have implications on switching mechanisms in molecular electronics. [1] Fan Zheng, V. P\'{e}rez-Dieste, J. L. McChesney, Yan-Yeung Luk, Nicholas L. Abbott, and F. J. Himpsel, Appl. Phys. Lett, to be submitted.   

Theoretical Investigation on the Electronic Structure of Alq3/Al Interface

Alq$_{3}$ [tris-(8-hydroxyquinolinato) aluminum] is one of the most widely used electron transport and emissive material in organic light-emitting devices (OLEDs). From the experimental observation of an extra gap state at the Alq$_{3}$/Al interface, a strong chemical interaction between the Alq$_{3}$ molecule and the Al surface was suggested. Contrary to the experimental studies, previous DFT calculations concluded that the interaction was physisorptive. One possible reason for the discrepancy between the theoretical and the experimental results is the complexity of the experimentally used electrode surfaces. In the present study, we investigated the effect of the surface roughness on the electronic properties of the Alq$_{3}$/Al interface by examining various possible electrode structures. We examined three structures for the Al substrate, the flat Al(111) surface, the Al(332) stepped surface, and the Al adatom adsorbed Al(111) surface. Alq$_{3}$ molecules are bound to Al substrates through their O atoms and about 0.3-0.6 electrons are transferred from the Al substrates to Alq$_{3}$. Upward configurations, in which molecular permanent dipole moments are directed to the vacuum side, reduce the work function by 1.0-1.5 eV, in reasonable agreement with experimental results. The characteristic of the molecular orbitals of Alq$_{3}$ were kept upon adsorption, which seems inconsistent with the gap state derived from the interfacial chemical interaction observed in the UPS and MAES experiments. Further details will be presented.   

Biosensor Consists of Na -- Doped Hydroxyapatite Thin Film

Hydroxyapatite (HAp) surface has an excellent ability of adsorption for functional biomolecules such as protein, DNA and so on. We have investigated the application of the HAp as a suitable material for biosensor. Thin film of the sodium -- doped HAp (Na -- HAp) is prepared in order to decrease the electric resistivity. We have studied variation of the resistance for the Na -- HAp thin films with the adsorption of the functional biomolecules. The sample were prepared by a pulsed laser deposition technique on porous alumina substrate. After the deposition, sample was post -- annealed in O$_{2}$ / H$_{2}$O atmosphere in order to crystallize the Na -- HAp. The powder X ray diffraction pattern shows the sample has a pure HAp structure. The gold comb electrodes were evaporated on the sample for the resistance measurement. The sample was set in pure water of 100 ml in a beaker. When Fetal Bovine Serum of 100 $\mu $l was dropped in the beaker, the Na -- HAp shows the drastic change of the AC resistance (at 120 kHz). This result shows that the Na -- HAp will be one of the most effective materials for the biosensor applications.   

Session P36: Focus Session: BCS-BEC Physics in Fermi Gases

Density Profiles of Strongly Interacting Trapped Fermi Gases

We study density profiles in trapped fermionic gases, near Feshbach resonances, at all $T \leq T_c$ and in the near-BEC and unitary regimes. The component profiles from noncondensed pairs, fermionic excitations and the condensate are also presented. As a consequence of noncondensed Cooper pairs, our profiles are in the unitary regime rather well fit to a Thomas-Fermi (TF) functional form, and equally well fit to experimental data. Our work lends support to the notion that TF fits can be used in an experimental context to obtain information about the temperature.   

Thermodynamics of ultracold fermions in traps in the strongly interacting regime

We discuss the entropy $S$, energy $E$ for trapped fermionic gases, over the entire range from BCS to BEC, and over all $T$ from below to above $T_c$. Our work, which is based on the conventional mean field ground state, shows that both ``bosonic" and fermionic excitations contribute to $S$, and that boson-fermion interactions are essential. Trap edge effects lead to low $T$ power law contributions for the fermions in the unitary and BCS regimes, while bosons contribute to $S$ with a $T^{3/2}$ dependence. Comparison with recent experiments by the Thomas group (cond-mat/0409283) shows very good quantitative agreement. This lays the groundwork for implementing thermometry in strongly interacting Fermi gases. \\ 1. Q.J. Chen, J. Stajic, and K. Levin, \textit{Thermodynamics of ultracold fermions in traps}, cond-mat/0411090; submitted to \textbf{Science}. \\ 2. J. Stajic, Q.J. Chen, K.Levin, \textit{Measuring condensates in fermionic superfluids via density profiles in traps}, cond-mat/0408104. \\ 3. Q.J. Chen, J. Stajic, S.N. Tan, K. Levin, \textit{BCS-BEC crossover: From high temperature superconductors to ultracold superfluids}, cond-mat/0404274.   

Feshbach Molecule Formation in Finite-Temperature Quantum Gases

An exciting development in the field of ultracold atomic gases is the ability to create diatomic molecules by adjusting a Feshbach resonance in the interatomic potential. An extraordinary application of this capability has been to dynamically traverse the BEC-BCS crossover in Fermi gases. While a great deal of attention has focused on equilibrium properties in the {\it{superfluid}} regime, a complete theoretical understanding of the dynamics of molecule formation in a {\it{normal}} gas is still lacking. In a recent article [Williams {\it{et al.}}, J. Phys. B: At. Mol. Opt. Phys. {bf{37}}, L351 (2004)], we presented coupled Boltzmann-like kinetic equations for the atoms and molecules. In this talk, we show that our theory can be used to understand why the molecular conversion efficiency increases as the temperature is lowered, as observed in a recent experiment [Hodby {\it{et al.}}, cond- mat/0411487].   

Phase Diagrams for a Fermi Gas with a Feshbach Resonance

We calculate the phase diagram for a system of Fermi atoms coupled to bosonic molecules by a Feshbach resonance. This work extends the recent work [Williams et al., New J. Phys. 6, 123 (2004)] on the phase diagrams for an ideal gas mixture to include the effect of the resonant interactions using the mean-field theory. We show the paths traversed in the phase diagrams when the molecular energy is varied either suddenly or adiabatically, and discuss the adiabatic phase diagrams obtained in recent experiments on the BCS-BEC crossover.   

Pairing of Opposite-Spin Fermions in the BCS-BEC Crossover Regime

We consider a pair of opposite spin fermions that interact with a degenerate Fermi gas, which is trapped in a spherical oscillator potential. We magnetically tune the S-wave scattering length in the negative region through a Feshbach resonance, producing an attractive interaction between fermions. We use the Hartree-Fock approximation to study the degenerate Fermi gas and pseudopotentials to mimic the interaction between the pair of fermions in a trap and a degenerate Fermi gas in the mean field approximation. We implement hyperspherical coordinates and a model potential for the pair interaction to study the spectrum and wavefunctions of the pair as a function of the magnetic field. In particular, we consider the strongly interacting regime, close to the collapse of the Fermi gas.   

Fermion-mediated BCS-BEC crossover in ultracold ${}^{40}$K gases

Feshbach resonance phenomena in atomic Fermi gases involves the multiple scattering of atoms from open channel states into a molecular bound state formed from neighbouring closed channel states. Current theories treat the closed channel molecule as a featureless boson, but it can be argued that this interpretation is inappropriate when applied to ${}^{40}$K, since the molecule shares a hyperfine spin state with the open channel states. Introducing a 3-fermion model, where a fermion is shared between the open and closed channel states, we explore how the nature of the bound state impinges on the mean- field characteristics of the system.   

Single particle excitations in the BCS-BEC crossover region

We present a theoretical study of the single particle excitations in the BCS-BEC crossover region of a trapped Fermi superfluid at $T=0$. We self-consistently solve the Bogoliubov-de Gennes coupled equations in a harmonic trap, including molecular bosons associated with a Feshbach resonance. We show that the single particle excitation gap $E_g$, which is the same magnitude as the order parameter in a {\it uniform} BCS superfluid, is much smaller than the magnitude of the order parameter ${\tilde \Delta}(r=0)$ at the center of the trap in the crossover region[1]. The excitation gap $E_g$ is determined by the lowest Andreev bound state localized at the edge of the trapped gas. We also calculate the rf-tunneling current spectrum and show how $E_g$ and ${\tilde \Delta}(r=0)$ appear in the spectrum. We compare our results with the recent experimental data for superfluid $^6$Li[2] as well as the recent theoretical work based on an LDA[3]. [1] Y. Ohashi and A. Griffin, cond-mat/0410220. [2] C. Chin et. at. Science {\bf 305}, 1128 (2004). [3] J. Kinnunen et. al. Science {\bf 305}, 1131 (2004).   

Dynamic projection on Feshbach molecules: a probe of pairing and phase fluctuations

We describe and justify a simple model for the dynamics associated with rapid sweeps across a Feshbach resonance, from the atomic to the molecular side, in an ultra cold Fermi system. The model allows us to relate equilibrium properties of the initial state to properties of the final state, such as the fraction of condensed molecules, the momentum distribution of normal molecules and the conversion efficiency as a function of ramping rate. We find that this `projection' onto molecules is a very sensitive probe of pairs (both condensed and non-condensed pairs) in the initial state, and arises ultimately from the short distance singularity of the Cooper pair wave function. We find that near the resonance, phase fluctuations sharply reduce the observed condensate fraction even at zero temperature. For very fast sweeps at low temperature, we predict a surprising nonmonotonic behavior of the molecule condensate fraction versus detuning of the initial state from resonance. In addition to probing the fermion pair condensate [1,2], this approach can detect noncondensed pairs, and possibly establish the presence of a phase fluctuation induced `psuedogap' phase in these systems. \newline [1] C. A. Regal, M. Greiner, and D. S. Jin, Phys. Rev. Lett. 92, 040403 (2004). [2] M. Zwierlein, {\it et.al.}, Phys. Rev. Lett. 92, 120403 (2004)   

Quantum phase transitions in a novel Fermi-Bose Hubbard model

We study a multi-band Fermi-Bose Hubbard model with on-site fermion-boson conversion and general filling factor in three dimensions. Such a Hamiltonian models an atomic Fermi gas trapped in a lattice potential and subject to a Feshbach resonance. We solve this Hamiltonian for paired fermions and bosons in the two state approximation at zero temperature. The problem then maps onto a coupled Heisenberg spin model. In the limit of large positive and negative detuning, the model correctly reproduces the quantum phase transitions in the Bose Hubbard and Paired-Fermi Hubbard models. Near resonance, the bosonic and fermionic Mott phases melt due to fluctuations between the two fields, giving rise to a total-number Mott state instead.   

Fermion superfluids of non-zero orbital angular momentum near resonance

The BEC-BCS crossover of fermion superfluids with near-resonance $s$-wave interactions has been extensively studied in the past year. We extend these studies to the pairing of Fermi gases near the scattering resonance of the $\ell\neq 0$ partial wave at $T=0$. Using a model potential which reproduces the actual two-body low energy scattering amplitude, we have obtained an analytic solution of the gap equation. We show that the ground state of $\ell=1$ and $\ell=3$ superfluids are orbital ferromagnets with pairing wavefunctions $Y_{11}$ and $Y_{32}$ respectively. For $\ell=2$, there is a degeneracy between $Y_{22}$ and a ``cyclic state".   

Quantum phase transitions across p-wave Feshbach resonance

We study a single-species polarized Fermi gas tuned across a narrow $p$-wave Feshbach resonance. We show the existence of a magnetic field-tuned quantum phase transition as detuning sweeps across the Fermi energy, between a $p_x$-wave BCS superfluid and a $p_x+ i p_y$ molecular superfluid in the BEC regime. The latter state, that spontaneously breaks time-reversal symmetry, furthermore undergoes a topological $p_x+ i p_y$ to $p_x+ i p_y$ transition at zero chemical potential, $\mu$. In two-dimensions, for $\mu>0$ it is characterized by a Pfaffian ground state exhibiting topological order and non-Abelian excitations familiar from fractional quantum Hall systems.   

Characteristics of Fermion Pairs with d- and extended s-wave Symmetries around BE-BCS Crossover

We study BE-BCS crossover features for electron pairs with $d_{x^2-y^2}$ and extended-s ($s^*$) symmetries on a quasi-2D square lattice. The pairing interaction is obtained from an extended Hubbard model with on-site repulsion U, and nearest-neighbor attraction V. We calculate various quantities for different filling f and V: Chemical potential $\mu$(V,f), gap $\Delta$(V,f), coherence length $\xi$(V,f). Tightly-bound BE pairs appear at some characteristic $V_b(f)$ at both small and large fillings for both symmetries. At the BE-BCS crossover, the quasiparticle distribution function $v_k^2$ for d-wave is strikingly different from that for $s^*$ wave: While for $s^*$ wave, the central peak in $v_k^2$ diminishes continuously, for $d_{x^2-y^2}$, it vanishes abruptly at the crossover, and redistributes around $(\pm \pi, 0), (0,\pm \pi)$ in the Brillouin zone. The Fourier transform, $v_r^2$ exhibits a ``checkerboard'' pattern. While the general features may be of relevance to the BE-BCS crossover in Fermi systems, the single-particle feature may be of relevance to the field of high $T_c$ cuprates near the pseudogap region. We also explore the density collective modes around the BE-BCS crossover using functional integral techniques within a 1-loop approximation.   

BCS to BEC Quantum Phase Transition in Spin Polarized Fermionic Gases

Recent experiments in cold fermionic gases have shown that $s$-wave magnetic field induced Feshbach resonances can be used to form diatomic molecules of $^{40}{\rm K}$ and $^6{\rm Li}$, which undergoe Bose-Einstein condensation (BEC) on the lower magnetic field side of the resonance. On the higher magnetic field side of the resonance, it has also been established that Cooper pairing takes place and a BCS condensate is formed. We discuss the possibility of a quantum phase transition in ultracold spin polarized fermionic gases which exhibit a $p$-wave Feshbach resonance. We show that when fermionic atoms form a condensate that can be externally tuned between the BCS and BEC limits, the zero temperature compressibility and the spin susceptibility of the fermionic gas are non-analytic functions of the two-body bound state energy. This non-analyticity is due to a massive rearrangement of the momentum distribution in the ground state of the system. Furthermore, we show that the low temperature superfluid density is also non-analytic, and exhibits a dramatic change in behavior when the critical value of the bound state energy is crossed.   

Session P37: Focus Session: Microfluidic Physics III: Surface Effects and Flows

Stick or Slip ?: Measuring slip lengths with nm resolution

Fluid dynamics within small channels draws great interest due to the development of microfluidic devices, yet details about flow immediately at a solid surface remain too vague. Previous attempts to measure surface flow rate were limited to a resolution of the optical wavelength. Here, by using a fluorescence resonance energy transfer (FRET) approach, we improve the resolution by 1-2 orders of magnitude. Two different flow systems, hydrodynamic flow and electrokinetic flow, were investigated with this technique.   

Large electrokinetic effects on slipping surfaces

We show, using extensive Molecular Dynamics simulations, that the dynamics of the electric double layer (EDL) is very much dependent on the wettability of the charged surface on which the EDL develops. For a wetting surface, the dynamics, characterized by the so-called Zeta potential, is mainly controlled by the electric properties of the surface, and our work provides a clear interpretation for the traditionally introduced immobile Stern layer. In contrast, the immobile layer disappears for non-wetting surfaces and the Zeta potential deduced from electrokinetic effects is considerably amplified by the existence of a slippage at the solid substrate. Potential applications of this effect in microfluidic devices will be discussed   

Models for Apparent Slip

A number of groups have reported microfluidic experiments consistent with liquid slip on solid surfaces. We present two physical models which do not require the breakdown of the no-slip boundary condition but lead to results consistent with slip. The first model considers the dynamic response of surface-attached bubbles in drainage experiments and the second considers electrical effects in pressure-driven flow experiments. In both cases, we evaluate the apparent slip length and compare them with the appropriate experiments.   

Wetting on nano-patterned surfaces observed with the Atomic Force Microscope

Recent technological advancements have allowed for precise nanometer-scale control of surface topology and chemical composition. Such surface engineering can be used to confine and manipulate tiny amounts of liquids or to enhance the liquid-repellency of a substrate. Despite the great potential of these methods, the actual behavior of liquids on the nanoscale is still to be elucidated experimentally. With this aim, we have used Atomic Force Microscopy (AFM) to investigate wetting of liquid alkanes onto chemically nanopatterned surfaces. In a first step, parallel, some tens nm-wide stripes with carboxylic acid termination (wettable) were created on the methyl-terminated surface of a self-assembled monolayer (octadecylthrichlorosilane, non wettable) through local electro-oxidation by a metallic AFM tip[1]. Noncontact mode AFM was used to image the condensation of liquids onto the nanopatterned surface. By finely controlling the amount of liquid condensed onto the striped surface we could follow morphological wetting transitions and estimate the size of the contact line and the magnitude of the line tension. [1] J. Sagiv and R. Maoz, Nano Lett. \textbf{3} 761(2003)   

Manipulating Liquids on the Tunable Nanostructured Surfaces

Recently demonstrated electrically tunable nanostructured superhydrophobic surfaces provide a promising new way of manipulating liquids at both micro and macro scale. Dynamic control over the interaction of liquids with the solid substrate is of great interest to many research areas ranging from biology and chemistry to physics and nanotechnology. In this work the influence of the nano-scale topography on the liquid-solid interaction is further investigated. The dependence of the superhydrophobic -- wetting transition on the topography of the nanostructured layer, its electrical properties, and its surface coating is discussed. The reversibility of this transition and its dependence on the geometry of the nano-size features are addressed. Several emerging applications of these surfaces, including lab-on-a-chip, chemical microreactor, and skin drag reduction are discussed.   

Effect of Korteweg Stress of Miscible Two Liquid Flow on Micro Fluidic Devices

In order to design the micro fluidic devices, it is important to investigate the dynamics such as mixing, molecular transformation and interfacial instability in both of miscible and immiscible multi-layer flow. In this study, miscible liquid two-layer flow, water and ethanol, in a Y-shaped microfluidic device, which consists of microchannels with 120 micro m in width and 35 micro m in depth, is experimentally investigated by particle image velocimetry (PIV) to clarify the flow characteristics. The obtained velocity distributions with a spatial resolution of 5.9 x 1.5 micro m\^{}2 around the miscible interface between water and ethanol varying flow rate and concentration of ethanol, indicate an imbalance in shear stress at interface. The difference of shear stress was compared with the Korteweg stress, which was generated by interfacial tension gradient due to a concentration gradient by diffusion in a miscible two-layer flow. The results indicate that the stress was balanced with the shear stress around the interface.   

A Microfluidic Tensiometer

Recent theoretical predictions indicate that a shift in surfactant transport mechanism from diffusion dominated to kinetically dominated occurs at highly curved interfaces where the radius of curvature is on the same order as feature sizes in microfluidic devices (10 to 100 microns). To date experimental evidence of this shift in transport mechanism has been lacking due to limitations on the degree of interface curvature imposed by traditional methods of surface tension measurement. We show that measurement of dynamic surface tension at highly curved interfaces is possible via a microfluidic tensiometer that uses glass micropipettes to control curvature dimension. We observe a dramatic decrease in the characteristic timescale for reducing the surface tension of an initially clean interface, compared with the timescale observed using a traditional pendant bubble method. We discuss the implications of this shift in timescale toward the determination of relevant physical properties of surfactant systems. The transport of surfactant molecules to and from liquid interfaces plays an important role in the formation and motion of droplets and bubbles in microfluidic devices and the results presented here offer new insight into the relevant mass transport timescales to be considered in such applications.   

Transport of nonconductive and conductive droplets in a parallel plate array

Electrowetting on dielectric technique is used to actuate conductive liquid droplets on electrodes patterned beneath a dielectric. Nonconductive liquids can be transported electrohydrodynamically inside channels. We show for the first time that it is possible to transport droplets of nonconductive liquids on dielectric surfaces, using modest voltages and frequencies ($<$100 V, $<$10 kHz). Ionic liquids, aqueous surfactants, buffers, and organic solutions can also be transported. Although conductive liquids show a significant change in liquid contact angle on application of potential, nonconductive liquids do not, suggesting a different mechanism of transport. The empirical criteria for moving droplets in a two-dimensional array are a liquid dielectric constant $\ge $ 4.3 and a molecular dipole moment $\ge $ 1.2 D. The transport mechanisms are discussed along with new microfluidic applications that these results suggest are now feasible.   

Analysis of Drop Shapes during Electrowetting on a Dielectric

Electrowetting refers to the electrostatic control of the interfacial energy of a liquid on a solid, primarily used for the transport of micro-liter volumes of drops on surfaces with embedded electrode arrays. In the present work, the drop is modeled as a two-dimensional lens-like conductor immersed in an infinite dielectric medium slightly above a planar conductor. A matched asymptotic expansion is used to approximate the electrostatic field surrounding the drop. The outer problem models the drop as a conducting circular segment resting on the conducting plane, each maintained at a separate constant potential. The inner problem corrects the region near the edge of the drop by modeling it as an infinite planar conducting wedge lying slightly above the conducting plane. By matching the inner and outer solutions, the charge density along the entire surface of the drop can be approximated, enabling the calculation of the total capacitance of the system. An energy minimization method similar to that of Shapiro \textit{et al.} [\textit{J. Appl. Phys.}, \textbf{93}, 5794 (2003)] is applied to the total energy consisting of the liquid/gas, liquid/solid and solid/gas surface energies, together with the electrostatic contribution, subject to the constraint that the drop volume remains constant. A modified form of the Young-Lippmann equation is thus derived that includes the contribution from the extra capacitance of the drop obtained via matched asymptotics.   

Simulations of electrowetting dynamics in free and and confined droplet geometries

Electrowetting has become popular for moving small amounts of fluids in confined spaces (e.g. [ J. Lee et al. Sensors and Actuators, A95 259 (2002) ]). Models are proposed in two cases: a quasi-steady thin film approximation for a free droplet on a surface, and a Hele-Shaw type model for the two-plate geometry. In the first case, a boundary integral method is formulated which leads to a very efficient numerical algorithm. In the second case, diffuse-interface methods are utilized. The phenomenon of contact angle hysteresis and its dynamical implications are addressed. In the case of the Hele-Shaw geometry, we compare our results to experimentally observed droplet motion.   

Bubble microstreaming: Transport and force actuation

The energy of acoustic waves can be focused onto the microscale through oscillating bubbles, setting up microfluidic flow. Rather than driving the flow directly through the bubble interface motion, we use small periodic oscillations that are rectified into a powerful, steady streaming flow. Substrate patterning allows for control of both the bubble positions and the direction of the flow. High transport speeds are obtained, while different flow patterns can be activated on the same subtrate. Directed transport is achieved by simple, localized patterning, without the need for confining microchannels. The flow field also allows for the simultaneous actuation of large forces onto transported objects, such as lipid vesicles or cells. Potential applications abound in MEMS and lab-on-a-chip systems handling biomaterials.   

3D Hydrodynamic Positioning using Micro-Steady Streaming Eddies

The eddy structure of low-intensity micro-steady streaming flows is useful for positioning particles and motile cells in predictable 3D locations that are dictated by the particle size, device design, and characteristics of the primary oscillating flow field. The flow field structure responsible for this positioning is characterized using tracer imaging methods and numerical simulations. Fluid-particle interactions are studied with a combined perturbation analysis and finite element method to understand the physics of microeddy-based hydrodynamic positioning and the implications for microfluidic device designs.   

Photosensitive chemical reactions in patterned microchannels

Through computer simulations, we study the behavior of a A/B/C ternary mixtures in which two immiscible components, A and B, undergo a photosensitive chemical reaction and produce third component, C. The reverse chemical reaction, namely consumption of the A and B species from the C components, is also possible. Initially, two parallel fluid streams, A and B, are driven by an imposed pressure gradient (Poiseuille flow) through the three dimensional microchannel. The microchannel is decorated with patches that have specific interactions with different components of the mixture. The presence of the patterned substrates enhances chemical reactions in the system since it diverts two initial parallel fluid streams and creates additional interfaces between A and B components. We consider the case where chemical reactions rates can be controlled externally by the light irradiation. We show that by applying time-dependent light irradiation, we can precisely control the distribution of each component within the channel, as well as to tune dynamically the properties of the patterned substrates.   

Session P39: Fermi Liquid and Strong Correlations

Critical Behavior of the Pauli Spin Susceptibility in A Strongly Correlated 2d Electron Liquid

We have performed measurements of the thermodynamic magnetization and density of states in a low-disordered, strongly correlated 2D electron system in silicon. We have found that the spin susceptibility of band electrons (Pauli spin susceptibility) grows by almost an order of magnitude as the electron density ($n_s$) is reduced, behaving critically near $n_s=n_\chi\approx8\times10^{10}$~cm$^{-2}$. This provides thermodynamic evidence for the existence of a phase transition. The density $n_\chi$ is coincident within the experimental uncertainty with the critical density $n_c$ for the metal-insulator transition in clean samples. The nature of the low-density phase ($n_s [Preview Abstract]

Thermodynamic Magnetization of Strongly Correlated 2d Electrons in Perpendicular Magnetic Fields

We will report measurements of thermodynamic magnetization ($M$) of strongly correlated 2D electrons in silicon in perpendicular magnetic fields. We see sawtooth oscillations of the magnetization as a function of the electron density, $n_s$. Near the integer filling factors, the slope $\partial M/ \partial n_s$ exceeds the maximum possible non-interacting value pointing to the existence of the regions with the negative thermodynamic compressibility. Comparing $\partial M/ \partial n_s$ on both sides of the spin gaps, we deduce the $g$-factor. The latter is found to be close to its bare value $g=2$ even at low electron densities, where the critical behavior of the spin susceptibility has been observed in similar systems. This indicates that it is the effective mass, rather than the $g$-factor, that is responsible for the giant enhancement of the spin susceptibility near the metal-insulator transition.   

Precursors of 1D behavior for $D>1$: evolution of the non-analytic correction to the Fermi-liquid behavior

The Fermi-liquid forms of the specific heat ($C(T)$) and static spin susceptibility ($\chi_s$) acquire universal non-analytic corrections[1] and the degree of non-analyticity increase inversely with the dimensionality. This predicts that the strongest non-analyticity in the specific heat should be found in 1D, however bosonization shows that $C(T)$ is linear in $T$ in 1D. We resolve this paradox by showing that the general argument, for non-analyticity in $D>1$ at the second order in the interaction, breaks down in 1D due to a subtle cancellation and the non-analytic $T\ln T$ term in 1D occurs at the \emph{third} order for electrons with spin. We obtain the same result by considering the RG flow of the marginally irrelevant operator in the sine-Gordon theory. For spinless electrons, the non-analyticities in the particle-particle and particle-hole channels cancel out and the resulting $C(T)$ is linear in $T$. The singularity in the particle-hole channel causes non-analyticity in the spin susceptibility $\chi_s \propto \ln \max \{|Q|,|H|,T\}$ present at the second order[2]. [1]A.V. Chubukov and D.L. Maslov, Phys. Rev. B 68, 155113 (2003). [2]I.E. Dzyaloshinskii and A.I. Larkin, Sov. Phys. JETP 34, 422 (1972)   

Quantum Correction to Conductivity Close to Ferromagnetic Quantum Critical Point in Two Dimensions: Ballistic Regime

We study a two-dimensional disordered itinerant electron system close to a ferromagnetic quantum critical point. In the ballistic regime we calculate the temperature dependence of longitudinal conductivity as correction to the classical Drude term. Near the quantum critical point this temperature dependence has an exponent one-third, which is distinct from the linear temperature dependence of Fermi liquids. Away from the quantum critical point the temperature dependence crosses over to the usual Fermi liquid type. The origin of this behaviour is due to the difference in the two regimes in the momentum transferred by the spin fluctuations to the electrons during elastic scattering. Away from the criticality the momentum transferred depends on the mass of the spin fluctuations, while close to the criticality it depends on the Landau damping which brings additional temperature dependence.   

Corrections to Fermi Liquid theory in 2D in a magnetic field

In this work, we consider a Fermi liquid in two dimensions in a magnetic field, and study the effects of the Zeeman splitting on thermodynamics. We derive the temperature dependence of the spin susceptibility $\chi_s (T)$ from the thermodynamic potential, and show explicitly how $2p_F$ scattering gives rise to a non- analytic temperature dependence of the susceptibility. We explain why small momentum scattering does not give rise to non-analytic $\chi_s (T)$. We discuss experimental implications of this result.   

Non-perturbative behavior of the Pomeranchuk quantum phase transition of a nematic Fermi fluid

A nematic phase of an electron gas was predicted by Oganesyan et. al.[1] using a mean field approach by tuning the quadrupolar Landau paremeter $F_2$ to the Pomeranchuk quantum critical point at $F_2\sim-1$. We will present a study of the behavior at this QCP beyond the perturbative results of [1] using non- perturbative methods of higher dimensional bosonization. We extend the results on the Fermion residue calculated by Castro Neto and Fradkin for the Landau fermi liquid phase[2] to this case and present a calculation of the Fermion Green's Function at the critical point. [1] Oganesyan, V., Kivelson, S. A., and Fradkin, E., Phys. Rev. B. {\bf 64} 195109 (2001) [2] Castro Neto, A. H., Fradkin, E. H., Phys. Rev. B. {\bf 51} 4084 (1995)   

Magnetic impurity in an itinerant Anti-Ferromagnet: Undressing the Kondo effect

We study the undressing of the Kondo effect in the aniferromagnetic phase of an itinerant metal. Expressed in terms of the quasiparticles of the AF, the Kondo interaction induces a new relevant operator that leads to incomplete screening and the ground state is no longer a spin singlet. We show how this comes about by expressing the exchange Hamiltonian in terms of the irreducible representation of the square lattice antiferromagnet and performing a renormalisation group calculation to obtain the ground state properties.   

Deconstruction of the Kondo Effect near the Antiferromagnetic Quantum Critical Point

The problem of a spin-1/2 magnetic impurity in a lattice near an antiferromagnetic transition of the host lattice is considered. Asymptotically near the critical point, a multichannel degenerate Kondo problem is realized; the number of channels depends on the symmetry of the lattice and the symmetry of the antiferromagnetic ordering vector. Besides its intrinsic interest, the problem is an essential ingredient in the problem of quantum critical points in heavy- fermions.   

Anomalous scaling at the quantum critical point in itinerant antiferromagnets

We show that Hertz $\phi^4$ theory of quantum criticality is incomplete as it misses anomalous non-local contributions to the interaction vertices. For antiferromagnetic quantum transitions, we found that the theory is renormalizable only if the dynamical exponent $z=2$. The upper critical dimension is still $d= 4-z =2$, however the number of marginal vertices at $d=2$ is infinite. As a result, the theory has a finite anomalous exponent already at the upper critical dimension. We show that for $d<2$ the Gaussian fixed point splits into two non-Gaussian fixed points. For both fixed points, the dynamical exponent remains $z=2$.   

Interacting electrons in 2D in the presence of van Hove singularities and phonons

We have recently extended the RG approach to interacting electrons\footnote{R. Shankar, Rev. Mod. Phys. {\bf 66} 129 (1994).} to include electron-phonon interactions\footnote{S.-W. Tsai, A. H. Castro Neto, R. Shankar, and D. K. Campbell, {\it ``Renormalization Group Approach to Strong-Coupled Superconductors''}, cond-mat/0406174}. We now apply this method to study van Hove singularities. We consider the 2D Hubbard model at half-filling and use the two-patch model. Without phonons, there is a spin-density wave instability due to ($\pi$,$\pi$) nesting. We first consider isotropic phonons, which suppress this instability channel. Angle-resolved photoemission spectroscopy data have suggested that electron-phonon interactions in the cuprates are highly anisotropic\footnote{T. P. Devereaux, T. Cuk, Z.-X. Shen, N. Nagaosa, Phys. Rev. Lett. {\bf 93}, 117004 (2004)}. There is an out-of-phase buckling mode of the oxygen atoms that couples strongly to electronic states in the anti-nodal direction and a breathing mode of the copper-oxygen bond that couples strongly to nodal electronic states. We study the effect of such phonons in our simplified model of interacting electrons. The retardation effects give important corrections to the imaginary-part of the electron self-energy. We also study the competition between the spin-density wave and the $s$- and $d$-wave superconducting instabilities.   

Renormalization-group treatment of the large-$N$ $t$-$J$ model

Renormalization-group techniques for interacting electrons \footnote{R. Shankar, Rev. Mod. Phys. {\bf 66} 129 (1994)} are applicable to systems where the interactions are small compared to the Fermi energy, $U < E_F$. This condition is not satisfied in some systems of interest, such as the cuprates. The limit of large $U$ has been studied by starting from the $t$-$J$ model and applying slave-boson techniques to project out double occupied states. The resulting action can then be solved by saddle point calculations. We start from the Fermi liquid solution\footnote{M. Grilli and G. Kotliar, Phys. Rev. Lett. {\bf 64}, 1170 (1990)} and employ the renormalization-group approach to treat the leading order fluctuations of the bosonic fields in the $1/N$ expansion. We use a recently developed method\footnote{S.-W. Tsai, A. H. Castro Neto, R. Shankar, and D. K. Campbell, ``{\it Renormalization Group Approach to Strong-Coupled Superconductors},'' cond-mat/0406174} that allows electron-electron and electron-boson interactions to be treated on an equal footing. With this starting point, the renormalization-group approach can be safely applied since the interaction terms involving the fluctuations are of order $1/N$, and therefore much smaller than the Fermi energy. We study the self-energy corrections and the instabilities of this system, including the energy scale for the transitions.   

Superconductivity in charge Kondo systems

I will present a theory of superconductivity in charge Kondo systems, materials with resonant quantum valence fluctuations, in the regime where the transition temperature is comparable to the charge Kondo resonance. I will discuss superconductivity induced by charge Kondo impurities, study how pairing of a superconducting host is enhanced due to charge Kondo centers and investigate the interplay between Kondo-scattering and inter-impurity Josephson coupling. I also will discuss the implications of above mentioned theory for Tl-doped PbTe, which has recently been identified as a candidate charge Kondo system.   

Luttinger-Liquid signature in scanning tunneling spectra of Li0.9Mo6O17

We present low-temperature scanning tunneling spectroscopy data from the quasi one-dimensional purple bronze Li$_{0.9}$Mo$_{6}$O$_{17}$. Our spectra show clearly a power-law behavior in density of states around Fermi-energy (-50meV $<$ E $<$ +50meV) with an exponent of $\alpha $ = 0.6 . Temperature dependent spectra between T = 5K and 50K are well-described using a model that involves tunneling into a Luttinger-Liquid at finite temperature. We do not observe any signature in the density of states near T = 24K where a insulator-to-metal transition has been reported. Finally we will discuss our data within the model of a zero bias anomaly (ZBA). However, this model does not describe the experimental data as well as the Luttinger-model does. (Oak Ridge national Laboratory, managed by UT-Battelle, LLC, for the U.S. Dept. of Energy under contract DE-AC05-00OR22725)   

Doping dependent isotope effects of the quasi-1D electron-phonon system: comparison with the high-temperature superconductors

The weak-coupling quantum phase diagrams of the one-dimensional (1D) Holstein-Hubbard and Peierls-Hubbard models are computed near half-filling, using a multi-step renormalization group technique. If strong enough, the electron-phonon interaction induces a spin gap. The spin gap, which determines the superconducting pairing energy, depends strongly on the band filling and decreases monotonically as the system is doped away from half-filling. However, the superconducting susceptibility exhibits a different doping dependence; it can vary non- monotonically with doping and exhibit a maximum at an "optimal" value of the doping. For a quasi-1D array of weakly coupled, fluctuating 1D chains, the superconducting transition temperature $T_c$ exhibits a similar non-monotonic doping dependence. The effect of changing the ion mass (isotope effect) on $T_c$ is found to be largest near half-filling and to decrease rapidly upon doping away from half-filling. The isotope effect on the spin gap is the opposite sign as the isotope effect on $T_c$. We discuss qualitative similarities between these results and properties of the high-temperature superconductors.   

Non-equilibrium quantum phase transition in an itinerant electron system

Recently there was much interest in quantum ferromagnetic transitions in one-dimensional itinerant electron systems. In particular, it was argued that spontaneous magnetization of the electrons might explain the 0.7 anomaly in quantum wires. We investigate the effect of an applied voltage bias on the ferromagnetic transition for one-dimensional itinerant electrons, and determine the critical behavior near the phase transition point.   

Session P40: Focus Session: Transport Properties of Nanostructures IV: Wires

Spin-flip scattering times from weak localization studies of Cu$_{93}$Ge$_4$Au$_3$ thin films

We have fabricated a series of Cu$_{93}$Ge$_{4}$Au$_{3}$ thin films by sputtering deposition technique. The Ge atoms were introduced to enhance the impurity scattering while the Au atoms were introduced to enhance spin-orbit scattering. The films were either 150 or 200 angstroms thick. The residual resistivities were tuned by adjusting the deposition rate and varied between 14 and 59 $\mu \Omega $ cm. Resistance measurements revealed ln$T$ behavior below 10 K or so, which could be ascribed to Kondo effect in addition to 2D electron-electron interaction effects. The electron dephasing times have been measured through the weak-localization-induced magnetoresistances. In particular, the spin-flip scattering times have been extracted from the total dephasing times. We will present the temperature and disorder behavior of the spin-flip scattering times and compare them with the Nagaoka-Suhl and recent theoretical calculations.   

Electrical Transport and 1/f Noise in Au Nanoparticle Films

We studied the transport properties of Au nanoparticle films deposited on interdigitated electrodes with electrode spacings ranging from 0.1 $\mu $m to 1 $\mu $m. $I-V$ characteristics are found to be nonlinear and strongly dependent on both the coating and size of the nanoparticles. Current is thermally activated at low bias voltages, exhibits a threshold behavior, and scales as $I\propto (V-V_{th} )^\zeta $ at low temperatures. To complement dc transport measurements, we have performed noise measurements on some of the films. All the films that were studied exhibit 1/f type noise at low frequencies. The magnitude of the 1/f noise is smaller in devices with a larger device area, indicating that the 1/f noise is caused by intrinsic processes. The noise amplitude is found to be strongly temperature dependent between 40-300 K, with a local peak at around 100 K, and weakly dependent below 40 K. The noise data could not be fit by a single activated process, which would have led to an Arrhenius type temperature dependence. At low temperatures, the normalized noise spectra scaled as ${S_I } \mathord{\left/ {\vphantom {{S_I } {I^2}}} \right. \kern-\nulldelimiterspace} {I^2}\propto (V-V_{th} )^\gamma $. The relationship between the scaling exponents \textit{$\zeta $} and \textit{$\gamma $} is consistent with our prediction of $\gamma =1-\zeta $.   

Quantum Coherence and Local/Nonlocal Resistance Measurements

Low temperature electrical properties of ferromagnetic nanowires are influenced by the interplay between disorder, quantum coherence, and magnetic correlations. Quantum coherence corrections to the conductance are of particular interest, and can be difficult to characterize experimentally. One of the ways of understanding these effects is by measuring the electronic coherence length of the system. Field dependences of local and nonlocal resistance fluctuations were measured for this purpose. Permalloy (Ni$_{0.8}$Fe$_{0.2})$ nanowires were made using standard electron beam lithography along with silver leads with various lead to lead distances. With these nanowires, local and nonlocal length dependent magnetic field correlations were measured and compared. Silver wire samples were also tested for the purpose of comparison. We present initial magnetic field correlation data.   

Role of surface anisotropy for magnetic impurities in electron dephasing and energy relaxation and their size effect

Recently the electron dephasing and energy relaxation due to different magnetic impurities have been extensively investigated experimentally in thin wires. It was shown earlier [1] that a magnetic impurity in a metallic host with strong spin-orbit interaction experiences a surface anisotropy of the form $H=K_d ({\bf n}{\bf S})^2$ which causes size effects for impurities with integer spin. The dephasing and the energy relaxation are influenced by the surface anisotropy in very different ways for integer spin having a singlet and for half-integer spin with a Kramers doublet ground state [2]. Thus for $S=1$ the dephasing is frozen out at low temperatures. That must also result in strong size effects and may resolve the puzzle between the impurity concentrations estimated from the measured electron dephasing and energy relaxation [3,4]. [1] O.~\'Ujs\'aghy, A.~Zawadowski, and B.~L.~Gyorffy, Phys.\ Rev. Lett. {\bf 76}, 2378 (1996). [2] O.~\'Ujs\'aghy, A.~Jakov\'ac, and A.~Zawadowski, to be published in Phys.\ Rev.\ Lett. [3] F.~Pierre, H.~Pothier, D.~Esteve, M.~H.~Devoret, A.~B.~ Gougam, N.~O.~Birge, cond-mat/0012038 and references therein. [4] G.~G\"oppert, Y.~M.~Galperin, B.~L.~Altshuler, and H.~Grabert, Phys.\ Rev.\ {\bf B66}, 195328 (2002).   

Time dependent universal conductance fluctuations in AuPd, Ag, and Au wires

Quantum transport phenomena allow experimental determinations of the phase coherence information in metals. We report quantitative comparisons of inferred coherence lengths from weak localization magnetoresistance measurements and time-dependent universal conductance fluctuation data. A detailed explanation of how these two measurements are performed and analyzed will be given. Strong agreement is observed in both quasi-2D and quasi-1D AuPd samples, a metal known to have high spin-orbit scattering. However, quantitative agreement is not seen in quasi-1D Ag wires below 10 K, a material with intermediate spin-orbit scattering. Possible explanations for this disagreement are discussed. An enhancement of the conductance fluctuations in Au has also been observed by depositing a thin layer of Al$_{2}$O$_{x}$, an oxide known to have a large number of mobile two level systems, over the sample. Preliminary results will be reported.   

Superconducting ultra narrow Al nanowires

We have successfully developed a technique\footnote{F. Altomare \textit{et al.},\textit{March Meeting 2001}}$^,$\footnote{F. Altomare \textit{et al.},\textit{March Meeting 2004}} for the fabrication of nanowires, of width comparable or smaller to 10~nm, using the (1$\bar{1}0 $) plane of a narrow MBE-grown ridge as a template. These wires are formed with direct connections to 4 terminal measurement pads. The versatility and reliability (yield exceeding 75\%) in the fabrication process, together with the small lateral size achievable, makes this technique uniquely suited for the study of diverse physical phenomena in nanowires composed of a variety of materials. In particular we have characterized AuPd wires as long as 20~$\mu$m and Al wires with length exceeding 10~$\mu$m. The Al wires superconduct at low temperature despite having a normal state resistance much greater than the superconducting quantum resistance($=\hbar/4e^2$). We will discuss the observed behavior of the normal-superconducting transition in applied magnetic field and the Current-Voltage characteristics of these wires. This work has been supported by NSF DMR 0135931 and 0401648.   

End States in One-Dimensional Chains

As a consequence of the lower dimensionality, a new kind of state is observed at the boundaries of one-dimensional atomic chains that are self-assembled by depositing gold on the vicinal Si(553) surface. Such “end states” can be thought of as zero- dimensional analogs to two-dimensional states that occur at a bulk surface. Scanning tunneling microscopy images taken at positive and negative polarities reveal contrast reversal at the end atoms that provides evidence for the formation of end states. To confirm this attribution, spatially resolved scanning tunneling spectroscopy along finite chains maps the density of states revealing the formation of quantized states. Further, we observe a transfer of spectral weight from the empty to the filled states over the end atoms. These end states lead to a breakdown in the simple particle in a box model for states along the chains. A comparison to a tight-binding model demonstrates how the formation of end electronic states transforms the density of states and the quantized levels within the chains. As a confirmation of the tight-binding model and the end electronic effects, calculated STM topography profiles at positive and negative biases reproduce the experimentally observed contrast at the end atoms. This work is done in collaboration with Daniel T. Pierce and is supported in part by the office of naval research.   

Gate Dependent One-dimensional Transport in In2O3 Nanowires

The gate-dependent one-dimensional transport of single- crystalline semiconducting In$_{2}$O$_{3}$ nanowire field effect transistors is studied at low temperature by measuring I-V and differential conductance. The In$_{2}$O$_{3}$ nanowires were synthesized by a laser ablation process to have a diameter of 10 nm and a length of 2 $\mu $m. Back gate was formed using a highly-doped silicon substrate with a gate oxide thickness of 0.5 $\mu $m. At a smaller positive gate bias, gaps at near zero source-drain bias were observed for both current and differential conductance spectra due to the absence of the density of states in the source-drain energy window. The transport can be explained by Fermi-liquid theory. On the other hand, when the Fermi energy of the nanowire moves up into the conduction band, the differential conductance of the semiconducting In$_{2}$O$_{3}$ nanowire exhibits zero-bias anomalies, following a power-law behavior similar to one-dimensional Luttinger-liquid. These results suggest that electron-electron interaction must be taken into consideration for the understanding of transport of nanowires at low temperature under a large gate bias.   

Nanowires and Nanoribbons of Charge-Density-Wave Conductor NbSe$_{3}$

We report synthesis of nanowires and nanoribbons of the charge- density-wave conductor NbSe$_{3}$ through direct reaction of Nb and Se powders under careful temperature control. The thickness of these nanostructures range from tens of nanometers to a few hundreds nanometers. Their morphologies and crystal structures were studied with scanning electron microscopy, x- ray diffraction and transmission electron microscopy. Four- probe resistivity measurements were conducted to characterize their electronic properties. Significant enhancement in depinning threshold fields was also observed in the nanowires in comparison to the values of the bulk crystals.   

Effects of Morphology on the Electronic Properties of Nano-lithographic Thin Pb Wires

Miniaturization of superconducting wires is being pursued aggressively for a number of applications as well as for studying the fundamentals of superconductivity in 1-dimensional systems. In this work, we have investigated the conductance of Pb wires with submicron widths and thicknesses as a function of temperature and wire morphology. Our wires are fabricated by electron beam lithography and thermal Pb deposition. In an effort to vary the morphology, we have deposited Pb upon substrates held at room temperature and at 77 K. We shall present from our latest measurements and discuss the effects of morphology upon the electronic properties of these wires, with particular emphasis on superconducting behavior. The authors gratefully acknowledge support from the NSF under grant~No. 0315662.   

Transport Measurements on Individual Branched Nanostructures

We have made electrical measurements on individual branched (``Y-junction") carbon nanotubes. After isolation on silicon substrates and identification via electron microscopy, photo and e-beam lithography were used to deposit metal electrodes ($e.g.$ Au/Ti) onto individual branches of the nanostructures, including 4-probe configurations across a branch point (Y-junction). Various post-processing procedures, such as rapid thermal annealing and electron beam welding, were employed in attempts to improve contact resistances. Four-probe I-V measurements at room temperature yield varied intrinsic conductivity in these nanostructures (resistances between $10^4$ and $10^7\Omega$). Transmission microscopy reveals a fishbone internal structure, which could be responsible for the low conductance. We also report on the construction of ``divining rod'' cantilevers out of these branched nanotubes, using an etch-well technique, toward potential SPM applications, and on similar attempts using inorganic ($e.g.$ ZnO$_2$) nanowires.   

Scattering Mechanisms in 3D Quantum Mechanical Simulations of a Silicon Quantum Wire

We examine the inclusion of scattering processes in MOS quantum wires at room temperature, where phonon effects are prevalent. In a manner similar to Green's function approaches, but more amenable to the site representation used in these methods, scattering is computed on a mode basis and then transformed to the site basis. Beginning with the corresponding matrix elements, we derive mode to mode scattering rates for the different phonon processes and then perform a basis transformation to switch from mode space to real space. The resultant transformation gives us a complex matrix with terms that represent both the elastic and inelastic contributions resulting from the inclusion of different phonon processes. This matrix is then added into the pre-existing tight-binding Hamiltonian and the relevant transport quantities are calculated. Here, we present results of the first implementation of separable phonon scattering rates in a three-dimensional, fully quantum mechanical, self-consistent device simulation. The processes included are acoustic deformation potential scattering, intervalley absorption and emission containing both $f$ and $g$ type processes.   

Electronic structure of single fullerene molecules in C60 nanocrystals grown on thin NaBr film

Since the discovery of C60 much attention has been devoted to utilizing its interesting electronic properties for the advancement of the controllable single molecule technologies. Here we report scanning tunneling microscopy and spectroscopy of C60 nanocrystals grown on thin sodium-bromide films supported on the NiAl(110) surface. STM allows us to study the electronic properties of the C60 crystal with submolecular resolution. Differential conduction spectroscopy reveals a variety of features corresponding to the vibronic excitation and charging of individual C60 molecules in these nanocrystals.   

Session P42: Focus Session: Magnetic Nanoparticles, Nanostructures & Heterostructures VII

The spatial profile of unpinned and pinned uncompensated spins in the antiferromagnetic layer of an exchange bias heterostructure determined with polarized neutron reflectometry

The spin structure of an antiferromagnet plays a critically important role as a means to establish a reference state in magnetic devices; however, the spin structure at the surface and interior of an antiferromagnetic thin film remains unknown. We have used the unique spatial sensitivity of polarized neutron reflectometry to measure the depth dependence of magnetization across the interface between a ferromagnet and an antiferromagnet. The net uncompensated magnetization near the interface responds to applied field, while uncompensated spins in the antiferromagnet bulk are pinned. A new model is proposed for exchange bias. This work was supported in part by the Office of Basic Energy Science, U.S. Department of Energy, BES-DMS, the University of California Campus Laboratory Collaborative program, and Laboratory Directed Research and Development program funds.   

Resonant X-ray Scattering Studies of Positively Exchange Biased Co/FeF2

We here report on our element sensitive resonant soft x-ray scattering studies on a positively exchange biased MgF$_{2}$/FeF$_{2}$/Co/Al sample at T = 20 K performed using circularly polarized incident X-ray beam. The data is analyzed in the Distorted Wave Born Approximation and the depth dependent magnetic density profiles of pinned and unpinned moments in both the ferromagnet and the antiferromagnet is quantitatively determined. Co and FeF$_{2}$ were found to be antiferromagnetically coupled and the net magnetization in the ferromagnet has a structure near the Co/FeF$_{2}$ interface. Diffuse scattering measurements indicated stripe-like domain structure with oppositely directed moments in Co and Fe at the interface that are correlated to interface roughness. Work of I.K.S is supported by DOE.   

Tailoring the spin direction of antiferromagnetic NiO thin films grown on vicinal Ag(001)

NiO is one of the most intensively studied antiferromagnetic materials to study the exchange bias effect. It was shown recently that the spin direction of a NiO thin film can be manipulated between out-of-plane and in-plane directions by growing NiO on MgO(001) and Ag(001). Here we report that we realized the manipulation of NiO spin direction\textit{ in the plane} of the film by growing NiO on a vicinal Ag(001) surface. X-ray Magnetic Linear Dichroism (XMLD) was used to measure the NiO spin directions. We found that the NiO film grown on vicinal Ag(001) surface has an in-plane uniaxial magnetic anisotropy which favors the NiO spin parallel to the steps for $<$100$>$ vicinal surface or perpendicular to the steps for $<$110$>$ vicinal surface. Moreover, we observed the XMLD effect from the NiO L$_{3}$ edge, which further support the magnetic origin of the XMLD effect in our sample.   

Influences of lateral domains and interfacial domain wall formation on exchange bias phenomena in TbFe/GdFe bilayers

We have studied the exchange bias (EB) in a ferrimagnetic TbFe/GdFe bilayer with antiferromagnetic interfacial coupling as a function of magnitude and angle of the cooling field. The magnetic configuration inside both layers has been followed by ultra sensitive magnetoresistance measurements. For increasing cooling field strength or angle from the GdFe easy axis we observe a continuous transition from negative to positive exchange bias. These effects are explained by the quenching of the magnetic configuration inside the random-anisotropy TbFe layer which leads to the presence of a frozen partial interface domain wall. The transition from positive to negative EB results from the continuous rotation of the direction of the interfacial pinning acting on the GdFe layer. Effects of the chirality of this domain wall on the EB have also been indentify. Finally the effect of lateral magnetic domains on EB have been investigated. The occurrence of multi-domain state is found to give rise to exotic and tunable hysteresis loops at low temperature that depend strongly on whether the domains are present in the TbFe or GdFe layers during cooling. The results are fully understood by taking into account both the presence of lateral domains and an interfacial domain wall.   

Exchange Bias of Single-Crystalline Fe$_x$Zn$_{1-x}$F$_2$/Co Bilayers

The exchange bias of polycrystalline Co films grown on epitaxial, 67 nm thick single-crystalline films of (110) Fe$_x$Zn$_{1-x}$F$_2$ was measured as a function of Fe concentration. A set of samples was grown with a pure, 1.0~nm thick FeF$_2$ layer at the interface, and another set was grown without the interface layer. Unlike previous measurements of \textit{twinned} Fe$_x$Zn$_{1-x}$F$_2$ films, the exchange bias of samples with the pure interface layer remains constant as the Fe concentration is decreased from $x=1.0$ to $x=0.35$. A decrease in samples without the pure interface layer was also observed, which can be explained by a weakening of the Co/Fe$_x$Zn$_{1-x}$F$_2$ exchange interaction as the Fe concentration is decreased. These results imply that, at least in highly anisotropic systems like FeF$_2$, the proposed domain state model does not explain the experimental data.\\ \\ Work supported by NSF grant 0400578.   

Parallel Versus Antiparallel Interfacial Coupling In Exchange-biased Co/FeF$_2$

The nature of exchange bias in FeF$_2$ (110) remains elusive due to it's nominally compensated surface. Other interesting phenomena include positive exchange bias and an enhancement of the the coercivity near $T_N$. In order to address these issues, soft x-ray dichroism absorption spectroscopy was used to investigate the direction of interfacial exchange coupling in a antiferromagnetic/ferromagnetic exchange-coupled Co (2.5 nm) /FeF$_2$ (68 nm) bilayer. The FeF$_2$ was epitaxially grown on MgF$_2$ (110) and the Co layer was polycrystalline. The sample was capped with a 2.0 nm layer of Pd to protect it from oxidation. For comparison, a nominally identical sample without Co was also grown. A small portion of interfacial Fe spins couples antiparallel to the ferromagnet, causing the positive exchange bias for cooling fields. A larger potion of interfacial spins, coupled more strongly and parallel to the ferromagnet, increases the degree of antiferromagnetic order and plays an important role in the observed coercivity increase at high temperatures.\\ \\ Work supported by DOE at SSRL and by NSF grant DMR-0400578 at WVU.   

Micromagnetic simulation to observe the reversal mechanism in exchanged biased system of NiFe/NiMn

Understanding the mechanism of exchange interaction between a ferromagnet (FM) and an antiferromagnet (FM) has been both a scientific and technological endeavor in recent years. However, the theoretical and computational efforts so far have rarely predicted important parameters such as exchange bias (H$_{E})$ and enhanced coercivity (H$_{c})$ for any particular system. In our attempt to explain the same, we have simulated the behavior of a FM (NiFe) in the presence of a polycrystalline AF (NiMn) through a moment-moment interaction. To incorporate the surface roughness of the AF grains, the surface spins were selected using a random number generator. This assigned a net moment to each AF grain at the interfacial surface. Our design incorporates about a quarter million cubes which has been the key factor to our understanding of the magnetization reversal mechanism. The time evolution of the FM moments is governed by the solution to the Landau Lifshitz Gilbert equation. The hysteretic behavior of the AF grain includes the effect of thermal excitation. Our results indicate that the reversal mechanism can be either domain nucleation or uniform rotation depending on field direction and training. The hysteresis loop yields an exchange bias of 120 Oe and enhanced coercivity of 100 Oe.   

Lateral length scales and local character of exchange bias.

Exchange bias (EB) is a ferromagnet (F) -- antiferromagnet (AF) proximity effect. EB manifests itself as a horizontal shift of a single hysteresis loop. In our studies, an untwinned 38--100 nm-thick layer of (110) FeF$_{2}$ is epitaxially grown on (110) MgF$_{2}$, followed by a 4--70 nm-thick layer of Co, Ni or Fe. Easy axis magnetization curves (SQUID and spatially resolved MOKE) for different cooling fields and remanent magnetizations for zero-field cooled samples exhibit negatively or positively shifted single or tunable double hysteresis loops (DHL). In the untwinned epitaxial FeF$_{2}$, the AF domains can be much larger than the grains, and, hence, as large as the F domains. When each F domain is in contact with only one AF domain, it does not average the direction and the magnitude of EB. In this regime, inhomogeneity of an AF-F sample, either structural or magnetic, can result in two subsystems formed upon cooling through the AF transition temperature. Each subsystem exhibits EB of the same magnitude but of the opposite sign, which gives rise to DHL. We conclude that when the domain size in the AF is larger than or comparable to that in the F, the local, non-averaging character of EB can be observed. Work supported by DOE and AFOSR.   

Bi-domain state in the exchange bias system FeF2/Ni

Independently exchange biased subsystems are observed in FeF$_{2} $/Ni bilayers after various field cooling protocols. For intermediate cooling fields a double hysteresis loop is found, while negatively or positively shifted single loops for small or large cooling fields respectively are encountered. Both the subloops and the single loops have the same absolute value of the exchange bias field. This suggests that the antiferromagnet breaks into two subsystems in such a way that the ferromagnet does not experience an average bias but is exchange biased in opposite directions with equal magnitude (`bi-domain'). This idea is confirmed by micromagnetic studies including the effect of the antiferromagnet. We also present experiments where thermally activated motion of these antiferromagnetic `domain' boundaries can be achieved.   

Angular Dependence of Exchange Anisotropy on Cooling Field in Exchange Biased Films

Exchange anisotropy in ferromagnet/antiferromagnet (FM/AF) films is usually introduced along the cooling field or FM magnetization direction. Here we investigate the angular dependence of the exchange anisotropy on the cooling field$^{1}$ with vector magnetometry. Three types of (FM=Fe,Ni /AF = FeF$_{2}$, MnF$_{2})$ samples have been studied where the AF layer is polycrystalline, epitaxial (110) and twinned (110). With a polycrystalline AF, the exchange field direction is always the same as the cooling field. With an epitaxial AF which has one spin axis, the exchange field direction is selected by the cooling field to be along the spin axis. With a twinned AF where there are two orthogonal spin axes, the exchange field direction is always along the bisector of the spin axes that encompass the cooling field. Transverse loops show that when the exchange field has a component perpendicular to the applied field, the magnetization reversals occur by coherent rotations in the direction of the perpendicular component. Our results demonstrate systematically the dependence of the exchange field direction on the cooling field direction. $^{1}$H. Shi and D. Lederman, Phys. Rev. B \textbf{66}, 094426 (2002). Work supported by NSF, DOE, Cal-IT$^{2}$ and NEAT IGERT.   

Interplay between Exchange Bias and Uniaxial Anisotropy

The effect of the relative orientation and magnitudes of the exchange bias and the uniaxial shape anisotropy has been systematically investigated in nanometer sized strips of NiFe/FeMn bilayers. The magnetic behavior of these patterned lines was characterized using magneto-optic Kerr effect for different orientations of the applied magnetic field. The samples exhibit peculiar magnetic behavior when the exchange bias, the uniaxial anisotropy, and the applied magnetic field are not collinear. In the case when the exchange bias and the uniaxial anisotropy are parallel, the shift of the hysteresis loop changes non-monotonically with the orientation of the applied magnetic field and exhibits a maximum loop shift that exceeds the value that would be expected from the interface coupling alone. Furthermore, when the applied magnetic field is perpendicular to the exchange bias, the magnitude and the orientation of the uniaxial anisotropy determines the magnitude and the sign of the loop shift. A simple modified coherent rotation model provides a quantitative prediction of the hysteretic behavior in these patterned exchange bias systems. These results show clearly that, in order to understand interfacial coupling, in addition to the loop shift, it is also necessary to characterize all other magnetic anisotropies.   

Temperature dependence of the training effect in exchange bias heterostructures

Recently, the training of the exchange bias (EB) effect in antiferromagnetic (AF)/ferromagnetic (FM) heterostructures has been considered in the framework of activated spin configurational relaxation [1]. The EB field, $\mu _{0}$H$_{e}$, is determined from hysteresis loops of the magnetization which are measured by SQUID-magnetometry after field-cooling the sample below the N\'{e}el temperature of the pinning layer. The evolution of $\mu _{0}$H$_{e}$ in terms of the number of consecutively cycled loops is derived from a discretized Landau-Khalatnikov (LK) equation. Here the time parameter is replaced by the loop index n. Mapping the LK equation onto an implicit sequence allows to describe the training effect, $\mu _{0}$H$_{e}$ vs. n for all n $\ge $1, of various EB heterostructures. In the limit n$>$1, our sequence approaches the empirical $\mu _{0}$H$_{e}$(n)$\propto $1/$\surd $n behavior. The best fit of the sequence to a data set $\mu _{0}$H$_{e}$ vs. n provides the essential fitting parameter $\gamma $ which combines properties of the free energy and the damping with the exchange coupling at the AF/FM interface. We study $\gamma $ vs. T by analyzing the T-dependence of the training effect in a CoO/Co bilayer. Various data sets of $\mu _{0}$H$_{e}$ vs. n are determined from hysteresis loops after in-plane field-cooling at $\mu _{0}$H=0.3T from T=320K to temperatures 5K$<$T$<$T$_{B}\approx $150K, respectively. $\gamma $ vs. T increases with increasing temperature which provides insight into the T-dependence of the free energy. [1] Ch. Binek, Phys. Rev. B \textbf{70}, 014421 (2004).   

Exchange biasing of (Ga,Cr)N thin films using a MnO layer

Recently, significant advances have been made in synthesizing various dilute magnetic semiconductors. To develop practical applications, and in particular MRAM type devices, it is important to formulate methods to manipulate the magnetic properties. In this study we report the observation of exchange biaing of ferromagnetic (FM) Cr-doped GaN films by an antiferromagnetic (AFM) MnO overlayer. The magnetic hysteresis loop shows a clear shift to negative magnetic field by $\sim $ 70 Oe when measured after field cooling, which is absent in single Cr GaN layers. Enhancement of the coercive field of the exchange biased Cr-dopedÊGaN, as compared to films without a MnO overlayer, is alsoÊfound. The observation of the exchange bias in this system is attributed to the exchange coupling at the FM Cr-GaN/AFM MnO interface.   

Session P43: Focus Session: Phase Complexity and Enhanced Functionality in Magnetic Oxides III

Magnetoelectric measurements of multiferroic thin film materials using microwave microscopy

We have developed a technique to quantitatively measure the magnetoelectric (ME) coupling effect in multiferroic materials using microwave microscopy operating at 1 GHz. The technique is used to measure the piezovoltage induced by an external magnetic field through the non-linear dielectric constant. The unique tip geometry of the microscope allows measurement of the ME coupling from a sample region as small as 1 micron cube without the use of a bottom electrode. AC magnetic field up to $\sim $ 10 Oe is used. We have demonstrated the measurement on different types of multiferroic films. In particular, PbTiO$_{3}$-CoFe$_{2}$O$_{4}$ nanocomposite thin films consisting of nanograins of PbTiO$_{3}$ embedded in the matrix CoFe$_{2}$O$_{4}$ was found to exhibit enhanced ferroelectric properties and robust ferromagnetism. The ME coefficient as large as 4 V/cmOe was observed in the nanocomposites at room temperature. Enhancement in the ME effect due to the mechanical resonance of the substrate was observed at $\sim $100 KHz.   

Large Second Harmonic Kerr rotation in GaFeO3 thin films on YSZ buffered Silicon

GaFeO$_{3}$ (GFO) simultaneously exhibits ferromagnetic and pyroelectric properties. Recently, magnetization-induced second harmonic generation (MSHG) have been investigated in single crystal GFO (Y. Ogawa et al., Phys. Rev. Lett. 92, (2004) 047401), where a remarkably large Kerr rotation has been found. We report on MSHG measurements from GFO thin films on YSZ buffered Silicon at room temperature (above $T_{C})$ and at 100 K (below $T_{C})$, where a Kerr rotation of $\sim $15 degrees in the ferromagnetic state is observed.   

All-Optical Subpicosecond Magnetic Switching in NiO(001)

All-optical switching scenarios have so far been demonstrated in atomic and molecular systems only, thus being relatively slow. Here we show, for the first time, the potential of intragap states at a solid surface, viz. antiferromagnetic NiO(001), a notoriously strongly correlated electron system of both high spin density as well as separated multiplet states, for \textit{magnetic} switching on the subpicosecond time scale. Employing a previously designed four-level scenario and computing the electronic structure from a high-level \textit{ab initio} quantum chemical method QCISD(T) with the additional inclusion of spin-orbit coupling yields: (1) The fine structure splitting of the $^3T_{2g}$ states of 70 meV found in experiment (optical second harmonic generation) is well reproduced, (2) Using gaussian pulses of 60-120 fs duration switching the magnetic state of the system back and forth is possible with high fidelity on time scales as short as 150 fs per single switch [1]. The decisive roles of the shape, frequency, and duration of the applied laser pulse are analyzed. The switched state can be reached by tuning the laser parameters to any of the multiplets within the gap. [1] R. G\'{o}mez-Abal, O. Ney, K. Satitkovitchai, and W. H\"{u}bner, PRL \textbf{90}, 227402 (2004).   

Optical Magnetoelectric Effects in Multiferroics

Multiferroics show peculiar magneto-optical properties: Optical refractive index and absorption change with the reversal of the propagation vector \textbf{k} of the electromagnetic wave. This magneto-optic effect is clearly distinct from the conventional magneto-optics like Faraday effect and named optical magneto-electric effect, because it can be considered as the high-frequency extension of the linear magneto-electric effect in multiferroics. We have recently succeeded in detecting the optical/x-ray magneto-electric effect in a polar ferrimagnet GaFeO$_{3}$, where spontaneous polarization \textbf{P}$_{0}$ and magnetization \textbf{M}$_{0}$ are parallel to the $b$ and $c$ axes, respectively. Optical magneto-electric effects are expected to show up for the electromagnetic wave with \textbf{k}//$a$, as the difference in absorption and refractive index with the sign reversal of the triple product of \textbf{P}$_{0}$, \textbf{M}$_{0}$, and \textbf{k}. X-ray magneto-electric absorption shows large enhancement at Fe 1$s$-3$d$ transition.$^{1}$ The obtained spectra are well explained by the interference between electric dipole and electric quadrupole transitions of Fe 1$s$ electrons in an FeO$_{6}$ cluster. Optical magneto-electric absorption of the order of 10$^{-3}$ was observed at around Fe intra-atomic d-d transition.$^{2}$ *Measurements of x-ray spectroscopy were performed at BL-1A, KEK-PF, Japan. $^{1}$M. Kubota et al., Phys. Rev. Lett. \textbf{92} (2004) 137401. $^{2}$J. H. Jung et al., Phys. Rev. Lett. \textbf{93} (2004) 037403.   

Enhanced Tc in a High Temperature Superconductor, Proximate to an Antiferromagnetic Insulator

Placing a high Tc superconductor physically close to an antiferromagnetic insulator brings fundamental scientific interests. Not only the non-superconducting component of the heterostructure plays the role of an artificially created barrier for a Josephson junction network, but also the heterostructure-interface may highlight the magnetic proximity effect or the strain-induced pressure effect on the superconductivity. Electrical transport and percolation behavior of a polycrystalline high Tc superconductor-antiferromagnetic insulator composite has been studied by resistivity, magnetic susceptibility, x-ray diffraction and the polarized optical microscopic experiment. We discovered that Tc can be significantly enhanced in a particular choice of an antiferromagnetic insulator.   

Anomalous Effect of Chromium on Enhancing Ferromagnetism of SrRuO$_3$

The ferromagnetism in SrRuO$_{3}$ is generally thought to be due to Stoner mechanism. It is also known that essentially all modifications, by A-site or B-site substitution, lower the Curie temperature of SrRuO$_{3}$. An anomalous effect, howerer, is found with Cr substitution. When Cr replaces Ru in SrRuO$_{3}$, the Curie temperature increases due to the ferromagnetic superexchange interaction between tetrahedral Cr$^{4+}$ and octahedral Ru$^{4+}$. This case provides the first example that the Stoner ferromagnetism of itinerant electrons can be augmented by superexchange ferromagnetism of localized electrons. On the other hand, the ferromagnetic enhancement effect of Cr is completely suppressed by the simultaneous La substitution of Sr, turning Cr into octahedral Cr$^{3+}$, which leads to antiferromagnetic Ru-Cr and Cr-Cr interactions. Magnetic, transport and XANES data are reported to depict these phenomena.   

Anomalous Transport in Ca$_{1-y}$Sr$_y$MnO$_3$ ($0\leq y\leq 0.75$)~$^{\ast}$

Electron-doped manganites such as Ca$_{1-x}$La$_x$MnO$_3$ have attracted considerable interest in recent years due to their inhomogeneous magnetic ground state, consisting of nanoscale ferromagnetic (FM) droplets and/or spin canted clusters within a G-type antiferromagnetic matrix.$^1$ The nominally undoped (Mn$^{4+}$) Ca$_{1-y}$Sr$_y$MnO$_3$ compounds [for which $T_N$ increases from 125 K (y=0) to 200 K (y=0.75)] have a small electron concentration, $n\sim 10^{18} {\rm cm}^{-3}$, associated with native defects. We report transport measurements on these materials that reveal an unusual transition at a characteristic temperature, $T^{\ast}\sim (0.5-0.8)T_N$. For $T [Preview Abstract]

Ferroelectric microdomains and magnetocapacitance in Ti-doped YMnO3

We have investigated microstructure giving rise to the magnetocapacitance (MC) in Ti-doped YMn$_{1-x}$Ti$_{x}$O$_{3}$ by electron microscopy, combining with conventional magnetic measurement We found characteristic diffuse scatterings elongated along both the [001] and [110] directions in x=0.175, which shows the maximum value of the MC. In addition, the microdomains with the size of 10-20nm consisting of ferroelectric (FE) phase appear in x=0.175. As the Ti concentration (x) is increasing, the paraelectric phase grows up at the expense of the FE phase and the paraelectric phase is dominated above x=0.2. We will discuss the relation between the MC and FE microdomains in this compound.   

Magnetic domains in the ferromagnetic and ferroelectric mixture of (La, Lu, Sr)MnO$_3$

Single crystals of (La$_{5/8}$Sr$_{3/8}$MnO$_3$)$_x$(LuMnO$_3$)$_ {1-x}$ (LSMO)(LMO) synthesized by the floating-zone method were studied by Magnetic Force Microscopy (MFM). Samples were mechanically cut and polished with the surface normal to the growth direction. Polarized optical microscopy shows that LSMO and LMO are separated in a stripe-like pattern due to the chemical immiscibility.\footnote{S. Park, et al., PRL 92, 167206 (2004)} MFM images show magnetic domains ($\sim$1$\mu$m) in the LSMO stripes and no magnetic signal in the LMO phase.   

Super-Colossal Magnetoresistance in Magnetic / Superconducting Oxides

I discuss the super-colossal magneto-resistive effect in complex magnetic/superconducting oxides. In nanometer scale manganite/cuprate heterostructures it is possible to change the magneto-resistance of the nano-heterostructure by several orders of magnitude through the application of a small parallel magnetic field. When consecutive manganite layers are coupled antiferromagneticaly, superexchange fields cancel out in the sandwiched cuprate atomic layer and the heterostructure is superconducting at low temperatures. When a small parallel field is applied to the heterostructure the magnetizations in the manganite layers align and the superexchange fields add up in the cuprate layers, thus destroying its superconductivity. This results in the ``super-colossal magneto-resistive effect''[1]. In my talk I will describe the microscopic theory that produces such effect. \newline [1] C. A. R. Sa de Melo, submitted (2004).   

Effect of adiabatic breathing and half-breathing phonons on striped ground states.

A spin-fermion model for high Tc cuprates is studied using numerical simulations. For certain dopings, stripes are observed in the ground state [1]. Different modes of adiabatic phonons are added to the Hamiltonian, among them the breathing and half-breathing modes. Diagonal and off-diagonal couplings are also considered. It is observed that increasing diagonal electron-phonon couplings tends to stabilize the stripes, while the off-diagonal terms destabilize them creating inhomogeneous ground states [2]. References: [1] C. Buhler, S. Yunoki and A. Moreo, Phys.Rev.Lett. 84, 2690 (2000). [2] Y. Yildirim and A. Moreo, in preparation.   

Soliton like excitations and inconmensurate phases in half doped manganites.

In half doped manganites, the Jahn-Teller effect, the antiferromagnetic coupling between Mn spins and the directionality in the hopping amplitude conspire to create an exotic spin, charge and orbital ordered phase (CE). In this phase the $x-y$ plains are coupled antiferromagnetically and into the layers the structure consists of zigzag chains with steps formed by three Mn ions. Along the chains the occupied Mn $e_g$ orbitals are ordered in the form $...(2x^2-r^2)\! - \! (x^2-y^2) \! - \! (3y^2-r^2) \! - \!(x^2-y^2)...$. Using a pseudospin notation, $|x^2-y^2>$=$\uparrow$ and $|3z^2 -r ^2>$=$\downarrow$, the order along the chain is described by the $x$-component of the pseudospin $\tau _x (i) \sim \cos (\frac{\pi}{4} i )$. This state is degenerated with a state with pseudospin order $\tau _x (i) \sim \cos (\frac {\pi}{ 4} i+\pi )$, and associated with this degeneracy, we expect there to exist soliton-like excitations. The solitons can have charge states $Q$=0, $\pm n \frac{e}{2}$. We study the dependence of the soliton properties on the Jahn-Teller coupling and Hubbard repulsion and we conclude that the solitons could be the low energy charged excitations of the CE phase. When a finite density of charge excitations exists, the ground state is an array of solitons, that develop an incommensurate charge, orbital and spin modulation.