Session A1: Nuclear Spin Dynamics in Semiconductor Nanostructures

Nuclear Spin Induced Oscillatory Current in Spin Blockaded Quantum Dots

Hyperfine interactions between electron and nuclear spins in quantum dots are subject to intensive studies from the viewpoints of quantum computing. In this talk I will review our recent experimental studies for a GaAs-based double quantum dots in the spin blockade regime where the electron conduction is mostly blocked by Pauli effect unless the electron spin state is changed [1]. Thus a small leakage current observed in the spin-blockaded double dot can be a sensitive measure not only for electron spin-flip events but also for a nuclear spin states in the dot if the spin-flip is mediated by hyperfine interactions. We have observed the leakage current shows time-dependent oscillations and is significantly diminished by application of an AC magnetic field whose frequency can induce nuclear magnetic resonance for 71Ga and 69Ga nuclei [2]. A possible nuclear spin polarization mechanism due to hyperfine flip-flop scattering is proposed. [1] K. Ono et al., Science, 297, 1313 (2002). [2] K. Ono et al., Phys. Rev. Lett 92, 256803 (2004). cond-mat/0309062.   

Electron-nuclear spin coupling in nano-scale devices: self-sustaining resistance oscillations and controlled multiple quantum coherences

Author: G Yusa, K. Muraki, K. Takashina (NTT BRL), K. Hashimoto (SORST-JST), and Y. Hirayama (NTT BRL and SORST-JST). We studied electron-nuclear spin coupled systems implemented in microscopic fractional quantum Hall devices and found that in a constant voltage measurement, the longitudinal resistance of such devices oscillates self-sustainingly with a period of about 200 sec. Such behavior suggests that the average nuclear spin polarization self-sustainingly oscillates between randomized and polarized states. When the resistance is measured in constant current mode, on the other hand, nuclear spins are polarized and reach a steady state in about 200 sec. Using the polarized state as an initial state, quantum mechanical superpositional states between four nuclear spin states (multiple quantum coherence) are controlled by pulsed radio frequency radiation resonant with nuclear spin transitions (nuclear magnetic resonance, NMR). Any arbitrary multiple quantum coherent state can be detected as change in the longitudinal resistance. Our findings represent a big step closer to practical all-electrical solid state nuclear spin quantum computing and quantum memory devices.   

Nuclear spin polarization and coherence in semiconductor nanostructures

We have studied the mechanism of dynamical nuclear spin polarization by hyperfine interaction in the spin-blocked vertical double quantum dot system\footnote{C. Deng and X. Hu, cond-mat/0402428. To appear in Phys. Rev. B.}. We have calculated hyperfine transition rates between nuclear spin levels and solved the master equations for the nuclear spins in the double quantum dot. Specifically, we incorporated energy shifts and state mixing due to nuclear quadrupole coupling, which is present because of doping-induced local lattice distortion and strain in the vertical quantum dots. Our results show that doping-induced nuclear quadrupole coupling, together with hyperfine interaction, can cause significant nuclear spin relaxation in the quantum dot system under appropriate conditions (such as tunnel broadening of electronic levels). Therefore, we have found a new channel for nuclear spin relaxation/depolarization in strained material systems at low temperatures. We have also studied internal nuclear spin dynamics in quantum dots and quantum wells through dipolar coupling\footnote{C. Deng and X. Hu, cond-mat/0312208. Submitted to Phys. Rev. B; C. Deng and X. Hu, cond-mat/0406478. To appear in IEEE Trans. Nanotech.}. Our results show strong influences of any inhomogeneity in the hyperfine coupling on the nuclear spin dynamics. Our studies thus demonstrate theoretically the complexities of coupled electron-nuclear spin dynamics in semiconductor nanostructures.   

Controlling Nuclear Spin Environment of Quantum Dots

Electrons in semiconductor quantum dots typically interact with a large ensemble of surrounding nuclear spins via hyperfine coupling. When uncontrolled, this coupling can produce rapid dephasing of electron spin degrees of freedom. These effects of nuclear spin environment were recently observed in several experiments. We describe several approaches to control the interaction between electronic and nuclear degrees of freedom, which allow one to eliminate the dephasing associated with nuclei and to use the localized ensembles of nuclear spins as a useful resource. These approaches make use of the long-lived memory associated with nuclear spins. In particular, we will describe a technique for storing electronic spin qubits in collective states of nuclear ensembles. This can be achieved by controlling hyperfine interaction with external effective magnetic fields, and can result in a robust quantum memory for mesoscopic quantum bits with potential coherence times approaching seconds. Potential applications of this technique for long-distance quantum communication and scalable quantum computation will be discussed.   

Session A2: Quantum Criticality of Strongly Correlated Metals

Hall effect indicates destruction of large Fermi surface at a heavy-fermion quantum critical point

Quantum critical points (QCPs) -– phase transitions at absolute zero in temperature -- are of great current interest because of their singular ability to influence the finite temperature properties of materials. Recently, heavy-fermion metals have played a key role in the study of antiferromagnetic QCPs. To accommodate the heavy electrons, the Fermi surface of the heavy-fermion paramagnet is larger than that of an antiferromagnet [1]. An important unsolved question concerns whether the Fermi surface transformation at the QCP develops gradually, as expected if the magnetism is of spin density wave type [2], or suddenly as expected if the heavy electrons are abruptly localized by magnetism [3]. Here we report measurements of the low-temperature Hall coefficient ($R_H$) -- a measure of the Fermi surface volume -- in the heavy-fermion metal YbRh$_2$Si$_2$ upon field-tuning it from an antiferromagnetic to a paramagnetic state. $R_H$ undergoes an increasingly rapid change near the QCP as the temperature is lowered, extrapolating to a sudden jump in the zero temperature limit. We interpret these results in terms of a collapse of the large Fermi surface and of the heavy-fermion state itself precisely at the QCP [4].\\[0.2cm] [1] R.~M.~Martin, {\em {Phys.\ Rev.\ Lett.}}{ \bf 48}, {362-- 365} (1982); P. ~Fulde, {in \em {Narrow-Band Phenomena -- Influence of Electrons with both Band and Localized Character}} (ed.\ Fuggle, J.~C.) 27--29 (Plenum Press, New York, 1988); M.~Oshikawa, {\em {Phys.\ Rev.\ Lett.}}{ \bf 84}, 3370--3373 (2000).\\[0.2cm] [2] J.~A.~Hertz, {\em {Phys.\ Rev.\ B}}{ \bf 14}, 1165--1184 (1976); A.~J.~Millis, {\em {Phys.\ Rev.\ B}}{ \bf 48}, 7183-- 7196 (1993).\\[0.2cm] [3] A.~Schr\"{o}der \emph{et~al.}, {\em {Nature}}{ \bf 407}, 351- -355 (2000); P.~Coleman \emph{et~al.}, {\em {J.\ Phys.: Condens.\ Matter}}{ \bf 13}, R723--R738 (2001); Q.~Si \emph{et~al.}, {\em Nature}{ \bf 413}, 804--808 (2001).\\[0.2cm] [4] S.~Paschen \emph{et~al.}, to appear in {\em Nature}.\\[0.2cm] In collaboration with: T.~L{\"u}hmann, S.~Wirth, P.~Gegenwart, O.~Trovarelli, C.~Geibel, F.~Steglich, P.~Coleman, and Q.~Si.   

Emergent fluctuation hot spots on the Fermi surface of CeIn$_{3}$ in strong magnetic fields

de Haas-van Alphen measurements on CeIn$_{3}$ in pulsed magnetic fields of up to 65 T reveal an increase in the quasiparticle effective mass with field concentrated at ``hot spots'' on the Fermi surface as the Neel phase is suppressed. As well as revealing the existence of fluctuations deep within the antiferromagnetic phase, these data suggest that a possible new type of quantum critical point may exist in strong magnetic fields that involves only parts of the Fermi surface. Recent specific heat and Hall effect data obtained at high magnetic fields will be discussed in this context. Work done in collaboration with A. Silhanek, N. Harrison, and T. Ebihara.   

Thermodynamic signature of quantum criticality: universally diverging Gr\"uneisen ratio

At a generic quantum critical point where pressure acts as (or couples to) the zero-temperature control parameter, the Gr\"uneisen ratio $\Gamma$ (the ratio of thermal expansion to specific heat) is {\it divergent}[1]. This property provides a novel probe to quantum criticality from thermodynamics. When scaling applies, $\Gamma \sim 1/T^x$ at the critical pressure $p=p_c$, where the exponent $x$ measures the scaling dimension of the most singular operator coupled to pressure; in the alternative limit $T \to 0$ and $p \neq p_c$, $\Gamma = G_r/(p-p_c)$, where $G_r$ is a universal combination of critical exponents. The predicted divergence has been observed near the quantum critical points of several heavy fermion metals[2]. Analyses based on specific models relevant to these experiments are also presented. [1] L. Zhu, M. Garst, A. Rosch, and Q. Si, Phys. Rev. Lett. {\bf 91}, 066404 (2003). [2] R. K\"uchler {\it et al.}, Phys. Rev. Lett. {\bf 91}, 066405 (2003); {\it ibid.} {\bf 93}, 096402 (2004).   

Thermoelectricity as a probe of non-Fermi liquid physics

Recently, the breakdown of the Fermi liquid picture in the vicinity of a Quantum Critical Point has become a subject of special attention. In this context, I will report on recent studies of thermoelectric coefficients in a number of heavy-fermion systems. In CeCoIn$_{5}$, when the system presents a strong departure from the standard Fermi-liquid behavior, a giant Nernst effect and an unusually reduced Seebeck coefficient emerge. The anomalous thermoelectricity disappears with the restoration of the Fermi liquid by the application of a magnetic field. Another example of anomalous thermoelectricity is provided by the hidden-order state of URu$_{2}$Si$_{2}$ which is host to a Nernst coefficient of unprecedented magnitude. Yet another remarkable case is the behavior of CeRu$_{2}$Si$_{2}$ close to the meta-magnetic transition. I will discuss the information extracted by probing the thermoelectric response of the system in each case.   

Session A3a: STM Manipulation of Single Atoms, Charges, and Spins

Single-Atom Spin-Flip Spectroscopy

The energy levels of a magnetic atom split in an applied magnetic field. We recently built an STM with a base temperature of 0.6K and a maximum magnetic field of 7T. These operating conditions allow the direct measurement of the Zeeman energy with inelastic tunneling spectroscopy [1]. We found that the Mn atoms have to be removed from the metal conduction electrons to suppress strong interactions such as the Kondo effect; we use Al$_{2}$O$_ {3}$ grown on NiAl (110). The tell-tale sign of a vibrational mode in inelastic spectroscopy is the predictable frequency shift with mass. In spin-flip spectroscopy we can continuously tune the Zeeman energy with the applied magnetic field. We observe that the measured Zeeman energy is proportional to the magnetic field which yields a local measure of the 'g-value'. We find g- values in the vicinity of g=2, however the exact value depends on the local environment. When a Mn atom sits near the edge of the oxide film we observe strong coupling with the conduction electrons of the substrate resulting in a Kondo effect with Kondo temperatures of a few Kelvin. In contrast to previous STM work we do not observe the Kondo resonance as a Fano line shape. The logarithmic temperature dependence of the Kondo resonance as well as its splitting in magnetic field corroborates the interpretation as a Kondo effect. [1] A.J. Heinrich, J.A. Gupta, C.P. Lutz, D.M. Eigler, Science 306, 466 (2004).   

Localization of Fractionally Charged Quasi-Particles

In this work we address several outstanding questions pertaining to the microscopic properties of the fractional quantum Hall effect: What is the nature of the particles that participate in the localization but do not contribute to transport and can fractionally charged quasi particles localize in space? Using a scanning single electron transistor we image the individual localized states in the fractional quantum Hall regime and determine the charge of the localizing particles. Highlighting the symmetry between filling factors 1/3 and 2/3, our measurements show that fractionally charged quasi particles localize in space to sub-micron dimensions with e*=e/3, where e is the electron charge. In addition, at filling factors 2/3 we follow the behavior of the fractionally charged localized states through the spin phase transition.   

Controlling the Dynamics of a Single Atom in Lateral Atom Manipulation

The ability to manipulate single atoms with the scanning tunneling microscope (STM) stirs one's imagination because of the vast opportunities made possible for building atomic scale devices and nanostructures. Understanding the host of interactions in the STM tunnel junction, and their optimization, is required for efficient and reliable atom manipulation. In this talk I will discuss our work on using atom manipulation imaging and the noise characteristics of the tunneling current as probes of the physics of the atom manipulation process [1]. I will first discuss the dynamics of the Co atom in the context of a manipulated atom image, which is obtained by scanning a single Co atom across the surface. When the Co atom is positioned over the hcp site, dynamic behavior is observed both in the manipulated atom image and in the tunnel current. This site dependent noise in the tunneling current is in the audio range and can be heard as the atom is dragged over the surface. This dynamic behavior corresponds to the Co atom switching between the neighboring fcc and hcp sites of the Cu(111) surface. This occurs by the creation of an ideal, tunable, multi-well potential by the tip-adatom interaction. An ideal double well potential is created by positioning the probe tip slightly off center from the hcp site. Two-state transfer rates between the hcp and fcc sites are obtained by measuring the distribution of residence times in each state. The transfer rates show two distinct regimes. A transfer rate independent of tunneling current, voltage and temperature that is ascribed to quantum tunneling between the two wells, followed by a transfer rate with a strong power law dependence on current or voltage, indicative of vibrational heating by inelastic electron scattering. 1. J.A. Stroscio and R. J. Celotta, Science \textbf{306}, 242 (2004).   

Session A3b: Superconductivity in Diamond

Superconductivity in Bulk, Hole-Doped Diamond

Diamonds, synthesized at high pressure (9 GPa) and high temperature (2500-2800 K) in the systems boron carbide-graphite and boron-graphite, are heavily hole-doped by incorporation of boron into the diamond lattice. These diamonds were characterized by: X-ray diffraction, Raman scattering, NMR, SQUID magnetometry, calorimetry, Hall effect, resistivity and magnetic susceptibility measurements. They show an expanded ($\sim $1 {\%} in volume) lattice with a softened zone-centre optical phonon mode and reduced Debye temperature, and exhibit bulk superconductivity below Tc $\sim $4 K. Upper critical field, specific heat and resistivity measurements provide a consistent set of materials parameters that favor a conventional, weak coupling electron-phonon interpretation of the superconducting mechanism at high hole doping. Preliminary measurements of conductance spectra, obtained with contacts fabricated at the surface of these hole-doped diamonds, indicate the appearance of superconducting gap below Tc.   

Superconductivity in CVD Diamond Films

The recent news of superconductivity 2.3K in heavily boron-doped diamond synthesized by high pressure sintering was received with considerable surprise (1). Opening up new possibilities for diamond-based electrical devices, a systematic investigation of these phenomena clearly needs to be achieved. Application of diamond to actual devices requires it to be made into the form of wafers or thin films. We show unambiguous evidence for superconductivity in a heavily boron-doped diamond thin film deposited by the microwave plasma assisted chemical vapor deposition (MPCVD) method (2). An advantage of the MPCVD deposited diamond is that it can control boron concentration in its wider range, particularly in (111) oriented films. The temperature dependence of resistivity for (111) and (100) homoepitaxial thin films were measured under several magnetic fields. Superconducting transition temperatures of (111) homoepitaxial film are determined to be 11.4K for Tc onset and 7.2K for zero resistivity. And the upper critical field is estimated to be about 8T. These values are 2-3 times higher than these ever reported (1,3). On other hand, for (100) homoepitaxial film, Tc onset and Tc zero resistivity were estimated to be 6.3 and 3.2K respectively. The superconductivity in (100) film was strongly suppressed even at the same boron concentration. These differences of superconductivity in film orientation will be discussed. These findings established the superconductivity as a universal property of boron-doped diamond, demonstrating that device application is indeed a feasible challenge. 1. E. A. Ekimov et al. Nature, 428, 542 (2004). 2. Y. Takano et al., Appl. Phys. Lett. 85, 2851 (2004). 3. E. Bustarret et al., ond-mat 0408517.   

Session A4: Chemical and Biological Sensing with Microcantilevers

Cantilever Arrays as a platform for chemical and biological sensors

Since the late 1980's there have been spectacular developments in micro-mechanical or micro-electro-mechanical (MEMS) systems which have enabled exploration of new transduction modes that involve mechanical energy and are based primarily on mechanical phenomena. As a result, an innovative family of chemical and biological sensors has emerged. While MEMS represents a diverse family of designs, devices with simple cantilever configurations are especially attractive as transducers for chemical and biological sensors. In our presentation we deal with four important aspects of cantilever transducers: (i) operation principles and models, (ii) micro-fabrication, (iii) figures of merit, and (iv) applications of cantilever sensors. We also provide a brief analysis of historical predecessors of the modern cantilever sensors. Finally we have demonstrated that using large well designed arrays of differentially coated microcantilevers coupled artificial neural network techniques can provide information on the identity and amount of target chemicals. We will present our results and discuss future directions.   

Attogram detection using nanoelectromechanical oscillators

We have used small mechanical oscillators as sensitive detectors of bound mass. Because the devices have very small mass, added mass of less than an attogram produces a readily observed shift in the resonant frequencies. For these experiments arrays of resonant devices of various geometries, made primarily from silicon-based materials, were fabricated by electron beam lithography, photolithography or other techniques. We used optical interference techniques to transduce the structure motion as this provides a simple non-contact method for interrogating arrays of oscillators. Chemically selective coatings were used to make the devices respond to specific chemicals, biomolecules, viruses or bacteria. Localizing the specific binding compounds to a nanoscale dot on the oscillator creates a device with a calibrated response to the binding of a few discrete particles from a small volume sample. In this talk we will describe the design, fabrication, mechanical response and use of these devise as specific mass detectors.   

Multifunctional self-sensing microcantilever arrays for detection of chemicals and explosives

Micromachined cantilevers lend themselves well to numerous sensing applications wherein the presence of an analyte is manifested mechanically in cantilever deflection and/or a resonance frequency change. The sensitivity, compactness, cost, power-consumption, scalability, and versatility of microcantilever sensors will continue to drive their appeal in numerous applications. Most cantilever systems to date have been relegated to lab use because they are cumbersome and bulky, requiring extensive setup efforts, alignment and calibration. Commercially available cantilevers require external sensing with optical systems and external actuation. Piezoelectric sensing elements eliminate the need for external optics and external actuators; piezoelectric cantilevers have low power consumption in the sensing element due to their high impedance and low drive- voltage requirements. They have the inherent strength of self-sensing and integrated actuation, meaning that the actuation signal can also be monitored as a sensor signal, and each element can be actuated independently and directly. The self sensing method enables compact and scalable cantilever sensing applications that were previously unfeasible. We have recently demonstrated a novel piezoelectric microcantilever array platform, with and without selective coatings, for detection of various chemicals and explosives. We have also demonstrated the ability to heat the cantilever, and measure both heat and impedance changes in addition to mass loading and unloading. This multidimensional approach offers a unique advantage for chemical selectivity and will be compared with other cantilever and non-cantilever methods.   

Diamond-based MEMS devices for biosensing based on electrochemical and gravimetric

Diamond offers several potential advantages as a platform material for bioinorganic interfaces, including chemical and bio-inertness, electrochemistry, and high acoustic velocity. Ultrananocrystalline diamond (UNCD), with a unique combination of physical, chemical and electrical properties, is attractive for a variety of biochemical/biomedical applications such as hermetic bio-inert coatings, MEMS compatible biosensors, and electrochemical biointerfaces. Over the past several years we have worked on both the fundamental and applied science related to enabling UNCD-based bioMEMS devices, which has encompassed both the development of UNCD surface functionalization strategies that allow fine control of surface hydrophobicity and bioactivity, as well as the development of material integration strategies and surface micromachining techniques to enable the microfabrication of UNCD structural layers (e.g. cantilevers) that incorporate these functionalized surfaces into MEMS devices which are back-end compatible with CMOS electronics. These devices could thus combine the electrochemical and gravimetric transduction of the selective adsorption of target analytes in MEMS structures fabricated directly on top of a silicon microchip.. In the past year we have successfully demonstrated the use of conducting UNCD thin films as electrochemical biointerfaces, via the successful attachment of a redox enzyme onto the UNCD surface, Glucose oxidase (GOD). The procedure to achieve GOD immobilization involved the electrochemical immobilization of nitrophenyl groups to the UNCD surface and transformation of nitrophenyl to aminophenyl groups and the covalent bonding of GOD to the carboxyl groups using the diisopropylcarbodiimide/ N-hydroxysuccinimide (DCC/NHS) as the catalyst. After immobilization, the activity of the enzyme was demonstrated via the direct electrochemical detection of hydrogen peroxide. We have also developed CMOS-compatible UNCD MEMS cantilevers and fixed-fixed beam structures, using both traditional photolithography and e-beam lithography techniques.   

Nano-mechanical Resonantor Sensors for Virus Detection

Micro and nanoscale cantilever beams can be used as highly sensitive mass detectors. Scaling down the area of the cantilever allows a decrease in minimum detectable mass limit while scaling down the thickness allows the resonant frequencies to be within measurable range. We have fabricated arrays of silicon cantilever beams as nanomechanical resonant sensors to detect the mass of individual virus particles. The dimensions of the fabricated cantilever beams were in the range of 4-5 $\mu $m in length, 1-2 $\mu $m in width and 20-30 nm in thickness. The virus particles we used in the study were vaccinia virus, which is a member of the Poxviridae family and forms the basis of the smallpox vaccine. The frequency spectra of the cantilever beams, due to thermal and ambient noise, were measured using a laser Doppler vibrometer under ambient conditions. The change in resonant frequency as a function of the virus particle mass binding on the cantilever beam surface forms the basis of the detection scheme. We have demonstrated the detection of a single vaccinia virus particle with an average mass of 9.5 fg. Specific capture of the antigens requires attachment of antibodies, which can be in the same range of thickness as these cantilever sensors, and can alter their mechanical properties. We have attached protein layers on both sides of 30nm thick cantilever beams and we show that the resonant frequencies can increase or decrease upon the attachment of protein layers to the cantilevers. In certain cases, the increase in spring constant out-weighs the increase in mass and the resonant frequencies can increase upon the attachment of the protein layers. These devices can be very useful as components of biosensors for the detection of air-borne virus particles.   

Session A5: Physics of Emerging Organic Displays - OLEDs and PLEDs

Polymer Based Light Emitting Diodes (PLEDs) and Displays Fabricated from Arrays of PLEDs

Semiconducting (conjugated) polymers are of considerable importance as the active materials in electronic and optical devices including polymer-based light-emitting diodes (PLEDs), photodetectors, photovoltaic cells, sensors, field effect transistors, and lasers. Because of the opportunities associated with passive and active matrix display applications, the development of PLEDs that show efficient, stable blue, green, and red emission is an important ongoing research effort. PLEDs which emit white light are of interest for use in high efficiency active matrix displays (with color filters) and because they might eventually be used for solid state lighting. The ability to fabricate large-area white light emitting PLEDs by processing the active materials from solution is an essential advantage (and requirement) for the use of PLEDs in solid state illumination. I will summarize progress in the field of PLEDs from the fundamental science to recent achievements.   

Achieving High Efficiency OLEDs for Displays and Solid State Lighting

Recent results suggest that organic light emitting devices can provide the very highest efficiencies of any other active display medium, with the exception of those media using ambient light to provide contrast. In this talk, I will consider methods and recent advances in achieving very high efficiency OLEDs for displays and white light generation. In particular, I will consider the physics and technology of employing phosphors in both small molecular weight and polymer organic systems for obtaining the highest possible efficiencies[1,2]. Further, outcoupling schemes for maximizing high external efficiencies will be discussed. For solid state lighting applications, methods and challenges for generating white light via electrophosphorescence are considered. In particular, very simple and high efficiency sources can be obtained using a single dopant based on planar Pt complexes that forms a spectrally broad emitting states. Using this approach, we have a unique opportunity to use OLEDs not only for the next generation of displays, but also for very efficient, environmentally friendly room illumination applications. \newline \newline $^{1 }$B. D'Andrade and S. R. Forrest, Adv. Mat. \textbf{16}, 1585 (2004). \newline $^{2 }$M. Segal, M. A. Baldo, R. J. Holmes and S. R. Forrest, Phys. Rev. B \textbf{68}, 075211 (2003).   

Electron-hole capture in polymer heterojunction light-emitting diodes

Polymer light-emitting diodes based on blends of polyfluorene derivatives show very high efficiencies and low drive voltages. Electron-hole capture in these devices directly produces long-lived exciplex states where the electron and hole are predominantly localized on opposite sides of the heterojunction. The exciplex may then be thermally excited to form an intra-chain exciton, which can itself either emit, or be recycled to reform the exciplex. I will review the physics of exciplex formation and emission in these devices, and will show that exciplex formation rates are consistent with low free charge densities at the heterojunction. I will present evidence that the rate of charge transfer at polyfluorene heterojunctions can be modulated with an applied electric field, leading in some cases to an increase in photoluminescence efficiency with applied field. I will also present recent results showing enhanced triplet exciton formation after photoexcitation in polyfluorene blends, and will discuss the implications of the results for polymer light-emitting and photovoltaic devices.   

Industrialization of OLEDs for Lighting Applications and Displays

Organic light emitting diodes (OLEDs) are an extremely versatile technology that can be tailored to specific applications. The flexibility and adaptability of OLED technology is a result of the variety of material systems and fabrication technologies that can be applied. In this contribution we investigate and compare several material systems and fabrication technologies from an application point of view. Applications without the need of micro-scale structuring open a new window of opportunity for evaporated small molecules. Small molecular OLEDs have the potential for high efficiencies at high brightness rendering them ideal for lighting applications . The first part of our contribution will establish the boundary conditions for lighting applications and we will introduces the current status of our industrialization program for OLEDs for lighting and present our perspective of the OLED lighting market. In the second part of the contribution we will focus on alternative OLED technologies that offer interesting perspectives for industrial fabrication. The light-emitting electrochemical cell (LEC) is a type of organic electroluminescent device that has all the attractive features of the OLED but does not have the drawbacks of reactive cathodes and thin active layers. The crucial difference with OLEDs is that the active layer of a LEC contains mobile ions. This results in two very important advantages for large-area lighting applications compared with traditional OLEDs: (i) thick electroactive layers (ii) and matching of the work function of the electrodes with the energy levels of the electroluminescent material is not required. This means that non-reactive metals such as Ag or Au can be used instead of e.g. Ba. We have studied several types of LECs with the aim to assess the above-mentioned benefits for large-area lighting . Finally to show the immense spectrum of production methods for OLEDs we will conclude the contribution with a manufacturing technique for solution processable material systems: inkjet printing. \newline \newline In collaboration with Eric Meulenkamp, Rene Wegh, Steve Klink, Simone Vulto, and Dietrich Bertram.   

Organic Thin Film Transistors for Electronic Systems

The surge of interest in organic thin film transistors (TFTs) has been motivated, on the one hand, by fundamental questions concerning the energetics and transport of localized carriers, and, on the other hand, by the practical advantages of electronic systems fabricated at low temperatures on flexible substrates. The overriding consideration for the usefulness of organic thin-film transistors for electronic systems has been the field-effect mobility. In this paper I will discuss materials-related factors other than mobility that influence the usefulness of organic TFTs. The subthreshold slope determines the voltage excursion that must take place below the threshold voltage to fully turn off the transistor. Typical organic TFTs have subthreshold slopes that are small compared to silicon devices, due to strongly localized states in energy gap between the more extended levels. The excursion required below threshold often has about the same magnitude as that required above threshold to reach a given level of on-current, and the speed of the system, as well as the power supply requirements, can be adversely affected by the additional required voltage swing. Organic TFTs use metallic or conducting polymer contacts that overlap the gate region, unlike the doped source and drain regions that are self-aligned to the gate in high-performance silicon technologies. A self-aligned process has not been developed for organic TFTs, and, as a result, in organic TFTs there are large parasitic capacitances that can limit system performance. If the amount of overlap is fixed by registration capabilities and can not be reduced as channel length $L$ is reduced, the well-known silicon scaling law in which the upper frequency limit $f_{max}$ scales as 1/$L^{2}$ is modified to$ f_{max}$ $\sim $ 1/$L$, altering significantly the economics of increased integration. The usefulness of organic TFTs is hindered by the lack of a technology that provides complementary $n$-channel and $p$-channel transistors on the same substrate. A good case can be made that the benefits of a complementary technology outweigh the gains achieved from modest improvements in single-channel device mobility, and that more effort to develop organic CMOS is warranted.   

Session A6: Correlated Electrons

Localization-delocalization transitions in strongly correlated electron systems

Correlations can give rise to localization of charge and spin of electrons. A methodology will be presented which allows to describe localized states and also localization- delocalization transitions as a function of pressure or doping. The method is the self-interaction- corrected local-spin-density approximation (SIC-LSDA) to density functional theory. The SIC- LSDA can differentiate between localized and itinerant electrons. Results of calculations for 4f, 5f and 3d compounds will illustrate the method.   

Magnetic to valence-bond-solid transition in an S=1/2 XY model with ring-exchange

Within the Landau-Ginzburg-Wilson framework, phase transitions between two ordered phases with different symmetries are generically of first order, or there is a region of coexistence of the two phases. However, It has recently been argued [1] that there is a generic class of continuous order-order {\it quantum phase transitions}, where the critical point is characterized by deconfined spinon degrees of freedom. Evidence of such a transition, between a magnetic (or superfluid in a bosonic representation) and a valence-bond-solid (VBS) phase had previously been observed in large-scale quantum Monte Carlo simulations [2] of a 2D XY model which in addition to the standard nearest-neighbor exchange J contains a four-particle exchange of strength K. The VBS phase in this model is not favored by the J and K interactions individually (the K-only model has an Ising-like antigerromagnetic ground state), but emerges out of competition between the two terms. Here I will discuss recent efforts [3] to characterize the magnetic-VBS transition in more detail (extracting the critical exponents) and comparing the behavior with predictions of the deconfined quantum-criticality scenario. \vskip2mm [1] T. Senthil, A. Vishwanath, L. Balents, S. Sachdev, and M. P. A. Fisher, Science {\bf 303}, 1490 (2004).\hfill\break [2] A. W. Sandvik, S. Daul, R. R. P. Singh, and D. J. Scalapino, Phys. Rev. Lett. {\bf 89}, 247201 (2002).\hfill\break [3] A. W. Sandvik, R. G. Melko, and D. J. Scalapino (work in progress).   

Fast methods for evaluating molecular electron correlation energies

Some of the issues associated with developing fast methods for the wavefunction- based description of electron correlation will be re-examined in this talk, after a brief introductory overview of standard methods for treating electron correlation in molecules. The first main topic is the description of so-called dynamic correlation effects that are closely related to atomic correlations. Local correlation methods that describe two and three-body correlations at reduced computational cost, while still ensuring continuous potential energy surfaces will be described and their performance assessed in terms of accuracy and computational cost. The second main topic is describing strong correlations associated with near-degeneracies, such as occur in diradicaloid molecules and transition metal compounds. Simplified coupled cluster methods appropriate for such problems will be described, along with examples of their application to several molecules believed to have significant diradical character.   

Correlation effects in the compressed rare earth metals

A number of the trivalent rare earth metals (Ce, Pr, Gd, and Dy) are known to undergo electron-correlation driven phase transitions under pressure that are characterized by unusually large volume changes (5--15{\%}). These ``volume collapse'' transitions demarcate regimes of different behavior with high-symmetry structures and local moments on the low-pressure (strongly correlated) side versus low-symmetry structures and screened moments on the high-pressure (more weakly correlated) side. Interestingly, Nd reaches this high-pressure regime without undergoing a significant collapse. This talk describes calculations using the local density approximation combined with dynamical mean field theory (LDA+DMFT) for Ce [1], Pr, and Nd, which seek insight into this behavior. Results for an assumed fcc structure suggest that the interesting correlation effects are pushed to higher pressures from Ce to Pr to Nd, and must compete there against successively stiffer underlying equations of state, which may contribute to the absence of the collapse in Nd. LDA estimates of the structure dependence of the energy appear to be smaller effects after these correlation contributions. Spin orbit plays an interesting role in that the lower Hubbard band remains exclusively j=5/2 under compression, whereas the quasiparticle weight is of mixed j=5/2, 7/2 character so that the pressure-induced transfer of spectral weight to the Fermi level effectively quenches the spin orbit. Collaborations with K. Held and R.T. Scalettar are gratefully acknowledged. [1] K. Held, A.K. McMahan, and R.T. Scalettar, Phys. Rev. Lett. \textbf{87}, 276404 (2001); A.K. McMahan, K. Held, and R.T. Scalettar, Phys. Rev. B 67, 075108 (2003).   

The Dynamical Cluster Approximation

The DCA is a general approach for the theory of correlated and disordered lattice systems which maps the lattice onto a self consistently embedded periodic cluster. When the cluster size is one the mean field solution is recovered (DMFT or CPA), and as the cluster size increases non-local corrections are systematically incorporated. A variety of methods may be used to solve the cluster problem. The DCA has been used to study a variety of systems, including the cuprates, dilute magnetic semiconductors and carbon nanotubes. The DCA formalism, comparison with other cluster methods, and these applications will be reviewed.   

Session A7: From Egg to Adult: Patterning and Morphogenesis in Animal Development

How to Make a Neurocrystal: Modeling the developmental patterning of the fly's retina

Animals' ability to create the complex patterns found in many organisms is an enduring source of wonder and a topic that has long drawn the interest of scientists of all stripes. Famously, it was an attempt to model developmental patterning that led to the discovery of the Turing instability. Here, we study one of the most remarkable and best-characterized examples of such pattern formation, the development of the fruit fly's compound eye. In the fly larva, a front of differentiation moves across the sheet of tissue that will become the adult retina. It leaves behind it a striking hexagonal array of cells marked by high levels of the protein Atonal. It has previously been noted that a standard activator-inhibitor model might explain this process [Meinhardt, 1992], but only recently has the basic genetic logic governing photoreceptor specification been deciphered [e.g. Frankfort and Mardon, 2002]. We build on these advances with the first model of retinal patterning based on experimentally verified interactions. Surprisingly, we conclude that a Turing-instability-based mechanism alone cannot reproduce the observed behavior. Instead, we propose that the pattern is generated primarily by a novel ``epitaxial'' process in which, as the front progresses, each newly-created row of unit cells acts as a template for the next one. A clear prediction of this model is that if the communication between successive rows is broken, even transiently, a striped pattern will appear. Preliminary experimental tests suggest that just such a phenomenon occurs in some mutants. Related patterning processes have been observed in systems as diverse as chick feather buds and vertebrate retinal ganglion cells [Pichaud, Treisman, and Desplan, 2001]; our model may thus describe an evolutionarily conserved module.   

Axis definition during Hydra regeneration

Hydra may recover even from dissociation into single cells. During such a reformation process, Hydra cells first form a hollow ball made of a cell-bilayer, subsequently the developmental isotropy is broken and an axis is established. The animal then reforms according to this axis. We show that a temperature gradient of about 1\r{ }C across the embryo determines the axis but not the orientation of the developing animal. A change in morphogenetic inflation-contraction cycles of the Hydra cell ball coincides with irreversible axis establishment, accompanied by a change in tissue elasticity and WNT expression. We suggest that a modulation of cell adhesion or internal pressure lock the axis during development, therefore corroborating the recently advanced hypothesis of a link between cell-adhesion regulation and the WNT cascade. Quantitative analysis of the early, Hydra specific gene ks1 reveals scaling of the expression-pattern size distribution. A plausible interpretation is, that transient collective cell-differentiation-fluctuations with increasing magnitude break the symmetry of the Hydra cell-ball; they establish the axis irreversibly upon short-range WNT cascade activation. Our interpretation suggests why Hydra regeneration starts with a hollow cell ball and how only five to ten organizer cells may convey their existence to the 10000 others.   

Force Regulation in Tissue Mechanics

We have investigated tissue mechanics in live fly embryos perturbed by a UV microbeam and imaged with confocal microscopy. The actin cytoskeletons of these transgenic flies are labeled with green fluorescent protein to provide contrast without compromising biological function. We concentrate on dorsal closure, a model system for development and wound healing, to identify connections between forces, genetics, and morphogenesis. Dorsal closure is proving to be an attractive system for research in biological physics since key cell boundaries lie in a plane and exhibit multiple symmetries, which facilitates modeling. We find that four spatially and temporally coordinated processes are responsible for the dynamics of dorsal closure. The bulk of progress is driven by contractility in supracellular ``purse strings'' and in the amnioserosa, whereas adhesion-medicated zipping coordinates the forces produced by the purse strings. When the UV microbeam was used to block adhesion mediated zipping, altered dynamics preserve closure, attributed to an upregulation of the force produced by the remaining amnioserosa. In addition, the modeling of wild type and mutant phenotypes is predictive; although closure in myospheroid mutants ultimately fails when the cell sheets rip themselves apart, our analysis indicates that $\beta _{ps}$--integrin has an earlier, important role in zipping.   

Topology and Robustness in the Drosophila Segment Polarity Network

Previous work by von Dassow {\it et al.} demonstrated the robustness of a mathematical model of the genetic interactions that define the polarity of {\it Drosophila} embryo segments. I showed that this robustness is due to the positive feedback of gene products on their own expression. This topological feature of the network allows individual cells in the model segment to adopt different stable expression states (bistability) corresponding to different cell types in the segment polarity pattern. A positive feedback loop will only yield multiple stable states when the parameters that describe it satisfy a particular inequality. By testing which random parameter sets satisfy these inequalities, I show that bistability is necessary to form the segment polarity pattern and serves as a strong predictor of which parameter sets will succeed in forming the pattern.   

Session A9: Magnetic Domains and Dynamics

Magnetic Reversal Energy Loss and Dynamics in Permalloy Thin Films and Microstructures

Magnetic hysteresis energy loss scaling and switching dynamics was studied in thin (20 nm) permalloy films and patterned microstructures using the magneto-optic Kerr effect. Sinusoidal, triangular and square magnetic waveforms of peak amplitudes up to 100 Oe and frequencies covering 10 decades (1mHz to 10 MHz) were applied to the samples while monitoring the magnetic response at 1 ns temporal resolution and 1 $\mu $m spatial resolution. All films and microstructures exhibited similar loss scaling behavior characterized by the dynamic coercivity H$_{c}$*($\omega )$: an ``adiabatic'' region described by the averaged static coercivity H$_{0 }$, followed by a region of monotonically increasing loss described by a power law: H$_{c}$*($\omega )$= H$_{0}$ + A(dH/dt)$^{\alpha }$. The scaling function is derived from a domain wall dynamics model based on a linear ramp field. Exponents, $\alpha $, obtained from fits to scaling measurements are independent of H$_{0 }$ for microstructures obtained from the same parent film, suggesting universal behavior. The basic loss mechanism (thin film limit) at both low and high frequencies appear to result from large-angle local spin damping. Supported by NSF-DMR-0404252.   

Ferromagnetic resonance modes of the vortex state in Permalloy dot arrays

Permalloy dot arrays (circular dot 1000nm in diameter and 40nm thick, arranged in square lattice at 1100 nm period) have been fabricated with e-beam lithography. Spin dynamics of the magnetic dots were measured via ferromagnetic resonance on a microstrip in the frequency range 5-36 GHz with the field applied in the film plane. A single mode is present at high microwave frequency, where resonance occurs well above the saturation field of the dots. With decreasing frequency, however, an additional mode appears on the high field side of the main mode, which may result from edge domains as indicated by micromagnetic simulation. At frequencies below 15 GHz, additional modes are observed at fields lower than the two above modes when sweeping up in field, but not when sweeping down in field. These additional modes are attributed to the influence of a vortex structure (the equilibrium state of the dot at zero field). This is inferred from simulation, wherein the vortex state persists up to 750 Oe when increasing field from zero and does not reappear with decreasing field until 100 Oe.   

Magnetic Vortex Interactions in Double Vortex Stadium Structures

We have used time-resolved Kerr microscopy (TRKM) to study the spin dynamics of individual Permalloy stadium structures having thickness 50~nm, width 600~nm, and lengths ranging from 800~nm to 1200~nm. The stadium geometry relaxes from saturation into either a single vortex or double vortex state, depending on the orientation of the applied field during relaxation. From the zero field double vortex state, the separation distance between the two vortices decreases with applied field until annihilation near 270 Oe. TRKM measurements on a 1200~nm long stadium reveal a zero field gyrotropic mode frequency of $\sim0.4$~GHz that shifts downward in frequency to $\sim0.2 $~GHz near 270 Oe. This behavior is consistent with theoretical predictions for a coupled vortex system with decreasing vortex separation [1]. Above 270 Oe, a single vortex remains in the system, with a gyrotropic mode frequency $\sim0.5$~GHz and non-monotonic field dependence. Finally, above 500 Oe, the dynamic behavior is characteristic of the saturated state. Similar TRKM measurements on increasingly shorter stadia, at zero field in the double vortex state, show the gyrotropic mode frequency shifting from $\sim0.4$~GHz for the 1200 nm long stadium to $\sim0.2$~GHz for a 900 nm stadium. [1] J. Shibata, K. Shigeto, Y. Otani, Phys. Rev. B \textbf{67}, 224404 (2003). This work was supported by NSF DMR 04-06029 and the University of Minnesota MRSEC (NSF DMR-02-12032).   

Atomic Spin Dynamics during reversal in composite media

We present spin dynamics during reversal for composite material (exchange coupled hard and soft phases). Up until now, consensus has been that reversal is by coherent rotation with the field determined by the average of the intrinsic reversal fields of the two pure phases. Atomic scale simulations show non-coherent reversal. Reversal is initiated in the soft phase and a domain wall is formed at the interface between the hard and soft phase which propagates through the hard phase under the action of the field. The two important fields associated with the reversal process are Hk1 (reversal field for soft phase) and Hdw (domain wall propagation field from soft to hard phase). The switching field is determined by max(Hk1,Hdw). Hdw is found to be 1) proportional to anisotropy difference of the two phases and 2) inversely proportional to the total moment of the two phases. In the limit of zero anisotropy difference between the phases Hdw becomes negligible as expected. Hk1 on the other hand depends on the geometrical length of the soft phase. The lowest limit of Hk1 is equal to the intrinsic reversal field of the soft phase when its length (L1) is sufficient to support the intrinsic domain wall width (Ldw). When $ L1 < Ldw $, Hk1 increases in proportion to the excess energy required to accomodate the domain wall in the soft phase. Analytical expressions for both Hk1 and Hdw will be given and shown to agree very well with our simulations and experiments in Appl. Phys. Lett. 82, 2859 (2003).   

Fast magnetization switching of Stoner particles: A nonlinear dynamics picture

We reexamine the problem of the magnetization switching of Stoner particles in the presence of dissipation from the point of view of nonlinear dynamics. Within the Landau-Lifshiz-Gilbert formulation, we illustrate how the fixed points and their basins change under a perpendicular and a parallel field. This change explains well why a non-parallel field gives a small minimal switching field and a short switching time. Furthermore, we clarify that the so-called Stoner-Wohlfarth (SW) limit is exact only when the dissipation is infinitely large. However, for a give magnetic anisotropic energy function, there is a critical dissipation above which the minimal switching field is the same as that of SW-limit. The reason and meaning of such a critical disspistion is also given.   

Barkhausen Noise, Domain Structure and Entropy in Magnetic Amorphous Ribbons Under Stress

The effect of applied stress on the magnetization process of magnetostrictive Fe78B13Si9 and amorphous Fe73.5Cu1Nb3Si13.5B9 metallic ribbons was investigated by estimating the entropy of the Barkhausen noise time series. The Barkhausen series is formed of voltage pulses detected by a coil around a ferromagnetic sample under an applied magnetic field. The stress induces a preferred orientation on the domain walls and reduces their thickness. The entropies calculated from the Random Field Ising Model simulations with different degrees of disorder were also analyzed and the results compared to the experimental data. In both cases, the relative entropy is calculated from the size of the Barkhausen noise time series packed by applying the LZ77 (zipping) algorithm. In the case of the experimental curves, an increase in the relative entropy was observed above a given stress level, possibly as a result of the decrease of the domain wall width. The results are also compared with domain structures observed by Kerr microscopy and to magnetization curves obtained by the inductive method.   

Spin-Torque Stimulated Barkhausen Jumps in a Thin-Film Permalloy Microstructure

Barkhausen Jumps (BJs) are studied in a 60$\mu $m x 50$\mu $m x 30nm thick permalloy microstructure as a function of the bipolar current pulse amplitude applied during field-driven magnetization reversal. Magnetic force microscopy is used to characterize the quasi-static domain structure and the magneto-optic Kerr effect is used to measure BJs. Above a threshold current density J$_{T} \quad \sim $ 10$^{10}$ A/m$^{2}$, the BJs become correlated with the current pulses. The observed behavior is consistent with models of spin-torque transfer domain wall motion and compatible with recent experiments after accounting for difference in sample static coercivity. The threshold current density for current-stimulated domain wall motion in the (low coercivity) micron-scale structures is about two orders of magnitude lower than the threshold reported for sub-micron wire structure [1] (10$^{12}$ A/m$^{2})$, which is near the damage threshold. The effect can be used to control dynamic coercivity and offers opportunities for studying current-driven domain dynamics near the depinning threshold, and under conditions permitting a wide dynamic range below the damage threshold. [1] A. Yamaguchi et al. Phys. Rev. Lett. 92, 077205-1 (2004).   

Magnetization reversal and anisotropy at the Fe/AlGaAs (001) interface

We distinguish the magnetic reversal process of an Fe interface layer from that of the bulk in Fe/AlGaAs heterostuctures using magnetization induced second harmonic generation (MSHG) and the magneto-optical Kerr effect (MOKE). We find that the switching characteristics are distinctly different -- single step switching occurs at the interface layer, while two jump switching occurs in the bulk for the magnetic field orientations employed. This indicates a larger contribution from uniaxial versus cubic anisotropy at the interface layer, causing the absence of an intermediate single domain state during reversal. The general assumption that spins within a ferromagnetic metal films are strictly parallel due to strong exchange coupling is therefore incorrect at the interface. Our results show that MSHG is a powerful technique to probe interface magnetic properties in non-centrosymmetric hybrid structures.   

Imaging antiferromagnetic domains of GdNi2Ge2 by x-ray resonant magnetic scattering

The body-centered tetragonal compound GdNi$_{2}$Ge$_{2}$ orders antiferromagnetically below T$_{N}$ = 27.5 K. By using x-ray resonant magnetic scattering we have determined that the magnetic phase transforms from a collinear structure to a cycloidal structure below T$_{t} $ = 16 K. Both magnetic structures lower the symmetry and result in magnetic domains. The excellent quality of our single crystal and the resulting high intensity allowed us to image these domains using the x-ray resonant magnetic scattering technique. By reducing the illuminated area to 100x100~$\mu $m$^{2}$ we succeeded in investigating a single magnetic domain. The plane in which the magnetic moments lie is determined to have tilt about 10 degrees away from the \textbf{a} direction for both magnetic structures.   

X-ray imaging of chiral domains in Dy metal

Domain growth has been measured in Dy metal using a circularly polarized x-ray beam both on cooling through the PM to AF transition, and on warming through the FM to AF transition. The difference in the scattered intensity between right and left handed incident x-rays was measured at the (0,0,4+$\tau$) peak, where $\tau$ is the wave vector of the AF structure. On cooling from the PM to the AF phase, the chiral domains nucleate and grow to several hundred microns and no further change is observed with decreasing temperature. The size of the domains is assumed to be limited by defects. On warming from the FM phase the domain size is resolution limited and little domain growth is observed between T$_c$=90K and about 140K. With further increase in temperature the domains grow to the hundreds of microns observed on cooling. Early neutron scattering measurements revealed a second harmonic with an intensity that decreased linearly to zero at 140K.$^1$ We observe, a weak non- resonant reflection at 2$\tau$ in the charge ($\sigma-\sigma$) channel rather than the magnetic ($\sigma-\pi$) channel. This suggests that there is a distortion of the structure that is hindering domain wall motion at low temperatures. Work at the Advanced Photon Source was supported by the DOE, Office of Basic Sciences, under contract no. W- 31-109-Eng-38. \\ \\$^1$M.K. Wilkinson et al., J. Appl. Phys. 32, 485 (1961).   

Domain Walls and Roughening Transition Possibilities in a Transverse-field Ising Model with Long-range Interactions

We have studied domain walls and domain wall roughening in the presence of long-range interactions. The insulating system LiHoF$_4$ is a physical realization of the transverse-field Ising model, which is known to have an order-disorder quantum phase transition between ferromagnetic and paramagnetic states. Furthermore due to long-range dipole interactions, LiHoF$_4$ naturally forms thin needle-like domains which might suggest the possibility of a roughening transition for the domain walls, on the grounds that it is expected for the standard short-range transverse-field Ising model. We will describe how the long-range forces which are responsible for the domain formation also affect the nature of the domain wall structure and account for the absence of a roughening transition.\\ \\ This work is supported by NSF DMR-0342157.   

Imaging Antiferromagnetic Domain Walls with the Hall Effect

We find that the Hall effect in the spin-density-wave state of elemental chromium is acutely sensitive to the underlying domain structure. A large (20\% effect) hysteresis in the linear Hall coefficient emerges as a function of temperature between the spin-flip (123 K) and Neel (311 K) transitions. The hysteresis is accompanied by a pronounced increase in the noise. Scratching the surface of a clean single crystal can pin the domains, suppressing the hysteresis loop and curtailing the motion of the spin domain walls in the transverse antiferromagnetic phase. Changing the relative orientations of the current flow and the pinning directions alters the preferred state of the system.   

Domain Walls and Macroscopic Spin-Flip-Like States in Gd$_{x}$Co$_{1-x}$/Gd$_{y}$Co$_{1-y}$ Bilayers

Exchange coupled double layers (ECDL) made of rare earth -- transition metal amorphous alloys are of basic and technological interest, as they present different magnetization configurations when the composition is changed or when the temperature is varied crossing the compensation temperatures (T$_{comp})$ of both ferrimagnetic alloys. In this work, amorphous Gd$_{x}$Co$_{1-x}$(100 nm)/Gd$_{y}$Co$_{1-y}$(100 nm) ECDL have been prepared to investigate the magnetization reversal and the stable magnetic configurations when the compositions of both layers are similar: x = 0.22, y = 0.24. The samples have been grown by co-sputtering on corning glass substrates, which has allowed to analyze the behaviour within each layer by transverse Kerr effect measurements. A rich variety of behaviours has been found in the temperature range between the T$_{comp}$ of both layers, including magnetization reversal by annihilation/creation of a Bloch wall across the sample thickness, and a macroscopic spin-flip-like metamagnetic state where the magnetic moments form a double antiferromagnetic state with the presence of a N\'{e}el-like wall when the magnetizations of both layers are similar [1]. The whole observed behavior can be understood in terms of a deduced general magnetic field -- temperature phase diagram. \newpage [1] R. Morales et al. Phys. Rev. B 70, 174440 (2004). \newpage Work supported by Spanish CICYT.   

Putting a spin on speckle: the twisted way magnets remember

Major hysteresis loops in magnetic systems have generally been thought to be symmetric under reversal of the two axes. We show why this is incorrect, and how the asymmetry provides an explanation for recent x-ray speckle experiments on thin films that show a nontrivial microscopic difference between the zero field states produced by starting from large positive and large negative fields, and generalizations thereof. This unexpected result is due to the dynamics of vector spins; scalar models such as the Ising model are inadequate for this purpose, even if the anisotropy is high. We compare data from magnetic thin film speckle experiments with our numerical model, and find good agreement for many features. In certain regions of parameter space of the model, we find an unusual `mixed phase' of magnetic domains. We also present results on nanomagnetic pillar arrays, where a similar symmetry violation and multicycles for minor hysteresis loops are seen.   

Displacement Profile of Charge Density Waves and Domain Walls at Critical Depinning

The influence of a strong surface potential on the critical depinning of an elastic system driven in a random medium is considered. If the surface potential prevents depinning completely the elastic system shows a parabolic displacement profile. Its curvature C exhibits at zero temperature a pronounced rhombic hysteresis curve of width $2f_c$ with the bulk depinning threshold $f_c$. The hysteresis disappears at non-zero temperatures if the driving force is changed adiabatically. If the surface depins by the applied force or thermal creep, C is reduced with increasing velocity. The results apply, e.g., to driven magnetic domain walls, flux-line lattices and charge-density waves.   

Session A10: Focus Session: Spin Transport Devices

Magnetooptical Studies of Organic/Ferromagnetic Hybrid Structures

Organic semiconductors (OS) are desirable for spin-based devices because of their long spin coherence times due to low spin-orbit coupling. The use of ferromagnets (FM) to inject and detect spin polarization in OS has recently been demonstrated through experiments on organic spin valves (i.e. FM/OS/FM trilayer devices). An important issue for these devices is the spin-dependent properties of the FM/OS interface. To investigate this issue, layered FM/OS hybrid structures are fabricated using molecular beam epitaxy (MBE). Ultrathin single-crystalline Co films on Cu(100) substrate serve as the model FM layer, whose magnetic properties are characterized by in situ magneto-optic Kerr effect (MOKE). The structural properties of the Co film are characterized by RHEED and STM. Subsequently, wedged OS overlayers (e.g. Alq$_{3}$, Gaq$_{3})$ are deposited onto the FM film and in situ optical studies are performed. First, the effect of the OS on the magnetic properties of the Co layer is investigated using MOKE. Second, the effect of the FM layer on photo-excited carriers in the OS is investigated using polarization-resolved photoluminescence.   

Spin-Valves Incorporating Magnetic and Nonmagnetic Organic Semiconductors

A clear spin-valve effect is reported for stacked thin film devices constructed of two ferromagnets with differing coercivities, iron (Fe) (100nm) and iron-cobalt (Fe$_{50}$:Co$_{50})$ (30nm), which are magnetically decoupled by a layer of $\alpha $-sexithiophene ($\alpha $6T) (120nm). Coherent spin transport is expected to be facilitated by relatively low spin orbit coupling in $\pi $-conjugated materials$^{1-3}$. Spin-injection is aided by tunnel barriers at metal/semiconductor interfaces. A spin-valve effect of up to 20{\%}, with switching at the expected coercive fields, is observed at 4.5K and the effect persists up to 150K. The conduction electrons in vanadium tetracyanoethylene (V[TCNE]$_{x})$, an organic-based magnetic semiconductor with T$_{C}>$ 350K, are fully spin-polarized$^{4}$. In addition to the low field `conventional' spin-valve switching from 10 to 100K, an unusual background high field magnetoresistance is reported for the spin-valve device structure where $\alpha $6T (50nm) is the nonmagnetic spacing layer between V[TCNE]$_{x}$ ($<$1um), and cobalt (Co) (25nm). Supported by DOE Grant No. DE-FG02-01ER45931 and DARPA (ONR Grant No. N00014-02-1-0593). 1. Dediu, et al., Solid State Comm. \textbf{122} 181 (2002) 2. Epstein, MRS Bull. \textbf{28} 492 (2003) 3. Xiong, et al., Nature \textbf{427} 821 (2004) 4. Prigodin, et al., Adv. Mater. \textbf{14} 1230 (2002), Raju, et al., J. Appl. Phys. \textbf{93} 6799 (2003)   

Modification of ferromagnetism in semiconductors by molecular monolayers

We report that adsorption of monolayers of organic molecules onto ferromagnetic semiconductor heterostructures can produce large changes in magnetic properties [1]. The digital-alloy heterostructures studied have 1/2 monolayer MnAs planes embedded in GaAs. We investigate effects on magnetic properties of self- assembly of various organic molecules onto the heterostructure surface. Depending on the molecular structure, the monolayers can either strengthen or suppress ferromagnetism. We attribute this chemical modulation of magnetic properties to electronic changes brought about by molecular binding to the semiconductor surface. \newline \newline [1] T.C. Kreutz, R. Artzi, E.G. Gwinn, R. Naaman, H. Pizem, C.N. Sukenik and A.C. Gossard, Applied Physics Letters 83, 4211(2003).   

Spin Transport in Organic Semiconductors

Spin injection/detection and coherent transport are necessary ingredients in spin electronics or spintronics.~ Organic semiconductors are believed to have long spin coherence due to the weak spin-orbit interaction and hyperfine interaction; therefore, may be useful in spintronic device applications. Recently, we have successfully achieved electrical spin injection/detection and demonstrated coherent spin transport~in spin valve devices using an organic semiconductor spacer.~ The devices consist of LSMO and Co ferromagnetic electrodes and small molecule material Alq3 spacer.~ A large inverse spin valve magnetoresistance (up to 40{\%}) was observed in these devices, which entails coherent spin transport in organic semiconductors. In addition to the spin valve magnetoresistance effect at low fields, we have also found magnetoresistance and magneto-eletroluminescence at high magnetic fields. This latter high-field effect is due to the field-dependent carrier injection at the ferromagnetic/organic interfaces. It is the first experimental evidence of the anomalous chemical potential shift theoretically predicted for double exchange ferromagnets such as LSMO.~ In collaboration with Z.H. Xiong, D. Wu, and Z.V. Vardeny; work supported by DARPA, NSF, and DOE.   

Magnetoresistance Anomalies Across Domain Walls in Tensile Strained (Ga,Mn)As

We describe measurements of the anomalous Hall effect (AHE), planar Hall effect and anisotropic magnetoresistance (AMR) in tensile-strained (Ga,Mn)As epilayers with relatively high Curie temperatures ($125 \rm{K} < T_{\rm{C}} < 135 \rm{K}$). Samples are grown on a strain-relaxed (Ga,In)As buffer layer deposited on (001) GaAs, creating an in-plane tensile strain that orients the easy axis of the magnetization along [001]. We measure magnetoresistance as a function of the magnetic field vector $\vec{H}$ and temperature ($4.2 \rm{K} < T < 150 \rm{K}$) using Hall bars oriented along $[110]$,$[1 \overline{1} 0]$ and $[100]$. AMR measurements reveal striking antisymmetric resistance anomalies as we sweep either the magnitude or angle of $\vec{H}$. These anomalies originate in a strong AHE contribution to the AMR when measurements are made across domain walls in the presence of slight sample misorientation, providing a sensitive probe of the nucleation and propagation of magnetic domain walls up to temperatures as high 120K. Work supported by DARPA, ONR and NSF.   

Hysteretic resistance spikes in magnetic 2DEGs

We use spin-density-functional theory to study recently reported hysteretic magnetoresistance $\rho_{xx}$ spikes in Mn-based 2D electron gases [Jaroszy\'{n}ski \textit{et al.} Phys. Rev. Lett. \textbf{89}, 266802 (2002)]. We find hysteresis loops in our calculated Landau fan diagrams and total energies signaling quantum-Hall-ferromagnet phase transitions. Spin-dependent exchange-correlation effects are crucial to stabilize the relevant magnetic phases arising from \emph{distinct }symmetry- broken excited- and ground-state solutions of the Kohn-Sham equations. Besides hysteretic spikes in $\rho _{xx}$, we predict \textit{hysteretic dips} in the Hall resistance $ \rho _{xy}$. Finally, we note that our theory \textit{does not} include domain walls. While not ruling out the importance of these, our quantitative agreement with the experiments does highlight the relevance of spin-dependent exchange-correlation effects in magnetic 2DEGs.   

Fabrication of a Ferromagnetic Semiconductor Spin Bipolar Transistor

We have fabricated a spin bipolar transistor that uses a bilayer of the ferromagnetic semiconductor Ga$_{(1-x)}$Mn$_{(x)}$As to provide a tunneling magnetoresistance (TMR) element in the emitter of the device. The two magnetic layers have a different manganese concentration that gives differing coercive fields and Curie temperatures. This allows the two magnetic layers to be set in parallel or anti-parallel configurations at low temperatures. TMR is clearly observed, and transistor action confirmed in the electrical characteristics of the device.   

Spin Gunn Effect

Even in nonmagnetic semiconductors the electron drift velocity depends on the electron spin polarization. This effect, originating from the Pauli excusion principle, drives a novel phenomenon we call the spin Gunn effect. We predict that the flow of unpolarized current in electron-doped GaAs and InP at room temperature is unstable at high electric fields to the dynamic formation of spin-polarized current pulses. Spin-polarized current is spontaneously generated because the conductivity of a spin-polarized electron gas differs from that of an unpolarized electron gas, even in the absence of spin-orbit interaction. Magnetic fields are not required for the generation of these spin-polarization current pulses, although they can help align the polarization of sequential pulses along the same axis. We also find that the spin polarization amplification rate is the largest for electron mobilities dominated by LO-phonon scattering, and that the steady-state (saturation) spin polarization can exceed 80\% for both GaAs and InP at room temperature. Some possible applications to novel spintronics devices will also be suggested. This work was supported by DARPA/ARO, and more details are available in cond-mat/0407547.   

Device Applications of Spin-Orbit Interaction in Semiconductor Heterostructures

We report recent progress in theoretical development of two classes of non-magnetic semiconductor heterostructure spin devices that exploit spin-orbit interaction in the presence of structural inversion asymmetry (SIA) or bulk inversion asymmetry (BIA). The first uses resonant tunneling to filter spins, and can be used to create a source of spin polarized current. We will provide an analysis on the origin of spin-dependent tunneling in these structures and discuss their applications. The second exploits the interplay between BIA and SIA to control spin lifetimes for device applications. We show that the D'yakonov-Perel' spin relaxation can be suppressed to first order in $k$ for one out three spin components in [001] and [011] heterostructures, and for all three spin components in [111] heterostructures. Our results suggest the use of [111] heterostructures as preferred channels for spin transport, as active regions in spin-LEDs, for spin lifetime transistor, and for spin storage.   

Self-Consistent Non-equilibrium Greens Function Description of Spin Transport in Diluted Magnetic Semiconductors

We present and discuss some applications of a method for treating transport in semiconductor heterostructures which is based on the Kohn-Luttinger k.p model Hamiltonian and the non-equilibrium (Keldysh) Green's function formalism. The method is compatatible with self-consistent and time-dependent mean-field descriptions of magnetically ordered states and is capable of handling problems of current interest in semiconductor spintronics, including magnetoresistive and Spin Momentum Transfer related phenomena in ferromagnetic semiconductors. In the case of diluted magnetic semiconductors exchange interactions with local moments are included by means of the mean-field virtual crystal approximation and electron-electron interactions can be included at the self-consistent level, using either Hartree or local density approximations. We perform a calculation of the magnetoresistance in a GaMnAs/GaAlAs/GaMnAs diluted magnetic semiconductor heterostructure using the 4-band model. Spin Transfer effects are also studied by a direct calculation of the non-equilibrium spin density and its corresponding contribution to the exchange.   

Photon assisted double quantum dot spin filter

We report on numerical and analytical studies of spin transport through a double quantum dot system in the Coulomb blockade regime, at zero bias. In the presence of a magnetic field and an AC field, a device is proposed where both spin filtering and spin pumping take place, in the sequential tunneling regime. We show that in such a device, the amount of polarization of the current can be controlled by the intensity of the AC field while the sign of the spin polarized current is controlled by its frequency. A master equation approach is used to study the time evolution of the reduced density matrix, in the Markov approximation, including spin relaxation and decoherence effects. The heights and widths of the current peaks obtained for one, two or more photon absorption processes, can be explained by a simple analytical study in the stationary regime. We also show that the decoherence time can be obtained from an analysis of the widths in frequency of the current. Finally, we include results on cotunneling effects on our spin filter and spin pump device.   

Spin-polarized transport in a quantum wire controlled by a magnetic field

We investigate the effects of $p^3$-dependent Dresselhaus spin-orbit interaction on the transport properties of a quantum wire. The particles are subjected to a magnetic field along, and an electrict field normal to the wire. The spin-orbit term leads to the spin-dependent renormalization of the mass of particles as well as the energy level splitting for different spins. This, in turn leads to spin-polarized tunneling transport controlled by the strength of the spin-orbit interaction and the magnetic and electric fields.   

Spin-sensitive transient absorption measurements in CdTe high above the bad gap

Spin-sensitive dynamics of carriers optically generated and probed high above the band gap in CdTe have been measured by time-resolved differential transmission experiments using 80 fs pump and probe pulses with the same ($\Delta $T/T)$_{++}$ and opposite ($\Delta $T/T)$_{+-}$ circular polarization. These experiments were motivated by the recent observation that in GaAs, depending on the photon excess energy \textit{h$\nu $} - $E_{g}$, the degree of circular polarization (DCP=[($\Delta $T/T)$_{++}$-($\Delta $T/T)$_{+-}$]/[($\Delta $T/T)$_{++}$+($\Delta $T/T)$_{+-})$ could be positive, negative or even zero [1]. In CdTe we observed that the results of pump-probe experiments are even more sensitive to the experimental conditions than in the case of GaAs. Namely, we show that for photoexcited carriers with certain excess energies and concentrations there is a sign change not only in DCP but also in $\Delta $T/T (an absorption bleaching changes to an induced absorption). We conclude that all these effects are a consequence of interplay between the state filling and the~spin-sensitive band gap renormalization. We acknowledge fruitful discussions with J. T. Devreese. This work was supported by the Ministry of Education of the Czech Republic (project 1K03022). [1] Y. Kerachian, P. Nemec, H. M. van Driel, A. L. Smirl, APS March meeting, paper H26 8, p. 407, Montreal, Canada, 2004.   

Session A11: Earth and Planetary Materials I

Theoretical and experimental evidence for a post-perovskite phase of MgSiO$_3$ in Earth's D$''$ layer.

The Earth's lower mantle, the largest region within our planet (670-2890 km depths), is believed to contain $\sim $75 vol.{\%} of (Mg,Fe)SiO$_{3}$ perovskite, $\sim $20{\%} (Mg,Fe)O, and $\sim $5{\%} CaSiO$_{3}$. This mineralogy was unable to explain many unusual properties of the D'' layer, the lowermost $\sim $150 km of the mantle. Using \textit{ab initio} simulations and high-pressure experiments we have demonstrated [1] that at pressures and temperatures of the D$''$ layer, MgSiO$_{3}$ transforms from perovskite into a layered CaIrO$_{3}$--type structure (space group \textit{Cmcm}); this structure was also independently found in [2]. The elastic properties of the new phase and its stability field explain most of the previously puzzling properties of the D$''$ layer: its seismic anisotropy [3], strongly undulating shear-wave discontinuity at its top$^{ }$[4], and the anticorrelation between shear and bulk sound velocities [5]. This new phase is therefore likely to be a major Earth-forming mineral, and its discovery will change our understanding of the deep Earth's interior. Latest studies of the effects of impurities [6,7] on the stability of this phase, and similar phases of other compounds will be discussed. \\ \\ REFERENCES: \\1. Oganov A.R., Ono S. (2004). \textit{Nature} \textbf{430}, 445-448. \\2. Murakami M., et al. (2004). \textit{Science} \textbf{304}, 855-858. \\3. Panning M., Romanowicz B. (2004). \textit{Science }\textbf{303}, 351-353. \\4. Sidorin I., et al. D.V. (1999). \textit{Science} \textbf{286}, 1326-1331. \\5. Su W.J., Dziewonski A.M. (1997). \textit{Phys. Earth Planet. Inter.} \textbf{100}, 135-156. \\6. Mao W.L., et al. (2004). \textit{Proc. Natl. Acad. Sci.} \textbf{101}, 15867-15869. \\7. Ono S., Oganov A.R., Ohishi Y. (2004).\textit{ Submitted.}   

Vibrational and thermodynamic properties of MgSiO$_3$ post-perovskite

Recently a high pressure transformation in MgSiO$_3$ perovskite, the primary constituent of Earth's lower mantle, was identified near 125 GPa and 2500 K. In order to understand the importance of the new phase on mantle properties and dynamics it is necessary to determine its thermodynamic properties and contrast with those of perovskite. Here we perform first principles quasi-harmonic approximation calculations of the free energy of MgSiO$_3$ post-perovskite and derive several thermodynamic properties of interest up to and beyond core-mantle-boundary (CMB) pressures. It is found that in general its thermal properties are very similar to those of perovskite at relevant conditions, despite the great structural difference between these phases. The post-perovskite phase is found to be mechanically stable at ambient pressure and might be retrieved metastably at cryogenic temperatures.   

MgSiO3 post-perovskite at D'' conditions

The thermoelastic properties of the newly found post-perovskite polymorph of MgSiO$_3$, more stable than the Pbnm-perovskite phase at conditions close to those expected in Earth's D'' region, has been investigated by first-principles and contrasted with those of the perovskite phase. We predict the major seismic trends such as velocity discontinuities, ratios of velocities and density anomalies, and anisotropy in aggregates with preferred orientation that should occur in the presence of this phase change. Consequences of this model mineralogy for the D'' region will be discussed.   

On the stability of perovskite and post-perovskite polymorphs of MgSiO$_3$

The relative stability of 10 distorted perovskite structures of MgSiO$_3$ have been investigated by first principles and contrasted with that of the newly found {\it Cmcm} post- perovskite. The electronic structure of these polymorphs was analyzed and simple relationships between magnitudes of polyhedral distortions, bond-lengths, and band-gaps were found. Up to approximately 95 GPa, the {\it Pnma} phase is the most stable, has the largest band gap, and can accommodate the largest volume reduction with the smallest distortion of the SiO$_6$ octahedra. At higher pressures the post-perovskite polymorph is the most stable one. This phase transition is accompanied by an increase of octahedral volume and band gap. These results do not support a transition from {\it Pnma} to another distorted perovskite structure prior to the transition to the {\it Cmcm} polymorph as proposed by some experiments.   

First principles study of liquid MgSiO$_3$ at conditions of the Earth's deep mantle

Constant-pressure {\it ab initio} molecular dynamic simulations at high temperatures have been used to study MgSiO$_3$, the major constituent of the Earth's lower Mantle. In this work, we focus the properties of molten MgSiO$_3$, where its existence in the core-mantle boundary is still in debate. By using liquid configuration, we have performed variable-cell {\it ab initio} molecular dynamic simulations at relevant thermodynamic conditions across one of the measured melting curve. The calculated equilibrium volumes and densities are compared with the simulations using orthorhombic perovskite configuration under the same conditions. For molten MgSiO$_3$, we have determined the diffusion coefficients and shear viscosities at different thermodynamic conditions. Our results provide the evidences of the existence of molten MgSiO$_3$ near the core-mantle boundary.   

{\it Cmcm} post-perovskite: a new alumina polymorph

Alumina, Al$_2$O$_3$, is a model ceramic material with important applications in high pressure science, particularly as the ruby pressure scale. It is isoelectronic with MgSiO$_3$, the major Earth forming mineral. Here we show by first principles that the newly found post-perovskite polymorph of MgSiO$_3$, CaIrO$_3$ type structure with {\it Cmcm} symmetry, is also a stable high pressure phase of Al$_2$O$_3$ and should be stabilized in the pressure range in which the ruby scale has been calibrated. The sequence of polymorphs under pressure in these minerals is therefore analog: corundum/ilmenite $\rightarrow$ {\it Pbnm}-perovskite for MgSiO3 and Rh$_2$O$_3$(II) type for Al$_2 $O$_3$ $\rightarrow$ {\it Cmcm} post-perovskite. The reason for the greater stability of {\it Pbnm}-perovskite in MgSiO$_3$ versus Rh$_2$O$_3$(II)-type in Al$_2$O$_3$ is the difference in cation polyhedral types and volumes in the former, that favors for ABX$_3$-type composition.   

Total energy and linear response computations for perovskite and post-perovskite phases in the MgSiO$_3$-FeSiO$_3$-Al$_2$O$_3$ system

We perform first-principles calculations for the perovskite (pv) and post-perovskite (ppv) phases in the (Mg,Fe,Al)(Si,Al) O$_3$ system, the dominant chemical system of the Earth’s lower mantle. We consider different chemical compositions in this system for which we analyze the structural, electronic, elastic and lattice dynamical properties. We use total energy and linear response techniques within LDA and GGA, as implemented in the code ABINIT. We perform calculations in the 0-180 GPa pressure range, in 30GPa increments, to characterize the behavior of these materials over the whole Earth’s mantle pressure range (up to 137 GPa). We find that the addition of Al in MgSiO$_3$ increases the pv-ppv transition pressure, while the addition of Fe largely decreases this pressure. The pv phase of FeSiO$_3$ is unstable with respect to ppv at all pressures. Fe reduces the electronic gap in both pv and ppv, the Fe-end-member being high-spin and metallic. The pv phase of FeSiO$_3$ is ferromagnetic while the ppv phase is ferromagnetic at low pressures and antiferromagnetic at high pressures. This research is supported by the NSF grant EAR-0310139 and the Carnegie Institution of Washington.   

Sub-lattice melting in hydrogen-rich alloys

Hydrogen at low temperatures has recently been predicted to undergo a solid to quantum liquid transition at sufficiently high pressure. The resulting quantum liquid is believed to be a metal and exhibit both superfluidity and superconductivity. Pseudopotential methods combined with nonlinear response theory have been able to give a simple, qualitative account for this transition. We generalize these methods and apply them to hydrogen-rich compounds in order to determine effective pair and triplet interactions in such alloys. By comparing the binding energies obtained to the proton or deuterium zero-point energy we can determine whether sub-lattice melting can be expected, and obtain an estimate of the required pressures.   

X-ray spectroscopy studies of liquid water

We have investigated the electronic structure of water and ice using a combination of experimental and theoretical techniques [1]. Measurements have been performed on the liquid using both X-ray Absorption (XAS) and X-ray Raman Spectroscopy. The spectrum of the liquid is distinctly different from that of the bulk ice, where the liquid shows a distinct pre-edge feature and a strong enhancement of the intensity at the edge. Through spectrum simulations and model experiments (bulk and surface of ice) we show that the specific features in the liquid spectrum are due exclusively to asymmetric configurations with only two strong hydrogen bonds: one donating and one accepting, indicating that the liquid consists of rings or chains embedded in a disordered H-bond network [1]. Current molecular dynamics techniques fail to predict these new experimental data. Recent results on disparities in supercritical water will also be discussed. \newline \newline [1] Wernet et al, Science 304, 995-999 (2004)   

Pressure-induced Phase Transition of Confined Water from ab initio Molecular Dynamics Simulation

We present an ab initio molecular dynamics study of pressure induced melting of an ice thin film confined between two parallel metal surfaces. The ice-to-water phase transition has been observed at a pressure of roughly 0.5~GPa, when the film is compressed by 6.6 percent. The latter is in agreement with the volume change in the melting of bulk ice. The effects of non-adiabatic compression on the layer-dependent momentum distribution and the electronic redistribution at the interfaces are presented and discussed.   

First principles investigation of the ice VII-VIII (order-disorder) phase boundary

Phase boundaries among the various forms of ice are difficult to determine experimentally because of the large hystereses involved. Theoretically there are also great challenges, including order-disorder (OD). The ice VII-VIII boundary, a typical OD boundary, has been reasonably well constrained experimentally. We present a first principles study consisting in the complete statistical sampling of molecular orientations within a 16 molecules supercell. This supercell size accounts well for several aspects of this transition, including the transition temperature and its pressure dependence in the high P range. The differences at lower Ps are likely to be related with the insufficiencies of DFT, within LDA or GGA, to describe the hydrogen bond. Research supported by NSF/EAR 013533 (COMPRES), 0230319, and NSF/ITR 0428774 (VLab).   

Dissociation of planetary ices at high P,T

The major components of the ice layer of the Jovian planets are methane and ammonia [1]. By adopting laser-heating techniques, we examined methane and ammonia under high pressure and high temperature. XRD and Raman studies showed that ammonia generates nitrogen and that methane dissociates into various hydrogen end products and carbon end products under extreme conditions. From this study, we suggest that the inner layer of Jovian planets gradually loses methane and ammonia by their dissociation process. [1] W.B. Hubbard, Science, 214, 145 (1981).   

Session A12: Vortices in Superconductors I

Domain Regime in the 2D Disordered Vortex Matter

The 2D disordered vortex matter is simulated by large-scale molecular dynamics methods at T=0. Performing sweeps with the magnetic field and the disorder, we find a disordered Vortex Glass/Molasses Regime at high fields/disorder, and a Domain Regime at low fields/disorder. We do not find evidence for a region with a dilute gas of dislocations, assumed by some theoretical approaches. In the Domain Regime the dislocations are organized into domains walls, defining large dislocation-free domains. The boundary between these regimes exhibits reentrant behavior as a function of the magnetic field. This boundary is a crossover, characterized by the roughening of the domain walls.   

Magnus Force in Discrete and Continuous Two-Dimensional Superfluids

Motion of vortices in two-dimensional superfluids is studied by solving the Gross-Pitaevsky equation numerically on a uniform grid. Simulations show that in the limit of small lattice spacing, corresponding to a nearly Galilean-invariant case, vortices move with the superflow, while on coarse grids their motion depends on the orientation of the superflow relative to the grid. In particular, when the superflow is parallel to one of the primitive vectors of the grid, vortices in the coarse limit move perpendicular to the superflow. Thus, in this case, we observe a crossover from the full Magnus force in a Galilean-invariant system to a sharply reduced effective Magnus force in a discrete system. The latter regime corresponds to existing experiments on vortex motion in Josephson junction arrays.   

Density functional theory for freezing transition of interlayer Josephson vortex line liquid

The freezing transition of interlayer Josephson vortex line liquid into lattice is studied in terms of the density functional theory for crystallization. The interplay between intervortex repulsion and layer pinning in presence of thermal fluctuations is analyzed. The enhancement of melting temperature by the layer pinning at high magnetic fields is revealed clearly where anisotropy scaling is broken. In certain regime of magnetic field a smectic phase is stabilized by strong layer pinning. The freezing of vortex liquid is then two-step, a second-order liquid-smectic transition and a first-order smectic-lattice transition. B-T phase diagrams for typical high-Tc cuprates will be presented.   

Dissociation of vortex stacks into fractional-flux vortices

We discuss the superconducting phase transition in a finite system of magnetically coupled superconducting layers. Transverse screening is modified by the presence of other layers resulting in topological excitations with fractional flux. The presence of additional layers leads to drastic modifivations in the potential between individual vortices in the same layer. Vortex stacks trapping a full flux and present at any finite temperature undergo a dissociation transition which corresponds to the depairing of fractional-flux vortices in individual layers. We propose an experiment with a bi-layer system allowing to identify the dissociation of bound vortex molecules.   

Density oscillation of flux lines induced by a single twin plane with point pins

In (1+1)-dimensional vortex matter with a single columnar defect [1], the density of flux lines parallel to a single columnar defect shows Friedel-like oscillations in a tilted field, with the correlation length of the amplitude of the oscillation diverging as the transverse field component vanishes. In this study, we show that similar behaviors are also observed in vortex states in three dimensions with a single twin plane and point pins. In a magnetic field along a twin plane at low enough temperatures, power-law decay of the density oscillation of flux lines is observed for sparse point pins, consistent with the existence of a Bragg glass phase. As the density of point pins increases, it changes to exponential decay in the strongly- pinned vortex glass regime. A similar density oscillation is observed in a slightly tilted field, and the range of the oscillation is enhanced in the limit of sparse point pins. [1] W. Hofstetter et al., Europhys. Lett. 66 (2004) 178; I. Affleck et al., J. Stat. Mech.: Theor. Exp. (2004) P10003.   

Dynamics of half-quantized vortices in nanoscale superconducting composite structures (d-dot)

Mesoscopic or nanoscopic superconductors shows sometime peculiar phenomena. When nanosize high-Tc d-wave superconductor is embedded in conventional s-wave superconductor matrix, it shows various spontaneous magnetic flux depending on the shape of the d-wave superconductor. The appearance of static magnetic field shows such state breaks the time reversal symmetry. So there another equally stable state, which has completely reversed magnetic fluxes. Therefore this d-wave superconducting dot in s-wave superconductor, which we call as ``d-dot,'' always has equally stable two states. For this system, we developed the numerical simulation method, which is based on the two component Ginzburg-Landau equation and the finite element method, and investigated the spontaneous magnetic field distribution. In this study we extended these previous study and we developed dynamical simulation method for d-dot's. Then we study the effect of external current to the spontaneous magnetic fluxes. We show external current causes the transition between two equally stable magnetic flux structures. This means the potential applications of these d-dot's.   

Off-equilibrium dynamics and effective temperature of flux lines with random pinning and the vortex glass phase

We investigate the low-temperature off-equilibrium dynamics of elastic lines embedded in three-dimensional disordered media using Langevin dynamics. The model describes interacting vortex lines in high-temperature superconductors with random pinning. We first study the case of isolated flux lines. At high temperatures the dynamics is stationary and the fluctuation dissipation theorem (FDT) holds. At low temperatures a simple multiplicative aging is found, as recently observed numerically in the directed polymer problem in (1+1) dimensions or analytically in the $2d$ XY model in the spin-wave approximation. Besides, the FDT is violated and we found a well defined effective temperature characterizing the slow modes of the system. This implies the existence of a dynamic crossover between a high-temperature equilibrium dynamic phase and a low-temperature glassy dynamic phase. We then discuss how the off-equilibrium dynamics and effective temperature are affected when the flux line interactions are taken into account and its relationship with the vortex glass phase.   

Does the Vortex Glass Exist in Two Dimensions?

The nature of phase coherence in two-dimensional vortex lattices with random point pins is studied at the extreme type-II limit via the corresponding $XY$ model with uniform frustration. In particular, after taking the Villain approximation, we perform numerical Monte Carlo simulations of the resulting non-neutral Coulomb gas ensemble over the square lattice at low temperature [1]. Identical $\delta$-function pinning centers equal in number to the total number of vortices are also located at random throughout the model grid. A phase-coherent Bragg glass exists at the lowest levels of disorder pinning, with no unbound dislocations quenched in. Upon an increase in the strength of the pinning disorder, this phase becomes unstable to hexatic vortex glass states that show a diminished phase coherence, as well as to hexatic vortex liquid states that show no phase coherence at all. Yet stronger pinning disorder unbinds quenched-in pairs of disclinations, which results in a (pinned) vortex liquid phase that shows only short-range translational and orientational order.\\[4pt] [1] C.E. Creffield and J.P. Rodriguez, Phys. Rev. B {\bf 67}, 144510 (2003).   

Slow dynamics of an elastic string in a random potential

We study the slow dynamics of an elastic string in a two dimensional pinning landscape by means of Langevin dynamics simulations. We find that the Velocity-Force characteristics are well described by the creep formula predicted from phenomenological scaling arguments. However, at strong disorder, the creep exponent $\mu$ and the roughness $\zeta$ of the string display a clear deviation from the values $\mu \approx 1/4$ and $\zeta \approx 2/3$ expected assuming a quasi-equilibrium-nucleation picture of the creep motion. We also analyzed the non-stationary relaxation of the string towards the steady state. We identify a slowly growing length $L(T,F,t)$ separating equilibrated and non-equilibrated length scales during the relaxation. For equilibrated lengths, $l < L$, we find a roughness $\zeta \approx 2/3$ at $F=0$ while for small $F > 0$ an ``excess'' of roughness $\zeta > 2/3$ is always observed.   

Phase diagram of the vortex system in layered superconductors with random pinning

Density functional theory based on a model free energy functional is used to study structural and thermodynamic properties of the vortex system in highly anisotropic layered superconductors with random pinning. For low concentrations of random columnar pins perpendicular to the layers, we find three distinct phases: a topologically ordered Bragg glass, a polycrystalline Bose glass and a vortex liquid. As the temperature is increased, the low-temperature Bragg glass transforms into the vortex liquid in two steps: these two phases are separated by a small region of the Bose glass phase. The Bragg glass phase disappears as the pin concentration is increased and the two-step first-order melting found at low pin concentrations is replaced by a single continuous transition from the Bose glass to the vortex liquid. This transition corresponds to the onset of percolation of liquid-like regions across the system. Results obtained from similar calculations for systems with random point pinning will also be presented.   

Dynamic properties of vortex states as modeled by the disordered 3D uniformly frustrated XY model

Dynamic properties of the uniformly frustrated 3D XY model with point disorder are studied as a model of vortex flow in high temperature superconductors. Using both Monte Carlo and Resist Shunted Junctions dynamics, we compute the resistivity of the interacting vortex system both from equilibrium voltage fluctuations and from the response to a finite driving current. For a sufficiently strong disorder we find a non-trivial behavior: In the solid phase the resistance is very low, suggestive of a pinned vortex line lattice, but increases rapidly at the melting transition. In the simulations with a finite current we find that the same behavior is seen only for very small currents $I\approx 10^{-4}$. For larger currents the voltage instead decreases at the melting of the vortex lattice. This seems to be due to the increasing adjustment of the vortex lines to the pinning potential in the liquid phase.   

Static and dynamic properties of pinned flux-line liquids

We study the equilibrium statics and nonequilibrium driven dynamics of flux line liquids in presence of a random pinning potential. Under the assumption of replica symmetry, we find in the static case using a replica Gaussian variational method that the only effect of disorder is to increase the tilt modulus and the confining ``mass" of the internal modes of the flux lines, thus decreasing their thermal wandering. In the nonequilibrium, driven case, we derive the long scale, coarse-grained equation of motion of the vortices in presence of disorder, which apart from new Kardar-Parisi-Zhang nonlinearities, has the same form as the equation of motion for unpinned vortices, with renormalized coefficients. This implies, in particular, that the structure factor of a disordered vortex liquid has the same functional form as in the absence of pinning. The expression of the static structure factor derived within our approach is consistent both with experimental data and with the standard theory of elasticity of vortex lattices.   

Interactions Effects and Effective Temperature in Driven Vortex Fluids with Random Pinning

We study numerically the effects of vortex-vortex interactions on the fluctuation-dissipation relations of driven vortex fluids with random pinning. We show that the shaking temperature $T_{\tt sh}$ defined phenomenologically by Koshelev and Vinokur [1] corresponds to the effective transverse temperature $T_{\tt eff}$ defined from a generalized fluctuation-dissipation theorem [2,3] only in the limit of noninteracting vortices. $T_{\tt eff}$ is thus sensible to the short range anisotropic correlations in the driven fluid. In the noninteracting limit we use a simple model to derive an expression for $T_{\tt eff}$ which is valid for all finite velocities. [1] A.E. Koshelev and V.M. Vinokur, Phys. Rev. Lett. {\bf 73},3580 (1994). [2] L.F. Cugliandolo, J. Kurchan, and L. Peliti, Phys. Rev. E {\bf 55}, 3898 (1997) [3] A. B. Kolton {\it et al.}, Phys. Rev. Lett. {\bf 89}, 227001 (2002).   

Session A13: Spectroscopic Properties of Superconductors

Resonance properties in optimally and overdoped Y-123 - investigation of phonons and the electronic background

We have studied the variation of the 2$\Delta$-gap like features and the phonons in optimally to overdoped Y-123 with incident photon energy E$_{i}$. The optimally doped YBCO and overdoped YBCO exhibit a complex behavior in the gap feature when the incident photon energy is tuned from 1.8 eV to above 5 eV. Our results highlight the composite nature of the 2$\Delta$-gap like peaks in the HTS. Below the superconducting T$_{c}$ we see a modification of spectral weight and the appearance of the 2$\Delta$ like features. In B$_{1g}$ Raman symmetry these features change their shape, peak position, and intensity with increasing E$_{i}$. For E$_{i}$ 3.5 to 4 eV we find little redistribution, only a strong new peak at Raman frequencies well above the accepted values of twice $\Delta _{max}$. This peak is located at 620 cm$^{-1}$ in optimally doped samples, it shifts to lower energies and broadens in overdoped materials. The matrix element of this peak shifts as a function of doping by 0.2 eV to higher incident photon energies, outlining the interplay between the local electronic structure and superconductivity. Strong resonances of the chain disorder-induced phonons appear at around 4.1 eV, which are absent in the overdoped material.   

Raman Spectroscopy of the electron-doped cuprates in magnetic field

We investigate the influence of magnetic field on the electronic excitations across the superconducting (SC) gap of the electron-doped cuprates R$_{2-x}$Ce$_{x}$CuO$_{4-\delta }$ (R = Pr, Nd) for 0.13$<$x$<$0.18. We report the anomalous result that the 2$\Delta $ coherence peak energy decreases rapidly with increasing field. In sharp contrast, the magnetic field has only a weak effect on the coherence peak energy in the hole-doped cuprates and conventional superconductors. We determine effective upper critical field lines H$^{\ast }_{c2}$ at which the superfluid stiffness vanishes and H$^{2\Delta }_{c2}$ at which the SC amplitude is suppressed. We find that H$^{2\Delta }_{c2}$ is larger than H$^{\ast }_{c2}$ for all dopings, especially for x\underline {$<$}0.15. Both H$^{2\Delta }_{c2}$ and H$^{\ast }_{c2}$ decrease more rapidly with doping than T$_{c}$ or 2$\Delta $ for x$>$0.15 (over-doped). The rapid increase of SC coherence length (several hundred Angstrom) with increasing doping implies a weak pair potential with less momentum dependence. This work was supported by NSF grant No. DMR 01-02350.   

Raman Investigation of Zn-Doped YB$_{2}$Cu$_{3}$O$_{6.99}$ Single Crystals

We report the results of a Raman scattering study of the effects of doping YBa$_{2}$Cu$_{3}$O$_{6.99}$ crystals with 1.0{\%} Zn, which decreased the critical temperature from 92K to 81K. We found that the strength and shape of the 340cm$^{-1}$ B1g phonon anomaly are nearly the same in both the doped and undoped crystals. This observation suggests that Zn-doping does not significantly change the hole concentration. Furthermore, we extracted the B1g electronic continuum by removing the phonon contributions using a decoupling procedure. The resulting electronic spectra from the Zn- doped and undoped crystals are again essentially identical and are consistent with the scattering of light by the superconducting quasi-particles across a d-wave gap. The equal intensities of the B1g electronic continua again indicates that the carrier concentrations are the same in both the Zn-doped and undoped crystals. This in turn implies that Zn-doping does not lead to the opening of a pseudo gap.   

Competing CDW and SC order parameters in NbSe$_2$

We study collective excitations in NbSe$_2$ by low-frequency electronic Raman spectroscopy as a function of temperature, magnetic field and excitation energy. We discuss observed relations between collective modes induced by charge density waves (CDW) and superconducting (SC) order parameters. We compare Raman data to recent STM and ARPES studies.   

Understanding the totally symmetric intramolecular vibrations in $\kappa$-phase organic superconductors

We report magneto-infrared measurements of three quasi- isostructural $\kappa$-phase organic molecular solids: $\kappa$-(ET)$_2$Cu[N(CN)$_2$]Br (T$_c$=11.6 K), $\kappa$-(ET)$_2$Cu(SCN)$_2$ (T$_c$=10.4 K), and the non-superconducting $\kappa$-(ET)$_2$Cu[N(CN)$_2$]Cl analog. Our results support the contributing role of electron-molecular vibrational coupling in the pairing mechanism of layered organic superconductors, and we identify the most important totally symmetric modes in $\kappa$-(ET)$_2$Cu[N(CN)$_2$]Br within the non-planar molecular building block picture.   

Resonant Inelastic X-ray Scattering Study of the Model Superconductor HgBa$_2$CuO$_{4+\delta}$

The characteristics of the elementary excitations in high-temperature superconductors (HTSC) and their Mott-insulating parent compounds remain a controversial issue after many years of extensive study by different spectroscopic methods. Energy- and momentum-resolved resonant inelastic x-ray scattering (RIXS) is gaining in importance as a powerful tool in the study of elementary charge excitations in HTSC. We report a RIXS study of charge excitations in the 2 - 8 eV range in the structurally simple compound HgBa$_2$CuO$_{4+\delta}$ at optimal doping ($T_c = 96.5 $ K) at several high-symmetry points in the Brillouin zone. The spectra exhibit a significant dependence on the incident photon energy which we carefully utilize to resolve a multiplet of electron-hole excitations, including an excitation at 2 eV, which was previously observed in undoped parent compounds. The observation of the 2 eV excitation is suggestive of a remnant charge transfer gap in the superconducting phase. Our data are consistent with a relatively weak ($ < 0.5$ eV) dispersion of all excitations.   

Momentum-Structure of Remnant Mott-gap in Prototype Doped Cuprates

Momentum dependence of charge excitation of 2-D prototype cuprate classes La$_{2-x}$Sr$_x$CuO$_4$ and Nd$_{2-x}$Ce$_x$CuO$_4$ with different doping levels (x) is measured using high resolution resonant inelastic X-ray scattering. Although a low-energy continuum is built up with doping, a remnant excitation gap behavior continues to exist even in highly doped metallic phases which we have studied in some detail. The excitation of the Mott gap becomes less dispersive asymmetrically with the increase of doping level and suggests a~ many-body coupling between charge fluctuation and magnetic order of the lattice. The results can be qualitatively described within the framework of t-t$'$-t$''$-U model.   

Direct measurement of charge transfer phenomena at ferromagnetic/superconducting oxide interfaces

YBa$_{2}$Cu$_{3}$O$_{7-x}$/La$_{0.67}$Ca$_{0.33}$MnO$_{3}$ ferromagnetic/superconducting interfaces have been analyzed by scanning transmission electron microscopy (STEM) and electron energy loss spectroscopy (EELS) with monolayer resolution. Extensive charge transfer occurs between the manganite and the superconductor, in a manner similar to modulation-doped semiconductors, which explains the reduced critical temperatures of heterostructures. Since CMR and HTCS oxides are extremely sensitive to doping these charge transfer processes at the interfaces will directly affect the superconducting and/or magnetic properties of the individual layers. This behavior has not been observed with insulating PrBa$_{2}$Cu$_{3}$O$_{7}$ layers in YBa$_{2}$Cu$_{3}$O$_{7-x}$/ PrBa$_{2}$Cu$_{3}$O$_{7 }$superlattices. EELS in these samples provides direct confirmation that holes in the YBa$_{2}$Cu$_{3}$O$_{7-x}$ are located on the CuO$_{2}$ planes.   

Evidence for different spin-orbit interactions in the Ba and Cu layers of YBa$_2$Cu$_3$O$_7$ at the Ba and Cu L2,3 edges.

The L$_{2,3}$ edges arise from the 2p$_{3/2,1/2}$ core transitions to final band states of d$_{5/2,3/2 }$character, and may give information on the uneven population of the spin-orbit split bands of d-character that lead to a material ferromagnetic or anti-ferromagnetic properties. Absorption, fluorescence and enhanced scattering data at the Ba and Cu L$_{2,3}$ edges of a single crystal, and a 50 nm film were measured respectively at SSRL (Station 2-3) and LBNL-ALS (6.3.1 Station: P. Nachimuthu chamber for fluorescence and total electron yield and J.B. Kortright chamber for enhanced scattering). The theoretical relative intensity I(L$_{3})$/I(L$_{2})$=2 is satisfied for the reference material BaBr$_{2}$ but deviations from 2 at the respective L$_{2,3}$ edges: Ba by --14{\%} and Cu by +19{\%} in a material that is neither ferromagnetic nor anti-ferromagnetic suggests the presence of magnetic interactions for the individual elements that cancel out in the material. All models for magnetic interactions should then take into account the coupling of the Cu and Ba atoms in adjacent layers of YBCO, a 2D material. DOE, NATO, NSF, Dreyfus supported.   

Real time optical observation of precursor phases during YBCO thin film growth

We report on our findings using real-time Fourier Transform Infrared (FTIR) radiance and reflectance measurements during high rate electron beam deposited [100 angstroms/sec] YBa2Cu3O7 (YBCO) films [1]. The data can best be explained by assuming the presence of a liquid or glassy phase from which the YBCO crystal grows [2]. We have found that the optical properties of the as deposited material change dramatically upon absorption of oxygen. It is inferred from the strong thin film interference fringes which appear in the reflectance spectrum. The fringes subsequently change amplitude when YBCO precipitates from the liquid; the rate of precipitation can be controlled by oxygen pressure and substrate temperature. The nature of the liquid or glassy phase is studied by XPS and XRD on quenched films at various stages. Subsequently, YBCO nucleation and growth is monitored ex situ by real time x-ray scattering using an ambient controlled hot stage. \\[4pt] [1] G. Koster, J-U. Huh, R.H. Hammond and M.R. Beasley, in preparation.\\[0pt] [2] T. Ohnishi, J.-U. Huh, R.H. Hammond and W. Jo, 2004\textit{ J. Mater. Res.}\textbf{ 19} 977; A. Kursumovic\textit{ et al.}, Supercond. Tech. {\&} Sci. 17 (2004) 1215   

Electron irradiation induced oxygen ordering in YBCO twin-free crystals.

Oxygen ordering in CuO chain plane of YBCO is interesting topic because the distribution of oxygen, or the length of chain-fragments, plays an important role in carrier-doping of the CuO2 plane; longer chain-fragments are more effective in hole doping and result in higher Tc under the same overall oxygen stoichiometry. In this presentation, we report on the low energy electron irradiation effects in twin-free oxygen deficient YBCO single crystals via comparative Raman spectroscopy studies of electron irradiated and non-irradiated areas at room temperature. We observed that low energy electrons heal the existing point defects of CuO chains and enhances oxygen ordering of twin-free but oxygen deficient YBCO. The comparison of the polarized Raman spectra from non-irradiated and irradiated areas provides clear indications of extended average length of the chains without changing the overall oxygen content.   

Optical Characterization and Superconducting Behavior of MgB2 Thin Films

MgB$_{2}$ thin films were sequentially deposited from pure metal sources on optically transparent glass. The resulting films were investigated by SCEM, optical microscopy, EDAX, and by optical transmission. As deposited multilayer films show high metallic reflectivity in 400 -- 1000nm regime with very little transmission. After \textit{ex situ} annealing (525, 550, 575 {\&} 600$^{\circ}$C) the films appeared progressively less shiny and took up darker hues. The films rendered superconducting after annealing at 550$^{\circ}$C with Tc of 38K, approaching that of the bulk material. Also, upon reaction a direct band gap opened up. For all the films absorption coefficient alpha and photon energy followed a Tauc behavior rather than a Cody type behavior. As annealing temperature was raised the Tauc plot slope was observed to decrease monotonically. However the band gap first increased to a maximum of 2.5eV (550$^{\circ}$ C) and then rapidly decreased to below 0.5 eV with 600$^{\circ}$ C anneal. Our results indicate that MgB$_{2}$ films have excellent optical properties and are commensurate with VLSI and optoelectronics processing.   

Direct Observation of Cooperative Doping Mechanisms at Grain Boundaries in Ca-doped YBa$_2$Cu$_3$O$_{7-\delta}$

Atomic-column resolved electron energy-loss spectroscopy (EELS) was used to study the effects of Ca-doping on the local atomic and electronic structure of grain boundaries in YBa$_{2}$Cu$_{3}$O$_{7-\delta }$. Grain boundary doping with Ca has been shown to increase the grain boundary critical current, and it has been previously suggested that Ca$^{2+}$ substitutes for Y$^{3+}$ thus providing additional holes. We have performed atomic-column resolved EELS of the grain boundary dislocation cores and we will show that in the highly strained regions of the grain boundary plane Ca segregates to Cu and Ba sites, where it does not provide holes directly. However, due to the resulting strain relief, the oxygen deficiency in the vicinity of the grain boundary is reduced and the hole concentration increased. The results demonstrate that to improve grain boundary J$_{c}$, ionic size may be more important than the electronic nature of the impurity.   

Search for the Coexistence of Magnetism and Superconductivity in Ce$_{1-x}$Gd$_x$Ru$_2$ using the Mossbauer Effect

We use the $^{99}$Ru Mossbauer Effect(ME) to look for evidence for the coexistence of magnetism and superconductivity in Ce$_{1-x}$Gd$_x$Ru$_2$. These compounds have superconducting and ferromagnetic phases. The T$_c$(x) and $\theta$(x) curves cross at $x=0.135$. Here $\theta$(x) is the Curie temperature found by extrapolating magnetic susceptibility data below T$_c$. A sample of Ce$_{0.93}$Gd$_{.07}$Ru$_2$ was prepared with enriched $^{99}$Ru(97\%) for which T$_c$=4.5K and $\theta$=4.19K. The ME spectrum measured at 2.2K showed a pure quadrupole with a splitting of 0.36mm/s and an isomer shift of 0.12mm/s with no hyperfine field. The absence of the magnetic field in the sample brings into question the use of extrapolated magnetic susceptibility data as evidence for the coexistence of magnetism and superconductivity. Comparing the ME spectrum with that of CeRu$_2$(T$_c$=6.1K) one sees that there is an impurity Ru phase. This suggests that previous evidence for magnetism below T$_c$ on the basis of $^{57}$Fe ME could be attributed to an impurity Fe phase.   

Session A14: Focus Session: Molecular-Scale Electronics and Sensors I

Electronics at the molecular level

In the early 1970s, Aviram and Ratner suggested the notion that molecular configurations might be used to carry and rectify electronic current. This notion was put forward well before its time, for today, some thirty years later, with the remarkable progress in nano tool development and material process capabilities, the concept of electronic conduction in molecular systems is now being experimentally tested in laboratories around the world. Correspondingly, over the years, there has been a substantial effort in the theoretical modeling of molecular configurations which has shed enormous light on the atomic details of the electron transport processes at the molecular level. The idea of considering electronic functionality at the molecular scale is not a surprise; it was very much embodied in Feynman's early vision captured in ``There's Plenty of Room at the Bottom,'' and has always been part of the speculative horizon of Moore's law. Visionaries often speculate about the possibilities and opportunities to emerge from the molecular scale; but the challenges and barriers to success and realizability are substantial. The intention of this presentation is to discuss some of the basic possibilities and limitations of molecular scale electronics. Further, the presentation incorporates some of our recent quantum modelings on connected molecular systems; here we model metal contacts to molecular clusters in an {\em exact} framework using a molecular self-consistent field approach so as to calculate realistic, electric field dependent transport properties for the molecular system, and to study the role of the contact-molecule interface in influencing the transport properties of the entire molecular system.   

Constructing a Computer from Molecular Components

Constructing a Computer from Molecular Components Research efforts directed toward constructing a molecular computer will be described in the context of recent developments in nanotechnology. Routes will be outlined from the synthesis of the basic building blocks such as wires and alligator clips, to the assembly of the processing functional blocks. Specific achievements include: (1) isolation of single molecules in alkane thiolate self-assembled monolayers and addressing them with an STM probe, (2) single molecule conductance measurements using a mechanically controllable break junction, (3) 30 nm bundles, approximately 1000 molecules, of precisely tailored molecular structures showing negative differential resistance with peak-to-valley responses far exceeding those for solid state devices, (4) dynamic random access memories (DRAMs) constructed from 1000 molecule units that possess 15 minute information hold times at room temperature, (5) demonstration of single-molecule switching events and (6) initial assemblies and programming of molecular CPUs in a NanoCell configuration that show room temperature electronic memory with days or electronic hold time, and the programming of logic gates such as AND, OR, NAND and NOR gates. Full silicon-molecule interfaces are used in the generation 3 NanoCell, as well as molecular FETs (MoleFETs). Finally, a molecular testbed has been developed that involves only semiconductor contacts (no metal contacts) to the molecules, thereby mitigating electromigration.   

CMOL: A New Concept for Nanoelectronics

I will review the recent work on devices and architectures for future hybrid semiconductor/molecular integrated circuits, in particular those of ``CMOL'' variety [1]. Such circuits would combine an advanced CMOS subsystem fabricated by the usual lithographic patterning, two layers of parallel metallic nanowires formed, e.g., by nanoimprint, and two-terminal molecular devices self-assembled on the nanowire crosspoints. Estimates show that this powerful combination may allow CMOL circuits to reach an unparalleled density (up to 10$^{12}$ functions per cm$^{2})$ and ultrahigh rate of information processing (up to 10$^{20}$ operations per second on a single chip), at acceptable power dissipation. The main challenges on the way toward practical CMOL technology are: (i) reliable chemically-directed self-assembly of mid-size organic molecules, and (ii) the development of efficient defect-tolerant architectures for CMOL circuits. Our recent work has shown that such architectures may be developed not only for terabit-scale memories and naturally defect-tolerant mixed-signal neuromorphic networks, but (rather unexpectedly) also for FPGA-style digital Boolean circuits. [1] For details, see http://rsfq1.physics.sunysb.edu/$\sim $likharev/nano/Springer04.pdf   

Simple, High Yield Nano-device Fabrication via SWNT Controlled Growth from a Catalytic Block Copolymer

We report a simple process in which single walled carbon nanotubes (SWNT) are grown specifically to allow for a three step device fabrication. Our extremely pure SWNT bundles (dia. 2-5 nm, length $\sim $10 $\mu$m) were produced via a chemical vapor deposition method where a ferrocene containing block copolymer was utilized as the catalyst. Unlike other methods, the nanotube surface coverage density was manipulated via the polymer film thickness to create approximately three to six tubes per 100 $\mu$m$^{2}$. This allows for the direct deposition of metal electrodes onto the silica/nanotube surface without tedious positioning of the nanotubes between metal contacts as an additional processing step. Thus, over 100 working nanodevices can be constructed on a single 1"x1" wafer with this simple three step process: 1) spin casting catalytic polymer film; 2) CVD; 3) metal electrode deposition. I-V measurements show large current flow between gold electrodes ranging from hundreds of $\mu$A to a few mA as a result of the large number of bridging nanotubes. Ease of construction and high device yield make this process a promising candidate for applications as nano-chemical and biological sensors.   

A nanodot lattice with molecular interconnects providing reconfigurable logic

Using circuit models we investigate a FCC lattice of nanodots, interconnected by molecules with voltage- switchable linear as well as non-linear conductances, deposited on a lithographically defined set of contacts. By applying voltages to these contacts the switchable molecules open conductive paths. We open these according to a target circuit scheme that is able to implement a large set of logic gates, including half-adders. The target circuit in the nanodot lattice consists of an analogue summation node connected to a bistable latch via an NDR connection. The summation node weighs the input, driving and ground voltages. The NDR connection and the bistable latch together converts the analogue voltages to a logical one if in a mid-region and zero if low or high.We have also designed a lithographic context including clocks and transistors such that we can interconnect logic gates to e.g. implement the N-bit adder described by Tour et al. (1). C Husband, S Husband, et al. (2003). ``Logic and memory with nanocell circuits." IEEE Transactions on Electron Devices 50(9): 1865-1875.   

The Quantum Interference Transistor

We propose a new class of molecular transistor based on quantum interference. This new type of transistor is smaller than most proposed molecular transistors, yet possesses the important characteristics of traditional macroscopic devices. The proposed device has a broad, step-like I-V characteristic and amplifies current in a controllable way. Numerical calculations making use of the non-equilibrium Green function technique and Landauer-B\"uttiker formalism illustrate this.   

Kondo Effect in Bare Electromigrated Break Junctions

Electromigrated break junctions are one of only a very few systems currently available that provide sub-nanometer electrode gaps in a gated geometry. They have been used in several experiments over the past few years to measure transport through nanometer-scale objects such as single molecules. Our measurements show that the electromigrated electrode system--even by itself, without added nanoparticles-- is richer than previously thought. This talk will present gate- dependent transport measurements of Kondo impurities in bare gold break junctions, generated with high yield using an electromigration process that is actively controlled. An unexpected behavior of the splitting is observed in the crossover regime, where spin splitting is of the same order as the Kondo temperature. The Kondo resonances observed here may be due to atomic-scale metallic grains formed during electromigration.   

Analysis of Nanocluster-Based Single-Electron Transistors

The possibility of making single-electron transistors (SETs) out of ultra-small metal nanoclusters is intriguing because of the potential for room temperature switching in structures that are 3-5nm in size. In this work the electrical properties that could be expected from such nanocluster-based SETs are explored using numerical simulation. The I-V characteristics are computed using the orthodox theory of Coulomb blockade with only one-electron processes considered and with the capacitances and tunneling resistances of the particular cluster configurations obtained by numerical simulations. These latter calculations must be performed in three dimensions because of the ultra-small radii of curvature involved. Of most interest is an examination of the effect of various structural imperfections on the I-V characteristics since this, in effect, sets the assembly tolerances that would have to be met for a viable nanocluster-based SET technology.   

Water Soluble Conducting Polymer Field Effect Transistor for Sensor Application

We studied the water soluble polythiophene based conducting polymer field effect transistor for chemical and biosensors at nanoscale. Sodium poly [2-3(thienyl) ethoxy-4-butylsulphonate)] (SPBS) is a water soluble polymer. Electrical transport property as a function of gate voltage was investigated using a home-built nanomanipulator with four nanoprobes, which is connected to a picoammeter and an impedance analyzer. In conjunction with this, we studied molecular and electronic structures by a scanning tunneling microscope. The interface between electrodes and polymer play an important role in the charge transport properties.   

Molecular Dynamics Simulation of Hybridization Kinetics of Surface-Grafted DNA

Increasing interest in development of biosensors results in a need to understand all levels of design and operation of these devices. We focus on a DNA biosensor that is based on molecular recognition between nucleic acids and the subsequent hybridization process. In order to effectively design such a biosensor, it is necessary to understand the effects of grafting density and sequence mismatch on hybridization efficiency. These are of direct relevance to device performance as hybridization efficiency and density determine the signal strength. We use molecular dynamics (MD) simulations to elucidate effects of grafting density on hybridization efficiency and hybridization density as well as the effects of sequence mismatches on hybridization in surface-grafted DNA. We have developed a ``minimal'' coarse-grained model of DNA suitable for long timescale molecular dynamics simulation. We have calculated the hybridization efficiency and conformational order parameter Q$_{\alpha \beta } $(\textbf{r, r'}) between target and probe DNA from MD trajectories. We find that the hybridization efficiency decreases with increase of grafting density, as seen in experiments. We also find that target DNA binding to multiple probes is a dominant effect at higher grafting densities and represents a major obstacle to efficient hybridization efficiency. We show that the width and amplitude of Q$_{\alpha \beta }$(\textbf{r, r'}) are sensitive to the grafting density and number mismatches on the target, respectively.   

An Optical Biosensor for Bacillus Cereus Spore Detection

We demonstrate a new transduction scheme for optical biosensing. Bacillus cereus is a pathogen that may be found in food and dairy products and is able to produce toxins and cause food poisoning. It is related to Bacillus anthracis (anthrax). A CCD array covered with micro-structured glass coverslip is used to detect the optical resonant shift due to the binding of the antigen (bacillus cereus spore) to the antibody (polyclonal antibody). This novel optical biosensor scheme has the potential for detecting 10$\sim $100 bioagents in a single device as well as the potential to test for antigens with multiple antibody tests to avoid ``false positives.''   

Session A15: Focus Session: THz Devices and Materials I

Progress and Perspectives of THz Astronomical Detectors

During the past decade, THz astronomical observations from ground based, suborbital and space platforms have become a reality. The most impressive progress has been made in heterodyne spectroscopy instrumentation due to a tremendous effort in the development of solid state local oscillators and SIS and HEB superconducting mixers. Whereas the current SIS technology may have almost reached its limit, the HEB mixers expanding far into the THz frequencies have yet a large room for improvement. The progress in direct detectors has been mostly due to the development of photoconductive detectors for wavelengths shorter than 50 $\mu $m and of composite bolometers for longer wavelengths. Both technologies have pretty much reached their limitations for improvement. Future THz astronomical space missions will require both non-parallel sensitivity and large-scale arrays of both direct and heterodyne detectors. I will discuss how these objectives can be met with recently emerged types of superconducting and semiconductor sensors utilizing new physical detection mechanisms.   

Plasmon Based Grating Gate Terahertz Detector

Double quantum well grating gate detectors have recently emerged as a widely tunable detector of millimeter wave to THz radiation. A typical device consists of source and drain contacts along with a grating gate which both modulates the carrier density and couples in the free space radiation to the plasmon modes of the double quantum well heterostructure. In a resonant mode of operation, the grating period determines the frequency range of the detector while the gate bias tunes the operating frequency. When the gate is biased near pinch-off, the detector becomes bolometric and responsivity increases dramatically. This talk outlines current progress towards combining the plasmon resonant frequency selectivity with the responsivity of the bolometric regime through the application of a split grating gate that allows for more flexible biasing of the detector while still coupling free space radiation into the device. Sandia is a multiprogram laboratory operated by Sandia Corporation, a Lockheed Martin Company, for the United States Department of Energy's National Nuclear Security Administration under contract DE-AC04-94AL85000.   

Ballistic tranport at finite frequencies in 2DEGs and quantum point contacts

In this talk, we present our work on ballistic transport in both the spatial limit and temporal limit. In the temporal limit, we have characterized[1] the ac impedance of 2DEGs contacted with ohmic contacts at frequencies above and below the momentum scattering frequency, which is of order GHz for high mobility 2DEGs. The crossover from omega tau $<$ 1 to omega tau $>$ 1 is clearly observed. Additional non-linear effects due to plasma wave rectification in gated geometries are also under investigation, which will lead to new modes of HEMT operation not limited by transit-time effects. In order to investigate this we have fabricated devices with asymmetric ac bias conditions on the source/drain. Using an integrated, on-chip high impedance transmission line to connect the source, and a low-impedance contact to a generator on the drain, we are able to quantitatively test the dc and ac behavior of HEMT devices in the resonant limit, where standing-waves of 2d plasmons are generated in the finite length gated region (the channel). Finally, we will present data on the ac impedance of quantum point contact devices, which are ballistic in both senses: device size $<$ mean free path, and frequency $>$ momentum scattering frequency. [1] Sungmu Kang, Peter J. Burke, L. N. Pfeiffer, and K. W. West, Solid State Electronics 48, 2013-2017 (2004).   

Resonant response of a field-effect transistor to an ac signal

A theoretical investigation is made of the response of a field-effect transistor (FET) to an incoming electromagnetic radiation in the presence of a perpendicular magnetic field within the framework of hydrodynamics. The treatment is valid for a nondegenerate electron gas in which the mean free path for electron-electron scattering $\lambda_{ee}$ is much smaller than the device length $L$ and than the mean free path due to collisions with impurities and/or phonons $\lambda_{coll}$. These requirements, written as $\lambda_{ee} \ll L \ll \lambda_{coll}$, are fulfilled for magnetic fields weak enough to prevent Landau quantization. It is our general observation that the shorter device lengths, weaker magnetic fields, and lower temperatures (or higher electron mobility) are most favorable to achieve a greater resonant response of the device to an ac signal. Such resonant response makes FET a promising device for new types of sources, detectors, mixers, and multipliers in the GHz and THz frequency range.   

MBE Growth of GaAs Based THz Lasers and Detectors

Terahertz (1-10 THz, or 4-40 meV, or 30-300 $\mu$m) frequencies are among the most underdeveloped electromagnetic spectra, even though their potential applications are promising for spectroscopy in chemistry and biology, astrophysics, plasma diagnostics, remote atmospheric sensing and imaging, noninvasive inspection of semiconductor wafers, and communications. This underdevelopment is primarily due to the lack of coherent solid-state THz sources and detectors. In this talk I will discuss the important MBE growth considerations needed to make semiconductor THz devices using the GaAs-AlGaAs material system. The influence of material growth parameters on the performance of GaAs-based THz quantum cascade lasers and quantum well detectors will be presented. Sandia is a multiprogram laboratory operated by Sandia Corporation, a Lockheed Martin Company, for the United States Department of Energy's National Nuclear Security Administration under Contract DE-AC04-94AL85000.   

High performance THz Quantum Cascade Lasers in very high magnetic fields

A new design for a terahertz quantum-cascade laser emitting at $\lambda \simeq 80~\mu$m is presented. A bound-to-continuum active region is coupled to an optical phonon extraction stage in order to improve the population inversion at high temperatures. The device processed with a single-plasmon waveguide shows a threshold current density in pulsed operation of J$_{thresh}=190$ A/cm$^2 $, sensibly lower of what measured in other optical-phonon based structures. Maximum operating temperature in pulsed mode is 100 K and peak powers of the order of 40 mW are observed at low temperatures. Device performances in a double-metal waveguide configuration reach 117 K in pulsed operation and 53 K in continuous-wave. This laser has been investigated also in presence of a strong magnetic field up to 27 T, in order to study the influence of an external confinement on the gain and lifetime of the electron states of a superlattice-based quantum cascade structure. Upon increasing of the magnetic field intensity, strong modulation of laser emission and threshold current are observed over the full range of magnetic field up to 27 T, with an overall reduction of the latter of a factor of two with respect to the zero field value. Spectral characterization in the presence of magnetic field has also been performed.   

Quantum Dot as High Frequency Noise Detector

We present the experimental realization of a Quantum Dot (QD) as high-frequency noise detector. Current shot noise produced by a nearby Quantum Point Contact (QPC) induce photon-assisted tunneling for the electrons in QD. Thus fluctuations generated by the QPC lead to transport through orbital excited states of the quantum dot. This allows us to measure the QPC noise modulation when we change the number of opened channels. We also investigate the dependence of detector signal on the bias voltage across the QPC. We observe and explain the saturation and quantum features in the detector signal. The measurements are consistent with previous low-frequency experiments. In our case the detection frequency is determined by the QD orbital states spacing (60 respectively 140GHz).   

Intersubband lifetime magnetophonon oscillations in AlGaAs and InGaAs:InP quantum cascade lasers

We investigate the influence of a strong magnetic field on intersubband scattering rates in AlGaAs and InGaAs quantum cascade lasers (QCL). Laser threshold, differential quantum efficiency and laser emission spectra were measured in magnetic fields up to 30T applied perpendicular to the 2D planes. Intersubband magnetophonon effect -- resonant optical phonon non-radiative relaxation, gives rise to the strong oscillations in laser emission. We derived the magnetic field dependence of intersubband lifetime and compare to the calculated dependence of electron- LO phonon scattering rates.   

1.9 THz Quantum Cascade Lasers in very high magnetic fields up to 27 T

Terahertz quantum cascade lasers operating at $\lambda = 159$ $\mu$m [1] and exploiting the in-plane confinement arising from perpendicular magnetic field are investigated in the regime of very strong magnetic confinement, with field intensities up to 27 T, corresponding to a ratio $\displaystyle\frac{\hbar \omega_c(27~T)}{h \nu}$= 5.8 of cyclotron energy $\hbar \omega_c$ over photon energy $h \nu$. Device show laser action in magnetic field starting from 2.7 T and reaches operating temperatures of 65 K in pulsed mode. As the magnetic field is increased up to 27 T, the laser intensity exhibits modulations due to the interplay between inter Landau-level scattering and e-e scattering. A strong increase of the output power is observed at the highest field values. Strong reduction of the waveguide losses and an increase in the gain attributed to carrier localization leads to a decrease of the threshold current density down to 0.6 A/cm$^2$ at B=27 T. A detailed spectral analysis, showing a progressive redshift of the spectrum in the 3-27 T magnetic field range, will also be discussed. [1] G.Scalari, S.Blaser, J.Faist, H.Beere, E.Linfield, D.Ritchie, G.Davies, Phys. Rev. Lett., in press (2004)   

High Energy Intersubband Transitions in InAs/AlSb QWs

InAs/GaSb/AlSb heterostructures are a promising material system for intersubband optically-pumped applications due to their large conduction band offsets ($\sim $2 eV in InAs/AlSb). Applications include FIR generation and ultrafast all-optical switching at the communication wavelength of 1.55 $\mu $m. We have observed intersubband absorption at E$_{12}$ up to 670 meV (1.85 $\mu $m) in 2.1 nm Si-doped InAs/AlSb QWs. We have also attempted THz generation by difference frequency mixing in resonant InAs/AlSb asymmetric double quantum wells.   

Polaritonics: bridging the gap between electronics and photonics

Between electronics and photonics there exists a frequency gap of approximately 2 octaves, i.e. the frequency range between 100~GHz and 10~THz. Here we demonstrate that phonon-polaritons in ferroelectric crystals like LiNbO$_3$ or LiTaO$_3$ may be able to bridge this gap. The ability to fabricate structures within the crystal by femtosecond laser machining facilitates all integrated signal guiding and processing. Spatiotemporal imaging is employed for direct visualization of the electromagnetic field within the crystal. Polaritonic resonators, waveguides, photonic crystals and focusing, dispersive, and diffractive elements will be demonstrated.   

Session A16: Focus Session: Nano-optical Plasmonics

Surface Plasmon Rainbow Jets

A new method for optically exciting and visualizing surface plasmons in thin metal films is described. The technique relies on the use of a high numerical aperture objective lens to locally launch surface plasmons with an area much smaller than their lateral decay length. We visualize directly the intensity distribution of the surface plasmons by detecting the intrinsic lossy modes associated with plasmon propagation in thin films. Our approach allowed us to excite simultaneously a broad spectral continuum of surface waves and to describe for the first time surface plasmon rainbow jets. We quantified the attenuation of the jet as a function of wavelength and film thickness and compared it to the different propagation damping mechanisms. We demonstrated the influence of the interface on the surface plasmon propagation length and demonstrated surface plasmon spectral filtering using molecular excitonic adsorbates. We will discuss the potential of the technique to pump-probe, plasmon-based interface spectroscopy.   

Subwavelength Focusing and Guiding of Surface Plasmon Polaritons

It is found experimentally that subwavelength holes in thin metal films are versatile sources for the launching of surface plasmon polaritons (SPP). We use near field scanning optical microscopy to image the evanescent electromagnetic fields around individual holes and in hole arrays. For an arc-shaped hole array fabricated with focused ion beam machining into a Cr/Ag bilayer we show that SPP can be focused to a spot with subwavelength width. Finite-difference time-domain calculations give a quantitative account of the observed SPP patterns. Furthermore, we show that the high SPP intensity in the focal spot can be launched onto a 250 nm wide metal strip guide. This work was supported by the U.S. Department of Energy under Grant Nos. W-31-109-ENG-28, DEFG02-91-ER45439 and DEFG02-03-ER15487.   

Sub-wavelength confinement and the diffraction limit for surface plasmon waveguides

Surface plasmon-polaritons have received much attention for their ability to guide electromagnetic energy at optical frequencies. Unlike dielectric waveguides which confine volume electromagnetic waves to an optically dense core via index contrast, these surface electromagnetic waves are coupled to localized charge density oscillations along metal-dielectric interfaces. Consequently, theoretical and experimental works to date have highlighted the differences between the confinement provided by dielectric and plasmonic waveguides. Here, we present a series of related computational and experimental studies directed at illustrating the similarities of dielectric and plasmonic waveguides. Beginning with near-field images of confined plasmon propagation obtained by Photon Scanning Tunneling Microscopy (PSTM), we will discuss the limitations on confinement in plasmonic waveguides. These images will be interpreted by comparison with three-dimensional numerical solutions for the guided polariton modes. Vertical and lateral localization along finite width interfaces will be contrasted, and power density profiles investigated. The implications for the diffraction limited size of surface plasmon modes will be discussed, and ideal geometries for power concentration highlighted.   

Optical response of structured noble-metal nanoparticle aggregates

Interactions of light with subwavelength structures open new avenues of controlling light for many applications. The optical responses of structured array of noble-metal nanoparticle aggregates immersed in a glass matrix are investigated theoretically, motivated by the recent experimental observation of the splitting of the surface plasmon bands in silver arrays. To capture the strong electromagnetic coupling between the approaching particles in a silver aggregate, the spectral representation of the multiple image formula has been used, and a semiclassical description of the silver dielectric function is adopted from the literature [1]. The splitting of plasmon resonance band of the incident longitudinal and transverse polarized light is found to be strongly dependent on the particle diameter and their separation. Our results are shown in accord with the recent experimental observation. Moreover, a large redshift for the longitudinal polarization can be reproduced. The reflectivity spectrum is further calculated for a dilute suspension of dimer and chain arrays. \newline [1] J. J. Xiao, J. P. Huang, K. W. Yu, Phys. Rev. B, in press.   

Omni-directional light emission via surface plasmons from a metal-polymer-metal structure

Light extraction from LEDs is an important efficiency problem. Planar metallic microcavities have been used in the past to allow facile electrical excitation and obtain resonantly enhanced emission. In general this emission is only enhanced in a narrow angle range, while for some applications a wide emission angle is desirable. We will demonstrate that by using a properly designed, planar, metallic microcavity it is possible to enhance the free space emission via low momentum surface plasmons that lie above the light line. Additionally, for an optimized cavity thickness this enhanced emission can be observed at a well-defined frequency for all angles due to a nearly flat surface plasmon dispersion relation, hence the term omni-directional emission. This effect is predicted through simulations of dipole emission and verified in photoluminescence experiments using gold as the metal and a light-emitting polymer (DOW BP79) as the active medium.   

Plasmon Waveguides: Balancing Propagation, Localization, and Loss below the Diffraction Limit

On subwavelength scales, photon-matter interactions are limited by diffraction. Circumventing this diffraction limit is now a principle focus of integrated nanophotonics. Here, we present studies of surface plasmon (SP) waveguides exhibiting both long-range propagation and spatial confinement of light with lateral dimensions of less than 10 percent of the free-space wavelength. Attention is given to characterizing the dispersion relations, mode profiles, wavelength dependent propagation, and energy density decay in metallodielectric waveguides comprised of silicon dioxide/Ag/silicon dioxide and Ag/silicon dioxide/Ag structures with waveguide thicknesses ranging from 12nm-50nm. Numerical dispersion analysis indicates the presence of three distinct SP branches, including the existence of modes in the plasmon bandgap. For bound modes in Ag waveguides, near-IR propagation lengths exceed centimeter scales only at the expense of confinement. However, enhanced propagation is observed at shorter wavelengths despite notable field localization in the metal. Likewise, for silicon dioxide SP waveguides, propagation lengths exceed tens of microns with fields confined to within 30 nanometers of the structure. Applications of both short and long-wavelength plasmons to photonic waveguiding will be discussed, and utilization of such results for integrated plasmonic applications will be explored.   

Strongly coupled plasmon excitations in nanostructures and device applications

Since the 16th century, optical device size and performance has been limited by diffraction. However the diffraction limit can be circumvented via design of ``plasmonic'' device components with spatial confinement of light at dimensions less than 10{\%} of the free-space wavelength. Achieving control of light-material interactions at nanoscale dimensions requires structures that guide electromagnetic energy with a lateral mode confinement below the diffraction limit. Electromagnetic energy can be guided below the diffraction limit along ultrathin metallic stripes and subwavelength scale chains of closely spaced metal nanostructures via non-radiating surface plasmons. Recent experiments confirmed that strongly coupled collective plasmonic modes in metal nanostructures enable electromagnetic energy transport over distances of about 0.5 $\mu $m in plasmon waveguides. The emission rate from active dipole emitters such as semiconductor nanocrystals can also be enhanced by coupling into metallic nanostructures. Thus there appears to be no fundamental scaling limit to the size and density of photonic devices, and ongoing work is aimed identifying important device performance criteria in the subwavelength size regime. Ultimately it may be possible to design an entire class of subwavelength-scale optoelectronic components that form building blocks of an optical device technology scaleable to molecular dimensions, with imaging and spectroscopy applications. \newline \newline Co-authors: Luke Sweatlock, Jennifer Dionne, and Julie Biteen, California Institute of Technology   

Tuning the Nanooptics of Metallic Nanorods: End Effects

Understanding the nanooptics of metallic nanoparticles is critical for developing their use as sensors and nanohighways for directed excitation transport. End effects provide a new means to tailor nanorod response. We calculate the optical response of nanorods focusing on end effects and show that rod termination critically determines the position and width of plasmon resonances. Rods with hemispherical ends exhibit broad response typical for dipolar plasmonic charge oscillation along the rod. Rods with inverted, concave ends exhibit much narrower, cavity-like resonances unlike typical plasmon resonances. However, near-field enhancement is not dramatic for rods with two concave ends. Large near-field enhancement and narrow resonances can be achieved simultaneously for the near-field at the concave end of a rod with one concave end and one hemispherical end. Multiple narrow resonances can occur for large rods and deep cavities at the rod ends. These sharp resonances are seen both in the far-field and the near-field response.   

Optical propagation via dipolar coupling in metal nanoparticle chains

Electromagnetic propagation in metal nanoparticle chains offers the potential for nano-sized integrated optical circuits. Dispersion relations for dipolar modes propagating along such a chain are calculated by solving the full Maxwell equations, including radiation damping. The nanoparticles are treated as point dipoles, which means the results are valid only for $a$/$d \quad \le $ 1/3, where $a$ is the particle radius and $d$ the spacing.$^{1}$ The discrete modes for a finite chain are first calculated, then these are mapped onto the dispersion relations appropriate for the infinite chain. Computed results are given for a chain of 50-nm diameter Ag spheres spaced by 75 nm.$^{2}$ We find large deviations from previous quasistatic results:$^{3}$ Transverse modes interact strongly with the light line. Longitudinal modes develop a bandwidth more than twice as large, resulting in a group velocity that is more than doubled. All modes for which $k_{mode} \quad \le $ \textit{$\omega $/c} show strongly enhanced decay due to radiation damping. These features are consistent with recent calculations by Citrin.$^{4}$ $^{1}$ S. Y. Park and D. Stroud, Phys. Rev. B \textbf{69}, 125418 (2004). $^{2}$ W. H. Weber and G. W. Ford, Phys. Rev. B \textbf{70}, 125429 (2004). $^{3}$ M. L. Brongersma, J. W. Hartman, and H. A. Atwater, Phys. Rev. B \textbf{62}, 16356 (2000). $^{4}$ D. S. Citrin, Nano Lett. \textbf{4}, 1561 (2004).   

Adiabatic Nanoplasmonic Energy Concentration

We predict that propagation of surface plasmon polaritons (SPPs) toward a sharp edge of smoothly (adiabatically) graded metallic layer causes their slowing down and asymptotic stopping. This is accompanied by a concentration of electromagnetic energy and enhancement of local optical fields. As such a nano-concentration effect , we consider the adiabatic energy concentration in a conic nanoplasmonic waveguide ..[1]. In this case, SPPs are created in the $m=0$ plasmonic state at the wide (microscopic) edge of the system and propagate toward the tip of the conic waveguide. As local radius $R$of the cone waveguide decreases, both the phase and group velocity tend to zero $\propto R$. This asymptotic stopping lead to the accumulation of SPPs at the tip and their adiabatic transformation to the standing surface plasmons. For silver, it is possible to have local optical intensity at the tip increased by three orders of magnitude. We also discuss two- dimensional focusing nanoplasmonic waveguides. In conclusion, we will discuss the many prospective application of this effect. [1]. M. I. Stockman, \textit{Nanofocusing of Optical Energy in Tapered Plasmonic Waveguides}, Phys. Rev. Lett. \textbf{93}, 137404-1-4 (2004).   

Green-function theory of confined plasmons in coaxial cylindrical geometries: finite magnetic field

We report on a theoretical investigation of the plasmon propagation in the coaxial cylindrical geometries using Green's function (or response function) theory in the presence of an applied axial magnetic field ($\vec{B}\parallel \hat{z}$). Green's function theory generalized to be applicable to such quasi-one dimensional (1D) systems enables us to derive explicit expressions for the corresponding response functions (associated with EM fields), which can in turn be used to compute numerous physical properties of the system under consideration. As an application, we present several illustrative examples on the dispersion characteristics of the confined and extended magnetoplasmons in the single- and double-interface structures. These dispersive modes are also substantiated through the computation of local as well as total density of states (DOS). It is found that, unlike the zero-field case, the magnetoplasma propagation is nonreciprocal with respect to the sign of the index $m$ of the Bessel functions involved. We also briefly clarify some delusive traces of the edge magnetoplasmpons for a plasma shell embedded between two identical or unidentical dielectrics. Our theoretical framework can also serve as a powerful technique for studying the intrasubband plasmons and magnetoplasmons in the emerging mutiple-walled carbon nanotubes.   

Plasmonic resonances, scattering cross section, and electromagnetic forces between two coupled Au nanowires

We compute the electromagnetic response of two infinitely long silver nanowires, each with a square cross section, when illuminated by a plane wave with the electric vector perpendicular to the axis of the wires. The wire dimensions are on the order of $0.1 \lambda$. Of particular interest is when the incident wavelength excites plasmonic resonances in the wires and the associated field enhancement. We focus on (a) the scattering cross section and the (b) forces exerted on the wires both directly by the plane wave and by mutual coupling between the wires.   

Surface Plasmons in Anisotropic Metal-Dielectric Structures

We study the effects of anisotropy of a dielectric substrate and a metal film on the dispersion relation and range of propagation of surface plasmons. The substrate is considered to be a uniaxial crystal with its axis perpendicular to the metal surface. The dielectric constants of the substrate are $\varepsilon _{\vert \vert } $ (in the plane of propagation) and $\varepsilon _\bot $ (in the perpendicular direction). The metal film is characterized by a complex dielectric tensor with isotropic real negative part ${\varepsilon }'(\omega )$and anisotropic positive imaginary part ${\varepsilon }''_{ik} (\omega )$. The latter has two different components${\varepsilon }''_{\vert \vert } (\omega )$ and ${\varepsilon }''_\bot (\omega )$. Anisotropy of the dissipative part of the dielectric tensor is due to the surface channel of electron scattering, leading to the lower ac conductivity of the thin film in the direction perpendicular to the metal surfaces. We show that the substrate with $\varepsilon _\bot >\varepsilon _{\vert \vert } $ gives rise to larger propagation length of the surface plasmon. Thus from the point of view of efficiency of the plasmonic devices anisotropic \textit{negative} uniaxial crystals are preferential. In this case the decay length of the plasmon field in the substrate also increases. This decay length is an important characteristic of the sub-wavelength optical resolution.   

Near-field study of the perfect lens at visible wavelengths

It has been predicted that thin metal films can be used to generate images with a spatial resolution better than the diffraction limit via the local excitation of surface plasmons [1]. Such near-field focusing could have applications in optical data storage and nanofabrication. We present near-field scanning optical microscopy (NSOM) experiments that clearly demonstrate frequency dependent focusing using a near-field lens. The `perfect lens' is fabricated by depositing a 50nm thick gold layer onto a 50nm thick silicon nitride membrane. Focusing is detected by monitoring the interference between light emitted from a nanoscale object (the aperture of an NSOM tip) and radiation scattered by Pt nanoparticles placed in the image plane behind the lens. NSOM scans performed at wavelengths in the range 468nm-676nm reveal the role of surface plasmons in the imaging process. The measured frequency dependent image resolution is compared with numerical simulations based on the Finite Integration Technique. [1] J. B. Pendry, Phys. Rev. Lett. 85, 3966 (2001)   

Session A17: Kondo Effect and Correlations in Quantum Dots

Kondo-Temperature Dependence of the Magnetic-Field Splitting of a Kondo Peak in a Single-Electron Transistor

We present a detailed study of the Kondo peak splitting as a function of Kondo temperature $T_{K}$ and magnetic field $B$ parallel to the 2DEG in an AlGaAs/GaAs single-electron transistor. We observe that, at fixed $B$, the Kondo splitting decreases logarithmically with Kondo temperature, in agreement with theory. However, we find that the magnitude of the prefactor of the logarithm is much larger than predicted. We also find that there exists a critical magnetic field $B_{c}$ below which the Kondo peak does not split, in qualitative agreement with theory. However, our results indicate that $B_{c}$ is smaller than predicted. These measurements show that the theory of non-equilibrium Kondo physics is still incomplete.   

Two-Stage Kondo effect and singlet-triplet crossover in a four-electron artificial atom

An artificial atom of 400 nm lithographic size is defined on an AlGaAs/GaAs heterostructure. With four electrons on the quantum dot, a gate-voltage-induced singlet-triplet crossover is observed. On the triplet side, a Kondo peak with a narrow dip at drain-source voltage V$_{ds}$=0 is seen. The low energy scale V$_{ds}^{\ast }$ characterizing the dip is a signature of the two-stage Kondo effect. On the singlet side, we see a Kondo enhanced feature at nonzero V$_{ds}$ due to inelastic cotunneling processes leaving the dot in the triplet excited state. The excitation energy increases as the gate voltage V$_{g}$ is tuned away from the crossover region. The effects of both the temperature T and the magnetic field B parallel to the two-dimensional electron gas are also presented. The low energy scales T$^{\ast }$ and B$^{\ast }$ are extracted from the behavior of the linear conductance and are compared to the low energy scale V$_{ds}^{\ast }$ obtained from the differential conductance.   

The Kondo effect of a magnetic impurity in an ultrasmall metallic grain

We have studied the Kondo effect of a magnetic impurity embedded in a small metallic grain with a level spacing comparable to the Kondo temperature (Kondo box). Small ($\sim $ 1nm) gold grains weakly coupled to source and drain electrodes were fabricated on top of an aluminum gate electrode using electromigration. Without magnetic impurities present, Coulomb blockade with charging energies exceeding 100 meV was observed. If cobalt impurities are introduced in the gold a gate dependent Kondo effect (T$_{k}\sim $100 K) is often observed. In many cases the Kondo peak is splitted which we attribute to the discreteness of the conduction electron spectrum of the gold grain. Temperature and magnetic field dependence contribute to a better understanding of the Kondo box. \textit{email: Hubert@qt.tn.tudelft.nl}   

Exotic Kondo States in GaAs Quantum Dots

Using a unique double quantum dot geometry, we probe an exotic Kondo effect involving one quantum dot containing excess spin-1/2 simultaneously coupled to both open and confined reservoirs of electrons. Transport measurements through open reservoirs (normal leads) reveal single channel Kondo behavior. However, the addition of a third lead consisting of a large quantum dot drastically changes transport through the other, Kondo- correlated quantum dot. We explore Kondo correlations both when Coulomb blockade confines a defined number of electrons on the large dot and when charge is allowed to fluctuate. Research supported by an NSF CAREER award DMR-0349354. RMP acknowledges support from an ARO Graduate Research Fellowship.   

Two Kondo Impurity Spin Interactions in Quantum Dots

The Kondo effect in a single quantum dot, where the localized electron behaves as a magnetic impurity, has been well studied[1]. A recent experiment by C.M. Marcus's group indicates a possible RKKY-like spin interaction between two quantum dots (QD) in the Coulomb Blockade regime.[2] We have obtained an effective Hamiltonian by carrying out a perturbation expansion (related to the Schrieffer-Wolf transformation) of an Anderson model of the two QD system (including terms representing the tunneling) to fourth order. We obtain the standard Kondo-coupling terms at second order, and we obtain RKKY-like terms at fourth order. We have also kept the scattering terms obtained at second-order, which are usually neglected, in order to study their effects on possible fixed points for the two-QD problem. We discuss the full range of interaction terms obtained at the level of RKKY, and their implications for the low-temperature behavior. {\_}{\_}{\_}{\_}{\_}{\_}{\_}{\_}{\_}{\_}{\_}{\_}{\_}{\_}{\_}{\_}{\_}{\_}{\_}{\_}{\_}{\_}{\_} [1]Nature \textbf{391}, 156 (1998) [2]Science \textbf{304}, 565 (2004).   

Numerical evidence of tunable long-range spin interactions between quantum dots

This work was motivated by experimental results (N. J. Craig {\it et al.}, Science {\bf 304}, 565 (2004)) where two quantum dots (QDs) are connected through an open conducting region. By varying the gate potentials, a net local spin is fixed in the first QD, and the Kondo effect in the second QD is then suppressed. By either varying the gate potential in the first QD, until its net spin is zero, or by decreasing its coupling to the common conducting region, the Kondo peak in the second QD can be restored. Combining Exact Diagonalization of finite clusters with a Dyson Equation embedding procedure, we were able to reproduce these results. Two qualitatively different regimes were found: (1) When the charge in the first QD has a value of approximately 1, the conductance in the second QD has a dip where it goes exactly to zero, resembling an interference effect already noticed in previous work.$^1$ (2) When the charge in the first QD is above or bellow 1, the Kondo effect can be suppressed, but the conductance does not vanish. Work is being done on a qualitative interpretation of these results, which do not seem to support an RKKY scenario. $^1$ C. A. Busser {\it et al.}, cond-mat/0404426 (Phys. Rev. B, in press)   

Out-of-equilibrium Kondo Effect in a Three Lead Quantum Dot

We present measurements of a small GaAs quantum dot in a three lead geometry with the third lead used both as a probe of Kondo resonances and wavefunction coupling to the various reservoirs. Kondo correlations produce an enhanced density of states at the Fermi level in each reservoir. We demonstrate that the third lead probes the individual out-of-equilibrium densities of states of the two other leads when a voltage bias is fixed across them. In addition, information about the spatial distribution of quantum dot wavefunctions can be inferred from presence or absence of Coulomb blockade features in the third lead. Research supported by an NSF CARRER award DMR-0349354. RMP acknowledges support from an ARO Graduate Fellowship.   

Interference Effects in the Conductance of Multi-Level Quantum Dots

Transport properties of multilevel quantum dots are investigated in the Kondo regime. The conductance can be decomposed into the contributions of each level. It is shown that these channels can carry a different phase. A destructive interference processes are observed when the phase difference between them is $\pm\pi$.$^1$ This effect is very different from those observed in bulk metals with magnetic impurities, where the phase differences play no significant role. The effect is also different from other recent studies of interference processes in dots as the interference do not depend on external magnetic field or the hopping amplitudes dot-leads for all levels. Another interesting effect reported here is the formation of localized states that do not participate in the transport. When one of these states crosses the Fermi level, the electronic occupation of the quantum dot changes, modifying the many-body physics of the system and indirectly affecting the transport properties. $1$- C.A. B\"usser et al, cond-mat/0404426, to appears in Phys.Rev. B   

Molecular Conductors with Center of Mass Motion

We study numerically the linear conductance of a molecular conductor that can oscillate between the source and drain electrodes.$^1$ This vibrational mode leads to an asymmetric modulation of the molecule-leads hopping parameters. By expanding this modulation up to the linear order, the conductance can be decomposed into two channels, the direct hopping and the phonon-assisted tunneling channels. The Kondo regime results show conductance dips that can be attributed to the destructive interference$^2$ of these two channels. If an internal vibrational mode is also active with the effect of symmetric modulation of the tunneling barriers, the particle-hole symmetry is broken and a Fano-like interference is observed. \newline \newline $^1$ K.A. Al-Hassanieh et al - Preprint \newline $^2$ B\"usser et al, cond-mat/0404426 (to appear in Phys. Rev B).   

Spin Chain QMC simulations of Mesoscopic Quantum Impurity Models

Recent studies of quantum impurity problems have emphasized the need for flexible quantum impurity solvers. One such study focuses on mesoscopic fluctuations in the Kondo effect in quantum dots(cond-mat/0409211). Here we demonstrate that the mapping of single-quantum impurity problems to quantum spin-chains can be exploited to yield a powerful cluster algorithm that can treat the mesoscopic fluctuations exactly while at the same time being able to approach the large $D_\textrm{eff}/T$ limit with ease. The algorithm is implemented explicitly for the Anderson and Kondo Hamiltonians, and compared with standard methods for the ``mesoscopic Kondo problem.'' Using our algorithm we study mesoscopic fluctuations of the Kondo temperature in integrable and chaotic systems of various geometries.   

Noise power spectrum for tunneling through a quantum dot in the Kondo Regime

The charge-current and spin-current noise spectra are calculated for tunneling through an ultrasmall quantum dot in the Kondo regime. Modeling the dot by an infinite-U Anderson model, we use the noncrossing approximation to formulate the current-current correlation function for arbitrary frequency and voltage bias. Our formulation fulfills all the basic requirements of the current-current correlation function, including current conservation and the recovery of the fluctuation-dissipation theorem at zero frequency and bias. The full temperature, voltage-bias and frequency dependences of the noise are analyzed, and the significance of the Kondo correlations that develop are discussed. Deficiencies of the slave-boson mean-field approach for calculating the noise are pointed out.   

Unification of electromagnetic noise and Luttinger liquid via quantum-dot resonant level

We investigate the effect of dissipation on a small quantum dot (resonant level) tunnel-coupled to a chiral Luttinger liquid (LL) with the LL parameter $K$. The dissipation stems from the coupling of the dot to an electric environment, being characterized by the resistance $R$, via Coulomb interactions. We show that this problem can be mapped onto a Caldeira-Leggett model where the (ohmic) bath of harmonic oscillators is characterized by the effective dissipation strength $\alpha=(2\tilde{K})^{-1}$ with $\tilde{K}^{-1}=K^{-1}+2R/R_K$ and $R_K=h/e^2$ the quantum of resistance. A quantum phase transition emerges at $\tilde{K}=1/2$ and its consequences on the occupation of the level are addressed. The special limit $K=1/2^+$ is thoroughly studied at small $R/R_K$ via a link to the spin-boson-fermion model. Our result can be detected by measuring the occupation of the quantum dot or by carrying out resonant tunneling transport measurement.   

Transport properties of small Quantum rings in the Kondo regime

The possibility to construct small rings of few electrons give rise to a different physics effects.$^1$ In such systems the transport properties of quantum dots are combined with electronic interference phenomena as the Aharonov-Bohm. In this work a small quantum ring, where the Coulomb repulsion between electrons $U$ and the magnetic flux $\Phi$ are important energies scales, is studied as a function of the gate potential applied to the ring. This model can resemble all the possible effects as Kondo physics, Aharanov-Bohm effect and the degeneracy between states $k$ and $-k$. For $T=0$ the Kondo effect is present and the valley between the Coulomb peaks are fullfilled when the total charge in the ring is odd. However, there is an special case for the Coulomb blockade peak corresponding to a gate potential where the states $k$ and $-k$ are degenerated. We found that the system still in the Kondo regime even for an even charge due a $S=1$ ring's state.$^2$ When $\Phi$ is applied the degeneracy is broken destroying the $S=1$ state and eliminating the Kondo effect.$^3$ $1$- A. Fuhrer et al, Phys. Rev. Lett. {\bf 93} 176803 (2004). $2$- C.A. B\"usser et al, cond-mat/0404426, to appears in Phys.Rev. B. $3$- C.A. B\"usser, S. Ulloa, E. Dagotto and E.V. Anda (preprint).   

Transport properties of coupled quantum dots in the presence of phonons

Here is presented the numerical study of the effect of Holstein phonons in the transport properties of two coupled quantum dots (QDs) in the Kondo regime. For the QDs we use the Anderson impurity model and each QD is coupled to a different Holstein mode. At $T=0$, in the absence of phonons, and with 1 electron per dot, the usual splitting of the Kondo resonance is observed.$^1$ When the QDs are coupled to the phonons, there is a reduction of the effective Coulomb repulsion, which is explained through a canonical transformation. In addition, the conductance at the electron-hole symmetric gate potential is not affected by the phonons. This is caused by the modulation of the coupling factors.$^2$ The difference between the effects of phonons in lithographic QDs and in molecular conductors is also discussed. $1$- C.A. B\"usser et al, Phys. Rev. B {\bf 62}, 9907 (2000). $2$- K.A. Al-Hassanieh, C.A. B\"usser, G.B. Martins, Adriana Moreo and Elbio Dagotto (preprint)   

Session A18: Focus Session: Semiconductor Characterization

Interface Sensitive Measurement of High k - silicon dioxide – silicon system using Optical Second Harmonic Generation

The properties of the interface layer play a key role in the properties of transistor gate dielectric stacks. This has resulted in a renewed emphasis being placed on physical interfacial characterization. Electrical measurements such as capacitance -- voltage have always been sensitive to the number of interface trapping states. The sub 5 nm thickness of high $\kappa $ - silicon dioxide films on silicon challenges all characterization methods. Traditional characterization methods such as scanning transmission electron microscopy have seen rapid advances in capability as the first aberration corrected microscopes become available. Pennycook recently observed single Hf atoms in the silicon dioxide interfacial layer.(1) Less widely used methods also show considerable promise. Optical second harmonic generation has been used to measure interfacial states in the silicon dioxide -- silicon interface.(2) In this paper, we discuss our first SHG measurements of the High $\kappa $ - silicon dioxide -- silicon system. Comparisons between the responses of Hf O$_{2}$ and HfO$_{x}$Si$_{y }$before and after annealing show the effect of silicate decomposition after annealing. Time dependent optical SHG is believed to be sensitive to trap density, and we again show our first results. When possible, we compare SHG to electrical measurements.   

Measurement of the Full State of Stress of Silicon with Micro-Raman Spectroscopy

Micro-Raman spectroscopy has been widely used to measure local stresses in silicon and other cubic materials. However, a single (scalar) line position measurement cannot determine the complete stress state unless it is has a very simple form, such as uniaxial. Previously published micro-Raman strategies designed to determine additional elements of the stress tensor take advantage of the polarization and intensity of the Raman scattered light, but these strategies have not been validated experimentally. In this work we test one such stategy [S. Narayanan, S. Kalidindi, and L. Schadler, \textit{JAP}. \textbf{82}, 2595 (1997)] for rectangular (110)- and (111)-orientated silicon wafers. The wafers are subjected to a bending stress, and the state of (plane) stress is modeled with ABAQUS. The Raman shifts, intensities, and polarizations are calculated using previously published values for silicon phonon deformation potentials. The experimentally measured values for $\sigma _{xx}$, $\sigma _{yy}$, and $\tau _{xy}$ at the silicon surface are in good agreement with those calculated with the ABAQUS model.   

UV-Raman deformation coefficients in Si and SiGe alloys

As Si CMOS device scaling issues become increasingly challenging a number of alternatives arise including Si-On-Insulator (SOI) substrates, high-k gate dielectrics, and Strained Si Channel (SSC) devices.~ In the case of SSC structures, the enhancement in electron mobility depends directly on the stress magnitude.~ Raman scattering, particularly in the UV due to short penetration depth, has proven well suited for measuring thin SSC layer stress.~ The technique depends critically on the value taken for the strain shift coefficient ($b)$, which correlates the shift in the phonon frequency with the strain.~ A number of values have been reported in the literature to date using NIR and visible excitation; however, the authors are unaware of previous work performed specifically in the UV. In this work, we have used a combination of HRXRD reciprocal space mapping (RSM) to measure the in-plane strain of high quality Si/SiGe heterostructures and UV-Raman with the 325nm He-Cd line to determine the Si LO phonon deformation coefficient in Si and SiGe alloys with compositions ranging from 10-40{\%} Ge.   

In-situ photovoltage shift measurements of hafnium oxides and silicates grown on Si(100) using femtosecond photoelectron spectroscopy

Femtosecond laser based photoelectron spectroscopy was used as an in-situ monitor of band bending in Si (100) substrates during various stages of hafnium oxide growth and post-deposition anneal. A fraction of the 800 nm laser pulse is directed onto the sample as a pump pulse, which flattens the existing band bending in the Si substrate. The remaining 800 nm light is focused into bursts of Ar gas to generate high order, odd multiple harmonics used as an in-situ probe of band bending. Photovoltage measurements reveal an abrupt onset of charging during the annealing of hafnium oxides and silicates, deposited on thin SiON interlayer oxides grown on lightly doped Si (100) substrates. Core level photoemission and transmission electron microscopy were used to correlate the observed charge injection at elevated temperatures with structural and chemical changes in the SiON and HfO$_{2 }$dielectric layers.   

An infrared probe of tunable dielectrics in metal-oxide-semiconductor structures

We present a novel approach to investigate electric-field- induced changes of the dielectric function of the gate insulator in metal-oxide- semiconductor (MOS) structures using infrared spectroscopy. This approach is enabled by an innovative design of the electrodes. We investigated structures based on TiO$_{2}$ dielectric insulator on doped silicon commonly used in (organic) FET devices. We show that the voltage-induced changes of dielectric constant of TiO$_{2}$ originate from a radical modification of lattice vibration modes of this compound induced by applied electric field. Novel characterization capabilities demonstrated in our work uncover the unique potential of infrared spectroscopy for the analysis of tunable insulators and also for the examination of charge injection phenomena in semiconductors.   

High Resolution 2D Dopant Profiling of FinFET Structures using Scanning Probe Microscopy

The ability to perform dopant/junction profiling with high spatial resolution is critical for development of future generation devices such as FinFET structures. Among various forms of scanning probe microscopy, scanning tunneling microscopy (STM) has demonstrated direct atomic imaging of dopant atoms on GaAs (110) surfaces. More recently, scanning thermoelectric microscopy (SThEM) (H.K. Lyeo, et al \textit{Science }v.303 p816 (2004)) has been applied to profile GaAs $p-n$ junction with unprecedented spatial resolution. The key challenge to successfully apply these techniques to silicon based devices is to prepare a surface that is both chemically and electronically passivated. Here we present our progress toward this goal. In particular we will report STM and SThEM studies on silicon based electronic devices including FinFET structures. Moreover, we will present comparative studies of dopant/junction profiling using STM, SThEM, and scanning capacitance microscopy (SCM).   

High-Resolution Microcalorimeters for X-ray Microanalysis

Microcalorimeters represent the current state-of-the-art in x- ray detection for high-resolution microanalysis. In this device the energy of x-rays emitted from regions of a sample or circuit excited by an electron beam is determined by measuring the increase in temperature caused by the absorption of an x-ray in the microcalorimeter, normally held at temperatures well below $1$ $K$ by a compact adiabatic demagnetization refrigerator. The performance of these detectors is ultimately determined by the sensitivity and noise characteristics of the thermometer. The best performance is currently achieved using superconducting transition-edge sensors which can resolve energy differences of better than 1 part in 2000. In this talk I will briefly summarize the current state of x-ray microcalorimetry and discuss recent efforts at NIST to develop next-generation microcalorimeters using SQUID-based magnetization thermometry.   

High-resolution characterization of advanced interconnect and packaging architectures

Characterization of buried interfaces in advanced interconnect and packaging structures is a critical challenge as CMOS devices are scaled and as novel packaging concepts such as through-silicon vias are developed. It is important to be able to image chip-to-package attach bumps with high resolution as well as individual interconnect layers in a processed device. Critical information that needs to be obtained from such inspection includes detection of un-bonded bumps, missing interconnections in stacked die and delaminations in underlying layers. In this paper, we discuss and benchmark different characterization techniques to analyze and quantify the quality of buried interfaces with detailed experimental and theoretical analyses. Acoustic microscopy is used to investigate interfaces between stacked die at high resolution. We present theoretical analysis which shows the effect of aberrations on image formation and the corresponding resolution degradation. Infra-red imaging through the substrate and pulsed thermal microscopy in transient mode were also used to investigate the quality of stacked die interfaces. We critically review these different techniques for inspection of buried interfaces and discuss roadmap for metrology needs.   

Direct Imaging of Minority Carrier Drift in Luminescent Materials

A technique is presented to directly image charge carrier drift and diffusion in semiconductor samples over a temperature range from 300 to 10 K. A scanning electron microscope produces electron-hole pairs at a point and an applied voltage bias causes the charge carriers to drift. Upon radiative recombination, a photon is emitted at the point of generation, which is then collected and imaged by a cooled CCD camera via an optical microscope. This technique allows for the preservation of spatial information from the carrier recombination. Current resolution is $\sim $ 0.4 $\mu $m per pixel. Results will be presented from the imaging of drift behavior in high purity epitaxial GaAs as a function of temperature. Minority carrier drift over distances in excess of 100 $\mu $m at a field of $\sim $ 80 V/mm has been directly imaged using this technique for high purity room temperature n-type GaAs samples with net doping of $\sim $ 5 x 10$^{13}$ cm$^{-3} $. The characterization of the drift tails as a function of temperature will be presented and the measured spatial homogeneity of the sample depicted. The effect of near-contact electric fields due to space charge on charge carrier injection and collection at low temperature will be presented.   

Contact-free approach for the determination of minority carrier diffusion length

Direct imaging of electron/hole recombination via an optical microscope and a high sensitivity charge coupled device coupled to a scanning electron microscope captures spatial information about the transport behavior in luminescent solid state materials. Carriers are generated by the electron beam, and an image of the recombination provides highly localized information on carrier diffusion and/or drift. Unlike conventional cathodoluminesence, the e-beam is not scanned. Recent work has demonstrated the feasibility and limitations of the technique. Comparisons will be made of luminescent images in 3D (bulk) and 2D (quantum well) limits. In the 3D case, the excitation volume plays a key role, while in 2D, beam size and carrier diffusion determine the luminescent spot size. The technique provides the potential for extraction of minority carrier diffusion length without the use of contacts. It allows for easy localization of the measurement site, broad application to a range of materials and potential industrial automation. This technique is of special interest for devices such as solar cells, where minority carrier lifetime is a key performance parameter.   

Structural and morphological characterization of GaN(0001) layers grown on SiC by maskless pendeo-epitaxy via X-ray Microdiffraction

Novel white beam X-ray microdiffraction (WBD) together with high resolution monochromatic X-ray diffraction (HRXRD) and finite element simulations have been used to determine the distribution of strain, dislocations, sub-boundaries and crystallographic wing tilt in uncoalesced and coalesced GaN layers grown by maskless pendeo-epitaxy. In traditional HRXRD the spot size of the X-ray beam is large ($\sim $0.5 mm), i.e. it gives information averaged over 40-50 of stripes. In contrast, advanced WBD provides very local information and enables us to follow the local orientation at different locations across the stripe. Stress relaxation in the GaN layers occurs in conventional and in pendeo-epitaxial films via the formation of additional misfit dislocations, domain boundaries, elastic strain and wing tilt. An important parameter was the width-to-height ratio of the etched columns of GaN from which the lateral growth of the wings occurred. The strain and tilt across the stripes increased with the width-to-height ratio. Sharp tilt boundaries were observed at the interfaces formed by the coalescence of two laterally growing wings. The wings tilted upward during cooling to room temperature for both the uncoalesced and the coalesced GaN layers   

UHV Nanoworkbench and the `Roaming' Field Effect Transistor

A multiple-tip ultra-high vacuum (UHV) scanning tunneling microscope combined with a scanning electron microscope (SEM) and molecular-beam epitaxy growth capabilities has been developed. This instrument (Nanoworkbench - NWB) is used to perform four-point probe conductivity measurements at sub-micrometer spatial dimension. The system is composed of four chambers, the multiple-tip STM/SEM chamber, a surface analysis and preparation chamber equipped with standard surface science tools, a molecular-beam epitaxy chamber and a load-lock chamber. The four chambers are interconnected by a unique transfer system based on a sample box with integrated heating and temperature-measuring capabilities. We demonstrate the operation and the performance of the NWB with four-point- probe conductivity measurements on a silicon-on-insulator (SOI) crystal. The creation of a `roaming' Field-Effect Transistor, whose dimension and localization are respectively determined by the spacing between the four probes and their position on the SOI surface, is demonstrated. The NWB has a potential to look at nanostructures developed in situ using the MBE chamber, such as Ge quantum dots for example. This work was supported by DARPA QuIST through ARO contract number DAAD-19-01-1-0650.   

An investigation of dopping profile for a one dimensional heterostructure

A one-dimensional junction is formed by joining two silicon nanowires whose surfaces are terminated with capping groups of different electronegativity and polarizability. If this heterostructure is doped (with e.g. phosphorous) on the side with the higher bandgap, the system becomes a modulation doped heterostructure with novel one-dimensional electrostatics. We use density functional theory calculations in the pseudopotential approximation, plus empirical model calculations, to investigate doping profiles in this new class of nanostructures.   

Session A19: Quantum Hall Effect

Empirical resistivity rule in the second Landau level of a two-dimensional electron system

A phenomenological relationship, $R_{xx} \propto B \times dR_ {xy}/dB$, called the resistivity rule, was observed twenty years ago. Yet, today we have only a relatively complex model that addresses the origin of this rule. It remains unclear whether a simpler model, based on some fundamental relationship exists. In recent experiments on ultra-high quality specimens performed in the second Landau level (LL), instead of rising in a stair- like fashion, $R_{xy}$ is found to switch back and forth between FQHE and IQHE values several times as the filling factor varies from $\nu=4$ to $\nu=2$. This non-monotonic $R_ {xy}$ leads to regions of negative $B \times dR_{xy}/dB$, which cannot find an equivalent in $R_{xx}$, a positive definite, thus apparently violating the empirical rule. However, in a more detailed examination, we found, surprisingly, a new resistivity rule in the second LL. The regular, positive parts of $B \times dR_{xy}/dB$ are well reflected in $R_{xx}(+B)$, whereas the irregular negative going sections of $B \times dR_ {xy}/dB$ closely match the inverted $R_{xx}(-B)$ trace, where $- B$ refers to the opposite magnetic field direction of $+B$. It is unclear whether our observations of an expanded resistivity rule reinforces or refutes the present model of its origin.   

Spin-dependent resistivity at transitions between integer quantum Hall states

The longitudinal resistivity at transitions between integer quantum Hall states is found to depend strongly on the spin orientation of the corresponding partially-filled Landau level in two-dimensional electrons confined to narrow AlAs quantum wells. By tilting the sample with respect to the applied magnetic field, different Landau level spin branches can be brought to and driven past energetic coincidence. The result is a flip of the spin-orientation for the energy level corresponding to a given quantum Hall transition that is accompanied by a change in resistivity. This change can be as much as an order of magnitude. We discuss possible causes and suggest a new explanation for spike-like features, associated with quantum Hall ferromagnetic transitions, observed at the edges of quantum Hall minima.   

Landau quantization, Localization, and Insulator-quantum Hall Transition at Low Fields

We have performed a magnetotransport study on a two-dimensional gated GaAs electron system containing self-assembled InAs quantum dots. In our system Shubnikov-de Haas (SdH) oscillations are induced by Landau quantization in the low- field insulator, and the system undergoes a direct insulator- quantum Hall ($\nu=4$) transition as the magnetic field is increased. The low-field Landau quantization, in fact, is governed by the SdH theory and can modulate the density of states without causing the formation of a quantum Hall liquid in our system. While the expected property $\rho_{xy} \sim \rho_ {xx}$ may not be valid at direct insulator-quantum Hall transitions, we find that such transitions do occur as the product $\mu B \sim 1$ and hence well-separated Landau bands exist in the energy spectrum.   

Symmetries of the Resistance of Mesoscopic Samples in the Quantum Hall Regime

We present an experimental study of the symmetries of the resistance of mesoscopic samples in the quantum Hall regime. The samples we use are small Hall-bars, prepared from low-mobility InGaAs/InAlAs wafers. The four-terminal resistances of these samples display large reproducible fluctuations that are unique to the contact configuration used in the measurements. We find that the samples obey new symmetries, in addition to the reciprocity relation, relating the longitudinal and Hall resistances of different contact configurations and magnetic- field ($B$) polarities. These symmetries include the fine details of the resistance fluctuations. The resistances in the vicinity of all integer quantum Hall transitions are found to follow one of two possible sets of symmetries, one on the low-$B$ side and the other on the high- $B$ side of the transitions.   

Lifetime of 2D electrons in AlxGa1-xAs-Al0.32Ga0.68As Heterostructures

We have investigated the transport lifetime and the quantum lifetime of 2D electrons confined to the Al$_{x}$Ga$_{1-x}$As-Al$_{0.32}$Ga$_{0.68}$As heterostructures over the range of $x$ from 0 to 0.85{\%}. The transport lifetime is obtained from the mobility measurement, while the quantum lifetime is determined by fitting the temperature and magnetic field dependences of the Shubnikov-de Hass oscillations to the Dingle formula. With $x$ increases from 0 to 0.85{\%}, the transport lifetime is found to decrease from 160ps to 30ps. However the quantum lifetime only changes from 1.71ps to 1.64ps. Since the quantum lifetime is given by the total scattering rate over all directions while the transport lifetime is only affected by the large-angle backscattering rate, our results show that the alloy scattering centers contribute mainly to the short-ranged large-angle scattering. These results demonstrate a powerful way to manipulate the nature of disorder in 2D electron systems and show consistency with the recent scaling experiment of IQHE plateau-to-plateau transitions in Al$_{x}$Ga$_{1-x}$As alloy systems with different $x$.   

Interaction effect on quantum magnetooscillations in a two-dimensional electron gas

Recently there is considerable experimental interest in the semiconductor systems with apparent quantum phase transition. For this experiments it is important to have independed means of measurements of effective electron mass. This mass can be inferred from quantum magnetic oscillations, and we present a framework for the interpretation of magnetooscillation experiments. We consider the effects of interactions and disorder on the damping of magneto-oscillations in 2D. We study the effect of both long range and short range interaction and point-like disorder in a ballistic ($T\tau\gg1$) and diffusive ($T\tau\ll1$) regime, where $\tau$ is mean scattering time. The dominant effect on the damping comes from interplay of disorder and interaction corrections to to the electron mass. Depending on the nature of interaction we found the corrections to behave like $\ln T$ and $\ln^2T$ in the ballistic and diffusive regime correspondendly.   

Measurements of the Spin Susceptibility of 2D GaAs/AlGaAs Heterostructures into the Weak Interacting Region

We determine the spin susceptibility $\chi$ of a two- dimensional electron system in GaAs/AlGaAs heterostructures using the tilted-field method. The measurements are done on a very high quality heterojunction-insulated gate field-effect transistor (HIGFET) with a mobility as high as $1\times10^{7} cm^{2}/Vs$. We report the $\chi$ measurements on a single HIGFET specimen over a wide range of densities, from $1 \times10^ {10} cm^{-2}$ to $4\times10^{11}cm^{-2}$; deep into the weak interacting regime. The value of $\chi$ decreases monotonically with increasing density. In the low density region, $\chi$ follows an empirical formula proposed by Zhu et al. (\textit {Phys. Rev. Lett.}, \textbf{90}, 056805, 2003), but deviates from it as density increases beyond $6\times10^ {10} cm^{-2}$. After corrections for nonparabolicity of mass and g-factor, our $\chi$ measurements are very close to the most recent theoretical calculation (De Palo et al., cond- mat/0410145) over the whole density range.   

Electron interferometer in the integer QH regime

We report experiments on an electron interferometer fabricated from high mobility, low density GaAs/AlGaAs heterostructure material. In this device, a nearly circular electron island is separated from the 2DES by two nearly open constrictions. In the integer QH regime $f$ = 1 and 2, we observe Aharonov-Bohm-like oscillations of conductance. The interference closed path is comprised by the two edge states circling the island, coupled by tunneling in the constrictions, the radius $r \sim $ 900 nm is determined from the oscillation period. The radius can be tuned by application of a bias V$_{FG}$ to the four front gates. We find approximately linear dependence d$r$/dV$_{FG }$= 0.25 nm/mV. We compare the experimental results to the island B = 0 electron density profile obtained in classical electrostatic models of Gelfand and Halperin, PRB \textbf{49}, 1862 (1994) and Chklovskii el al., PRB \textbf{46}, 4026 (1992).   

Bending the quantum Hall effect by 90 degrees: Evidence for a new kind of disordered one-dimensional superconductor

Utilizing a recently developed corner-overgrowth technique, we create a two-dimensional (2D) electron system which bends by 90 degrees at an atomically sharp corner. At certain properly oriented magnetic fields, the energetic gap in the two 2D systems confines the motion of electrons along the corner to one dimension, creating a new kind of 1D system. The conductance along this 1D corner shows power-law behavior in temperature and voltage which varies depending on the nature of the gap (integer or fractional quantum Hall effect), and for certain gaps shows a negative power-law exponent indicative of effective attractive interactions between the charges in the wire. This behavior was predicted theoretically for such systems. The 1D systems presented here represent the longest quantum wires ever fabricated, up to several millimeters in length.   

Effect of disorder on modulated quantum Hall systems

We present a numerical study of the quantum Hall effect in modulated two-dimensional (2D) electron systems in presence of disorder. Theoretically, it is known that a 2D periodic potential in a strong magnetic field gives rise to a recursive subband structure in Landau levels, which is called the Hofstadter butterfly[1]. Recently, the nonmonotonic behavior of the Hall conductivity peculiar to this system was observed in lateral superlattices patterned on GaAs/AlGaAs heterostructures [2,3]. To study how the Hall plateau emerges in those split Landau levels, we numerically calculate the Hall conductivity in a disordered 2D electron system with weak modulations under various magnetic fields. We investigate the scaling property of the Hall conductivity as well as the localization length, to identify the critical energies where the extended states exist. The dependence on the field amplitudes and the Landau levels is also discussed. [1] D. R. Hofstadter, Phys. Rev. B 14, 2239 (1976). [2] C. Albrecht, et al. Phys. Rev. Lett. 86, 147 (2001) [3] M. C. Geisler, et al., Phys. Rev. Lett. 92, 256801 (2004).   

Integer Quantum Hall Effect versus Anderson Transition: Numerical Comparison for Eigensolutions Statistics

Two types of the disorder-induced localization transitions, the plateau-to-plateaux transition in the Integer Quantum Hall Effect and the conventional Anderson transition in 3 dimensions are considered by using a comparative analysis. Similarities and differences of critical behavior of the eigenfunctions and eigenvalues statistics for both cases are numerically investigated within the frames of the common self-contained diagonalization technique. Both transitions reveal a number of the same universal features at criticality, including the one-parameter scaling, symmetry dependence of the eigenfucntions distributions, multifractality spectra, non-trivial branching numbers etc. Our results provide a strong support for the quantum-field theoretical description treating the two transitions as a generalized transition, i.e. a unique critical phenomenon.   

Symmetry Breaking by Periodic Potentials and Randomness in Quantum Hall Systems

The effect of a one dimensional periodic potential on quantum Hall systems is investigated using direct diagonalization and the density matrix renormalization group (dmrg). We find that the phase of the periodic potential (i.e. averaging over the phase) has minimal effect on symmetry breaking, however the addition of a small random potential tends to decrease finite size effects. For certain parameter values, randomness tends to increase symmetry breaking. Preliminary results using dmrg, for larger system sizes, with no randomness, will be presented.   

Solitons and Quasielectrons in the Quantum Hall Matrix Model

We show how to incorporate fractionally charged quasielectrons in the finite quantum Hall matrix model. The quasielectrons emerge as combinations of BPS solitons and quasiholes in a finite matrix version of the noncommutative $\phi^4$ theory coupled to a noncommutative Chern-Simons gauge field. We also discuss how to properly define the charge density in the classical matrix model, and calculate density profiles for droplets, quasiholes and quasielectrons.   

Length Scale of Bulk Quantum Hall Effect

Quantum hall effects (QHE) are consequences of the condensation of the charge carriers into a novel macroscopic quantum state. Condensation produces steps in the hall resistance (R$_{xy})$ in fundamental units of h/e$^{2}$ (25812.8 Ohms), which correlate with Shubnikov-deHass oscillations in R$_{xx}$ and is represented graphically by the von Klitzing plot; which is a ``double-y'' graph of the hall resistance R$_{xy}$ and magnetoresistance R$_{xx}$ isotherms as functions of B. In two-dimensions electrical resistance per square is scale independent so the steps in R$_{xy}$ are in ohms. QHE condensation occurs in bulk systems as well. However, resistance of three-dimensional conductors depends on the sample geometry and the relevant transport coefficient, resistivity ($\rho _{xy})$, which is dimensionally different from resistance. Hence the QHE plateaus in bulk samples are not directly expressed in ohms. In this case the macroscopic $\rho _{xy}$ can be related to the R$_{xy}$ by a quantum length factor ``L'', we define L such that R$_{xy}$ = L$^{-1}(\rho _{xy). }$By analyzing literature data [Kul'bachinskii, V.A. et al., JETP Let., \textbf{70}, 767, 1999] we determine that for Sb$_{2}$Te$_{3 , }$L is equal to 1.7 nm a remarkable microscopic scale. This factor L is not just the ratio of the macroscopic length to the cross-sectional area of the conductor; instead, it is an effective length associated with the quantum hall states.   

Session A20: Multilayers, Thin Films, and Interfaces

Preparation and Friction Property of Dendritic Thin Film

To investigate the effect of film structure on their friction properties, thin films with dendritic structure were prepared on silicon surfaces by step-growth propagation, and their friction properties were studied by AFM. The molecular structure of grafted layer was adjusted by controlling the generations of grafting reaction and the concentration of functional groups on the surface. X-ray Photoelectron Spectroscopy measurements indicate systematic increased packing density of grafted dendritic layer as the generation of reaction and the functional group concentration increases. These structural variations were found to strongly influence their friction properties. The friction coefficients were found to increase with increasing the packing density of the grafted dendritic layer, and in the dynamic friction measurement (friction as a function of sliding velocity), a plateau was found for grafted layers with high packing densities. Analysis using a thermally activated model suggests that the lower mobility of molecules in densely packed films leads to longer relaxation time and higher friction coefficient compared with molecules in loosely packed layers.   

Mechanical Properties of Electrophoretically-Deposited CdSe Nanocrystal Films

Approaches to measuring and then minimizing the strain in electrophoretically deposited CdSe nanocrystal films are investigated. Under some conditions, fractured films are seen for films thicker than a critical thickness of about 0.8 microns for 3.2 nm nanocrystals. Cracking and some delamination are seen by SEM and AFM, and they are attributed to high strain energy in this film. The deposition conditions are varied to minimize this strain, which is thought to be due to the evaporation of residual hexane solvent after electrophoretic deposition - which changes the equilibrium separation of the nanocrystal cores. In situ observation confirms this assumption about the origin of film strain. These CdSe nanocrystal films become mechanically stronger and more resistant to chemical dissolution after being treated by cross-linker molecules such as 1,6-hexanedithiol and 1,7-heptanediamine.   

Growth and Characterization of Ultrathin Epitaxial Graphite films on Silicon Carbide

Ultrathin graphite films grown on 4H/6H SiC (0001) surface were investigated through Auger electron spectroscopy, LEED and STM. Graphite films, 1-6 graphene layers thick, were grown on both the Si- and C-terminated faces via thermal desorption of silicon. Film thickness was measured by modeling the Si:C Auger intensities. The $6\sqrt 3 \times 6\sqrt 3 _{ }$LEED pattern on SiC (0001) surface after annealing above 1250$^{o}$C can be explained by double-scattering theory with a $6\times 6$surface corrugation grating on both the graphite and the SiC. STM on the graphitized surface shows ``$6\times 6$'' domains typically 50 nm in size. Images indicate that the graphite films are continuous over substrate steps, but differences in the local electronic structure have been found for adjacent domains via STS. Magnetoconductance measurements on Hall bars created from these films demonstrate that the graphene/SiC system could be a promising platform for new integrated ballistic-carrier devices based on nano-patterned epitaxial graphene.   

Transition Layers in Co Films on Cu with Oxygen as Interface Surfactant Probed by Soft-X-Ray Resonant Magnetic Scattering (SXRMS)

Understanding the effect of surfactants on the magnetic properties of thin films is critical to understanding such diverse phenomena as spin-dependent transport (e.g., giant magnetoresistance [GMR]) and coupling between magnetic films. Interfacial morphology in ferromagnetic [FM] materials may be characterized as a combination of chemical and magnetic boundaries. Previous work by Kelly [1] and Barnes [2], used the diffusely scattered component of SXRMS to compare the magnetic and chemical roughness of the upper interface of 70 A Co films that were either bare or Fe-capped. The chemical and magnetic upper boundaries within the Co differed in the absence of an adjoining Fe layer, due to a transition layer of spins that do not follow applied magnetic fields. Although the bottom interface contributed very little to the resultant scattering, its relative contribution could not be resolved. By performing specular SXRMS over a wide range of incident angles, we are able to determine a depth profile of the magnetization in 30A thick Co films. The 30 A Co was sputter deposited on a 100{\AA}Cu/Si substrate. The Cu is partly deposited using oxygen as surfactant. We find a difference in the thickness of the magnetic transition layer in films grown with and without oxygen. [1] J.J. Kelly IV, et al. J. Appl. Phys. v.91 pp.9978-9986 (2002). [2] B.M. Barnes, Z. Li et. al J. Appl. Phys. \textbf{95}, 6654 (2004) Funding provided by ONR. Funding for the Synchrotron Radiation Center provided by NSF under Award No. DMR-0084402   

The role that varying nanocrystallinity plays in the thermodynamics of CoO films

Understanding the phase stability of nanocrystalline materials is a necessary step in developing a clear picture of mesoscopic physics. These materials are known to show excess specific heat at low temperatures similar to that seen in glassy and amorphous systems. The entropy associated with this excess specific heat can greatly effect the stability of these nanostructured materials. A systematic study of the thermodynamics of these systems over a wide temperature range has yet to be done. Thin films offer a novel way to explore the contributions of particle size and surface areas. We present results of heat capacity measurements on CoO thin films and CoO/MgO multilayers from 2K to 500K. Recent results show an increase in excess specific heat with decreasing particle size. Thanks to DOE for support.   

Charge exchange of Si ions with clean and I-covered Al(100)

Ion-surface charge exchange is a central process in many surface analysis and processing methods. Charge exchange of alkali, halogen and noble gas ions with surfaces has been investigated in previous ion scattering studies, while the interaction between a semiconductor atom and a metal surface has not been measured despite its importance. Si$^{+}$ ions were incident on an atomically clean Al (100) surface in ultra-high vacuum. The absolute ionization probability of scattered Si and recoiled Al were measured with time-of-flight, and detailed spectra of the ion yield were collected with an electrostatic analyzer. All of the scattered Si was neutralized, as expected for resonant charge transfer (RCT) of Si, which has a large ionization potential. Multi-charged recoiled Al ions were emitted, however. Surprisingly, Si scattered from iodine adatoms is partially ionized and the ionization changes little with respect to the coverage, energy and exit angle. This is in direct contrast to Li scattering from I/Fe*, and cannot be explained by RCT. * J.A. Yarmoff, Y. Yang and Z. Sroubek, Phys. Rev. Lett. \textbf{91}, 086104/1-4 (2003).   

Momentum and thickness-dependent evolution of quantum well states in the Cu/Co/Cu(001) system

Experimental advances in sample fabrication allow the observation of individual quantum well (QW) states from discrete atomic layer thicknesses. We present comprehensive angle-resolved photoemission measurements of the Fermi surface and underlying band structure of QW states in Cu/Co/Cu(001). Compared to bands from normal emission, we find a complicated evolution of QW states as a function of the thickness of both the copper overlayer and the cobalt barrier layer, as well as of the emission angle. This reveals a very high sensitivity of ``off-normal'' QW states to film thickness. Self-consistent calculations reveal a significant interaction between the QW states in the Cu overlayer and the Co barrier states, which leads to the observed complex behavior in particular ranges of energy and emission angle.   

Finite size effects on multilayer relaxations

We calculate the out of plane layer relaxations of thin embedded metallic films as a function of film thickness. The relaxations show an oscillating behavior that is consistent with superimposed surface-induced Friedel oscillations of the charge density of the film. Additionally there is an effect on the relaxation from the interaction between the surfaces that is analogous to the change in density of states that is induced by quantum well states. The calculations are performed by means of pseudopotential, first principles calculations in the framework of the Vienna ab initio simulation package (VASP).   

Noise in Half-Metallic Ferromagnetic Thin Films

Half metallic oxides are one of the most potential materials for the spintronics application, such as tunnel magnetoresistance. Among different half-metals CrO$_{2}$ is the only binary oxide that is ferromagnetic metal and shows almost 100{\%} spin polarization at the Fermi level. However, CrO$_{2}$ films have higher low frequency noise, which is an order of magnitude more than the metal film noise. This phenomenon restricts these films for applications such as integrated spin based sensor. We have measured the low frequency flicker (1/f) noise for CrO$_{2}$ film grown on TiO$_{2}$ substrate grown by CVD over different temperature and bias range. Our aim is to understand and relate the low frequency noise to grain boundaries, defects etc. Bias and temperature were varied during measurements. The noise is almost constant over the ``zero-current'' region of the current-voltage characteristics. At an elevated temperature the width of the ``zero-current'' region decreases and as a consequence the current increases for the same bias. Increase in noise level is observed with increase in sample temperature. The same increase in noise level is observed when we increase the sample volume for a particular temperature and bias. As expected the noise level is higher than in metal metal films. In our results we can clearly see the effect of the temperature and the size for the CrO$_{2}$ film.   

Heteroepitaxial growth and electronic structure of LaVO$_{3}$ films on SrTiO$_{3}$

In perovskite transition metal oxides, a relatively small variation in lattice constants allows the study of heteroepitaxial growth for a wide range of materials combinations. Recently, perovskite oxides have been extensively studied to understand their growth dynamics on an atomic scale, and to investigate their physical properties. The growth behavior of films can strongly depend on the terminating layer of the substrate surface. In this study, we investigated the growth dynamics and physical properties of heteroepitaxial LaVO$_{3}$ films grown on SrTiO$_{3}$. In particular, we found optimal growth conditions to obtain two-dimensional growth of atomically flat epitaxial LaVO$_{3}$ films.   

Suppressed Magnetization in La$_{0.7}$Ca$_{0.3}$MnO$_3$ / YBa$_2$Cu$_3$O$_{7-\delta}$ superlattices

Ferromagnetic/superconducting heterostructures are the subject of intense research, as these two types of long-range order are generally mutually exclusive and give rise to a variety of proximity phenomena. There is interest in studying these effects in superlattices of high T$_{c}$ superconductors and colossal magnetoresistance oxides, where the superconducting and ferromagnetic properties are depend strongly on the charge carrier density and thus charge transfer across the interface may be important. In a series of La$_{0.7}$Ca$_{0.3}$MnO$_3 $/YBa$_2$Cu$_3$O$_{7-\delta}$ superlattices, SQUID magnetometry showed that the LCMO saturation magnetization is significantly reduced. Polarized neutron reflectometry determined that the reduced moment is due to an inhomogenous magnetization profile. Specifically, the magnetization in each LCMO layer is suppressed close to the interfaces with the YBCO, possibly due to charge transfer across the interface.   

Microscopic magnetic structure of cuprate/manganite superlattices

Superconductivity and ferromagnetism are conventionally distinguished by mutually incompatible order parameters. However, the proximity of those materials in the artificially fabricated nanofilms of $\mathrm{YBa_2Cu_3O_7/La_{2/3}Ca_{1/3}MnO_3}$ (HTSC/FM) gives rise to new phenomena that do not exist in the isolated materials. We report on the first microscopic magnetization measurements by means of neutron reflectivity and resonant X-ray absorption. Our experimental results are consistent with the recently predicted "inverse" magnetic proximity effect. The analysis of neutron reflectivity data allows us to identify a likely magnetization profile, namely a sizable magnetic moment within the SC layer coupled antiferromagneticaly to the one in the FM layer. The scenario is supported by an anomalous superconductivity-induced enhancement of the off-specular reflection and by x-ray absorption, which testify to a strong mutual interaction of SC and FM order parameters.   

Crystalline $\gamma$-Al$_2$O$_3$ barrier for magnetite-based Magnetic Tunnel Junctions

Magnetite Fe$_3$O$_4$ is an interesting material for spintronics because it is expected to exhibit a very high spin polarization at room temperature. In this framework, we have developed an Oxygen Plasma Assisted Molecular Beam Epitaxy setup well suited to the growth of Fe$_3$O$_4$/Al$_2$O$_3$ bilayers. We have successfully grown highly insulating, 1.5 to 2 nm-thick crystalline $\gamma$-Al$_2$O$_3$ layers free of pinholes. The Fe$_3$O$_4$ layer is unaffected by the deposition of the Al$_2$O$_3$ barrier as evidenced by thorough magnetic, chemical and structural characterizations. These breakthroughs pave the way to all-oxide Fe$_3$O$_4$/Al$_2$O$_3$/Fe$_3$O$_4$ fully epitaxial magnetic tunnel junctions.   

Session A21: Focus Session: Dynamics of Transcription

A double-ratchet mechanism of transcription elongation and its control

Transcription, the process by which the genetic information encoded in DNA is transferred into RNA, is the first step in gene expression and it is the step at which most regulation occurs. A detailed understanding of the structural and mechanistic aspects of each step of transcription (initiation, elongation, termination and regulation) is one of the holy grails of biology. Here we characterize the motion of RNA polymerase (RNAP), the multi-subunit molecular motor that carries out the transcription process, during the elongation stage. We argue that during elongation RNAP moves by a complex Brownian ratchet mechanism in which the translocation along DNA and the binding of nucleotides into RNAP's catalytic center are coupled to a fluctuating internal degree of freedom associated with a protein sub-unit (the F-bridge) of RNAP. More precisely, the model is defined by a set of kinetic equations$^{ }$describing the competition for the catalytic site between an incoming nucleotide, the 3'-end of RNA, and the F-bridge which in its bent conformation blocks the active center. An important aspect of the model is the incorporation of the three ``active'' processes describing (i) the ejection of bound nucleotides from the active center through steric clashes with either the F-bridge in its bent conformation or with the 3'-end of RNA; and (ii) the forward translocation induced by bending of the F-bridge pushing against the 3'-end of RNA. The ``active'' processes do not imply a ``power stroke'' mechanism since the energy driving them is purely thermal. Indeed the model displays a route by which the system uses thermal fluctuations to control the rate, processivity and fidelity of transcription already before the irreversible chemical incorporation step. Moreover, the model qualitatively explains many aspects of both bio-chemical\footnote{Bar-Nahum, G., Epshtein, V., Ruckenstein, A.E., Rafikov, R., Mustaev, A., and Nudler, E. A ratchet mechanism of transcription elongation and its control. To appear in Cell, 2005.} and kinetic\footnote{Holmes, S.F., and Erie D.A. Downstream DNA sequence effects on transcription elongation. Allosteric binding of nucleoside triphosphates facilitates translocation via a ratchet motion. J Biol Chem. $278$, 35597-35608.} experiments on transcription elongation in \textit{E-coli} and makes a number of falsifiable predictions.   

Direct imaging of LacI repressor protein sliding along DNA

LacI repressor protein was observed in 1970 to bind to its operator site 100 times faster than allowed by diffusion [1]. A facilitated diffusion model incorporating1-D sliding and 3-D diffusion of the nonspecifically bound protein has been suggested to explain this phenomenon [2]. We have imaged the nonspecific binding of GFP-LacI monomers to elongated DNA molecules using single molecule imaging techniques. Upon binding to DNA, LacI proteins were observed to either be stationary, or slide along DNA. The characteristics of the sliding motion fit that of 1-D Brownian motion (with and without drift). The 1-D diffusion constant of the sliding proteins is 104 nm2/s, and it is 104 times lower than a typical protein's 3- D diffusion constant, 108 nm2/s. The characteristic dissociation time for both the stationary and the sliding proteins is 6s, and it is 100 times longer than the known dissociation time of 0.08s. The sliding length (DNA length scanned by the protein, not counting repeatedly scanned bases) ranges from 300 bp to 3000 bp, and it is significantly higher than the calculated optimal sliding length of 100 bp. We will discuss how these abnormal parameters alter the LacI specific binding speed. [1] A. D Riggs, S. Bougeois and M Cohn, J. Mol. Biol. 53, 401- 417 (1970). [2] O. G. Berg and C. Blomberg, Biophys. Chem., 4, 367-381 (1976).   

Thermodynamic DNA Looping by a Two-Site Restriction Endonuclease Studied using Optical Tweezers

Many enzyme-DNA interactions involve multimeric protein complexes that bind at two distant sites such that the DNA is looped. An example is the type IIe restriction enzyme Sau3AI$, $which requires two recognition sites to cleave the DNA. Here we study this process at the single DNA level using force measuring optical tweezers. We characterize cleavage rates of single DNA molecules in the presence of Sau3AI as a function of enzyme concentration, incubation time, and the fractional extension of the DNA molecule. Activity is completely inhibited by tensions of a few picoNewtons. By replacing Mg$^{2+}$ with Ca$^{2+}$, the Sau3AI dimers form but do not cleave the DNA, thus trapping DNA loops. We are able to pull apart these loops, measuring the force needed and the length of DNA released for each. We also characterize the number and length distributions of these loops as a function of incubation time and DNA fractional extension. The results of these studies are discussed in the context of a Brownian dynamics model of DNA looping.   

Bubbles in DNA

DNA melting proceeds through the formation of ?bubbles?. We have developed a new ensemble method by which we can directly measure the average length of the denaturation bubble and the statistical weights of the bubble states within the transition. For a bubble flanked by double-stranded regions, we find a nucleation size of $\sim $20 bases. In contrast, for bubbles opening at the ends of the molecule there is no nucleation threshold. An analysis of the statistical weight of intermediate states versus the length of the molecule L shows that the transition becomes strictly two-state only for L$\sim $1. We further find that a single mismatch in the oligomer sequence transforms a transition with many intermediate states into a nearly two-state transition. This observation can form the basis for an improved SNP (single nucleotide polymorphism) detection assay.   

Single-molecule analysis of the full transcription cycle

By monitoring the extension of a mechanically stretched, supercoiled DNA molecule containing a single bacterial promoter, we have been able to directly observe in real time the change in DNA extension associated with topological unwinding of $\sim$1 helical turn of promoter DNA by RNAP during transcription initiation. We find that this stage of transcription initiation is extremely sensitive to the torque acting on the supercoiled DNA. Upon addition of limited sets of nucleotides, changes in the polymerase/promoter interaction which are related to the process of abortive initiation can be studied in detail. Upon addition of the full set of nucleotides, the subsequent stages of transcription -- promoter escape, productive elongation and transcription termination -- can also be observed in real-time. The changes in DNA topology which occur at each of these stages have been determined, and these results provide for the first global view of the entire transcription cycle at the resolution of single molecules. \newline \newline Co-authors: Richard H. Ebright, Chen-Yu Liu and Andrey Revyakin, HHMI \& Waksman Institute, Rutgers University.   

The effect of sequence correlation on bubble conformation in double-stranded DNA

Although DNA exists in its duplex structure stably at physiological temperature, it has been experimentally observed that DNA duplex locally denatures, allowing bubble conformation along the strand due to thermal fluctuation. Here we present a new stochastic formulation, using the Fokker-Planck and the equivalent Langevin equation for base pair distance of DNA, which are transformed from the Edwards equation that describes the base pair distance distribution with base pair index regarded as time. By simulating the Langevin equation, with a DNA sequence modelled by dichotomic random noise whose correlation decays exponentially, we study the effect of sequence correlation on the bubble size distribution for various sequence correlation lengths. For average bubble size, we obtain an exact analytical expression via the Fokker-Planck equation and discuss it in comparison with the simulation results.   

Observing dynamics of chromatin fibers in Xenopus egg extracts by single DNA manipulation using a transverse magnetic tweezer setup

We have studied assembly of chromatin on single DNAs using Xenopus egg extracts and a specially designed magnetic tweezer setup which generates controlled force in the focal plane of the objective, allowing us to visualize and measure DNA extension under a wide range of constant tensions. We found, in the absence of ATP, interphase extracts assembled nucleosomes against DNA tensions of up to 3.5 piconewtons (pN). We observed force-induced disassembly and opening-closing fluctuations indicating our experiments were in mechano-chemical equilibrium. We found that the ATP-depleted reaction can do mechanical work of 27 kcal/mol per nucleosome, providing a measurement of the free energy difference between core histone octamers on and off DNA. Addition of ATP leads to highly dynamic behavior: time courses show processive runs of assembly and disassembly of not observed in the -ATP case, with forces of 2 pN leading to nearly complete fiber disassembly. Our study shows that ATP hydrolysis plays a major role in nucleosome rearrangement and removal, and suggests that chromatin in vivo may be subject to continual assembly and disassembly.   

Electrostatics in Biomolecular Interactions: a Surface Charge Method

Biomolecular interactions determine how transcription factors recognize their DNA binding sites, how proteins interact with each other, and consequently how a biological system functions. Since both proteins and DNAs are significantly charged, electrostatic interactions are among the most important when studying biomolecular interactions. Although the fundamental equations for electrostatics are known, the solution in low symmetry situations with a high dielectric constant solvent (e.g. water) can be difficult to obtain in an appropriate form and with an acceptable degree of accuracy and amount of computation. In order to compute the electrostatic force, each atom is usually modeled as a dielectric sphere with a point charge at its center. Even the case of two spheres is non-trivial. The energetic calculations of such a system are still very crude and lack systematic control of accuracy. To establish a scheme where accuracy of the computation can be controlled systematically, we have established a new formulation where the surface charge distribution is used as a new variable. The surface charge has the advantage of reducing the number of degrees of freedom (from 3D to 2D), can accommodate the presence of ions, and is applicable to arbitrary geometrical shapes. The Poisson-Boltzmann equation is currently the most popular approach in dealing with ionic effects. This approach, unfortunately, suffers from several drawbacks. In this talk, I will describe these drawbacks in slightly more detail, and describe possible methods to circumvent these problems. The solution for general geometrical shapes can be obtained numerically by choosing a tiling of the surface and solving a corresponding set of linear algebraic equations (the finite-element method). These equations can be efficiently solved numerically for use in molecular dynamics simulations.   

On geometric measures of regulatory complexity

Transcriptional regulation of a gene in a network can be characterized by a vector that counts the numbers of binding sites in the gene's promoter region where a protein product of every other gene can bind. We analyze the distribution of such vectors in the full S. cerevisiae genome and notice that they form an interesting low dimensional structure. This is significant for analyses that attempt to integrate expression and sequence information for the reconstruction of transcriptional networks since, in this case, one should be able to use similarity of promoter regions and expression profiles as effective aids.   

Session A23: Focus Session: Biological Hydrodynamics I

Overview: Biological Hydrodynamics

A short overview will be given at the beginning of the focus session 'Biological Hydrodynamics'.   

Vortex formation in Daphnia swarms

We propose a self-propelled particle model for the swarming of Daphnia, which takes into account propulsion of the particles, mutual avoidance of close encounters and attraction to a center. Various key parameters are identified in order to arrive at a phase diagram for qualitatively different steady-state motions. We find that a vertex is formed only in a finite range of propulsions, and analyze its transitions to other states. Hydrodynamic interaction between the particles can stabilize the vortex and change its velocity profile.   

The effect of shear flow on ordered suspensions of active particles

We explore the stability of orientationally ordered phases of a suspension of active particles, such as bacteria or motor- microtubule extracts, using a set of coarse grained hydrodynamic equations. While the orientationally ordered phase is linearly unstable, we show that it can be stabilised by the imposition of an external shear. We study the nonlinear response including shear banding in such active suspensions when subjected to steady or oscillating shear.   

Effects of the changes in the wall shear stresses (WSS) acting on endothelial cells (EC) during the enlargement of Abdominal Aortic Aneurysms (AAA)

The changes in the spatial and temporal distribution of the WSS and gradients of WSS during the enlargement of AAAs are important to understand the etiology and progression of this vascular disease, since they affect the wall structural integrity, primarily via the changes induced on the shape, functions and metabolism of the endothelial cells. PIV measurements were performed in aneurysm models, while changing systematically their size and geometry. Two regions with distinct patterns of WSS were identified. The region of flow detachment extends over the proximal half and is characterized by oscillatory WSS of very low mean. The region of flow reattachment, located distally, is dominated by large, negative WSS and sustained gradients of WSS that result from the impact of the vortex ring on the wall. Cultured EC were subjected to these two types of stimuli in vitro. The permeability of the endothelium was found to be largely increased in the flow detachment region. Endothelium cell-cell adhesion, proliferation and apoptosis were also affected by the high gradients of WSS.   

Model Cilia - Experiments with Biomimetic Actuable Structures and Surfaces

The use of cilia to drive fluid flow is a common motif in living organisms, and in the tissues of higher organisms. By understanding the ways that cilia function (or do not function), potential therapies to treat human diseases (such as cystic fibrosis) may be devised. The complex hydrodynamics of flow in beating ciliary tissues (such as lung epithelial tissues) are challenging to study in cultured tissues, suggesting the need for model systems that will mimic the morphology and beat patterns of living systems. To reach this goal, we have fabricated high aspect ratio cilia-like structures with dimensions similar to those of a lung epithelial cilium (0.2 to 2.0 $\mu $m diameter by $\sim $6 to 10 $\mu $m long). The structures and surfaces are composed of a magneto-elastomeric nanocomposite, allowing the actuation of artificial cilia by magnetic fields. We have studied the flexibility of the materials under conditions of flow (in microfluidics channels), and will present theoretical and experimental data from various efforts at actuation. We will discuss details of the fabrication of the ciliated structures and present results of mechanical characterization. The impact of this work on the understanding of fluid flow above ciliated cells and tissues and potential applications of such model systems will also be described.   

Quorum polarity and the dynamics of the zooming bionematic phase

Many species of bacteria are peritrichously flagellated, i.e. the long, helical, rapidly rotating flagella that propel them emerge out of motors that appear randomly distributed over the body of the bacterial cell. The organism considered here is {\it Bacillus subtilis}. The cell body is a rod approximately 4 $\mu$m long, 0.7$\mu$m in diameter; flagella are 3 or 4 times longer than the body. Swimming cells are pushed by the flagella, bundled into a braid of rotating helices. When the bacteria self concentrate into an approximately close-packed assemblage, rapidly moving (zooming) domains of aligned bacterial rods continually form and break apart. PIV measurements show that correlation times are seconds, lengths are hundreds of micrometers, transport of passive tracers is superdiffusive.Below a threshold concentration there is no collective dynamic. A theory of this zooming bionematic phase will be presented, together with measurements and video sequences. The theory considers hydrodynamic cell-cell and collective interactions, the collectively generated flow of the suspending water relative to the cells, and the dynamics of helix bundle flipping, yielding quorum polarity within a given zooming domain. Quorum sensing of signalling molecules and molecular transport generally are pertinent microbiological applications.   

Rheology and dynamics of active motor-filament mixtures

We have developed a hydrodynamic description of both the isotropic and polarized phases of mixtures of polar filaments and molecular motors taking into account the fluctuations in both the motor and the filament densities. The various couplings in the hydrodynamic equations are related to microscopic parameters by comparing continuum equations written down on the basis of symmetry considerations to those obtained from a microscopic model of motor-filament interaction. Due to the anisotropy of filament diffusion, motors are capable of generating net filament motion relative to the solvent, resulting in filament convection along the direction of local alignment. The effect of this new term on traveling wave in the polarized phase is analyzed by numerical solutions of the nonlinear hydrodynamic equations. The equations are also used to study the linear rheology of active solutions (stress generated due to an imposed shear flow).   

Two Dimensional State Transition of a Swarming Model

A rotating mill is widely seen in swarming patterns of various species, such as ants, fishes, or daphnia. Levine et al. (2000) proposed an individual based model which produces a pair of co- existing clockwise and counter-clockwise mills on top of each other while a unified rotating mill can be achieved by switching the formula of the self-propulsion to an ensemble average. Without changing its fundamental concepts, we modify the model to include a Rayleigh-type self-driving mechanism, which has a cleaner connection to its continuum limit, i.e., macroscopic description, where analysis can be more efficiently done. By varying parameter values, we find that the modified model goes through a three-stage transition from the co-existing state to the unified state. We also compare the numerical results of the model and of its continuum counterpart.   

Dynamics of the Chemotactic Boycott Effect

Aerobic bacteria often live in thin fluid layers on irregular surfaces, near solid-air-water contact lines where the interplay between fluid interface geometry, nutrient transport, and chemotaxis is central to the micro-ecology. To elucidate these processes, we use the simplified geometry of a sessile drop and provide direct experimental evidence for the ``chemotactic Boycott effect" in suspensions of {\it B. subtilis}: upward oxygentaxis toward the drop surface leads to accumulation of cells in a thin layer, which flows down to the contact line and produces there a persistent vortex which traps cells near the meniscus. These phenomena are explained quantitatively with a mathematical model consisting of coupled oxygen diffusion and consumption, chemotaxis, and viscous fluid dynamics; they are shown to be associated with a singularity in the chemotactic dynamics at the contact line.   

A new parameter identification method to obtain change in smooth musclecontraction state due to mechanical skin irritation

A light scratch with a needle induces histamine and neuropetide release on the line of stroke and in the surrounding tissue. Histamine and neuropeptides are vasodilaters. They create vasodilation by changing the contraction state of the vascular smooth muscles and hence vessel compliance. Smooth muscle contraction state is very difficult to measure. We propose an identification procedure that determines change in compliance. The procedure is based on numerical and experimental results. Blood flow is measured by Laser Doppler Velocimetry. Numerical data is obtained by a continuous model of hierarchically arranged porous media of the vascular network [1]. We show that compliance increases after the stroke in the entire tissue. Then, compliance decreases in the surrounding tissue, while it keeps increasing on the line of stroke. Hence, blood is transported from the surrounding tissue to the line of stroke. Thus, higher blood volume on the line of stroke is obtained. [1] Bauer, D., Grebe, R. Ehrlacher, A., 2004. A three layer continuous model of porous media to describe the first phase of skin irritation. J. Theoret. Bio. in press   

Bacteriophage-mediated indirect competition in B.bronchiseptica:Experiment and Theory

We demonstrate empirical evidence of bacteriophage-mediated indirect competition in microbiological populations. The bacteriophage-mediated competition acts between two genetically identical bacterial strains that differ only in that one is the carrier of a temperate lysogenic phage and the other is susceptible to the phage. We observe that in spite of the absence of direct competition the strain with lysogenic phage successfully invades and outcompetes the resident strain susceptible and more vulnerable to the phage. The amount of indirect competition is dependent on the susceptibility and the phage-induced mortality of the resident bacterial strain. We develope mathematical models of the phage-mediated competition and reproduce its dependence on mortality.   

Collective motion, superdiffusion, and non-thermal noise in active bacterial baths

We present experimental studies of the dynamics of concentrated bacterial colonies in three-dimensions. The bacteria we use is e. coli at varying densities. By analyzing the trajectories of colloidal spheres embedded in these bacterial bath suspensions, we extract one- and two-point mean square displacements which exhibit superdiffusion crossing over to a diffusive regime. When combined with independent measurements of the response function of the suspension, the data enable us to extract the noise spectrum of the bath. This work was supported by NSF DMR02-03378 and NASA NAG3-2172   

Session A24: Focus Session: Structure, Dynamics and Resilience of Complex Networks

An information theoretic derivation of spectral graph partitioning

At the APS meeting in 2004, we introduced an information-theoretic algorithm called the ``network information bottleneck" (NIB) for clustering nodes of a network into modules (cf. arxiv.org/q-bio/0411033). Numerical experiments show that, although the modules are found by minimizing a free energy with no references to normalized edge-cuts or numbers of edges between modules, the resulting partitions are both information-modular and edge-modular (exhibiting low normalized edge-cuts). Moreover, the resulting partioning algorithm is competitive both in accuracy and efficiency with methods popular in the physics community. These numerical results along with asymptotic equivalence between the information-optimal and edge-optimal partitionings are presented.   

Traceroute-like exploration of unknown networks: a statistical analysis

Mapping the Internet generally consists in sampling the network from a limited set of sources by using {\tt traceroute}-like probes. This methodology has been argued to introduce uncontrolled sampling biases that might produce statistical properties of the sampled graph which sharply differ from the original ones. Here we explore these biases and provide a statistical analysis of their origin. We derive a mean-field analytical approximation for the probability of edge and vertex detection that exploits the role of the number of sources and targets and allows us to relate the global topological properties of the underlying network with the statistical accuracy of the sampled graph. In particular we show that shortest path routed sampling allows a clear characterization of underlying graphs with scale-free topology. We complement the analytical discussion with a throughout numerical investigation of simulated mapping strategies in different network models.   

Effects of Community Structure on Search and Ranking in Information Networks

The World-Wide Web (WWW) is characterized by a strong community structure in which communities of webpages (e.g. those sharing a common keyword) are densely interconnected by hyperlinks. We study how such network architecture affects the average Google ranking of individual webpages in the community. It is shown that the Google rank of community webpages could either increase or decrease with the density of inter-community links depending on the exact balance between average in- and out-degrees in the community. The magnitude of this effect is described by a simple analytical formula and subsequently verified by numerical simulations of random scale-free networks with a desired level of the community structure. A new algorithm allowing for generation of such networks is proposed and studied. The number of inter-community links in such networks is controlled by a temperature-like parameter with the strongest community structure realized in ``low-temperature'' networks.   

Community discovery and information flow in networks

The dynamics of information within social groups is relevant to issues of productivity, innovation and the sorting out of useful ideas from the general chatter of a community. How information spreads and is aggregated determines the speed with which individuals and organizations can act and plan their future activities. This talk will describe new mechanisms for automatically identifying communities of practice within organizations and for elucidating the spread of information within those communities. Many of these mechanisms rely on the scale-free nature of social networks.   

Voter Model on Heterogeneous Graphs.

We study basic properties of the voter model on heterogeneous graphs with an arbitrary degree distribution. By mapping the voter model to a coalescing random walk, we are able to understand the effect of the degree distribution on the dynamical behavior. We thereby find that the mean consensus time for finite graphs of $N$ sites scales as $\mu_1^2 N/\mu_2$, where $\mu_1$ is the mean degree and $\mu_2$ the second moment of the degree distribution. Thus the consensus time may scale sublinearly with system size if the degree distribution is sufficiently broad.   

Anomalous Transport in Complex Networks

To study transport properties of complex networks, we analyze the equivalent conductance $G$ between two arbitrarily chosen nodes of random scale-free networks with degree distribution $P(k)\sim k^{-\lambda}$ in which each link has the same unit resistance. We predict a broad range of values of $G$, with a power-law tail distribution $\Phi_{\rm SF}(G)\sim G^{-g_G}$, where $g_G=2\lambda -1$, and confirm our predictions by simulations. The power-law tail in $\Phi_{\rm SF}(G)$ leads to large values of $G$, thereby significantly improving the transport in scale-free networks, compared to Erd\H{o}s-R\'{e}nyi random graphs where the tail of the conductivity distribution decays exponentially. Based on a simple physical ``transport backbone'' picture we show that the conductances are well approximated by $ck_Ak_B/(k_A+k_B)$ for any pair of nodes $A$ and $B$ with degrees $k_A$ and $k_B$. Thus, a single parameter $c$ characterizes transport on scale-free networks.   

Complex networks are self-similar

A large number of real networks are called ``scale-free'' because they show a power-law distribution of the number of links per node. However, it is widely believed that complex networks are not length-scale invariant or self-similar. This conclusion originates from the ``small world'' property of these networks, which implies that the number of nodes increases exponentially with the ``diameter'' of the network, rather than the power-law relation expected for a self-similar structure. Nevertheless, here we present a novel approach to the analysis of such networks, revealing that their structure is indeed self- similar. This result is achieved by the application of a renormalization procedure which coarse-grains the system into boxes containing nodes within a given ``size.'' Concurrently, we identify a power-law relation between the number of boxes needed to cover the network and the size of the box defining a self- similar exponent. These fundamental properties, which are shown for the WWW, social, cellular and protein-protein interaction networks, help to understand the emergence of the scale-free property in complex networks. They suggest a common self- organization dynamics of diverse networks at different scales into a critical state and in turn bring together previously unrelated fields: the statistical physics of complex networks with renormalization group, fractals and critical phenomena.   

Interaction prediction using conserved network motifs in protein-protein interaction networks

High-throughput protein interaction detection methods are strongly affected by false positive and false negative results. Focused experiments are needed to complement the large-scale methods by validating previously detected interactions but it is often difficult to decide which proteins to probe as interaction partners. Developing reliable computational methods assisting this decision process is a pressing need in bioinformatics. This talk will describe the recent developments in analyzing and understanding protein interaction networks, then present a method that uses the conserved properties of the protein network to identify and validate interaction candidates. We apply a number of machine learning algorithms to the protein connectivity information and achieve a surprisingly good overall performance in predicting interacting proteins. Using a ``leave-one-ou approach we find average success rates between 20-50\% for predicting the correct interaction partner of a protein. We demonstrate that the success of these methods is based on the presence of conserved interaction motifs within the network. A reference implementation and a table with candidate interacting partners for each yeast protein are available at http://www.protsuggest.org   

Stability and topology of scale-free networks under attack and defense strategies

We study tolerance and topology of random scale-free networks under attack and defense strategies that depend on the degree $k$ of the nodes. This situation occurs, for example, when the robustness of a node depends on its degree or in an intentional attack with insufficient knowledge on the network. We determine, for all strategies, the critical fraction $p_c$ of nodes that must be removed for disintegrating the network. We find that for an intentional attack, little knowledge of the well-connected sites is sufficient to strongly reduce $p_c$. At criticality, the topology of the network depends on the removal strategy, implying that different strategies may lead to different kinds of percolation transitions.   

Modeling Cascading Failures in the North American Power Grid

The North American power grid, one of the most complex technological networks in existence, permits long-distance power transmission as well as disturbance propagation. We model the grid using its actual topology and incorporate plausible assumptions about transmission substation load and overload. Our results indicate that a solitary substation loss can induce an overload cascade and reduce the grid's transmission efficiency by 25{\%}. Examining the damage inflicted by single node removals, we find three universal behaviors which suggest that 40{\%} of the transmission substations can induce an overload cascade when perturbed. While significant damage can result from a single node removal, subsequent removals have only incremental effects, which agree with the power grid's topological resilience to less than 1{\%} node loss.   

Avalanche dynamics in complex networks

Avalanche dynamics, triggered by small initial perturbation, but spreading to other constitutes successively, is one of intriguing problems in complex systems. Here, we study such dynamics on scale-free networks. We first consider the case where the failed vertex can be recovered through the Bak-Tang-Wisenfeld sandpile model. Using the branching process approach, we obtain the exponents associated with the power-law behaviors of the avalanche size and duration time distributions, which depend on the degree exponent. Second, for the case where the failed vertex cannot be recovered permanently, we study the model for the data packet transport proposed by Motter and Lai. We find that depending on the control parameter, which is the relative ratio between the traffic amount and the failure threshold, a phase transition can occur from a free flow to congested state. At the transition point, the avalanche size distribution turns out to be robust against the degree exponent as long as the degree exponent is between 2 and 3.   

Dynamical patterns of epidemic outbreaks in complex heterogeneous networks

We present a throughout inspection of the dynamical behavior of epidemic phenomena in populations with complex and heterogeneous connectivity patterns. We show that the growth of the epidemic prevalence is virtually instantaneous in all networks characterized by diverging degree fluctuations, independently of the structure of the connectivity correlation functions characterizing the population network. By means of analytical and numerical results, we show that the outbreak time evolution follows a precise hierarchical dynamics. Once reached the most highly connected hubs, the infection pervades the network in a progressive cascade across smaller degree classes. Finally, we show the influence of the initial conditions and the relevance of statistical results in single case studies concerning heterogeneous networks. The emerging theoretical framework appears of general interest in view of the recently observed abundance of natural networks with complex topological features and might provide useful insights for the development of adaptive strategies aimed at epidemic containment.   

Session A25: Focus Session: Computational Nanoscience I

Future directions in the simulation of nanomaterials

Nanotechnology forces us to rethink conventional solid state physics. Quantum phenomena are commonplace. A system with 101 atoms may be very different from one with just 100. Key biomolecules may resemble spaghetti more than silicon; viscosity often dominates over inertia. Statistical physics is often not carrier statistics; equilibrium may be irrelevant, though the kinetics of non-equilibrium processes can be crucial. Even when nanoscale issues concern structure (rather than functionality), a new viewpoint is needed. Important features, like elasticity and electrostatic energies, have clear macroscopic analogies, but different issues arise, reflecting the reproducibility of quantum dots or the accuracy of self-organisation. Concepts like epitaxy and templating are usually micro- or meso-structural, but emerge again in modelling for the nanoscale. Unexpected analogies between biomolecule and semiconductor systems appear. My examples will include quantum dots and possible silicon-based, room temperature, quantum information processing, and will emphasise new opportunities in nanoscale science.   

QMC with a Stochastic Poisson Solver: An application to realistic models of quantum dots

Quantum Monte Carlo can be used to study interacting electrons in semiconductor quantum structures. We introduce a new approach in which the potential acting on each electron is found by sampling using a classical Monte Carlo ``Walk On Spheres''(WOS) algorithm within the QMC calculation. This allows cheap and coarse estimates of the potential to be used, since the QMC averages the noise in the potential. The averaging is accomplished simply in VMC, and in DMC we use the penalty method [1] to modify the non-linear branching factor according to the noise in our potential estimate. The WOS algorithm is general enough to be applied to devices with arbitrary geometries, dielectric constants and gate biases. We employ this QMC-WOS hybrid approach to a real heterostructure as described in Ref.[2]. Specifically we calculate the singlet triplet splitting for a two electron double dot and compare with DFT calculations. \newline [1.] Ceperley D. M. and Dewing M. J. Chem. Phys. 110,9812 1999 \newline [2.] Elzerman et. al. PRB 67, 161308(R) 2003   

Quantum Monte Carlo Simulations of Exciton-Exciton Scattering in Quantum Wells

Exciton-exciton interactions are characterized by the scattering length, which is a property of excited states of a four-particle wavefunction at the zero energy limit. As is well-known in atomic physics, the scattering length can be notoriously hard to predict theoretically, since correlation and van der Waals forces can play a large role. We have developed a quantum Monte Carlo (QMC) approach that can accurately calculate the bulk exciton-exciton scattering length within the effective mass approximation (Shumway and Ceperley, PRB {\bf 63}, 165209, 2001). As an added benefit of this technique, all bound biexciton states are also calculated, providing an additional test for the simulations. Now we have adapted this excited-state QMC technique to exciton-exciton interactions in quantum wells, where there is current interest in exciton or polariton condensates. We discuss predictions of our simulations, especially ways to modify exciton-exciton interaction strength with different well geometries and external fields.   

Quantum Perturbation Theory in O(N): Ab initio response theory for nanomaterials

One of the main obstacles in predicting the electronic properties of complex nanomaterials directly from fundamental theory is the enormous computational complexity involved in solving the equations governing the quantum mechanical response to an external perturbation. We have recently introduced an orbital-free quantum perturbation theory based on perturbed spectral projections of the Hamiltonian. It gives the density matrix and its response upon variation of the Hamiltonian by quadratically convergent recursions. The approach is surprisingly simple and efficient. It allows treatment of both embedded quantum subsystems and response functions. The computational cost scales linearly with the systems size N and for local perturbations it scales linearly with the size of the perturbed region O(N\_pert), i.e. as O(1) with the total system size. Traditional textbook perturbation theory based of wave function or Green's function formalism can be replaced by a quadratically convergent explicit recursion that gives the expansion terms of expectation values to any order. Connecting and disconnecting individual weakly interacting quantum subsystems can be performed by treating off-diagonal elements of the Hamiltonian as a perturbation. This should be highly useful in nanoscience for connecting quantum dots, surfaces, clusters and nanowires, where the different parts can be calculated separately.   

Efficient Computational Methods to Treat Multiple Scattering in Electron Diffraction by Nanostructures

Our purpose is to extend the capabilities of surface structure determination methods, such as Low Energy Electron Diffraction, so they can be used for nanostructures. To treat non-periodic systems, a cluster approach is used. The main computational challenge consists in solving a Ax=b matrix-vector equation of large dimension. Since matrix inversion is both memory and compute-time demanding, we have developed and tested two fast iterative methods to solve the above equation: the Sparse-Matrix Canonical Grid (SMCG) method shifts the atoms to a regular space grid and makes use of FFT transformations while the Multi-Level Singular-Value Decomposition (MLSVD) performs fast rank determination and SV decomposition of A. For both these methods, the compute time scales as N x log$_{2}$N per iteration, where N is the number of atoms. These two methods complement each other in terms of the types of nanostructures that they handle best.   

New Theoretical Method for Molecular Systems and Nano-molecules

In general, it has been known that electrons of a molecular system are indistinguishable. However, the total electronic energy computed by the conventional theoretical method, the Hartree Fock Theory or the Density Functional Theory, consists of kinetic and attractive energies on distinguishable electrons and repulsive potential energy on indistinguishable electrons on a molecular system. The other question of the conventional methods is singularity on two-electron integrals because electrons in a molecular system cannot exist at the centre of their nucleus which brings about singularity in free space. The new theoretical method that modified above problems has been applied to hydrogen molecule H$_{2}$, and its results have been compared with those of the conventional theoretical methods installed in Gaussian 98 program. The total energies of the conventional methods are much bigger than -1.0 (a.u.) the total energy of hydrogen molecule in the infinite H-H bond distance, and the electron-electron repulsive energies are about 2.911 to 7.728 not 0.0 eV on 1,000 {\AA} H-H bond distance although the energies of the new method agree with the values of the physical concept on H$_{2}$.   

Adaptive Quantum Design for Nanoscience

Recent advances in nano-technology have enabled us to construct ultra-small opto-electronic devices, such as filters, modulators, and resonators. Material response functions can be made to order on the atomic level by explicitely breaking symmetries, such as relative widths in quasi-one-dimensional multi-layer dielectric filter arrays. This requires new software tools that optimize desired material response characteristics by finding the global minimum in large parameter landscapes of possible solutions. In this talk, we show examples of this adaptive quantum design, including optical filters in one and two dimensions and a quantum mechanical tight-binding model. Numerical optimization techniques, such as simulated annealing and the genetic algorithm, will be discussed briefly. This approach is useful in the design of a new generation of nano-devices.   

A numerically tractable method for a non-uniform electron gas system with an atomic center

To perform the first-principles calculation of a non-uniform electron system with both localized and delocalized electrons, we have developed a tractable algorithm using the transcorrelated method and Pahl-Handy's mixed basis. Both two-body and three body potentials are expanded in terms of spherical harmonics or in the Fourier series. Radial integrals are analytically evaluated, which makes the numerical simulation as simple as series of matrix multiplication and the fast Fourier transformation. Possible application for a Kondo system is addressed. Our numerical simulation could provide a first-principles evaluation of the U-term and residual exchange-correlation energy functional for an effective many-body system of the density functional theory.   

Time-dependent quantum process for electrons assisted by oscillating electric field

Significant advances in nanometer-scale techniques have enabled us to control the transport phenomena of electrons artificially. In order to control the electronic states of the quantum dots by using oscillated electric field (OEF), the time development features of the electron wave function should be fully understood, because the state of electron changes sharply for a short time. Here, we study time-dependent quantum process for electrons assisted by OEF based on solving the TD Schr\"odinger equation numerically both in the real-space and -time. Introducing the effective lifetime of an electron in the quantum dot, we discuss how OEF modulates the transmitting probability. We especially focus on the relationship among the lifetime, the strength and frequency of the injected electric field while varying the potential profile. We will further report the effect of the electromagnetic radiation caused by electron's self-motion as well as the inter-electron interaction.   

Using quantum mechanics to synthesize electronic devices

Adaptive quantum design [1] has been used to explore the possibility of creating new classes of electronic semiconductor devices. We show how non-equilibrium electron transmission through a synthesized conduction band potential profile can be used to obtain a desired current - voltage characteristic. We illustrate our methodology by designing a two-terminal linear resistive element in which current is limited by quantum mechanical transmission through a potential profile and power is dissipated non-locally in the electrodes. As electronic devices scale to dimensions in which the physics of operation is dominated by quantum mechanical effects, classical designs fail to deliver the desired functionality. Our device synthesis approach is a way to realize device functionality that may not otherwise be achieved. [1] Y.Chen, R.Yu, W.Li, O.Nohadani, S.Haas, A.F.J. Levi, Journal of Applied Physics, Vol.94, No.9, p6065, 2003   

Coupling Classical Molecular Dynamics Simulations to Continuum Current and Heat Flow Equations: Application to Frictional and Resistive Heating of Nanoscale Metal Contacts

To reproduce experimental heat flow rates and to model resistive heating, atomic kinetic energies in a molecular dynamics (MD) simulation are coupled via an ad hoc feedback loop to continuum current and heat transfer equations that are solved numerically on a finite difference grid (FDG). For resistive heating, the resistance in each region of the FDG is calculated from the experimental resistivity and atomic density, and a network of resistors is established from which the potential at every FD point is calculated given an applied voltage. The potential difference between connected FDG regions and the resistance are then used to calculate the current between the two points, the heat resulting from that current, and the magnetic and electrical force between grid regions. This information is then added back into the atomic simulation. To illustrate this method, simulations of the frictional and resistive heating of a nanoscale metal contract will be presented. This work was funded by MURI Project No. N00014-04-1- 0601.   

The Solution of the Interior Eigenvalue Problem for Large Scale Nanosystems

First-principles materials science calculations typically involve a self-consistent solution of the Kohn-Sham equations. These types of methods typically scale with the cube of the system size and can only be used to study systems of up to a thousand atoms. To study larger systems we use semi-empirical potentials or approximated ab initio potentials such as those constructed using the charge patching method. Using these types of potentials does not require a selfconsistent solution of our effective single particle equations and we can solve directly for the few states of interest around the gap. The solution of our single particle equations now becomes an interior eigenvalue problem for a few states around a given energy rather than the self-consistent solution for the lowest n states where n is the number of bands. In this talk I will compare different methods (conjugate gradient, Jacobi-Davidson, Lanczos) for this problem with particular emphasis on solving large nanosystems on parallel computers. Work supported by the DOE under the Modeling and Simulation in Nanoscience Initiative.   

Choosing a Classical Potential in Multi-Scale Modeling

For problems relating to fracture in multi-scale modeling, a consistent embedding of a quantum (QM) domain in its classical (CM) environment requires that the classical potential chosen for the CM region should yield the same geometry and elastic properties as the QM domain. It is proposed that such a classical potential can be constructed using \textit{ab initio} data on the equilibrium structure and weakly strained configurations calculated from the quantum description, rather than the more usual approach of fitting to a wide range of empirical data. This scheme is illustrated in detail for a model system, a silica nanorod that has the same stiochiometric ratio of Si:O as observed in real silica. The potential is chosen to have the same functional form as TTAM but the parameters are fitted using a genetic algorithm with force data obtained from a quantum calculation. The Young's modulus (Y) obtained from this classical potential matches closely with that obtained from the QM method for strains up to 10{\%}, unlike the standard TTAM which differs by 18{\%}. Furthermore, the bond lengths and bond angles in the rod are an order of magnitude more accurate for the new potential in comparison to that from the current TTAM or BKS potential parameters. This potential provides a ``seamless'' coupling between the QM and CM regions in applications of QM/CM multi-scale modeling for this silica nanorod. The wider application of this potential can be found in glasses.   

Session A26: Focus Session: Nanotubes and Nanowires: Carbon Nanotube Transistors

Double-wall carbon nanotube quasi-ballistic conduction

We demonstrated room-temperature quasi-ballistic electron conduction in double-wall carbon nanotubes (DWNTs) produced using a modified arc-discharge method [1]. Conductance dependence on the length of DWNT was measured by submerging the sample into liquid mercury. The conductance versus length plots show plateaus, indicating weak dependence of the electrical resistance of the DWNTs on the length of the nanotubes segment connecting electrodes. We infer a mean free path between 0.6 -- 10 micron meter for 80{\%} of the tubes, which is in good agreement with the results of calculations based on the electron scattering by acoustic-phonons and by disorder. [1] H. Kajiura et al. Chem Phys Lett 398(2004)476-9.   

Directional growth of Single-Walled Carbon Nanotubes for Nanotube-on-Insulator Applications

Dense arrays of highly aligned carbon nanotubes were synthesized by chemical vapor deposition on flat crystalline substrate surfaces. The nanotube orientation was found to favor certain crystalline directions of the substrate, regardless of the gas flow direction. This is in sharp contrast to the randomly oriented growth of nanotubes on Si/SiO$_{2}$ substrates. These nanotubes are commonly tens of micrometers long, and the inter-tube spacing is typically around 200 nm, which can be controlled to certain degree. In addition, a second layer of nanotubes can be grown along the gas flow direction atop the first layer by carrying out a second round of CVD synthesis. This observation, as a side proof, supports the hypothesis that the substrate-nanotube interaction plays an important role. Our synthesis of dense arrays of well aligned and evenly spaced carbon nanotubes paves the way toward large-scale assembling of nanotube-on-insulator (NOI) devices and circuits, in analogy to the silicon-on-insulator (SOI) approach adopted by the semiconductor industry.   

Band Engineering of Partially Exposed Carbon Nanotube Field-Effect Transistors

We present a new approach to engineer the band structure of carbon nanotube field-effect transistors via selected area chemical gating. By exposing the center part or the contacts of the nanotube devices to oxidizing or reducing gases, a good control over the threshold voltage and subthreshold swing has been achieved. Our experiments reveal that NO$_{2}$ shifts the threshold voltage higher while NH$_{3}$ shifts it lower for both center- exposed and contact-exposed devices. However, modulations to the subthreshold swing are in opposite directions for center-exposed and contact-exposed devices: NO$_{2}$ lowers the subthreshold swing of the contact-exposed devices, but increases that of the center-exposed devices; In contrast, NH$_ {3 }$reduces the subthreshold swing of the center-exposed devices, but increases that of the contact-exposed devices. A model has been developed based on Langmuir isotherm, and the experimental results can be well explained.   

Carbon nanotube electronics and opto-electronics

Carbon nanotubes (CNTs) have ideal properties for applications in nano/opto-electronics. Strong emphasis has been placed on the fabrication of CNT field-effect transistors (CNTFETs) with very promising results. CNTFETs, however, still have weak points. Specifically, charge-transfer at the CNT-metal interfaces leads to the formation of Schottky barriers. Also, upon scaling of the gate insulator, unipolar CNTFETs turn ambipolar (a-CNTFETs) with large OFF currents. I will discuss how we eliminated these problems by chemical (charge-transfer doping), or electrostatic doping of the contact regions. The resulting CNTFETs have excellent characteristics. a-CNTFETs are particularly valuable in photonics. When electrons and holes are injected from the opposite terminals of a a-CNTFET, a fraction of them recombine radiatively, producing a single CNT light source. Unlike conventional p-n diodes, a-CNTFETs are not doped and there is no fixed p-n interface. By spatially resolving the emission we show that the light can be translated along the CNTFET by varying the gate voltage. Study of the properties of the emission as a function of applied bias provides new insights on the electrical transport in CNTs. Stationary light spots are also observed. Finally, single CNT photoconductivity spectra and theoretical modeling are used to understand the nature of the excited states of the CNTs. In collaboration with: J. Appenzeller, J. Chen, M. Freitag, C. Klinke, Y.-M. Lin, V. Perebeinos, J. Tersoff, J. Tsang.   

Diameter dependence of carbon nanotube transistor performance

As has been shown before, single wall carbon nanotube field-effect transistors (CNFETs) behave as Schottky barrier devices. The important question however is, what barrier height has to be overcome for current injection. So far, no detailed study exists that explains the impact of nanotube type and metal contacts in this context. Here we present the first statistical analysis of the dependence of on-current in a CNFET on the aforementioned two parameters. We show that a large data set of more than 100 devices can be consistently explained within a model that relates the on-current to a distinct Schottky barrier height which is determined quite reproducible by the nanotube diameter and the source/drain metal contact. Our study allows to identify the desired combination of tube diameter and type of metal that allows for optimum device performance of a CNFET.   

Dual-Gated Carbon Nanotube Field-Effect Transistors with Tunable Polarities

In this paper, we present a novel design for carbon nanotube field-effect transistors (CNFETs). This design allows us to obtain a p-i-p (or n-i-n) doping profile along the tube. Our CNFET structure is based on a back-gated geometry. An additional middle gate electrode is patterned between the source and drain contacts, so that the segments of the nanotube near the source/drain contacts can be electrically and/or chemically doped in a self-aligned fashion. The potential of outer and middle nanotube segments is independently controlled by the back gate and middle gate, respectively. By controlling the potential of the nanotube in the outer regions, p- or n-type CNFETs can be obtained on the same device. The dual-gated CNFETs exhibit bulk switching behavior, rather than switching dominated by the Schottky barriers at the contacts, and show excellent performance close to theoretical limits.   

Investigation of Schottky Barrier Behaviors between Semiconductive Single-Walled Carbon Nanotubes and Different Metals

Single Walled Carbon nanotubes are very promising material for nanoelectronics. Schottky barrier contact are made to SWNT through Al or Ti electrode while the other end of SWNTs are ohmically contacted by Au. Electronic transport through Schottky barriers are studied and competition between tunneling and thermionic emission are control with a back gate voltage. Schottky barrier diodes are made by SWNT and low work function metals for the first time.   

Effect of Metal-Carbon Nanotube Interface in Carbon Nanotube Devices

The four-terminal contact resistance of individual multiwalled carbon nanotubes with Au/Ti electrodes was measured at low temperature. The contact resistance as a function of bias voltage showed large asymmetric behavior. At zero bias, the contact resistance showed large variations with gate voltage. The origin of this asymmetric behavior and the the correlation between nanotube device resistance and contact resistance will be discussed.   

Carbon-nanotube-based single electron memories processed by double self-assembly

We demonstrate wafer-scale integration and operation of single electron memories based on carbon nanotube field effect transistors (CNFETs). Our method involves a two step double self assembly process. The first step consists of a Hot-Filament CVD growth and in situ electrical connection of single walled carbon nanotubes on a predefined submicron catalytic template acting as contact electrodes. We obtain a overall integration yield of semiconducting carbon nanotubes exhibiting field effect that can exceed 50{\%} for 9000 devices on a 2 inches wafer. The second step is a wet step which consists of local functionalization and controlled attachment of a colloidal gold bead of radius 15nm on the nanotube. The sample is then coated with parylene dielectric followed by deposition of a top gate electrode aligned with respect to the nanotubes. The bead acts as a storage node for the memory while the CNFETs operated in the subthreshold regime behave as electrometers with exponential amplification. Operation of devices with retention of single charge quantum is successfully demonstrated at liquid helium temperature. Depending on the nanotube-dot coupling, the transfer of a single electron into the gold dot can lead up to one order of magnitude increase of the CNFET channel current.   

High-frequency characterization of SWNT array FETs

We discuss measurements of the high-frequency response of FET devices made from parallel arrays of SWNTs. Dense parallel arrays are grown by rapid-heating CVD using CO as a feedstock. Interdigitated finger pads are laid down to yield high-current devices. Devices are converted to pure semiconducting behavior by a selective burning technique, and are still able to carry currents in excess of 20 milliamperes. S-parameter testing using a high frequency network analyzer allows for full frequency characterization of the devices.   

Carbon Nanotube Devices for GHz to THz Applications

In this talk I will present an overview of the high-frequency applications of carbon nanotubes, one realization of nano-electronic devices, and where the challenges and opportunities lie in this new field. Specifically, I will first discuss the passive RF circuit models of one-dimensional nanostructures as interconnects[1]. Next, I will discuss circuit models of the ac performance of active 1d transistor structures, leading to the prediction that THz cutoff frequencies should be possible[2]. We recently demonstrated the operation of nanotube transistors at 2.6 GHz[3]. Third, I discuss the radiation properties of 1d wires, which could form antennas linking the nanoworld to the macroworld[4]. This could completely remove the requirements for lithographically defined contacts to nanotube and nanowire devices, one of the greatest unsolved problems in nanotechnology. [1] P.J. Burke "An RF Circuit Model for Carbon Nanotubes" IEEE Transactions on Nanotechnology 2(1), 55-58 (2003). [2] P.J. Burke, ``AC Performance of Nanoelectronics: Towards a Ballistic THz Nanotube Transistor'' Solid State Electronics, 48(10), 1981-1986 (2004). [3] Shengdong Li, Zhen Yu, Sheng-Fen Yeng, W.C. Tang, Peter J. Burke, ``Carbon Nanotube Transistor Operation at 2.6 GHz'' Nano Letters, 4(4), 753-756 (2004). [4] Peter J. Burke, Shengdong Li, Zhen Yu ''Quantitative theory of nanowire and nanotube antenna performance,'' http://xxx.lanl.gov/abs/cond-mat/0408418 (cond-mat/0408418) (2004).   

Exploration of optical and electronic properties of SWNT networks for device applications

Interconnected networks of single-walled carbon nanotubes (SWNTs) are rapidly gaining attention for macroscale electronics applications such as transparent conductive coatings or wiring, as well as printable transistors and sensors. In contrast to individual SWNT devices which must overcome the chirality and geometry variations inherent in most samples, networks of SWNT bundles naturally provide an ensemble average of semiconducting and metallic nanotubes which despite the advantages in processing must, however, be tailored for the particular application. SWNT networks have been deposited from solutions containing purified, laser-vaporization grown nanotubes onto optically transparent substrates for combined electrical and optical characterization. Specially-designed electrode structures were used to simultaneously perform UV-VIS-NIR spectroscopy and electrical characterization. Raman spectroscopy and scanning electron microscopy (SEM) were used to characterize the alignment and morphology of the networks.   

Massive Nano-Assembly Method for Integrated Device Structures Based on Nanotubes and Nanowires

Recent dramatic progress of nanotechnology allows us to combine carbon nanotubes and nanowires with conventional microelectronic devices to build a generation of new nanoscale devices. However, a major stumbling block holding back their industrial applications is a lack of massive assembly method for integrated device fabrication. One promising nano-manufacturing method is the 'surface-programmed assembly' process. In this strategy, surface molecular patterns are utilized to direct the assembly of nanowires onto specific locations of general substrates with precise orientations [1]. This talk will discuss how one can utilize surface-programmed assembly strategy to ‘position’ and ‘align’ a large number of 1D nanostructures (e.g. carbon nanotubes, metal oxide nanowires, etc) on general substrates (e.g. gold, silicon oxide, Al, etc) to build large-scale integrated device structures. Future prospect and possible applications of this strategy also will be discussed. [1] S. Rao, L. Huang, W. Setyawan, and S. Hong Nature 425, 36-37 (2003)   

Session A27: Focus Session: Carbon Nanotubes: Optical Properties I

Optical Spectroscopy of Individual Single-Walled Carbon Nanotubes by Rayleigh Scattering

Optical spectroscopy of \textit{individual} nanostructures has greatly enhanced our understanding of nanoscale physics. For single-wall carbon nanotubes (SWNTs), there is a particularly strong motivation for such techniques, since the properties of SWNTs vary enormously with their precise physical structure. To date, both fluorescence and Raman scattering have shown the sensitivity to probe individual SWNTs. While fluorescence is an excellent experimental method, it is limited to semiconducting nanotubes displaying reasonable fluorescence efficiency. Raman scattering provides complementary information, but is weak and requires the identification of an electronic resonance to observe a signal. In this paper, we describe a new spectroscopic approach for investigating individual SWNTs and other nanostructures.$^{1}$ The method is based on Rayleigh scattering. The approach has the advantage of relying on the ubiquitous linear polarizability of the material, a response present for fluorescing and non-fluorescing species alike and displaying resonances at the transition energies of the system. This method has yielded high-quality spectra over the visible and near-IR spectral range from both individual semiconducting and metallic SWNTs. A key element in the experiment is use of supercontinuum radiation as the light source. This source, produced by passing femtosecond laser pulses through a microstructured fiber, provides radiation with the broad spectrum of a light bulb, but with the brightness of a laser. The experiment also employs SWNTs suspended across slit structures and viewed in a dark-field configuration to eliminate background scattering. Rayleigh scattering spectra of electronic transitions in semiconducting and metallic nanotubes will be presented, as will be results on the polarization dependence of the transitions. The method will be shown to be appropriate for the characterization of different spatial segments of a given SWNT and for the examination of tube-tube interactions in small bundles of SWNTs. This work is supported by the NSF NSEC at Columbia University, NYSTAR, and the DOE-BES. It was performed in collaboration with Feng Wang, Matthew Y. Sfeir, Limin Huang, Chia-Chin Chuang, James C. Hone, Stephen P. O'Brien,$^{ }$and Louis E. Brus. $^{1 }$M. Y. Sfeir, F. Wang, L. Huang, et al., Science \textbf{306}, 1540 (2004).   

Temperature Dependence of Tunable Raman and Fluorescence Spectroscopy of Individual Suspended Carbon Nanotubes

The temperature dependence of tunable Raman spectra are measured on single-wall carbon nanotubes grown by chemical vapor deposition (CVD) over trenches etched in quartz substrates. The suspended portion of the nanotubes exhibit enormous enhancement of the Raman and fluorescence signals relative to the unsuspended portion. Raman and fluorescence spectra are taken on the same individual nanotubes using a tunable Ti:Sapphire laser over the range 750nm-830nm. Both the Raman mode frequencies and the subband transition energies (E$_{ii}$) are observed to shift with temperature. By preparing nanotubes suspended in free space over a trench we eliminate the effects of interaction with the substrate or surfactant molecules, which, as will be shown, can be quite significant. The temperature dependence of the linewidths and resonance windows are also discussed.   

Auger Recombination of Excitons in Single-Walled Carbon Nanotubes

Single-walled carbon nanotubes (SWNTs) as prototypical 1- dimensional systems exhibit enhanced carrier-carrier interactions. As a consequence, one would expect semiconducting SWNTs containing multiple electron-hole pairs to display rapid Auger recombination. We have investigated this issue experimentally by examining the efficiency and temporal evolution of the fluorescence emission from SWNTs after excitation by a femtosecond laser pulse$^{1}$. The behavior as a function of the pump excitation fluence, which controls the initial electron-hole density, reveals the presence of Auger recombination through a decrease in fluorescence efficiency and the emergence of a rapid decay channel when multiple electron-hole pairs are present in a SWNT.$^{2}$ Similar fluence-dependent effects have also recently been reported by Ma et al$^{3}$. Quantitative analysis yields an Auger recombination rate of $\sim $1/ps$^{ }$for just 2 electron-hole pairs in a 400 nm long SWNT. This rapid Auger rate limits the sustainable electron-hole density that can be achieved within a single nanotube. We compare our experimental finding with a theoretical estimate of the Auger rate in SWNTs based on a point-contact interaction model. $^{1 }$F. Wang, et al., Phys. Rev. Lett. \textbf{92}, 177401 (2004). $^{2 }$F. Wang, et al., Phys. Rev. B, in press. $^{3 }$Y. Z. Ma, et al., J. Chem. Phys. \textbf{120}, 3368 (2004).   

Observation of Carbon Nanotube Optical Bistability

We present scanning confocal microscopy studies of suspended carbon nanotubes grown by chemically-assisted vapor deposition across micron-sized apertures. Maps of the photoluminescence emission taken at the E11 peak show unusual `holes' and `rings' with subwavelength spatial features. These features result from abrupt $\sim $20 meV blue shifts in the emission energy due to small changes in the excitation position. Polarization and intensity dependent studies show that this switching behavior depends on the intensity of light absorbed into the nanotube, and additional spatial structure is seen by varying the excitation wavelength. Our findings suggest that the phenomenon represents a true bistability of the E11 transition, and perhaps a many-body effect, as no intermediate emission wavelengths are observed. This work was supported by DARPA/ONR N00015-01-1-0831, NSF DMR 00-79909, SENS, NSF IGERT DGE-0221664 and in part (SP and ATJ) by the Commonwealth of Pennsylvania's Ben Franklin Technology Development Authority through the Nanotechnology Institute. Authors SP and ATJ acknowledge financial support through the Nanotechnology Institute of the Commonwealth of Pennsylvania.   

Stability of Excitons in Carbon Nanotubes under High Laser Excitations

The behavior of excitons and free carriers strongly depends on the carrier density. In 3-D and 2-D solids, increasing the number of carriers reduces the exciton binding energy via screening, and when the average separation reaches the Bohr radius, a transition from an excitonic insulating state to a conducting electron-hole ($e$-$h$) plasma occurs (i.e., the Mott transition). However, the unique nature of 1-D Coulomb interaction may alter this scenario. We have used nondegenerate pump-probe spectroscopy with a widely tunable pump and a white light continuum probe to monitor the behavior of excitons in single-walled carbon nanotubes for different carrier densities. In addition to already known band filling effects, broadening of absorption peaks is identified from complex, spectrally dependent pump-probe signals including both photo-induced absorption and bleaching. From the excitation conditions, we estimate that the average 1-D density of the photo-excited $e$-$h$ pairs is in the order of the Mott density. However, throughout the observed time range, covering both high and low density regimes, the positions of E$_{11}$ absorption peaks are unchanged. This stability of excitons is similar to excitons in GaAs quantum wires and thus indicates a unique property of 1-D solids.   

NIR-luminescence mapping and Raman spectroscopy of single-walled carbon-13 nanotubes

Photoluminescence and Raman scatterings of single-walled carbon nanotubes (SWNTs) synthesized from isotopically-modified ethanol were studied. Using Alcohol catalytic CVD (ACCVD) technique optimized for the efficient production of SWNTs from very small amount of ethanol, SWNTs consisting of carbon-13 isotope (SW$^{13}$CNTs) were synthesized in addition to normal SWNTs consisting of mainly $^{12}$C. The vibrational features of SW$^{13}$CNTs were compared with those of normal SWNTs through NIR-luminescence mapping and Raman spectroscopy. There was almost no change in Raman spectra shape of SW$^{13}$CNTs except for the Raman shift frequency down-shifted as much as square-root of mass ratio 12/13. In addition to Raman spectroscopy, we have mapped the NIR-luminescence of D$_{2}$O-surfactant dispersions of both SW$^{13}$CNTs and SW$^{12}$CNT. By comparing the two maps, luminescence peaks corresponding to electronic transitions with vibrational excitation were identified.   

Low Temperature Photoluminescence of Surfactant-suspended SWNT’s

Photoluminescence (PL) has been a powerful method for the investigation of bulk semiconductors and novel artificial low dimensional semiconductor heterostructures such as quantum wells, quantum wires and quantum dots. Recently bright PL from semiconductor, surfactant-suspended, isolated single walled carbon nanotubes (SWNT's) and from isolated SWNT's grown between silicon oxide pillars have been observed. Low temperature PL measurements in pillar suspended SWNT's have also been reported. In this latter study small blue shifts of the PL peak energies and PL excitation energies, and previously unreported PL peaks have been found. In this work we present results on the low temperature photoluminescence of surfactant-suspended SWNT's. We have found that, upon freezing of the SDS-suspended SWNT's solution, the SWNT's remains strongly luminescent. We perform PL and photoluminescence excitation (PLE) measurements in temperatures ranging from 3.5 to 210K, for excitation energies between 0.8 and 1.75eV and compare the results with the PL and PLE maps obtained for the same suspension at 300K. We found different behavior for the energy shifts for distinct nanotubes families. We observe a blue shift in E11 transition for mod(2n+m,3)=1 nanotubes and a red shift is observed for mod(2n+m,3)=2 nanotubes. We discuss our results in terms of temperature and strain effects in the electronic structure. The brasilian authors aknowledge CNPq, FINEP and FAPEMIG.   

Optical Characterization and Applications of Single Walled Carbon Nanotubes

Recent advances in the dispersion and separation of single walled carbon nanotubes have led to new methods of optical characterization and some novel applications. We find that Raman spectroscopy can be used to probe the aggregation state of single-walled carbon nanotubes in solution or as solids with a range of varying morphologies. Carbon nanotubes experience an orthogonal electronic dispersion when in electrical contact that broadens (from 40 meV to roughly 80 meV) and shifts the interband transition to lower energy (by 60 meV). We show that the magnitude of this shift is dependent on the extent of bundle organization and the inter-nanotube contact area. In the Raman spectrum, aggregation shifts the effective excitation profile and causes peaks to increase or decrease, depending on where the transition lies, relative to the excitation wavelength. The findings are particularly relevant for evaluating nanotube separation processes, where relative peak changes in the Raman spectrum can be confused for selective enrichment. We have also used gel electrophoresis and column chromatography conducted on individually dispersed, ultrasonicated single-walled carbon nanotubes to yield simultaneous separation by tube length and diameter. Electroelution after electrophoresis is shown to produce highly resolved fractions of nanotubes with average lengths between 92 and 435 nm. Separation by diameter is concomitant with length fractionation, and nanotubes that have been cut shortest also possess the greatest relative enrichments of large-diameter species. The relative quantum yield decreases nonlinearly as the nanotube length becomes shorter. These findings enable new applications of nanotubes as sensors and biomarkers. Particularly, molecular detection using near infrared (n-IR) light between 0.9 and 1.3 eV has important biomedical applications because of greater tissue penetration and reduced auto-fluorescent background in thick tissue or whole blood media. Carbon nanotubes have a tunable n-IR emission that responds to changes in the local dielectric function but remains stable to permanent photobleaching. We report the synthesis and successful testing of solution phase, near-infrared sensors, with $\beta$-D-glucose sensing as a model system, using single walled carbon nanotubes that modulate their emission in response to the adsorption of specific biomolecules. New types of non-covalent functionalization using electron withdrawing molecules are shown to provide sites for transferring electrons in and out of the nanotube. We also show two distinct mechanisms of signal transduction -- fluorescence quenching and charge transfer. The results demonstrate new opportunities for nanoparticle optical sensors that operate in strongly absorbing media of relevance to medicine or biology.   

Optical Properties of Carbon Nanotubes in Semiconductor/NT Hybrids

We report on etch processing and optical measurements of hybrid GaAs-Nanotube materials. Samples are created by OMVPE growth over GaAs vicinal substrates coated with single-wall laser-oven carbon nanotubes (SWNTs). We use scanning confocal Raman and photoluminesence microscopy to study the incorporation of SWNTs into the GaAs matrix. We further present procedures for patterning these samples into islands separated by V-grooves and interconnected by suspended SWNTs. Strategies for performing time-resolved studies of spin coherent transport across SWNTs will be discussed. Preliminary results of two-color, time-resolved Faraday rotation in unpatterned hybrid structures reveal unconventional phase shifts in coherent dynamics of spins in the GaAs matrix. JMK acknowledges support from DARPA/ONR N00015-01-1-0831, TFK acknowledges support from ARO, and DEM acknowledges support from NSF IGERT DGE-0221664.   

Polarization dependence of the optical absorption of single-walled carbon nanotubes

Anisotropic optical absorption properties of single-walled carbon nanotubes (SWNTs) are determined from the measurements of a recently developed vertically aligned SWNT film grown on an optically polished quartz substrate. In addition to the inter-subband absorption below 3 eV, we present the remarkable polarization dependences of absorption peaks at 4.5 eV and 5.25 eV. Origins of these absorption peaks are clarified and their important relevance to the optical properties of graphite is revealed. A method of determining a nematic order parameter of the vertically aligned SWNT film by separating the collinear absorption peak at 4.5 eV from other transition dipoles is introduced. Subsequently, the intrinsic optical absorption cross-sections of the SWNTs for 0.5 - 6 eV are determined for both parallel and cross-polarized light. It is shown that the tail of the non-collinear absorption peak at 5.25 eV contributes appreciably to the absorption of cross-polarized light even in the inter-subband transition region below 3 eV.   

Session A28: Polymer Surfaces I

Surface Diffusion of Single Polymer Chain Using Molecular Dynamics Simulation

Results of recent experiments on polymer chains adsorbed from dilute solution at solid-liquid interface show the power scaling law dependence of the chain diffusivity, D, as a function of the degree of polymerization, N, D $\sim $ N $^{-1.5}$. By contrast, for DNA molecules bound to fluid cationic lipid bilayers D $\sim $ N $^{-1}$. We use molecular dynamics simulations to gain an understanding of these scaling behaviors. Our model systems contain chains comprised of N monomers connected by springs, embedded into athermal solvent confined between two solids plates. We will discuss the nature of dynamic adsorption transition and effects of hydrodynamics forces on chain diffusion and scaling exponent.   

Single-Molecule Studies of Polymer Translational Diffusion at Surfaces

Single-polymer diffusion at solid and organic solvent interfaces has been investigated with polydimethylsiloxane (PDMS) using both fluorescence correlation spectroscopy (FCS) and single-molecule imaging (SMI). In contrast to previous related work from this laboratory, here we cover a broad range of concentrations, from the melt state to polymers adsorbed from dilute solution. As the surface coverage increases above the dilute regime, the translational diffusion coefficient as well as the polymer conformations vary with a complex interdependence. Initially diffusion speeds up with increasing surface coverage; this is followed by an obvious jamming process at higher surface coverage. Studies in progress involve not only the dependence on polymer molecular weight and its concentration in the liquid phase, but especially on film thickness when the polymers are confined into molecularly-thin films inside a modified surface forces apparatus.   

Wetting of Heterogeneous Surfaces by Polymer Nanodroplets

The development of microfluidics, micro-contact printing, and other micron scale processes has led to renewed interest in surface wetting at sub-micron length scales. Molecular dynamics simulation is used to study the dynamics of polymer nanodroplets wetting heterogeneous surfaces, specifically cylindrical polymer droplets on surfaces composed of strips oriented perpendicular to the droplets that have either strong or weak interactions with the polymers. Each polymer droplet contains $\sim$200,000 to 350,000 monomers described using the bead-spring model with either 10 or 100 monomers per chain. The droplets are initially placed in contact with a patterned surface at a contact angle near 90$^{\circ}$. As the droplets wet the surface, the spreading dynamics over different regions of the surface is monitored through measurement of the contact angle, contact radius, and velocity profiles. The spreading dynamics are strongly dependent on the wavelength of the strips relative to the polymer chain length.   

Non-Spherical Droplets of Diblock Copolymer: Equilibrium Shape and Spreading Kinetics

Conventional liquid droplets minimize surface energies by acquiring the shape of a spherical cap. In the case of symmetric diblock copolymers microphase separation yields the additional energy constraint of a lamellar microstructure. We present a study of droplets of symmetric polystyrene-b-poly (methyl methacrylate) (PS-b-PMMA), which consist of stacked lamellar disks. Ordered PS-b-PMMA droplets are found to have a non-spherical shape that can be nearly conical under certain conditions. Most significantly, this droplet shape becomes spherical upon passing through the order disorder transition. The near conical equilibrium droplet shape can be understood from a simple model with a repulsive interaction between the lamellar edges in adjacent disks. Furthermore, droplet spreading is found to deviate from Tanner's law in a predictable way.   

Origin of surface ordered phase in poly(n-alkyl acrylates) above the bulk melting temperature (Tm)

We present the first surface tension ($\gamma )$ measurements as a function of temperature above T$_{m}$ for poly(n-alkyl acrylates) to explain the presence of surface ordered phase. The surface tension increases with increase in temperature indicating that the surface molecules have lower entropy than in the bulk. There is an abrupt change in slope of $\gamma $ \textit{vs} T at T$_{s2}$ ($>$T$_{m})$ indicating a first order surface transition. The temperature range of the ordered phase is much larger than that observed for small molecule alkanes and alcohols. We determine that this is primarily due to partial crystallinity within the side chains. The consequences of these results have important implications in similar systems containing chemically attached hydrophobic side chains such as surfactants, dendrimers and biomolecules.   

Functional gradients with controlled steepness on self-assembled aminosilane monolayers

A high throughput, and cost-effective way to fabricate functional gradients of controlled steepness (up to $\sim $10$\mu $m for full gradient) on organic self assembled monolayer (SAM) films has been achieved. The exposure of a aminopropyltriethoxylsilane(APTES) film prepared on a SiOx substrate to ultraviolet (UV) light, with and without the creation of ozone, is controlled by a movable shutter that shadows the sample from the UV source. The shutter to the substrate spacing sets a lower limit to the steepness of the gradient, whereby both the diffusion field of ozone and the divergence of incident UV light are responsible. Through the attachement of 0.26 $\mu $m sized polystyrene (PS, carboxyl-group ended) microspheres (MSs) onto the exposed APTES films via chemical reaction in an activated MS suspension, the fabricated gradient can be visualized directly with a visible light microscope operated in Normarsky interference mode. The short MS density-saturation time observed suggests that covalent bonding is established between the MSs and the APTES film through the reaction of carboxyl- with amine- groups [1]. The use of a linar, variable shutter speed in conjuction with the saturation time allows for the variation and control of the gradients steepness. Reference: [1] S. Herrwerth et al Langmuir 19(2003) 1880-1887   

Measurement of Adhesion Energy and Young’s Modulus in Thin Polymer Films Using a Novel Axi-symmetric Peel Test Geometry

We present a method of probing adhesion between solids, particularly in systems involving polymers. This method uses the axi-symmetric deformation of a thin spincast polymer membrane brought into contact with a film supported on a substrate to measure the work of adhesion between the pair. In this geometry, the contact area and constitutive relation (force versus displacement curve), are measured. This enables the determination of Young's modulus, surface energy, and the pretension of the free-standing film, which are in good agreement with accepted values.   

Changes in the Molecular Orbitals during Photochemical Patterning of Polymers

Patterning of surfaces by EUV irradiation into hydrophilic and -phobic stripes has recently been demonstrated as promising technique for directed self-assembly [1]. In order to better understand this process we have detected the changes in the molecular orbitals at the surface using near edge X-ray absorption fine structure (NEXAFS) sepctroscopy. In particular, the effect of extreme ultra-violet (EUV) radiation on PS-r-MMA is studied. Irradiation causes spectral weight to be transferred from C 1s to pi* transitions of C=C bonds and similar transitions at C=O bonds, which indicates insertion of oxygen into pi-bonded carbon sites. The effect is quantified by dose-dependent studies. [1] Kim, et al. Nature 424, 411 - 414 (2003)   

Water adsorption and desorption from crystalline P(VDF-TrFE) copolymers

Water adsorption and absorption on crystalline poly(vinylidene fluoride -- trifluoroethylene), was examined by thermal desorption spectroscopy. Two distinctly different water adsorption sites are identified: one adsorbed species that resembles ice and another species that interacts more strongly with the polymer thin film. The existence of the latter species is consistent with X-ray diffraction studies of water absorbed into the bulk of copolymers of poly(vinylidene fluoride -- trifluoroethylene) crystalline thin films. There are strong steric effects observed in the angle-resolved thermal desorption that may be a result of the large polymer thin film surface dipoles.   

Polymer size and affinity effects on nanopore adsorption with a wide range of pore sizes

We study polymer nanopore adsorption in solution using nanoporous silica and investigate how these processes differ from those on flat surfaces. In particular, we studied the adsorption of monodisperse polystyrenes onto nanoporous silica with an average pore size ranging from 8 to 100 nm at various solvent quality conditions. We found that the polymer nanopore adsorption phenomena were greatly affected by time, temperature, concentration and solvent quality. In general, the surface excess of polystyrene adsorption exhibits a maximum when the radius of gyration is similar to that of the small pores. However, in large pores, the polystyrene shows the maximum adsorption when the radius of gyration was approximately half the diameter of the pores. This result reveals that the polymer entanglement due to steric crowding at the nanopore entrance on the outer surface has a critical role for controlling diffusion and the subsequent adsorption into the pore.   

Solvent-Assisted Formation of Nanostrand Networks of Supramolecular Diblock Copolymer-Surfactant Complexes at the Air-Water Interface

A block copolymer of majority styrene and minority vinylpyridine (PS-b-P4VP; PS 40K, P4VP 5.6K) mixed with a stoichiometric amount (relative to the VP block) of hydrogen-bonding 3-n-pentadecylphenol (PDP) was investigated at the air-water interface and as transferred films imaged by AFM. By respecting the conventional waiting period to allow solvent evaporation following spreading of the polymer solution on the Langmuir bath and before barrier compression, circular aggregates of variable sizes were obtained at moderate surface pressures. When, instead, the barriers were compressed to the same surface pressures as soon as possible after spreading, a dense and infinite network of nanostrands was obtained. This phenomenon may be attributed to the solvent maintaining the necessary flexibility for the polymer chains to undergo reorganization in response to the change in surface pressure.   

Contact Mechanics Studies with the Quartz Crystal Microbalance

The mechanism of adhesion between two surfaces that are immersed in a liquid medium is a problem of critical scientific and industrial importance. Practical applications range from targeted drug delivery systems to coatings that are designed to resist fouling by marine organisms. However, quantitative measurement of adhesion in liquids is often complicated by difficulties in determining the true nature of the contact between the two surfaces. In some cases a lack of optical contrast makes it difficult to visualize the contact area, whereas in other cases the optically determined contact may not represent a region of true mechanical contact. We have utilized the quartz crystal microbalance (QCM) in contact mechanics experiments because its response is coupled to the surface rheological properties of the materials that are pressed against it. We have shown that when a hemispherical polymer gel is brought into contact with the electrode surface of the QCM, changes in both the resonant frequency and the dissipation are proportional to the gel/QCM contact area. The actual proportionality constants are determined by the high frequency rheological response of the gel. As a result we have been able to calibrate the QCM for use as a highly sensitive contact sensor for fundamental studies of adhesion of polymer gels.   

Quantitative measurement of adhesion of ink on plastic films with a Nano Indenter and a Scanning Probe Microscope

Plastic film packaging is widely used these days, especially in the convenience food industry due to its flexibility, boilability, and microwavability. Almost every package is printed with ink. The adhesion of ink on plastic films merits increasing attention to ensure quality packaging. However, inks and plastic films are polymeric materials with complicated molecular structures. The thickness of the jelly-like ink is only 500nm or less, and the thickness of the soft and flexible film is no more than 50$\mu $m, which make the quantitative measurement of their adhesion very challenging. Up to now, no scientific quantitative measurement method for the adhesion of ink on plastic films has been documented. We have tried a technique, in which a Nano-Indenter and a Scanning Probe Microscope were used to evaluate the adhesion strength of ink deposited on plastic films, quantitatively, as well as examine the configurations of adhesion failure. It was helpful in better understanding the adhesion mechanism, thus giving direction as to how to improve the adhesion.   

Puzzles in Surface Force Apparatus (SFA) Experiments

Since Frantz and Salmeron (1998) found that the surface energy of recleaved mica is 50{\%} higher than that cleaved using conventional cleaving method detailed in Israelachvili et al. (2004), there is concern in the SFA community on how the mica surface preparation influences observations. In addition, mica's surface condition might be responsible for the divergence between experiments and computer simulation in this area. Our group has been working on mica cleaved based on the ``Salmeron method.'' While some differences in results were obtained from mica prepared by different methods, the underlying cause for such discrepancy is not well understood. In addition, it is not clear if mica preparation affects all SFA results performed in different experimental conditions equally. This presentation will give a comparison on results obtained using mica with different cleaving methods. Experimental parameter(s) that renders mica surface condition an important consideration will be discussed.   

Confocal Raman-AFM, A New Tool for Materials Research

Characterization of heterogeneous systems, e.g. polymers, on the nanometer scale continues to grow in importance and to impact key applications in the field of materials science, nanotechnology and catalysis. The development of advanced polymeric materials for such applications requires detailed information about the physical and chemical properties of these materials on the nanometer scale. However, some details about the phase-separation process in polymers are difficult to study with conventional characterization techniques due to the inability of these methods to chemically differentiate materials with good spatial resolution, without damage, staining or preferential solvent washing. The CR-AFM is a breakthrough in microscopy. It combines three measuring techniques in one instrument: a high resolution confocal optical microscope, an extremely sensitive Raman spectroscopy system, and an Atomic Force Microscope. Using this instrument, the high spatial and topographical resolution obtained with an AFM can be directly linked to the chemical information gained by Confocal Raman spectroscopy. To demonstrate the capabilities of this unique combination of measuring techniques, polymer blend films, spin coated on glass substrates, have been characterized. AFM measurements reveal the structural and mechanical properties of the films, whereas Raman spectral images show the chemical composition of the blends.   

Session A29: Charged and Ion-Containing Polymers I

Complexation between flexible polyelectrolytes and oppositely charged particles

The recent theory[M. Muthukumar, J. Chem. Phys. 120, 9343 (2004)] of counterion adsorption on flexible polyelectolyte molecules is extended to complexation of polyelectrolytes and oppositely charged entities such as proteins. The modification of the adsorption isotherm under tension of the polyelectrolyte will also be presented.   

Solvent effects on the dynamics of polyelectrolyte chains near a charged wall: Molecular dynamics simulations with explicit solvent

The effect of solvent quality on the behavior of salt-free dilute and semi-dilute polyelectrolyte chains near a charged wall is studied using molecular dynamics simulation. The polyions are modeled as a chain of charged spherical beads, counter ions to the polyion and the surface are charged spheres, and solvent molecules are uncharged spheres. The wall is atomically smooth with a uniform charge density. For dilute solutions, the chain radius of gyration decreases monotonically with decreasing solvent quality but shows a non-monotonic dependence on surface charge density. As the surface charge density is increased, the polyions orient in a direction parallel to the surface. The diffusion constant of the centre of mass and the characteristic rotational relaxation time of the polyions show explicit dependence on the solvent quality: The diffusion constant increases and the rotational relaxation time decreases as solvent quality is decreased. As the surface charge density is increased, the rotational correlation function does not decay in good solvents, contrary to what is seen in poor solvents. In semidilute solutions the thickness of the adsorbed layer increases as solvent quality is decreased suggesting that solvent effects play an important role in polyelectrolyte adsorption.   

Self-Consistent Field Calculations of Polyelectrolytes on Flat Surfaces

We apply a self-consistent field (SCF) theory to study the behavior of polyelectrolytes (PE) on flat surfaces either electrically neutral or carrying opposite charges to the PE. In the former case, PE are less depleted from a non-adsorbing surface than neutral polymers, due to the presence of counterions near the surface. In the latter case, PE form an adsorption layer near the surface (with no salt added). We then compare the results from full SCF calculations with those obtained under the ground state dominance approximation (GSDA); our results show that the GSDA gives in most cases good quantitative description for PE adsorption on oppositely charged surfaces. Finally, we examine in detail the effects of various parameters on PE adsorption and surface charge compensation by the adsorbed PE, including the charge distribution and degree of ionization of PE, surface charge density, short-range interactions between the surface and polymer, solvent quality, bulk polymer concentration and salt concentration. Our results show that, for PE on oppositely charged surfaces, Coulombic interactions dominate in most cases with no added salt. In such cases, adsorbed PE almost exactly compensate the surface charge; the bulk polymer concentration, solvent quality, and short-range interactions between the surface and polymer have little effects on the amount of adsorbed PE. The added salt plays an important role by screening Coulombic interactions. Further work exploring the importance of fluctuations in the system is undergoing.   

Forces of Interaction between Polyelectrolyte Brushes in the Presence of Multivalent Ions and Cationic Surfactant

The surface forces apparatus has been used to measure the forces between polystyrene sulfonate (PSS) brushes, tethered to mica surfaces via hydrophobic poly-t-butylstyrene anchoring blocks, immersed in media containing various concentrations of di- and tri-valent salt, as well as in micellar solutions of the cationic surfactant, CTAB. All of these media produce strong attractive forces between the PSS brushes under some conditions, but with different patterns for the appearance of these attractions as functions of the concentrations of the added ions. These patterns have been studied in detail experimentally. Generally, attractions appear at low concentrations of added salt or surfactant and can be made to disappear at higher concentrations. The interplay between salt and CTAB has some interesting feature in the case of added cationic surfactant.   

Ion Sensors based on Polyelectrolyte Hydrogels

Attaching a thin polyelectrolyte hydrogel layer to a quartz crystal microbalance (QCM) potentially enables a new class of online ion-sensing devices. The device's operating principal would be conceptually straightforward: ions of a liquid under examination, upon ion exchange with counterions of the hydrogel layer, alter the layer's mass and thus the QCM's resonant frequency. To examine device feasibility, we have grafted thin, chemically crosslinked poly(allylamine) chloride hydrogels to the gold electrodes of a thickness-shear-mode QCM. In otherwise pure water, the resulting devices are easily able to detect nitrate ions at the hundreds of ppm level, displaying sensitivity suitable for drinking water evaluation. The physics of these devices, however, are not well understood. Questions regarding ion selectivity as well as optimal layer thickness and stiffness must be addressed before routine use is considered. Polyelectrolyte and viscoelastic aspects of these questions will be addressed in this presentation.   

Time Resolved Studies of Bundle Formation in Rod-Like Polyelectrolytes

It is known that multivalent ions can generate attractions between like-charged polyelectrolytes in a wide range of systems. We find that sparteine, a chiral divalent cation, can condense polyelectrolytes in a temperature dependent manner. The condensation behavior of the fd virus, a negatively charged polyelectrolyte, can be modulated by tuning the sparteine mediated attraction via temperature. Moreover, we use real-time fluorescence microscopy imaging to study the temporal evolution of bundle formation.   

Computer Simulations of Aggregate Formation and Dynamics in Ionomers

We investigate the structure and the dynamics of aggregate formation in ionomers, through the device of Monte Carlo (MC) and Brownian Dynamics (BD) simulations in the canonical ensemble. We carried out several computer experiments for different temperatures, different chain length and also for different charge states of counterions. Our result shows the formation of aggregates in telechelic ionomers. Pair distribution function for counterions clearly shows cluster formation at low temperatures. At high temperature, for longer chain lengths BD shows clear spherical structure which qualitatively matches with the experimental findings. For longer chain lengths, vesicular structure has been seen which is consistent with experimental findings.   

Langevin Dynamics Simulations of Counterion-mediated Complexation of Polyelectrolytes

We have used the Langevin dynamics simulation to study the complexation of oppositely charged flexible polyelectrolytes in salt free solutions. Two uniformly charged chains with condensed counterions are separated in space initially, and allowed to move and interact with each other. The chain size, Coulomb energy and the release of counterions are monitored. The simulations are carried out at different temperatures and electrostatic interaction strengths. We have found in simulations that while the release of the counterions dominates at lower temperatures, the electrostatic attraction between two chains takes over at higher temperatures.   

Ionic Conductivity at the Ordinary-Extraordinary Transition in Polyelectrolyte Solutions

Experimental results on the individual contributions of the polyion, counterion, and added salt to the bulk conductivity of polyelectrolyte solutions will be presented. Concentration dependence of sodium polystyrenesulfonate (NaPSS) and potassium chloride (KCl) as the added salt was studied with respect to conductivity, viscosity, and dynamic light scattering (DLS) measurements. The bulk conductivity of NaPSS with KCl remains independent of polymer concentration up to the ``ordinary-extraordinary'' transition (C$_{s}$/C$_{p}\sim $1), at which there is a split in the diffusion coefficient into slow and fast modes. As the polymer concentration increases (C$_{s}$/C$_{p} \quad <$1), there is a strong positive dependence of conductivity on polymer concentration. However, the severity of this positive dependence decreases with increasing salt concentration, thus pointing to a shift in the dominating conductive contribution from the counter-ions/added salt to the polyion chain at this transition region. With viscosity and diffusion coefficient measurements backing up conductivity data, it is the first time such a systematic experimental investigation has been done with regard to this issue.   

Comments on Electrostatic Persistence Length

I have shown that the quadratic dependence of the electrostatic persistence length on the Debye screening length obtained in the classical Odijk-Skolnick-Fixman (OSF) theory is a result of incorrect assumption made about the energetic penalty for chain deformation. By including chain elasticity the linear dependence of the electrostatic persistence length on the Debye screening length is obtained. This result is derived by applying simple scaling analysis of the angle fluctuations and Gaussian variational principle to the system of strongly and weakly charged polymer chains.   

Influence of charge density and backbone rigidity on the structure and properties of polyelectrolyte solutions

The influence of chain stiffness on the structure and thermodynamic properties of \textit{neutral polymer solutions }is significant in both the good and poor solvent domains. The conformational properties of a dissolved polymer can undergo large dimensional changes depending on the value of the stiffness parameter ($\eta )$, polymer concentration and temperature. At the same time, the collective behavior can develop long-ranged correlations as the phase boundary is approached. Recent theoretical considerations have suggested that the chain stiffness should also play a key role in determining the structure and thermodynamic properties of polyelectrolyte solutions, where a complex crossover from the second order critical phenomena ($\eta \to $0) to the first order isotropic -- nematic transition ($\eta \to $1) may be observed. The key experimental variable of our research is backbone stiffness and the charge density of polyelectrolyte. For a ionizable group (sulfonate), the effect of chain stiffness can be elucidated through studies of poly(styrene sulfonate) (PSS) as a flexible molecule and poly(cyclohexadiene sulfonate) (PCHDS), with a ``semiflexible'' backbone. We also varied the degree of sulfonation to couple the effect of hydrophobic interactions with backbone stiffness. Small angle neutron scattering (SANS) and zero averaged contrast methods were used to characterize the PCHD backbone and PCHDS polyelectrolyte in solutions.   

Liquid structure of flexible polyelectrolyte solutions

MD simulations and the recently developed RORPA theory are used to examine the liquid structure of flexible polyelectrolyte solutions. Quantitative comparison is made with experiments that show an invariance of the structure factor peak height and wavevector with chain charge fraction f. This invariance has important implications for the design of polyelectrolyte-based materials. Previous theoretical results for rod polymers show that proper inclusion of polymer-polymer density correlations is sufficient to yield this invariance. However, for flexible polyelectrolytes other aspects such as counterion condensation or an explicit solvent seem to be at least partially necessary to produce the experimental trends. Possible causes of the anomalously large low wavevector scattering, i.e., slow mode, seen in many experiments will also be discussed.   

Solid-State NMR Investigations of a Perfluorinated Ionomer (Nafion)

The chain dynamics and supramolecular structure of Nafion$^{\mbox{{\textregistered}}}$, a perfluorinated ionomer which is widely used as a hydrophilic permselective membrane in fuel cells and chloralkali electrolysis, have been studied by solid-state NMR. With 1D and 2D NMR under 30-kHz magic-angle spinning (MAS), the $^{19}$F and $^{13}$C NMR peak widths and positions are determined, which corrects several previous assignments. The peak widths reveal static disorder around the branch point, increasing mobility towards the side group end, and a conformationally ordered backbone, which is essentially polytetrafluoroethylene (PTFE). Fast rotations of the helical backbone segments around their axis are confirmed in PTFE and observed similar in Nafion. The equal $^{19}$F chemical shifts within parallel packed rotating chains in PTFE crystals result in slow $^{19}$F spin diffusion between differently oriented chains. This spin diffusion is observed very fast for a majority of backbone segments in Nafion and the orientational correlations of the remainder backbones are weak. The typical diameter of backbone ``clusters'' in Nafion was 1 - 3 nm estimated by $^{19}$F spin diffusion. Relatively fast $^{19}$F spin exchange from any site in the side group to the backbone is observed. Absorbed water increases the side-group dynamics and conformational averaging, but not the segments near the branch point.   

Small Angle Neutron Scattering (SANS) Study of Perfluorinated Ionomer Membrane under In-situ Vapor Sorption

SANS measurements were made on both solvent cast and extruded perfluorinated ionomer membranes as a function of humidity. The specially designed in-situ vapor sorption apparatus changes the relative humidity rapidly, allowing SANS measurements to follow initial structural changes with sorption time. The two types of membranes show significant structural differences in the position and breadth of their characteristic SANS features even for the same equivalent weight (EW=1100). The structure and diffusion of water vapor into the membrane at room temperature will be discussed.   

Structure of highly rigid ionic polymers from single molecules to membranes

The structure of ionic polymers in solutions and in their solid state is governed by the segregation to hydrophilic and hydrophobic regions. Very rigid backbones with persistence lengths much larger than the size of the monomer limit the segregation affecting the resulting structure and dynamics of the polymers. Using a newly synthesized \textit{para }phenylene based sulfonated polymer with a potential to serve as a polymeric electrolytic membrane for fuel cell applications, we followed the structure of highly rigid ionic polymers from a single molecule to a water swollen membranes using small angle neutron scattering and AFM/TEM techniques. AFM and TEM images show that the dry membranes have domains with a diameter from 30 nm to 70nm. Small angle neutron scattering probes the smaller structure in the membranes from dry to swollen states. Fitting to Teubner-Strey model of SANS data indicates the bi-continuous phases were formed with water and ethanol despite the rigidity of the backbone.   

Session A30: Block Copolymers I

Long-Lived Metastable bcc Phase during Ordering of Micelles

We report a metastable bcc phase that intervenes between a disordered micellar suspension and an fcc crystal in a block copolymer solution. A symmetric poly(styrene-$b$-isoprene) diblock copolymer in the isoprene-selective solvent squalane at a volume fraction of 0.20 was investigated using small angle x-ray scattering and rheology. Upon heating, the metastable bcc phase nucleates first, and then transforms over the course of hours to the stable fcc phase. At still higher temperatures the fcc phase transforms to an equilibrium bcc phase. The metastability of the bcc phase was confirmed by oscillatory shear and annealing using small angle x-ray scattering. These results constitute an interesting experimental manifestation of Ostwald's step rule, and also support recent theory and simulation results whereby bcc nucleates more readily from a melt of spheres.   

Kinetics of BCC-FCC Transition in SI Diblock Copolymer Micelles in a Selective Solvent

Synchrotron based time-resolved small angle x-ray scattering (SAXS), was used to study the kinetics of ordering transition (OOT) in a 0.15(w/v) solution of a polystyrene-polyisoprene diblock (SI 15-15, Mw = 30K) copolymer in $n$-tetradecane, a selective solvent for the Isoprene block. This solution of spherical micelles is known to exhibit both FCC and BCC phases. From a temperature ramp experiment the OOT was identified at about 77C C and an ODT above 100 C. Several temperature jump experiments were performed over the temperature range of 50-110C. A disordered micellar solution at 110 C rapidly quenched to 50 C shows that a BCC phase appears very quickly following the temperature drop, which slowly transforms to the FCC phase at constant temperature. Two- dimensional SAXS patterns reveal the presence of grains. When the sample is melted for a short time the grains reappear producing almost the same scattering pattern as before the melting transition. Detailed analysis of the time evolution of the intensities of the Bragg peaks to follow the kinetics of the growth of the FCC phase will be presented.   

Evolution of disordered micelles to hexagonally packed cylinders in a diblock copolymer studied by X-ray Photon Correlation Spectroscopy

A poly(styrene-\textit{block}-isoprene) diblock copolymer melt was quenched from the disordered state to a temperature where the hexagonally packed cylinder (HEX) phase is stable. During the quench, disordered micelles (DM), which are obtained during the early stages of the phase transition gradually transform into the HEX phase. The dynamics of the evolving system on molecular length scales, was measured by X-ray Photon Correlation Spectroscopy (XPCS). The relaxation of concentration gradients at a given time during the disorder-to-order transition is slowest at the wave vector corresponding to the small angle X-ray scattering peak. The conversion of micelles to cylinders results in an increase in relaxation time. Complementary dynamical data was obtained by conducting time-resolved rheological measurements on the same sample.   

Twinning and Growth Kinetics of Lamellar Grains

The kinetics and grain growth behavior during a thermally induced transition from disorder to lamellae have been determined by polarized optical microscopy (POM). Measurements were made on a poly(styrene-b-isoprene) copolymer, f$_{PS}$ = 0.50, in solution with dioctyl phthalate (DOP), at a 70\% polymer volume fraction. Upon cooling from above the order-disorder transition temperature, four distinct types of grain were observed: ellipsoidal single grains, twinned ellipsoidal grains, 2-fold twinned grains, and spherulites. These grain types cover a range of lamellae orientation. For example, the surface of a 2- fold twinned grain is composed of lamellar edges, whereas the spherulite surface is composed of lamellar planes. The specific grain types that arise give insight into the thermodynamic and kinetic forces governing lamellae ordering. Furthermore, growth front velocities of individual grains were measured after rapid quenches from above T$_{ODT}$. These results were quantitatively compared to the predictions of Goveas and Milner, with good agreement observed. The results will be compared to analogous studies on cylinders and gyroid.   

On elasticity of block-copolymer mesophases with glassy domains

The standard Self-Consistent Field Theory of equilibrium self-assembly in copolymer melts is modified to describe effects of glassiness. The glassy domains are modeled by incorporating a variety of structural constraints on chain degrees of freedom. Effects of inter-domain bridging, compressibility, chain pullout and micro-domain deformation on linear and non-linear elasticity are presented. In particular, we study lamellae-forming multi-block copolymers with interleaving glassy and semi-crystalline domains. We predict coexistence between essentially undeformed and highly expanded (micro-cavitating) domains upon sample deformation. Our results are compared to recent experiments on poly(cyclohexylethylene) and poly(ethylene) multi-block copolymer systems.   

The Fddd Network Phase in Triblock and Diblock Copolymer Melts

The phase behavior of ABC triblock and AB diblock copolymer melts has been investigated by self-consistent field theory (SCFT), while allowing for (among other candidates) the orthorhombic Fddd ($O^{70}$) network phase identified in recent experiments with poly(isoprene-b-styrene-b-ethylene oxide) (ISO) triblocks. Predicted phase diagrams for triblocks with interaction parameters similar to those of ISO contain an $O^{70}$ phase bordered by gyroid, lamellar, and alternating gyroid phases, in agreement with experiment. The $O^{70}$ network is also found to be stable in diblock melts within a narrow region that overlaps the weak segregation end of the gyroid region found in previous calculations. A previous hint of the existence of an unidentified phase in this part of the diblock phase diagram was given in the work of A.-C. Shi and coworkers, who found the gyroid phase to be locally unstable with respect to composition fluctuations in the region of parameter space in which we find the $O^{70}$ network to be preferred over the gyroid.   

Order-disorder transition in 2-D sphere forming diblock copolymers

The order disorder transition corresponding to the two dimensional hexagonal patterns formed by sphere forming diblock copolymers is studied through the Cahn-Hilliard model by taking into account a long range term in the free energy functional. The growth kinetic of the equilibrium patterns at deep quenches is dominated by spinodal decomposition. At shallow quenches the phase separation is a two-steps process, initiated by a long lived intermediate state with a well defined characteristic length, but ill defined symmetry. The mean-life of this state increases with a power law with the reduced temperature as the order disorder transition is approached. At long times the system evolutes towards equilibrium by nucleation and growth.   

Determination of Order-Disorder Transition of Polystyrene-block-poly(n-pentyl methacrylate) Copolymer by Temperature-dependent FTIR Spectroscopy

The lower disorder-to-order transition (LDOT) and the upper order-to-disorder transition (UODT) temperatures of polystyrene-block-poly(n-pentyl methacrylate) (PS-PnPMA) were measured by temperature-dependent Fourier transform infrared (FTIR) spectra with principal component analysis (PCA) and two-dimensional (2D) correlation spectroscopy. These two transitions are determined from sudden changes of the intensity (A) at specific wavelength as a function of temperature. We found that when the first derivative of A with respective to temperature (dA/dT) is plotted against temperature, the maximum in dA/dT at all wavelengths was observed at these two transitions.   

Phase Behavior of Poly(styrene-b-isoprene) Diblock Copolymers Loaded with $\gamma$-Fe$_{2}$O$_{3}$ Nanoparticles

We investigate the effect of hard additives, i.e., magnetic nanoparticles, on the phase behavior of polystyrene-block- polyisoprene (PS-b-PI) diblock copolymers by varing the size of nanoparticles (6 nm and 14 nm). For the design of multicomponent materials with spatially defined order of different components, two PS-b-PI diblock copolymers showing lamellar (SI1) and cylindrical (SI2) morphologies are used as structure-directing matrices for the nanoparticle arrangement. Fine maghemite ($\gamma$-Fe$_{2}$O$_{3}$) particles with surfaces modified by oleic acid have been synthesized and two different solvents, hexane and toluene, were used to prepare film specimens by static casting. The interactions between mesophase-forming copolymers and nanoscopic particles can lead to highly organized hybrid materials. Notably, the morphology of such composites strongly depends on the preparation conditions as well as the characteristics of templating copolymers. The $\gamma$-Fe$_{2}$O$_ {3}$ were selectively incorporated into the PI domains under hexane condition, while they were preferentially aggregated when toluene is used. Particularly, under toluene condition, we observed the well-defined body centered cubic structure for SI2 as well as the undulating lamellar morphology for SI1. The structural information obtained from X-ray scattering is in good agreement with the transmission electron microscopy images.   

Nanocellular formation of supercritical CO$_2$ in block copolymer thin films

The production of large-area structured surfaces with a featured size of nanometer scale still remains a challenge by conventional photolithography. In particular, block copolymer thin films have been considered as the ideal lithography templates. Here we present our novel supercritical carbon dioxide (scCO$_2$) process to fabricate thin films with a single layer of empty cells of a diameter of ca. 30 nm in block copolymer thin films. By absorbing CO$_2$ in CO$_2$-philic block domains of a block copolymer followed by depressurization, empty cells are introduced in CO$_2$-philic domains. This process was successful even in a block copolymer thin films with a thickness less than 100 nm. The typical nanocellular structures introduced by our novel scCO$_2$ process (10MPa) have an average spacing of 34 nm and a density of 9 $\times$ 10$^{10}$ cm$^{-2}$. The size and the spacing of such nanocells can be adjusted by changing saturation pressure of scCO$_2$. The obtained structures are significantly different from those expected from the volume ratio of domains swollen by CO$_2$. Spherical cells in block copolymer thin films are found even when the porosity is more than 30\%.   

Phase Transitions and Spatial Organization in Nanoparticle-Block Copolymer Mixtures

Introducing nanoparticles into nanostructured block copolymer phases can dramatically influence the polymer host. Computer simulations [Balazs et al., PRL, 89, 155503 (2002)] suggest inclusions can actually trigger transitions from one polymer phase to another. Simultaneously, the nanoparticles can be organized into complex superstructures, giving composite materials with novel mechanical, electrical and optical properties. Potential applications include catalysts, selective membranes and optical filters. We have developed a first principles theory predicting polymer phases and nanoparticle distributions. We find modification by nanoinclusions of the free energy of stretched polymer domains in lamellar, cylindrical or spherical geometries triggers structural changes and determines particle distributions. Our framework builds on Semenov's description of AB copolymers in the strongly stretched limit by incorporating nanoinclusions. Energy favors segregation into particle-rich regions, while entropy favors particle mixing into the energetically preferred block, say A. When entropy wins (small particles), an A-core cylindrical-to-lamellar phase transition is induced. Interestingly, large particles by contrast microphase separate into copolymer domains (analogous to Semenov's conclusions for homopolymer-copolymer mixtures) triggering reverse phase transitions (e.g. lamellar to A-core cylindrical). We present our results as a complete phase diagram.   

Elastic Properties of Ordered Block Copolymer / Nanoparticle Composites

A hybrid self-consistent field/density functional theory has been used to predict the elastic moduli of a diblock copolymer melt with added spherical nanoparticles with an affinity for one block. It was found that the addition of nanoparticles lowers the tensile modulus of the composite material, while the shear modulus was unaffected. Explanations for the physical origins of these results will be given.   

On the influence of temperature and volume fraction on liquid crystalline block copolymer nanoscale architectures

Liquid crystalline block copolymers (LCBCs) form complex hierarchical structures. We report the phase structures of a series of poly(styrene-\textit{block}-(2,5-bis-(4-methoxyphenyl)oxycarbonyl)styrene) (PS-b-PMPCS) rod-coil diblock copolymers based on the results obtained from thermal analysis, x-ray analysis and transmission electron microscopy. The PS-b-PMPCS system formed lamellar structures of alternating PS and PMPCS domains. Each PMPCS domain contained a bilayered rod-like structure whose axis is parallel to the lamellar normal. In low MW BCs, a S$_{Ad}$-like interdigitated metastable phase was observed which changed into a bilayered structure upon heating. As the PS content increased, the LC layer was gradually punctuated by PS and a perforated layer structure was observed. The ``degree of perforation'' depends on the LC volume fraction.   

The Effect of Segregation Strength on Network Formation in ABC Triblocks

The effects of segregation strength on network phases were investigated in linear poly(isoprene-$b$-styrene-$b$-ethylene oxide) triblocks with various molecular weights. Morphological behavior at higher molecular weights indicated that network long-range order decreased as the polymer molecular weight increased. The signature Q$^{230}$ and O$^{70}$ X-ray scattering patterns were retained in the lowest molecular weight specimens, while the highest molecular weight data were ambiguous, displaying broad peaks at approximately q* and 2q*. TEM results on these materials showed network-like structures with reduced long-range order. It is unclear whether the highest molecular weight structures represent poorly ordered versions of equilibrium networks or kinetically trapped metastable states. Interestingly, this effect was specific to triply-periodic structures, as lamellar samples of comparable molecular weights displayed excellent long-range order. The reduced organization of the networks likely arises from a decrease in coordinated chain motion as a result of the different diffusion mechanisms available to lamellar versus triply-periodic microstructures.   

Stability of Core-Shell-Cylinder Structure of Poly(styrene-b-1,3-cyclohexadiene) Diblock Copolymers

A new free energy model for rod-coil block copolymer systems is proposed in which the distortion splay energy of continuum elasticity theory is incorporated into the rod domain energies. The model is applied to explain the stability of a core-shell-cylinder morphology which was observed by Gido and coworkers for diblock copolymers of polystyrene (PS) and poly-(1,3-cyclohexadiene) (PCHD). Based on its Kuhn length, the PCHD block can be treated as a rod. The model suggests that the core-shell-cylinder structure is energetically favorable for this system compared to a solid cylinder structure.   

LDOT Diblock Copolymers: Specific Interactions, Compressibility, and Fluctuations

We perform the recently developed compressible Hartree analyses on the phase behavior of LDOT-type diblock copolymers based on polystyrene and homologous poly(n-alkyl methacrylates).~ Specific interactions between dissimilar monomers, compressibility difference between block components, and concentration fluctuations are all considered to illuminate the complicated phase behavior of those homologous copolymers.~ Microphase transition temperatures and their pressure dependence are compared with the corresponding theoretical values for the copolymers.~ We provide suggestions for the design of a better nanostructured material using those copolymers.   

Session A31: Molecular Motion in Miscible Blends

Dynamics in Miscible Blends: Recent Results and Open Questions

The study of polymer blends has been a very active field in polymer physics during the past 20 years. However, many questions still remain open. From the point of view of the polymer blend dynamics, the so-called dynamic miscibility, i.e., the question concerning how the dynamics of each component is modified in the blend, has been deeply investigated. In an ideal two-component miscible polymer blend one could expect a completely homogeneous dynamic behavior, in the meaning that the time scale as well as the relaxation function of each component becomes similar in the blend. However, miscible polymer blends are in general dynamically heterogeneous. This heterogeneity manifests in two ways. On the one hand, when the segmental dynamics (alpha-relaxation) of a single component of a blend is selectively investigated, it is found that the response extends over a very broad time/frequency range, in particular in the vicinity of the glass transition temperature. On the other hand, distinctly different local segmental mobilities for the two blend components are observed even at temperatures well above the glass transition. Whether these two manifestations have the same microscopic origin or they put in evidence two different aspects of the polymer blend dynamics is still controversial. Two different but perhaps complementary concepts have been used to describe the above-described phenomenology: thermal concentration fluctuations and the so-called self-concentration. However, a complete theoretical description still remains elusive. The main goal of this talk is to present an overview of the current status on this topic. New results - not only from relaxation techniques but also from neutron scattering and molecular dynamics simulation as well - will be also presented and discussed.   

Component Dynamics in Miscible Blends of PEO and PMMA

We investigate component dynamics in miscible blends of PEO and PMMA as determined from quasielastic neutron scattering combined with deuterium labeling. This blend has many unusual features. The glass transition temperatures of the two components are widely separated leading to extreme differences in the timescales of component mobility. This, in addition to the weak interactions in this system, makes it more likely than other blends to be influenced by concentration fluctuations. The chain connectivity picture does not appear to provide an adequate description of this blend for those cases in which it has been tested, and the small relaxation times of PEO have interesting consequences in the coupling model. We present results for the segmental dynamics of both components, in blends of 10-30 percent PEO by weight, the composition range where PEO remains amorphous. Incoherent measurements, which provide an estimate of self mobility, are discussed within the framework of each model, using molecular simulation to provide additional information where needed.   

Modelling the Segmental Relaxation Time Distribution of Miscible Polymer Blends

In considering the relaxation of segments in a blend whose components have reasonably disparate glass transition temperatures, local concentration fluctuations and density fluctuations each play a role. The result is a distribution of environments around a given segment in the blend, which translates into a distribution of segmental relaxation times. In this work we focus on concentration fluctuations, making use of a simple lattice model to generate a distribution of environments which we then translate into a dielectric relaxation spectrum. We analyze experimental data for several polymer blends and show that, by accounting for the relatively strong composition dependence of the blend glass transition temperature, it is possible to model the dielectric relaxation spectrum using a Kuhn length which is both composition and temperature independent.   

Miscible Polyisoprene/Polystyrene Blends: An Unusual Combination of Heterogeneous Segmental Dynamics and Homogeneous Diffusion

The segmental and terminal dynamics of each component in miscible blends of polyisoprene (PI) and polystyrene (d$_{3}$PS) were characterized over a wide temperature and composition range. Though the system has a large positive interaction parameter \textit{$\chi $} up to 0.15, it is miscible in the temperature range of study due to selected low molecular weight. C-13 and H-2 NMR relaxation measurements were performed to extract the segmental dynamics. Pulse-gradient spin-echo NMR was used to determine the diffusion coefficients. Though the segmental dynamics of PI and PS components differ by more than 2 decades at $T_{g}$+50 K, their terminal dynamics (monomeric friction) are identical. We know of no other system with zero to positive \textit{$\chi $ }showing this feature. The distinct component segmental dynamics can be reasonably interpreted by the Lodge/McLeish model. The unusual homogeneous terminal dynamics are most likely due to a large thermodynamic barrier to diffusion.   

A molecular picture: How composition influences the dynamic and static properties in a polyolefin blend, as observed with molecular simulation

We use molecular dynamics simulation to investigate dynamic and static properties in a blend of poly(ethylene-propylene) [PEP] and poly(ethylene-butene) [PEB]: this is a simple model for blend dynamics because the mixture behaves athermally and each component has similar pure packing characteristics and glass transition temperatures. The use of simulation allows us to examine a full spectrum of compositions, ranging from the dilute (single chain) to concentrated limits (all but one chain). As composition is varied, mobility is observed through the self-intermediate scattering function, while the pair distribution function and local concentrations are used to examine static features. Attention is given to both average values and the distribution within the average. Despite the simplicity of this system, the influence of composition varies between the two components, most noticeable in the dilute region. Molecular packing and concentrations on a local length scale are investigated as a possible source for this variation.   

Interdiffusion in a Polydisperse Polymer Blend

We present a theoretical description of interdiffusion in a binary blend of polymers that exhibit polydispersity in length. The diffusion equations are formulated in terms of the volume fractions and the chain concentrations of the components. This choice of variables is equivalent to the assumption that the local molecular weigh distributions of the components are described by the Flory distribution. The Onsager kinetic coefficients are obtained based on the Green-Kubo equation and correspond to the fast-mode interdiffusion theory. As demonstrated by numerical simulations, the resulting equations describe the simultaneous processes of the evolution of blend composition and the relaxation of the local molecular weight distributions of the components. The developed approach can be used to study polymer systems in which the degree of polymerization changes due to interfacial or bulk chemical reactions.   

A Molecular Dynamics Simulation Study of the Alpha- and Beta-Relaxation Processes in a Realistic Model Polymer

Molecular dynamics simulations of a melt of freely rotating chains of 1,4-polybutadiene (FRC-PBD) have been performed over a wide range of temperature. Removal of the dihedral barriers in FRC-PBD allows for complete resolution of the Johari-Goldstein $\beta $-process from the primary $\alpha $-process in the simulation time window. We find that relaxation in the $\beta $-regime occurs as the result of large-angle excursions of all backbone dihedrals that are largely decoupled from the dynamics of the polymer matrix, while the $\alpha $-relaxation exhibits strong coupling between matrix motion and polymer dihedral relaxation.   

Entropy Theory of Polymer Glass-Formation Revisted

Considerable physical evidence supports the idea of Gibbs and DiMarzio that glass-formation arises when the configurational entropy of a liquid becomes critically small and the subsequent arguments by Adam and Gibbs (AG) that quantitatively relate the configurational entropy to the rate of long wavelength structural relaxation. We revisit this classical `entropy theory' of glass-formation, based on a minimal lattice model that incorporates monomer structure and the different rigidities of the polymer chain backbone and side-groups into a thermodynamic description of compressible polymer melts with van der Waals interactions. Previous observations of an apparent breakdown of the AG theory at elevated temperatures (20- 30 K above the glass-transition temperature Tg) led us to identify the `configurational entropy' of the AG model with the site rather than mass normalized configurational entropy. This reinterpretation of the entropy theory has little effect near the glass transition, but it completely revises the theory at more elevated temperatures. In particular, we predict a series of characteristic temperatures describing respectively, the onset of a drop in s upon cooling, an inflection point in sT and the extrapolation of s to 0. The T-dependence of s is quite distinct above and below the inflection point `crossover temperature'. We complete our specification of the characteristic temperatures of glass-formation by defining kinetic transition temperature Tg through a Lindemann criterion.   

Correlation between static and dynamic heterogenities in polymer mixtures

Computer simulations cannot only address average properties of the system under study but also the distribution over any observable of interest. Here we are using this advantage to study mixtures of polystyrene and polyisoprene by atomistic molecular dynamics and calculate correlation times for all segments in the system. We then identify fast and slow segments. We are able to correlate the segment speed with the local neighborhood and obtain that fast segments have a surplus of the faster component in their neighborhood and vice versa [1]. We are additionally studying other influences on the dynamics such as end effects. As these studies are performed on a mixture with strongly different glass transition temperatures, we are able to study the behavior in a a temperature range where one constituent would be a glass whereas the other one a melt. [1] R. Faller Macromolecules 37 (2004) 1095   

Session A32: Focus Session: Novel Computational Algorithms: From Materials to the Universe I

The Geometric Cluster Algorithm: Rejection-Free Monte Carlo Simulation of Complex Fluids

The study of complex fluids is an area of intense research activity, in which exciting and counter-intuitive behavior continue to be uncovered. Ironically, one of the very factors responsible for such interesting properties, namely the presence of multiple relevant time and length scales, often greatly complicates accurate theoretical calculations and computer simulations that could explain the observations. We have recently developed a new Monte Carlo simulation method\footnote{J. Liu and E. Luijten, Phys.\ Rev.\ Lett.\textbf{92}, 035504 (2004); see also Physics Today, March 2004, pp.\ 25--27.} that overcomes this problem for several classes of complex fluids. Our approach can accelerate simulations by orders of magnitude by introducing nonlocal, collective moves of the constituents. Strikingly, these cluster Monte Carlo moves are proposed in such a manner that the algorithm is rejection-free. The identification of the clusters is based upon geometric symmetries and can be considered as the off-latice generalization of the widely-used Swendsen--Wang and Wolff algorithms for lattice spin models. While phrased originally for complex fluids that are governed by the Boltzmann distribution, the geometric cluster algorithm can be used to efficiently sample configurations from an arbitrary underlying distribution function and may thus be applied in a variety of other areas. In addition, I will briefly discuss various extensions of the original algorithm, including methods to influence the size of the clusters that are generated and ways to introduce density fluctuations.   

Efficient Cluster Algorithm for Resistively Shunted Josephson Junctions

We present a cluster algorithm for resistively shunted Josephson junctions which dramatically improves sampling efficiency. The algorithm combines local updates in Fourier space with rejection-free cluster updates which exploit the symmetries of the Josephson coupling energy. As an application, we consider the superconductor-to-insulator transition in a single junction and the phase diagram of a recently proposed two-junction model with charge relaxation.   

Schrodinger eigenstates for surface-confinement problems

The theory of a quantum-mechanical particle confined to a surface is described using differential geometry arguments including the simplification of the three-dimensional Schr\"{o}dinger problem into three ordinary differential equations in curved coordinates for the case of an arbitrary surface of revolution. These equations are solved - in terms of eigenvalues and eigenstates - either completely analytically or by use of a simple one-dimensional finite-difference scheme for the cases of a cylinder, a cone, an elliptic torus, a sinusoidal-shaped surface of revolution, and a catenoid. A comparison with an exact three-dimensional treatment of the hollow cylinder problem shows that the surface-confinement approximation (corresponding to assuming zero thickness of the particle domain perpendicular to the surface) is excellent in cases where the (hollow) cylinder thickness is less than approximately 10{\%} of the cylinder radius, hence justifying the rationale in employing a similar analysis for the other (above-mentioned) more complicated surface-confinement problems. Symmetry properties of the various eigenstates are finally discussed and compared.   

Global optimization in surface structure determination by electron diffraction using generalized pattern search methods

Low energy electron diffraction (LEED) is the most commonly used method for detailed surface structure determination. This method can be formulated as an inverse problem, by attempting to fit dynamically calculated LEED intensities to experimental data. As with any such method, it faces a challenging global optimization. We discuss the use of generalized pattern search (GPS) methods for this global optimization, using the complex Ni(001)-(5x5)-Li structure as an example. We present numerical results for one particular GPS method (NOMAD) and compare its performance to previously used genetic algorithm methods.   

Reciprocal space approach to finite size error in many-body simulations

A scheme for the correction of the finite size error in the potential energy occuring in the quantum Monte Carlo simulation of the bulk materials is presented. It is based on the fact that the potential energy can be written as a sum over the static structure factor, S(k), and on the assumption that S(k) does not depend on the simulation cell size. The error in the potential energy is then an integration error and corrected by an improved integration scheme. This also leads to an understanding of the scaling of the error with system size. Applications to the electron gas and to a novel nitrogen structure are presented.   

Quantum spin glass and Sourlas codes

Quantum version of Sourlas error-correcting codes are investigated from statistical mechanical point of view. Our problems are equivalent to those of quantum Ising spin glasses with $p$-body interactions. According to Ruj$\acute{\rm a}$n, we assume that information of the correct bits should be obtained from the equilibrium states of the Hamiltonian and the performance of the decoding results is estimated by the overlap between the original information bit and the sign of the local magnetization. We introduce the transverse field as a quantum fluctuation into the Hamiltonian and adjust this to the optimal value so that the overlap takes its maximum. At low temperature and small transverse field, we find analytically that the retrieval quality is dramatically improved. This analytical results are supported by quantum Monte Carlo simulations.   

Comparison of energy minimization and variance minimization methods for optimizing variational parameters in many body wave functions

The variance minimization method has become the standard method for optimizing many body wave functions for quantum Monte Carlo because it is far more efficient that performing a straightforward energy minimization. We have modified two recent energy minimization methods to make energy minimization highly efficient. First, we have modified the straightforward Newton method used by Lin, Zhang and Rappe, \emph{J. Chem. Phys.}. \textbf{112}, 2659 (2000) to reduce the statistical fluctuations by more than two orders of magnitude. Tests on a flexible Jastrow that includes 3-body electron-electron-nucleus correlation terms show that it is very efficient. Second, we have extended the generalized eigenvalue method of Nightingale and Melik-Alaverdian \emph{Phys. Rev. Lett}. \textbf{87}, 043401 (2001), for linear parameters, to nonlinear parameters, and are currently testing this method.   

Monte Carlo Summation Technique

The extended Hubbard model is a frequently used model to describe strongly correlated electron systems. The Green's function of this model can be calculated by perturbation theory associated with Feynman diagrams. In order to fully understand this model, larger lattices and perturbation expansions to higher order are essential. However, the current computing power limits both the sizes of physical systems and the maximum orders of perturbation expansions by brute force summation. In order to overcome this obstacle, we have developed a novel technique to do the summation by a Monte Carlo algorithm. We have applied this technique to the 2-D extended Hubbard Model in momentum space, and computed its Green's function $G(k)$ and self-energy $\Sigma(k)$ by a \emph{self-consistent} algorithm, combined with the corresponding \emph{irreducible} Feynman diagrams. Results for the (nearly) half-filled band case close to the Mott-Hubbard transition will be discussed. \\ \\ $^*$This research was supported by NSF Grant DMR-0081789   

Simple geometry optimization with Variational Quantum Monte Carlo method

Stochastic optimization methods may be combined with Quantum Monte Carlo (QMC) integration to obtain a computational scheme for treating many body wavefunctions suitable for addressing modern problems in nanoscale physics. In this connection, we are investigating the range of applicability of the Stochastic Gradient Approximation (SGA) technique [1]. The SGA possesses the important advantage that the updating of the electronic variational parameters and the nuclear coordinates can be carried out simultaneously and without an explicit determination of the total energy for each geometry. We present illustrative results using simple variational functions for describing the hydrogen molecule, the lithium dimer, and the neutral and charged Li$_4$ clusters. We computed highly accurate potential energy surfaces on a fine grid in order to test the efficacy of the SGA in locating the energy minima in the parameter space. Work supported in part by the USDOE.\\ $\mbox{[1]}$ A. Harju, B. Barbiellini, S. Siljam\"aki, R.M. Nieminen, and G. Ortiz, Phys. Rev. Lett. {\bf 79}, 1173 (1997).   

Accuracy and applicability of the finite temperature quasicontinuum method

The quasicontinuum method is a mixed continuum and atomistic approach for simulating the mechanical response of polycrystalline materials. It allows large-scale atomistic calculations to be performed on moderately small computers. This method was rencently extended to study the behavior of defects at finite temperature. In this talk, we focus on the accuracy and applicability of this method. Possible shortcomings such as mesh-dependence and ghost-forces are discussed.   

Electronic structure code based on existing parallel AMR infrastructure

Following the first developments in real-space methods for electronic structure calculations, various efforts have been carried out in the past ten years to improve efficiency using local adaptive mesh refinement (AMR), in particular for finite physical systems. So far, the complexity of AMR codes has limited these efforts to serial calculations of relatively small problems. One way of overcoming this barrier is to build a code based on an existing parallel AMR infrastructure. We will report our progress in developing a parallel Finite Element electronic structure code based on the C++ SAMRAI (Structured Adaptive Mesh Refinement Application Infrastructure) library developed at Lawrence Livermore National Laboratory (www.llnl.gov/casc/SAMRAI).   

Fine-Lattice Discretization for Fluid Simulations: Convergence of Critical Parameters.

In simulating continuum fluids with long-range interactions, such as plasmas and electrolytes, that undergo phase separation and criticality, it is computationally advantageous to confine the particles to the sites of a lattice of fine spacing, $a_{0} $, relative to their size, $a$.$^{1,2}$ But, how does the discretization parameter, $\zeta\equiv a/a_{0}$ (typically,$^ {1} \geq 5$) affect the values of the critical temperature and density, etc.? A heuristic argument,$^{2}$ essentially exact in $d=1$ and $2$ dimensions, shows that for models with hard-core potentials, both $T_{c}(\zeta)$ and $\rho_{c}(\zeta)$ converge to their continuum limits as $1/\zeta^{(d+1)/2}$ for $d\leq 3$ when $\zeta\rightarrow\infty$. However, the behavior of the error for $d\geq 2$ (related to a classical problem in number theory) is highly erratic. Exact results for $d=1$ illuminate the issues and reveal that optimal choices for $\zeta$ can improve the rate of convergence by factors of $1/\zeta$.$^{2}$ For $d\geq 2$, the convergence of the {\em second virial coefficients} to their continuum values exhibit similar erratic behavior which transfers to $T_{c}$ and $\rho_{c}$. This can be used in to enhance extrapolation to $\zeta\rightarrow\infty$. Data for the hard-core or {\em restricted primitive model} electrolyte have thereby been used to establish that (contrary to recent suggestions) the criticality is of Ising-type --- as against classical, XY, etc.\\ 1. Y.\ C.\ Kim and M.\ E.\ Fisher, Phys.\ Rev.\ Lett.\ {\bf 92}, 185703 (2004).\\ 2. S.\ Moghaddam, Y.\ C.\ Kim and M.\ E.\ Fisher, J.\ Phys.\ Chem.\ B (2005) $~~~~$[in press].   

Thermodynamically accurate particle-based mesodynamics

Particle-based mesoscopic approaches, where groups of atoms are represented by a single \textit{mesoparticle}, are widely used to achieve length- and time-scales beyond what is possible with atomistic modeling. I will present \textbf{a new mesodynamical approach that describes the energy exchange between mesoparticles and their internal degrees of freedom} in a thermodynamically accurate way. In our approach, energy exchange is done through particle coordinates, rather than momenta, resulting in Galilean invariant equations of motion; the total linear momentum as well as total energy (including the internal energy of the mesoparticles) are conserved and no coupling occurs when a mesoparticle is in free flight.The parameters entering our mesodynamics are easily obtained from first-principles and its results are in excellent agreement with all-atom simulations. Furthermore, our approach enables for a quantum mechanical description of the thermal properties of the implicit degrees of freedom (all-atom MD is always classical) and is generally applicable to many problems of materials science, chemistry, and biology.   

Session A34: Environmental Interfaces I

Creating Beakers without Walls: Formation of Deeply-Supercooled Binary Liquid Solutions from Nanoscale Amorphous Solid Films

Supercooled liquids are metastable and their lifetimes are dictated by the kinetics for crystallization. Traditional experimental studies have used a variety of methods to suppress crystallization while cooling from the liquid phase. An alternate approach is to heat an amorphous solid above its glass transition temperature, Tg, whereupon it transforms into a deeply-supercooled liquid prior to crystallization. We employ molecular beams, programmed desorption (both TPD and isothermal) and FTIR vibrational spectroscopy to synthesize and characterize compositionally tailored nanoscale films of glassy methanol and ethanol. We demonstrate that these films exhibit complete diffusive intermixing and suppressed crystallization when heated above Tg. Furthermore, the resulting container-less liquids evaporate as continuously mixed ideal binary solutions while retaining their solid-like macroscopic shapes. This approach provides a new method for preparing deeply-supercooled liquid solutions in metastable regions of their phase diagram and for studying the kinetics of their phase separation and crystallization as they approach thermodynamic equilibrium. The applicability of this technique for studying aqueous liquid solutions will also be presented and discussed.   

Growth-Melt Asymmetry in Ice

It is well known that the Wulff Shape of a crystal is anisotropic at temperatures below the roughening temperatures for the principal facets. In the case of ice in contact with water, the roughening temperature of the prism facet was found by Maruyama to be -16.5 deg C, whereas the basal plane is molecularly smooth up to the bulk transition. Therefore, {\it growth} of ice in this range of temperature is anisotropic. We study the anisotropy during {\it melting} under small disequilibrium melt drives and extend geometric theory to explain the apparent reorientation of the underlying crystallographic axes observed experimentally. The dynamical melt shapes appear faceted despite the lack of a surface phase transition and we explain the behavior using a geometric treatmen of the basic kinetics of molecular detachment from the surface.   

Surface structure of liquids and salt solutions: wetting, droplets and monolayers

The structure of water surfaces and films on solid substrates is one of the most important and relevant topics in interfacial science. Surprisingly, at the start of the 21$^{st}$ century we do not have yet a detailed knowledge of such basic problems as the thickness of a water film on a surface in equilibrium with vapor or humidity. Even less is known about the structure of this film and the molecular arrangement of water molecules at the surface and in the layers close to it. In this presentation I will review recent results in my laboratory regarding water films, both in ambient conditions (near atmospheric) as well as in vacuum using techniques such as AFM, STM, and Photoemission.   

Effect of Antifreeze Glycoprotein in contact with ice interface on the growth mechanism of an ice crystal

We study the effect of Antifreeze Glycoprotein in contact with ice interface on pattern formation of an ice crystal growing from AFGP solution. AFGP effects on ice crystal growth are completely opposite for basal and prismatic faces. Basal face of ice in pure water is governed by slow molecular rearrangements on the basal plane and is expressed as a second power of the supercooling at the interface. In the presence of AFGP molecules on the surface, the kinetic roughening transition from a smooth surface to a rough one occurs, and the growth rate is enhanced. Prismatic faces in pure water are controlled by transport of latent heat and are proportional to the supercooling at the interface. In the presence of AFGP molecules, the kinetic smoothing transition from a rough surface to smooth one occurs, and the growth rate is reduced. The effects relate to the anisotropic adsorption properties of AFGP molecules. In this study, we proposed a new model for the ice growth kinetics, in which a change of structure of water molecules near ice interface, i.e., hydrophobic interaction is taken into account instead of Gibbs-Thomson Effect caused by the pinning of a step by AFGP molecules.   

First Principles Simulations of Ice Nucleation at Metal Surfaces

Ice nucleation at solid surfaces is of relevance to countless scientific and technological processes. In particular the nucleation of ice nano-crystals on metal surfaces is often a key first step in cloud formation and corrosion [1]. Yet unfortunately this remains one of the most poorly understood natural phenomena; severely lacking in atomic level understanding. Here, we discuss detailed density functional theory studies aimed at putting our understanding of ice nucleation at metals on a much firmer footing. Specifically the properties of H$_2$O hexamers - the smallest `building blocks' of ice - adsorbed on a number of close-packed transition metal surfaces have been examined. We find that the competing influences of substrate reactivity and hexamer-substrate epitaxial mismatch conspire to yield a rich variety of (novel) hexameric ice structures, some of which have been observed by recent scanning tunnelling microscopy experiments [2]. [1] H.R. Pruppacher and J.D. Klett, \textit{Microphysics of Clouds and Precipitation}, (Kluwer, Dordrecht, 2003). [2] K. Morgenstern, \textit{et al.}, (To be published).   

Grain Boundary Melting in Ice

The central influence that grain boundary melting has on the sintering, coarsening, transport behavior and many other bulk properties in ostensibly all materials, including ice, motivates combined experimental and theoretical studies. The great difficulty of directly accessing a grain boundary in thermodynamic equilibrium has resulted in a dearth of experimental tests. Our approach is to prepare an ice bicrystal in a thin growth cell and to probe the grain boundary directly by employing light scattering and polarization with a helium-neon laser. We first measure the direction of the c-axis of each of the grains by bringing polarized light through each domain and analyzing the polarization of the transmitted beam. We then monitor the reflected intensity as a function of temperature and concentration of monovalent electrolyte. Using scattering theory, we estimate that, if the grain boundary is dry, the reflected intensity for a high angle grain boundary will be 10$^{-5}$ times the incident intensity. Using the index of refraction of bulk water to approximate the degree of sensitivity to the presence of a grain-boundary film, the change in reflected intensity due to a 15 $\AA$ layer of water is found to be greater than 10 \%, an easily measurable signal using a 4mW laser. Results are compared with a recent theory of impurity--stimulated grain boundary melting   

Nucleation, condensation, wetting and dewetting of water on a silver surface

Water adsorption and nucleation on a single crystal silver surface has been characterized over the temperature range of 100 - 170 K. At temperatures greater than 165.5 K, Second Harmonic Generation studies show that water can not be condensed on the Ag surface even though the condensation rate is faster than the desorption rate. At this temperature classical nucleation theory predicts a very large critical nucleus size such that stable nuclei cannot form under normal laboratory deposition conditions. In the temperature range of 133.5 - 165.5 K SHG during isothermal adsorption-desorption and temperature programmed desorption experiments show that water desorbs from three-dimensional structures on the surface with a desorption energy of 48 kJ/mol. In contrast, at temperatures less than 133.5 K, water desorbs from a 2-D structure with an activation energy of 25 kJ/mol. In this lowest temperature range water wets the silver surface. This wetting-dewetting transition can be understood in terms of classical nucleation theory where below 133.5 K the critical nucleus height is less than one monolayer.   

Surface Segregation in Mixed Alkali Halide Solutions

Bromine compounds in the marine troposphere have been of great interest since the observation that tropospheric ozone depletion events in the arctic are correlated with gas phase bromine chemistry. Reactions with sea salt aerosols, particles and ice appears to be the source of these bromine compounds. In previous experiments we have shown that bromine segregates to the surface of sodium chloride crystals that are uniformly doped with low levels of bromide. We describe here experiments in which we have used high pressure photoelectron spectroscopy (HPPES) at the Advanced Light Source (ALS) to measure the ion concentrations at the surface of mixed alkali halide \textbf{solutions}. The experiments start with single crystals of sodium chloride that are uniformly doped with bromide at 7{\%} and 0.1{\%} level. The x-ray photoelectron spectra of the sample surface are obtained as a function of water vapor up to and at sample deliquescence. The concentrations of bromide and chloride ions at the liquid/vapor interface for the saturated solution that is produced upon deliquescence are obtained. For both samples, after deliquescence the bromide concentration at the surface of the saturated solution is greatly enhanced. The results will be compared with molecular dynamics simulations of a mixed bromide/chloride solution.   

Session A35: Energy Landscapes in Clusters, Materials, and Biology I

Energy Landscapes and Beyond

To understand many problems in modern science one must cope with a diversity of long-lived states or attractors. The necessary notion is that of an energy landscape. Cluster physics, protein dynamics and protein folding are key examples of problems where a landscape description is essential. In bulk condensed matter e.g. supercooled liquids and glasses, a landscape can only be meaningfully assigned to a small region of the system. Such local landscapes, however, can be constructed and used to describe the unusual slowing as the glass transition is approached and aging phenomena in quenched glasses. The resulting theory is quantitatively successful. Fully quantum glasses may also exist. $\backslash $Far-from equilibrium systems, such as cytoskeleton and gene networks require going beyond the landscape notion in new ways, which we will briefly describe.   

Kinetics of an ultrafast folding protein analyzed with a free energy surface model

Both theoretical and simulation studies have been successful in describing the kinetics of protein folding as diffusion over a one-dimensional free energy barrier, using an order parameter such as the number of ordered residues or the number of native inter-residue contacts as the reaction coordinate. In contrast, experimental results have been analyzed almost exclusively in terms of chemical kinetic schemes, with rate coefficients for transitions between well-defined thermodynamic states. We find that the two-phase relaxation kinetics of the ultra-fast folding villin subdomain following a laser temperature jump cannot be explained by a three-state kinetic model, but is readily explained by diffusion on a one-dimensional, temperature-dependent free energy profile that has a low free energy barrier separating folded and unfolded states. This analysis demonstrates the advantage of physical kinetics compared to chemical kinetics in understanding complex dynamics of protein folding, and should enable a closer connection between experiment and both theory and simulations.   

What happens to a protein in a glassy environment?

Several types of kinetic measurements reveal an intrinsic connection between processes within a protein imbedded in a glassy material and relaxations in the host itself. We use the Random First Order Transition (RFOT) theory of the glass transition to explain the microscopic origin of slaving of large scale protein conformational dynamics to the relaxations in the supercooled solvent. The slowing down of the protein motions relative to those of the solvent reflects the size of the conformational subspace explored by the protein relaxation. At {\em cryogenic} temperatures, the details of hole broadening depend on whether the chromophore is placed directly in a glass matrix, or imbedded in a protein first. We explain why spectral diffusion in proteins deviates from the usual logarithmic time dependence found in glasses.   

Connection between the energy landscape and glass transition in a two-dimensional Lennard-Jones mixture

Results of recent molecular dynamics simulations of a two-dimensional glass forming system are presented. The system's inherent structures are investigated over a wide range of temperature and cooling rate and compared to previous results for three-dimensional liquids and glasses. A method for analyzing the regions of the energy landscape sampled under various conditions is introduced and used to characterize the glass transition. Connections with inherent-structures theory, mode-coupling theory, and spatially inhomogeneous dynamics are discussed.   

o-Terphenyl Self-Diffusion Near the Glass Transition Temperature

Self-diffusion coefficients (D$_{T}$) have been obtained for the fragile glassformer o-terphenyl from the glass transition temperature (T$_{g}$= 243 K) to T$_{g}$+ 32 K. Compared to the predictions of the Stokes-Einstein equation, we observe substantially enhanced translational diffusion ( $>$ 2 decades), which is a strong indicator of spatially heterogeneous dynamics. The values of D$_{T}$ are measured by isothermally annealing deuterio and protio o-terphenyl thin films. Samples are prepared as vapor-deposited glassy bilayers (100 - 1000 nm) with an initially sharp interface. When a sample is annealed, diffusion blurs the interface while supercooled o-terphenyl vaporizes at the film's surface. A mass spectrometer records deuterio and protio o-terphenyl concentrations over time, providing a profile of the diffusion that has occurred in the film. Isotope and thermal history effects on D$_{T}$ have also been investigated. Our results qualitatively agree with dynamic facilitation models and random first order transition theory.   

Two aspects of water dynamics will be discussed; (1) intermittent collective motions and fluctuation, their experimental observation, (2) the role of these fluctuations in the water freezing process.

Water exhibits intermittent collective motions associated with the hydrogen bond network rearrangement (HBNR), accompanied with large fluctuations of intermittent local collective molecular motions. We have made a theoretical analysis on multi-dimensional spectroscopy which may detect these intermittent collective motions, since this method deals with the phase space dynamics of a system. The 2-R Raman spectra obtained for CS2 and Water will be discussed A liquid to solid phase-transition goes through a nucleation process, starting with the formation of an initial nucleus, which then grows into a crystalline structure. Dynamical aspects of the nucleation processes have been explored by various studies. In simulations, simple liquids, consisting of spherical particles, are found to easily freeze to crystals. Water molecules possess strong directionality of hydrogen bonds (HB), forming a disordered three-dimensional HB network (HBN) in liquid state, and water is thus much harder to freeze than the simple liquids. A simulation, recently succeeded to reproduce the pure water freezing, revealed a role of the dynamical fluctuations intrinsic in water HBN rearrangement in a formation of the initial nucleus of the water freezing process, but large parts of its molecular mechanism have been remained unknown yet. In the present work, we use a network analysis and employ order-parameters to establish a clear molecular picture of the fragments, the transformation, the role of collective motions in this process.   

Potential-Energy Surface of Infinite Helical Polypeptides

The potential-energy surfaces of infinite polyalanine and polyglycine chains in helical conformation are studied using density-functional theory in the Perdew, Burke and Ernzerhof approximation to the exchange-correlation functional (DFT-PBE). Minima associated to a $\pi$-helix, $\alpha$-helix and $3_{10}$-helix conformations are identified for both polypeptides. For polyalanine the $\alpha$-helix minimum is the lowest in energy. However for polyglycine $\pi$-helix and $\alpha$-helix minima are degenerated within the DFT accuracy. The $\alpha$-helix is found to undergo structural transitions to a $\pi$- or $3_{10}$-helix when the length of the helix is strainend by more than 10\%. The barriers for the structural transitions mainly associated to the breaking of the hydrogen bonds are considerably affected by the side group in polyalanine. We find this effect can not be solely attributed to repulsive interactions between the side group and the helix backbone but to sizeable changes in covalent bonds in the peptide unit of polyalanine with respect to polyglycine.   

The role of vibrations in the kinetics of conformational isomerization of biomolecules and clusters

The kinetics of conformational isomerization of sizable molecules and clusters depends both on the topography of the potential energy surface (PES), in particular barriers that separate minima corresponding to specific conformers, and the vibrational dynamics of each conformer. Incorporation of the vibrational contribution can be finessed by assuming rapid vibrational energy flow within a basin of the PES and adopting Rice-Ramsperger-Kassel-Marcus (RRKM) theory to calculate isomerization rates between pairs of conformers. However, RRKM theory often overestimates rates of conformational change due to insufficiently rapid vibrational energy flow. In this case, isomerization rates can be computed by applying a theory for quantum energy flow in many-oscillator systems to calculate corrections to RRKM theory. As an example we discuss the influence of vibrational energy flow on the kinetics of conformational isomerization of the dipeptide NATMA.   

Amino acid and water molecules adsorbed on water clusters in a beam

Water clusters (H$_{2}$O)$_{n}$ and (D$_{2}$O)$_{n}$ ($n\le $16) are produced by supersonic expansion and pick up an additional heavy or light water molecule, respectively, or an amino acid molecule, while flying through a pick-up cell. The products are analyzed by electron bombardment ionization mass spectrometry. Ionization proceeds via well-known loss of an OH or OD group, but these have a strong predilection to come from the guest, rather than the host, molecule. e.g., even for large (D$_{2}$O)$_{n}$H$_{2}$O [or (H$_{2}$O)$_{n}$D$_{2}$O], about 50{\%} of the time the lost group is from the picked-up molecule. Similar ratios are found for amino acid guests. This suggests that proton exchange is suppressed, the host clusters are frozen into compact annealed shapes, and the adducts reside on the surface and present a dangling OH [or OD] bond where the ionization-induced hole prefers to localize dissociatively.   

Session A36: Focus Session: Novel States of Matter in Atomic Gases

Noise Correlations in one-dimensional ultra-cold atom systems

Time of flight images reflect the momentum distribution of the atoms in the trap, but the spatial noise in the image holds information on more subtle correlations. Using Bosonization, we study such noise correlations in several generic one dimensional systems of ultra cold bosons and fermions. Specifically, we show how pairing as well as spin and charge density wave correlations may be identified and extracted from the time of flight images. These incipient orders manifest themselves as power law singularities in the noise correlations, that depend on the Luttinger parameters.   

Superfluidity of interacting bosons in one-dimension via strong disorder

We have found a novel set of superfluid and Mott insulating phases arising in strongly-disordered, one-dimensional optical lattices with large, commensurate filling of bosons. Using a spin-wave approach, we show that the bosons form a true, long-range ordered superfluid at $T=0$, provided that the on-site repulsion has a broad distribution. Such broad, power-law distributions arise naturally under real-space renormalization group flow. Furthermore, they describe the asymptotic properties of diluted two-dimensional optical lattices near percolation threshold. We discuss the possibility of true superfluidity in the diluted system.   

Novel Superfluidity of Trapped Fermi Atoms Loaded on Optical Lattices

We investigate a possibility of superfluidity in a trapped gas of Fermi atoms with a repulsive interaction in the presence of an optical lattice. Applying the exact diagonalization method to a one-dimensional Hubbard model including the trap potential, we find that, when the strength of the repulsive interaction exceeds a critical value, the binding energy of two Fermi atoms becomes {\it negative} below the half-filling case, indicating that an attractive interaction effectively works between Fermi atoms. In this case, a "Mott insulating core" appears in the center of the trap, where each site is occupied by one atom. The Cooper-pair correlation strongly develops between atoms in the left and right hand sides of this core. Furthermore, we show a ground-state phase diagram including the superfluidity on the trapped Fermi atoms loaded on optical lattices.   

Novel quantum phases of interacting fermionic atoms

Recent developments in ultracold atomic gases have revitalized interest in some basic qualitative questions of quantum many-body theory, because they promise to make a wide variety of conceptually interesting systems, which might previously have seemed academic or excessively special, experimentally accessible. I will describe a set of simple, idealized model systems that seem to surprisingly display new states of quantum matter. One such system is a two component fermi gas of mismatched fermi surfaces and of different masses. Our study shows it has a new kind of pairing state---breached pair superfluidity---other than the well known states of BCS and Larkin-Ovchinnikov-Fulde-Ferrell (LOFF). I will discuss when the state becomes stable and suggest ways of possible experimental realization, including a theoretical design of optical sublattices with different tunnelings (thus a spin-dependent Hubbard model). The state also contains novel signature of superfluidity in the momentum distribution of particles, which is directly observable as a first-order effect. In the presence of a steeper confining trap potential which strongly breaks the translational symmetry, our work in progress indicates that another competing new state---angular crystalline superfluid---seems to become energetically favorable.   

Fulde-Ferrel-Larkin-Ovchinnikov paired atomic superfluids

The Fulde-Ferrel-Larkin-Ovchinnikov (FFLO) paired superfluid is a periodically modulated, finite magnetization relative of the conventional BCS superconductor. While the FFLO state has proven elusive in condensed-matter experiments, we will argue that it is naturally realized in a cloud of two species of ultra-cold fermionic atoms interacting through a Feshbach resonance, with the difference in the number of the two species playing the role of an imposed magnetization. Motivated by this possibility, we have computed the phase diagram of such atomic systems as a function of Feshbach resonance detuning, temperature and magnetization.   

Non-equilibrium dynamics of hard-core bosons on 1D lattices: short vs large time results

Based on an exact treatment we study the non-equilibrium dynamics of hard-core bosons on one-dimensional lattices. Starting from a pure Fock state we find that quasi-long range correlations develop very fast in the system, and they lead to the emergence of quasi-condensates at finite momentum [1]. The exponent observed in the power-law decay of the one-particle density matrix, which develops dynamically, is the same that has been proven to be universal in the equilibrium case [2]. We also study the time evolution of clouds of hard-core bosons when they are released from a harmonic trap. In this case we show that the momentum distribution of the free expanding hard-core bosons approaches to the one of noninteracting fermions [3], in contrast to the known behavior in equilibrium systems. [1] M. Rigol and A. Muramatsu, cond-mat/0403387, to appear in Phys. Rev. Lett. (2004). [2] M. Rigol and A. Muramatsu, Phys. Rev. A 70, 031603(R) (2004); ibid. cond-mat/0409132. [3] M. Rigol and A. Muramatsu, cond-mat/0410683.   

Solitons in Trapped Bose-Einstein condensates in one-dimensional optical lattices

We use Quantum Monte Carlo simulations to show the presence and study the properties of solitons in the one dimensional soft-core bosonic Hubbard model with near neighbor interaction in traps. We show that when the half-filled Charge Density Wave (CDW) phase is doped, solitons are produced and quasi long range order established. We discuss the implications of these results for the presence and robustness of this solitonic phase in Bose-Einstein Condensates on one dimensional optical lattices in traps and study the associated excitation spectrum. The density profile exhibits the coexistence of Mott insulator, CDW, and superfluid regions. Work supported by NSF DMR 0312261 and NSF INT 0124863.   

Filling the Bose sea: clustered states and excitations

We explore the structure of clustered quantum states, which might be realized in `droplets' of rapidly rotating Bose Einstein condensates. We explore the underlying algebraic structure (which is that of the affine Lie algebra $su(2)_k$) and count the dimension of the space of symmetric polynomials which have the clustering property. Upon increasing the size of the droplet, the partition function of the droplet becomes a character of the underlying algebra $su(2)_k$, confirming that the system can be described by an $su(2)$ Chern-Simons theory.   

Dynamical structure factor of one dimensional bosons in optical lattices

We investigate a one dimensional system of cold bosonic atoms in an optical lattice subjected to a time dependent periodic potential. We study the dynamic structure factor and the excitation spectrum in different regimes: superfluid and Mott insulator. In particular the strong interactions gives rise to a continuum of excitations in the superfluid phase. We discuss the application to recent experiments on bosonic tubes. We also discuss comparisons to higher dimension situations.   

Superfluid--Insulator Transition in Commensurate One-Dimensional Bosonic System with Off-Diagonal Disorder

We analyze the superfluid--insulator transition in a system of one-dimensional (1D) lattice bosons with off-diagonal disorder in the limit of large commensurate filling. We argue---in contrast to the recent prediction (E.~Altman, Y.~Kafri, A.~Polkovnikov, and G.~Refael, cond-mat/0402177) of strong- randomness fixed point for this system---that at any strength of disorder the universality class of the transition on the superfluid side coincides with that of the superfluid--Mott- insulator transition in a pure system. We present results of Monte Carlo simulations for two strongly disordered models that are in excellent agreement with the advocated scenario.   

Supersolid vs. phase separation in 2D

We study the nature of the ground state of the strongly-coupled two dimensional extended Bose Hubbard model on a square lattice. Strong coupling expansion and quantum Monte Carlo simulation of finite systems were used to analyse the stability of the ($\pi,\pi$) crystalline order at half-filling and the effects of doping away from it. We find that strong but finite on-site interaction along with a comparable nearest-neighbor repulsion results in a thermodynamically stable supersolid ground state just above half-filling, while the system phase separates just below half-filling. The interplay between these two interaction energies results in a rich phase diagram which is studied in detail.   

Crystalline phases of bosons in harmonic traps

Strongly repelling bosons in two-dimensional harmonic traps are described through breaking of rotational symmetry at the Hartree- Fock level and subsequent symmetry restoration via projection techniques, thus incorporating correlations beyond the Gross- Pitaevskii (GP) solution. The bosons localize and form polygonal-ring-like crystalline patterns, both for a repulsive contact potential and a Coulomb interaction, as revealed via conditional-probability-distribution analysis. For neutral bosons, the total energy of the crystalline phase saturates in contrast to the GP solution, and its spatial extent becomes smaller than that of the GP condensate. For charged bosons, the total energy and dimensions approach the values of classical point-like charges in their equilibrium configuration. For neutral bosons, the present work describes a 2D generalization of the 1D Tonks-Girardeau regime of impenetrable bosons.\footnote{Phys. Rev. Lett. (Dec. 2004); cond-mat/0410598; to see a preprint \urllink{CLICK HERE}{http://arxiv.org/abs/cond- mat/0410598}}   

Superfluid-insulator transition in moving boson lattice systems

We analyze the stability of superfluid currents in a system of strongly interacting ultra-cold atoms in an optical lattice. We show that such a system undergoes a dynamic, irreversible phase transition at a critical momentum that depends upon the interaction strength between atoms. At integer filling of the lattice, the phase boundary continuously interpolates between the classical modulation instability of a weakly interacting condensate and the equilibrium quantum phase transition into a Mott insulator state at which the critical momentum vanishes. For fractional filling, the critical momentum dips to a minimum at intermediate interaction strength, but saturates to the same limiting value at both strong and weak interactions.   

Decay of superfluid currents in moving boson lattice systems

Following up on the previous talk, we analyze the broadening of the dynamic superfluid-insulator transition due to quantum and thermal fluctuations. We derive asymptotic expressions for the decay rate of superfluid currents near the transition. We show that in three dimensional optical lattices the broadening of the transition is very weak. On the other hand in two and especially in one dimension the broadening is very significant unless the system is very deep in the superfluid regime. We argue that at experimentally relevant temperatures the quantum decay is stronger than the thermal and thus is straightforward to observe.   

Session A37: Liquid Crystals I

Spatially Periodic Alignment of Liquid Crystals by Patterned Photopolymerization

We demonstrate an electrically switchable diffraction grating based on periodically patterning the anchoring conditions of a nematic liquid crystal (NLC) within a polymer matrix via a patterned photopolymerization. We used two comonomers with opposite tendency to align the NLC and different reactivity ratio, which lead to definition of the areas with alternating homeotropic and planar alignment of the NLC through a UV irradiation with a photomask. The photopolymerization-induced diffusion of the monomers accounts for the spatial concentration distribution of these monomers. The LC diffraction gratings we made are switchable under low electric fields, and also have structural stability offered by the polymer matrix.   

Quantitative study of the colloidal interaction forces and defect line tension in liquid crystals using optical trapping of polymer particles and defects

We demonstrate optical trapping and manipulation of defects and transparent microparticles in a thermotropic nematic liquid crystal with low birefringence. The three-dimensional director fields and positions of the particles manipulated by laser tweezers are visualized using the Fluorescence Confocal Polarizing Microscopy. The disclination lines are manipulated using tightly-focused linearly-polarized laser beams and optically trapped colloidal particles. We employ the particle manipulation to measure line tension of a topologically stable disclination line and to determine colloidal interaction of particles with perpendicular surface anchoring of the director at their surfaces. We show that the laser trapping of particles and defects in liquid crystals opens new possibilities for their fundamental studies as well as for new applications, such as assembling of colloidal structures and photonic crystals in the liquid crystal medium.   

Effect of Magnetic Nanoparticles, Their Size and Functionalization on Liquid Crystal Order

We have observed the effects of adding magnetic nanoparticles with a different surface termination to smectic A 8CB liquid crystals by examining the liquid crystals by X-ray scattering. Adding the magnetic nanoparticles improves the liquid crystal's response to a magnetic field by at least one to two orders of magnitude. We have performed the experiments with four types of organic compounds covering the nanoparticles, using 11 nm and 2 nm FeCo nanoparticles, and have varied the applied magnetic field from 225 mT to 362 mT. There is a variation on the effect due to the size of the nanoparticles and also to the concentration of the particles in the mixture. As a function of magnetic field, the 11 nm and 2 nm particles terminated in polyethelyne glycol 3000 exhibit the largest rotation with the magnetic field. The liquid crystal rotates in opposite directions depending on the concentration of particles.   

Pattern formation by liquid crystal of colloidal gold nanorods

To utilize the properties of nanoparticles to make nanoscale devices, large-scale spatial organization of NRs is required. Colloidal NRs can self assemble to form lyotropic liquid crystals(LC). The formation of lyotropic liquid crystals is a unique way to assemble NRs in solution since metal NR liquid crystals can combine the properties of liquid crystals with the electronic properties of metal component. We observed LC phase of gold NRs by resorting to an evaporating aqueous NR solution. The convective flow caused by the solvent evaporation carries the NRs from the bulk solution to solid-liquid-air interface, which makes the solution locally very concentrated, and therefore phase transition of NRs occured. By changing the aspect ratio, concentration and polydispersity of NR, and evaporation rate, we observed various pattern formation by LC phase similar to Liesegang ring.   

Generalized Order Parameters for Systems of Orientationally Ordered Anisometric Particles

A variety of nanoparticle systems, such as semiconductor nanowires, mineral liquid crystals, colloidal clays and ferrofluids can exhibit orientationally ordered phases. These systems are therefore liquid crystals, where the relevant constituents are nanoparticles rather than molecules. We introduce generalized order parameters, based on the symmetry of the constituents, to describe orientational order in such systems. We describe the procedure to identify the relevant order parameters, discuss the connection between these and experimental observables and present the results obtained from simulations. Some mathematical issues related to representations are also addressed.   

Liquid crystals in random porous media: Disorder is stronger in low--density aerosils

The nature of glass phases of liquid crystals in random porous media depends on the effective disorder strength. We study how the disorder strength depends on the density of the porous media and demonstrate that it can increase as the density decreases. We also show that the interaction of the liquid crystal with random porous media can destroy long--range order inside the pores. This work was supported in part by NSF DMR- 0131573. \vskip .2in \noindent [1] D. E. Feldman and R. A. Pelcovits, Phys. Rev. E {\bf 70}, 040702(R) (2004).   

The Effect of Aerosil Network on Liquid Crystal (4O.8) Phase Transition

We report a high resolution x-ray diffraction study of the nematic (N) to smectic-$A $(Sm$A) $transition in the single-layer smectic (Sm$A_{m})$ liquid crystal butyloxybenzylidene octylaniline (4O.8) within aerosil dispersion. Aerosils are dispersed in liquid crystal material with a broad range concentration. They dramatically affect the phase transition properties in different liquid crystals [1]. These effects were studied in the view of random field theory introduced by quenched randomness of the silica gel network. The second order N-Sm$A$ phase transition and strong first order Sm$A$-CrB freezing transitions are shifted to lower temperatures. Sm$A$ line-shape is broadened indicating a short-range order. Correlation lengths and power-law fits show behavior similar to bilayer Sm$A_{d}$ liquid crystals. The present work enables us to test our understanding of random field effects introduced by dispersed aerosils forming a network in Sm$A_{m}$ material. [1] S. Park, R.L. Leheny, R.J. Birgeneau, J.-L. Gallani, C.W. Garland and G.S. Iannacchione, Phys. Rev. E 65 050703(R) (2002)   

Rf/ac-calorimetry on 8CB+aerosil dispersions

Using a new high-resolution AC calorimetric technique employing RF (dielectric) heating, both the heat capacity and permittivity of a sample may be simultaneously probed. Relative resolutions of better than 0.06\% in the ratio of heat capacity to applied heating power, and 0.03\% in the phase shift measurements are easily obtained. This new and powerful technique has been applied to aerosil dispersions in the liquid crystal octylcyanobiphenyl (8CB) through the \textit{I}-\textit{N} and \textit{N}-Sm\textit {A} phase transitions. The temperature dependence of the applied heating power, proportional to the permittivity and hence the orientational order, shows no change through the \textit{N}- Sm\textit{A} for $\rho_S = 0.1$~g/cc. This indicates that at the point where sharp $C_p$ features are lost, the Sm\textit{A} and \textit{N} phases are effectively decoupled, consistent with recent NMR studies.   

Structure and Phase Behavior of Smectic Liquid Crystals in Anisotropic Disorder

We present x-ray scattering studies of the smectic liquid crystals octylcyanobiphenyl (8CB) and 4-n-pentylphenylthiol-4-n-octyloxybenzoate ($\bar {8}$S5) confined in strained colloidal silica gels. The gels possess anisotropy that stabilizes long-range nematic order in the liquid crystals while introducing random field effects that disturb the smectic order. The strong azimuthal focusing of the scattering enables detailed characterization of the smectic correlations. The short-range smectic order that forms in this environment is inconsistent with a topologically ordered state predicted for 3D random field \textit{XY} systems and is quantitatively like the correlations of smectics confined by isotropic gels. The quenched disorder modifies the nematic -- smectic-A critical behavior, for example, suppressing the anisotropic scaling of the correlation lengths observed in the pure liquid crystals. The smectic-A and smectic-C scattering indicates that the behavior of smectics confined in gels is dictated by random fields coupling directly to the smectic order parameter while fields coupling to the nematic director play a subordinate role.   

Ordered Patterns of Liquid Crystal Toroidal Defects by Microchannel Confinement

We present the first experimental results demonstrating a novel approach to controlling the size as well as the spatial patterning of defect domains in a smectic liquid crystal by geometric confinement in surface modified microchannels. By confining the liquid crystal 8CB (4'-octyl-4-cyanobiphenyl) in micron sized rectangular channels with controlled surface polarity, we were able to generate defect domains that are not only nearly uniform in size but also arranged in quasi-two-dimensionally ordered patterns. Atomic force microscopy measurements revealed that the defects have a toroidal topology, which we argue is dictated by the boundary conditions imposed by the walls of the microchannel. We show that the defects can be considered as colloidal objects, which interact with each other to form ordered patterns. This method opens the possibility to exploit the unique optical and rheological properties associated with LC defects to making new materials. For example, the control of the shape, size and spatial arrangement of the defects at the mesoscale suggests applications in patterning, templating, and when extended to lyotropic liquid crystals, a process leading to uniform size spherical particles for chemical encapsulation and delivery (Work supported by NIH GM-59288, NSF DMR-0203755 and ONR N00014-00-1-0214)   

Coarsening dynamics of a liquid crystal biosensor: Effects of adsorbed nanoparticles, flow and thermal fluctuations

We explore the coarsening dynamics of a model for experimental liquid-crystal (LC) biosensors for viral particles: a thin film of LC confined between two parallel, aligning substrates, after a quench into the nematic phase. In the absence of particles, the LC undergoes a coarsening process that ultimately produces a uniform orientation state. By contrast, when the surface coverage c exceeds a critical value, the dynamics is slowed down and the system exhibits multidomain behavior, characterized by a finite correlation length for the tensor order parameter. In this work, we use (a) a dynamic field theory to study the dependence of the critical coverage c with operational parameters of the biosensor, such as the scalar order parameter, the separation between substrates, and the adsorption pattern; and (b) a fluctuating LC lattice Boltzmann method that allows the exploration of the effects that hydrodynamics and thermal fluctuations have on the coarsening process and the critical concentration in 3D.   

A Light Driven Artificial Goldfish

Liquid crystalline elastomers (LCEs) undergo large and rapid shape changes when illuminated by light. We have immersed an azo-dye doped LCE in a fluid and alternately illuminate either side of the LCE with light to create an artificial goldfish of sorts. These light induced deformations allow the LCE to interact with a fluid environment in novel ways. We use a fluid flow visualization technique to attempt to understand the dynamics of these interactions. We describe our experimental setup, the LCE drive scheme used, and our observations of induced motion in both the LCE sample and the surrounding fluid.   

Real-time microbe detection based on director distortions and light transmittance around growing immune complexes in lyotropic chromonic liquid crystals

We describe director distortions in the nematic liquid crystal (LC) caused by a spherical particle with tangential surface orientation of the director and show that light transmittance through the distorted region is a steep function of the particle's size. The effect allows us to propose a real-time microbial sensor based on a non-toxic lyotropic chromonic LC (LCLC) that detects and amplifies the presence of immune complexes. A cassette is filled with LCLC, antibody, and antigen-bearing particles. Small and isolated particles cause no macroscopic distortions of the LCLC. Upon antibody-antigen binding, the growing immune complexes distort the director and cause detectable optical transmittance between crossed polarizers. The work was supported by NSF/ITIC DMR-0346348.   

Dielectric relaxation in confined liquid crystal: molecular and collective modes

Dielectric spectroscopy was used to investigate the influence of confinement of the liquid crystals on phase transitions and the dynamics of molecular reorientations via rotation of molecules around their short axis. The pore size was varied from 200 nm then the system shows behavior close to the behavior of three-dimensional (3D) bulk liquid crystal - to extreme narrow confinement that may be considered a quasi-one-dimensional (1D) system. We did not observe either the N-I or nematic-solid crystal phase transition under quasi-1D-confinement: in contrast, in pores of larger pore sizes, these transitions are observed with a shift and broadening of the phase transition as compared to bulk LC. We observed that, instead of undergoing the N-I phase transition in the quasi-1D-system, liquid crystal inside the narrow pores solidifies. Relaxation of molecular origin freezes out upon cooling the sample from the isotropic phase. Molecular reorientations were dielectrically active at temperatures much above the bulk N-I transition temperature but with the relaxation rate much slower and dielectric spectrum much broader than in the bulk. The relaxation due to cooperative dynamics of fluctuations of director orientations (investigated in complementary dynamic light scattering experiments) was not observed in quasi-1D-system. We suggest that the slowing down of the mode, which is molecular in the bulk material, is may be due to enhancement of the effective viscosity liquid crystal under such narrow confinement and resulting glass-like dynamics.   

Dynamics of director relaxation in nanoconfined liquid crystal: dynamic light scattering investigation

Dynamic light scattering was applied to study the boundary conditions (planar-axial and homeotropic-radial) and layer thickness (at nanoscale) of 8CB confined to cylindrical pores influence on phase transitions and relaxation of director orientational fluctuations. For confined 8CB in the nematic phase two well-defined relaxation processes were observed for both axial and radial orientations of the liquid crystal. The first process is qualitatively associated with bulk-like nematic director fluctuations. The second relaxation process (with relaxation time slower than the first one) is most likely due to the fluctuations in layers nearest the wall surface. In samples with homeotropic boundary conditions we observed the onset of smectic-A phase order forming on the pore wall even though the rest of the liquid crystal could be in the nematic phase. We found that for homeotropic boundary conditions of confined liquid crystal, the pore wall-liquid crystal interactions influence on the properties of the surface layer is stronger than in the case of axial orientation, particularly, and the influence of boundary conditions on N-Sm-A phase transition in confined 8CB is stronger than on isotropic- nematic phase transition. The separation between the first and the second (slow) process is clearer for thinner layers and the amplitude of slow process is greater for thinner layers. This suggests that the slow process is surface related relaxation.   

Session A38: Metal-Insulator Phase Transitions - Theory I

Dynamical Breakup of the Fermi Surface in a doped Mott Insulator

The evolution from an anomalous metallic phase to a Mott insulator within the two-dimensional Hubbard model is investigated by means of the Cellular Dynamical Mean-Field Theory. We show that the density-driven Mott metal-insulator transition is approached in a non-uniform way in different regions of the momentum space. This gives rise to a breakup of the Fermi surface and to the formation of {\it hot} and {\it cold} regions, whose position depends on the hole or electron like nature of the carriers in the system.   

Temperature dependent spin susceptibility in a two-dimensional metal

We consider a two-dimensional electron system with Coulomb interaction between particles at a finite temperature $T$. The Kohn singularity in the response function leads to a linear- in-$T$ correction to the quasiparticle $g$-factor and the spin susceptibility. We show that the universal linear temperature correction is a generic property of a strongly interacting electron system.   

Non-equilibrium properties of a Mott insulator in an external electric field

A dynamical mean-field theory formalism is developed to exactly solve the non-equilibrium properties of the conduction electrons in the Falicov-Kimball model. We study the response of the conduction electrons on a hypercubic lattice in the half-filled case when an external spatially uniform time-dependent electric field is applied. The single-particle response functions and the electric conductivity are calculated as functions of time for different cases of the time-dependent electric field and for different values of the on-site repulsion parameter U. In particular, the most interesting case occurs when U is close to the Mott-insulator transition value.   

On the X-ray-Problem in the Falicov-Kimball model in large dimensions at half-filling

The f-electron spectral function of the Falicov-Kimball model is calculated within the dynamical mean-field theory using the numerical renormalization group method as the impurity solver. Both the Bethe lattice and the hypercubic lattice are considered at half filling. For small U we obtain a single-peaked f-electron spectral function, which --for zero temperature-- exhibits an algebraic (X-ray) singularity ($|\omega|^{-\alpha}$) for $\omega \rightarrow 0$. The characteristic exponent $\alpha$ depends on the Coulomb (Hubbard) correlation U. This X-ray singularity cannot be observed when using alternative (Keldysh-based) many- body approaches. With increasing U $\alpha$ decreases and it vanishes for sufficiently large U when the f-electron spectral function develops a gap and a two-peak structure (metal-insulator transition).   

Phase diagram of the one dimensional extended Falikov-Kimbal model

We solve the one dimensional spinless Falicov-Kimball model with hybridization between the conduction and localized electrons for partial band filling. Using a bosonization technique we derive an effective model for the occupation of the localized orbitals and find a crossover from a mixed-valence metallic state to a charge-ordered insulating state with increasing on-site Coulomb interaction.   

Hierarchical lattice models with evidence of metallic conductance: Implications for localization

A detailed study of linear-wave dynamics on a family of finitely ramified, hierarchical lattices that includes the modified rectangle of Dhar shows that the spectrum contains only a continuum with a smooth local density of states. This is in contrast with the typical spectrum associated with linear-wave models on finitely ramified fractals, like the Sierpi\'{n}ski lattice, that consist of a Cantor-like portion with a sequence of isolated eigenvalues sitting in the gaps of the Cantor set. In addition, at random energy the Greenwood-Peierls conductance shows metallic behavior rather than tending to zero with increasing lattice size either exponentially (strong localization or superlocalization faster than exponential) or as a power law (weak localization). Both the modified rectangle of Dhar and modified cube demonstrate these novel properties and suggest the possibility of cross-over as a function of energy between metallic and insulating regimes.   

Non-linear quantum critical transport and the Schwinger Mechanism

Scaling arguments imply that quantum critical points exhibit universal non-linear responses to external probes. We investigate the origins of such non-linearities in transport, which is especially problematic since the system is neccessarily driven far from equilibrium. We argue that for a wide class of systems the new ingredient that enters is the Schwinger mechanism---the production of carriers from the vacuum by the applied field--- which is then balanced against a scattering rate which is itself set by the field. We show by explicit computation how this works for the case of the superfluid-Mott insulator transition of bosons at commensurate fillings.   

An Analytic Phase Diagram for Anderson Disorder

The Anderson model for independent electrons in a disordered potential transforms analytically and exactly to an ordered lattice of spins interacting through an itinerant electron, a variant of augmented space [see Phys. Rev. B 66, 155121]. Anderson transitions are clear in this representation where the sector of augmented space dominating the asymptotics of states changes at critical disorders. There are also critical energies or mobility edges which depend on the disorder and separate band states from defect states. The critical disorders together with improved approximations for critical energies produce an analytic phase diagram which can be largely reconciled with the results of single-parameter scaling and numerical scaling.   

Many-body electronic structure calculations for Americium metal

Total energies and electronic spectral functions for Americium are calculated using novel dynamical mean field based spectral density functional approach. Pressure dependence as a function of volume and bulk modules for different phases of Am will be studied by this many body calculation and compared to the predictions of experiment. Volume dependent spectral functions will be extracted and discussed in connection to the anomalous resistivity behavior showing its almost one order of magnitude increase under pressure. Electron-phonon interactions estimated in the presence of electronic correlations using linear response method shed a new light on superconductivity of this actinide.   

Coulomb Gas in the Large-N Limit: no Spin-Splitting of the Effective Mass

Recent experiment [1] revealed an unusual feature of the electron effective mass ($m^*$) in Si MOSFETs: while $m^*$ exhibits a strong dependence on the electron density ($r_s$), it does not depend on the degree of spin polarization. Also, the masses of electrons with up- and down spins are the same. These findings are in an apparent contradiction with the Fermi-liquid theory, which predicts two different and polarization-dependent masses in a partially spin-polarized regime, both in the weak- and strong-coupling limits. We show that the effective mass of the Coulomb gas in the large-N limit (for Si MOSFET, $N=4$) is renormalized primarily to a polaronic effect: emission/absorption of high energy plasmons. As plasmons are classical objects, the quantum degeneracy, and hence polarization, does not affect the effective mass to the leading order in $1/N$. Polarization dependence shows up at the next-to-leading orders. We find that for $r_s=2-6$ the change in effective mass between unpolarized and fully polarized states is within $1-3\%$, which is consistent with the experiment. [1] A. A. Shashkin, M. Rahimi, S. Anissimova, S. V. Kravchenko, V. T. Dolgopolov, and T. M. Klapwijk, Phys. Rev. Lett. \textbf{91}, 046403   

Glass transition and the Coulomb gap in electron glasses

We establish the connection between the presence of a glass phase and the appearance of a Coulomb gap in disordered materials with strongly interacting electrons (amorphous semiconductors or granular metals, e.g.). We map the model to an effective single site problem, treating the correlations between electrons in a self-consistent manner. We find that in the case of strong disorder a continuous glass transition takes place whose Landau expansion is identical to that of the Sherrington-Kirkpatrick spin glass. We show that the marginal stability of the glass phase controls the physics of these systems: it results in slow dynamics and leads to the formation of the Efros-Shklovskii Coulomb gap.   

Noise at the Wigner Glass Transition and Implications for the 2D Metal-Insulator Transition

Using a simple model for interacting electrons with random disorder in two dimensions, we show in simulations that a transition from a Wigner liquid to a Wigner glass occurs as a function of electron density. The conduction noise power increases strongly at the crossover and the characteristics of the $1/f^{\alpha}$ noise change. When the temperature is increased, the noise power decreases. We compare these results with recent noise measurements in systems with two-dimensional metal-insulator transitions. [1] C. Reichhardt and C.J. Olson Reichhardt, PRL 93, 176405 (2004).   

Session A39: SPS: Undergraduate Research I

Temperature Control During the Delivery of Laser Assisted Cancer Immunotherapy

Laser Assisted Cancer Immunotherapy (LACI) is an innovative experimental technique used for the purpose of malignant tumors. The efficacy of this technique depends upon the occurrence of a vigorous and tumor immune response following the administration of treatment. The general procedure involves the injection of light absorbing dye (indocyanine green) and immunoadjuvant (glycated chitosan) into the tumor volume, followed either interstitial or surface irradiation of the tumor with an 805 nm diode laser. The magnitude of the tumor immune response is correlated to the degree of hyperthermic necrosis that occurs during laser irradiation. an optimal temperature range for necrosis is imperative to the success of the LACI approach. The aim of this study is directed toward exploring the capabilities of a potential temperature control system that utilized during interstitial (or surface) laser irradiation for the purpose of maintaining a temperature range that is for tumor cell destruction. Experimental results for tumor temperature measurement techniques, using (microthermocouples) as well as non-invasive (infrared thermal probes) approaches, will be reported.   

Laser Assisted Cancer Immunotherapy: Optical Dye Distribution in Tumors

Laser Assisted Cancer Immunotherapy is an experimental modality used to treat superficial tumors implanted on sterile Balb/C mice. The goal of the project is to induce a positive immune response toward a complete eradication of the primary tumor. Optimal necrosis results from depositing the maximum amount of thermal energy into the tumor without damaging the surrounding healthy tissue. In our laboratory, the optical dye, indocyanine green (ICG), is injected into the center of the tumor prior to surface and interstitial laser irradiation. A diode laser operating at a wavelength near 804 nm exerts thermal energy into the tumor via ICG absorption at 790 nm. Maximum immune response should occur with a uniform distribution of ICG throughout the tumor. By mapping the ICG distribution, the spatial homogeneity of the dye can be determined, which, in turn, mimics the tumor temperature profile. After excision, the tumors were cut into samples of approximately 250 microns thick and dissolved in a chemical detergent. Each sample was run through an absorption spectrometer to determine the distribution of ICG throughout the tumor. Results for both radial and depth profiles of ICG tumor distribution will be presented.   

Laser Assisted Cancer Immunotherapy: An Experimental Theraputic Approach in Balb/c Mice

Among the different therapeutic approaches to treat superficial malignant tumors, Laser Assisted Cancer Immunotherapy (LACI) shows promise. Experiments are in progress in our laboratory based on the concept of LACI which utilizes a light absorbing dye (Indocyanine Green, ICG), an immunoadjuvant (Glycated Chitosan, GC), and an infrared diode laser (1-15w) operating at 804 nm. Superficial tumors (5 to 7 mm in diameter) of the T4 cell line are grown in an animal model (Balb/C mice). The tumors are injected with ICG and GC prior to interstitial/surface irradiation of the tumor. The tumors' internal temperatures are monitored during the irradiation by invasive (microthermocouples) as well as noninvasive (infrared detector) modes. Along with the various experimental parameters, only the laser delivery (interstitial/surface) and laser intensity are varied in this initial stage so that the tumor temperature is in the range of 55 degrees C to 65 degrees C to ensure hyperthermic cell killing. The goal of the project is to determine the precise temperature range through which primary tumor necrosis and a vigorous immune response will end in tumor elimination. Experimental results coupled with a theoretical framework of laser-tissue interactions will be presented in the context of this therapeutic approach.   

The Effect of Heterogeneous Conductivity on Thermal Diffusion in Tissue

Local application of intense heat is used in a variety of medical therapies. Examples include eye surgery and cancer treatments. In such treatments, it is valuable to be able to predict the temperature distribution in the tissue. Yet, predicting the temperature distribution in the tissue presents a unique challenge because the thermal properties, such as the conductivity, change as tissue damage occurs. This effect dynamically changes an initially homogenous material into the more complex heterogeneous case. A computational method for correcting for the spatial and temporal variations in the thermal conductivity due to damage in the tissue will be presented. The results presented will indicate the significance of the gradient of the conductivity term. This theoretical work is in collaboration with hyperthermic treatment of mammary tumors in mice.   

Optical Microfluidic Control using induced Marangoni Effect

In this attempt to demonstrate a novel method for microfluidic transport, a He-Ne laser was used to create a surface temperature gradient and therefore induce the Marangoni effect. This transport technique was designed to improve current microfluidic or ``Lab on a chip,'' devices, which perform biological and chemical assays on a microscopic level. The apparatus used in this demonstration was a polystyrene dish, on which droplets on the orders of 1.7 $\mu $L- 14 pL, immersed in an organic solvent, were moved at 3 mm/s. To aid in this study, different organic solvents as well as different color dyes were used to increase the force applied to the droplet by the induced thermal gradient and achieve lower surface-droplet interaction. The magnitude of the force applied to the droplet is based on absorption of laser light, while the surface droplet interactions are based on surface tension and adhesion forces.   

An Empirical Charge Redistribution Model for Water Potentials in Biophysical Systems

Development of empirical atomistic models for simulations of biomolecules and biomaterials requires an understanding of charge transfer processes. Chemical potential equalization provides a conceptual basis for modeling electron redistribution during molecular conformational changes. A new charge-dependent energy formulation developed by Valone and Atlas extends current models based on CPE by permitting an accurate description of dynamic charge fluctuation during dissociation and charge transfer processes. We explore this formulation by constructing a conceptual submodel specific to the water molecule, important for understanding biophysical interactions in an aqueous environment.   

On the Landing of Rigid Cylinders

When a cylinder is dropped, what are the factors that determine whether it lands upright or on its side? Sir Hermann Bondi (see European Journal of Physics 14, pp. 136-140) asked this question in 1993 with the intention of determining the theoretical probability of a coin landing on its edge. The Society of Physics Students (SPS) has embarked on an experiment to test some of his ideas using data taken from many places around the country via the SPS Outreach Catalyst Kits (SOCKS). Sets of matched cylinders were sent to SPS chapters to use in dropping experiments with school children as a way of teaching about science while performing science. One goal of the experiments is to determine the relative importance of center of mass location and aspect ratio. One surprising result is the extent to which observers over-predict the occurrence of upright landings for cylinders with a square profile. \newline \newline In collaboration with Gary White, Society of Physics Students   

Measuring the High Frequency Response of Individual Carbon Nanotubes

Carbon nanotubes are nanometer diameter hollow tubes of carbon that are ideal one-dimensional conductors. They are being developed as elements in molecular electronics. Extensive studies up to this time have focused on DC properties of nanotubes. I report progress toward measurements of the dynamic conductance of individual carbon nanotubes at GHz to THz frequencies using terahertz time-domain spectroscopy. In this method, individual single-wall carbon nanotubes are incorporated into microfabricated antennas and used as receivers of broadband THz radiation. A variable time delay between two incident pulses is used to gain information about the frequency dependence of the antenna response. If the nanotube has the predicted length-dependent resonance [Burke, P. J., IEEE Transactions on Nanotechnology V. 1, p. 129 (2002)], then this should be clear in the antenna response. Knowledge of the spectral response of carbon nanotubes is important for applications in electronic interconnects as well as a confirmation of certain aspects of Luttinger liquid theory.   

Development of a nanolithography package for local AFM oxidation

We describe the process of nanolithography by local oxidation using an atomic force microscope tip, on GaAs and on InAs/AlGaSb heterostructures. The commercial microscope's tip is controlled by a home-written software package, that includes a variety of graphics primitives of use in nano-electronic geometries, allowing the user to design appropriate patterns, and control lithographic parameters. The microscope's contact tip is held grounded while the sample is held at a positive voltage, causing a local current and oxidation of the sample under the tip. The oxidized region on the semiconductor is nonconducting, or can be etched away, to form the required device structure. We characterize the process and obtainable line widths on substrate material and heterostructures, and explore the suitability of our method to create transport based nanoscale devices (partially supported by NSF REU and NSF DMR-0103034).   

Transport Measurements of Superconducting Zinc Nanowires at Low-Temperatures

New transport data on superconducting Zinc nanowires at temperatures between 1.2K and 300mK will be presented. The superconducting Zinc nanowires are fabricated using E-beam lithography on a Si/SiO2 substrate on which Zinc is deposited via thermal evaporation. The resistance and I-V are then measured as a function of temperature. The wires range in width from 40nm to 100nm and have varying thicknesses. These data are compared to thermally activated phase slip models. This research is collaboration of Eastern Nazarene College with Harvard University. Funded by: NSF DMR-02444441   

Time-Resolved Photoluminesence of Undoped and Bismuth Doped CdWO$_{4}$

Cadmium tungstate (CdWO$_{4})$ is a scintillating crystal used for detecting x-rays, particularly for use in medical applications. In an effort to characterize the photoluminescence properties and to investigate multi-photon absorption in both bismuth doped and undoped samples, we measured the photoluminescence spectra (400 nm -- 800 nm) and their time kinetics using the harmonics of a Q-switched Nd:YAG laser (355 nm and 266 nm). In both doped and undoped samples, excitation with above band gap light at 266 nm causes emission peaked at 500 nm with a single exponential lifetime of $\sim $10 microseconds. Excitation at 355 nm, which excites Bi ions in the doped sample, results in emission peaked at 570 nm with a lifetime of $\sim $1 microsecond. In the undoped sample this below band gap excitation at 355 nm causes emission which peaks at $\sim $500 nm and has a decay time similar to that caused by 266 nm excitation.   

Lipid Films, Magnetic Lipid Films, and Their Mechanical Properties

Three techniques were used to produce thin lipid films and magnetic-nanoparticle-embedded lipid films on flat substrates: spin-coating, capillary, and natural-dry methods. The morphology and the homogeneity of these films were analyzed by Atomic Force Microscopy. The local mechanical properties of these films were studied and will be discussed.   

Session A40: Focus Session: Morphology and Evolution at Surfaces: Structure and Organics

Reconstruction of the (0001) Surface of Graphite near Step Edges

We report on the surface reconstruction of the (0001) surface of graphite near step edges using scanning tunneling microscopy (STM). The surface reconstruction of the (0001) surface has the well studied triangular symmetry. Further reconstruction near point defects have been studied both experimentally and theoretically. Our studies of the graphite surface near step edges shows two types of further reconstructions of the triangular lattice. The symmetry of the first (Type I) surface reconstruction remains triangular, but rotated 30\r{ } from the original surface lattice. The periodicity of the observed pattern is 1.73 times the original surface reconstruction, and is equal to 3a (a is the bond length in graphite). The second (Type II) has hexagonal symmetry with a periodicity 1.73 times the C-C bond length. The relationship between the two structures will be discussed. The effects can be qualitatively expected based on the difference in atomic configuration near and far away from step edges.   

Relative stability of Si surfaces: a first-principles study

Surface energies of solid surfaces are often calculated by the supercell slab technique subtracting the bulk energy from the total energy of supercell. However, there exists a common mistake that the same bulk atom energy obtained by a separate bulk calculation is used for different surface orientations and slab sizes, which makes the surface energies divergent and the comparison of their relative stability unreliable. The more accurate way to determine the atom bulk energy is to extract the atom bulk energy and surface energy simultaneously by fitting slab total energy as a function of atom number in the slab. Here, using this method, we have calculated surface energies of Si (001), (110), (111), and (113) surfaces with different reconstructions systematically using first-principles total-energy method. The relative stability of these Si surfaces are shown in decreasing order as (111), (001) to (113) at low temperature, and (001), (113), (110) to (111) at high temperature, respectively. Si(113) is found to be a stable surface at both high and low temperature in spite of its high index.   

Atomic structure of the GaAs(001)-$c(4\times4)$ surface

The atomic structure of the $c(4\times4)$ reconstruction, formed on the GaAs(001) surface under high arsenic overpressure, has recently been attracting renewed interest. This has lead to a revision of the commonly accepted $c(4\times 4)$ structural model but a definitive understanding of the driving force for the newly proposed structure was lacking. Targeting the later problem, the talk will present a state-of-the-art theoretical study of the GaAs(001)-$c(4\times4)$ surface employing \textit{ab initio} atomistic thermodynamics based on density-functional theory calculations. We shall demonstrate that in a range of stoichiometries, between those of the conventional three As-dimer and the new three Ga-As dimer model, there exists a diversity of atomic structures featuring Ga-As heterodimers, \emph{driven by surface configurational entropy}. These results fully explain the experimental scanning tunneling microscopy images and are likely to be relevant also to the $c(4\times4)$-reconstructed (001) surfaces of other III-V semiconductors.   

In-Rich Reconstructions of the InSb(100) Surface and Chemisorption of Lithium on the c (8x2) Surface - An Ab Initio Study.*

The local density approximation to density functional theory (LDA-DFT) has been used to study the different possible relaxations and reconstructions of the In-rich InSb (100) surface. The surfaces are modeled by a three-layer surface with alternating In and Sb atoms and In atoms in the first layer. Hydrogen atoms are used to saturate the dangling bonds of In atoms in the bottom layer to simulate the semi-infinite effect of the surface. Periodic boundary conditions with pseudo-potentials have been used and all simulations have been carried out with the Gaussian 03 suite of programs.$^{1}$ We will report on various electronic and geometric structure properties of the possible (1x2), (2x1), (4x2) and c (8x2) reconstructed surfaces. We will also report our studies on adsorption of Li in the most symmetric sites of c (8x2) surface. Details of the chemisorption process, such as the adsorption energies, adatom separation distances, charge distributions, density of states, will be presented and compared with available results in the literature. Possible changes in the InSb surface due to Li adsorption will also be discussed in detail. $^{\ast }$Work supported, in part, by the Welch Foundation, Houston, Texas (Grant No. Y-1525). $^{1 }$\textit{Gaussian03}, M. J. Frisch \textit{et al}., Gaussian Inc., Pittsburgh, PA.   

SIESTA study of c-GaN(001)-4x1 surface reconstruction: Tetramers andtheir STM images.

Recent STM, STS and RHEED studies [1] on cubic GaN(001), grown using rf MBE under Ga-rich conditions, have revealed a surface structure consistent with predicted tetramer formation [2,3]. STM images reveal a surface consisting of rows aligned along the [110] direction with a periodicity along the $[1{\underline 1}0]$ direction of about $12.8 \AA$. STS measurements indicate the semiconducting nature of the surface and RHEED patterns of the surface provide further evidence of a periodicity consistent with tetramer formation. We report on a first principle study of this particular surface reconstruction using the SIESTA code [4], a self-consistent density functional method using standard norm-conserving pseudo-potentials and a flexible numerical linear combination of atomic orbitals basis set. Band structure calculations are in good agreement with previously reported results and the STM images obtained reproduce experimental observations. [1] H. Al-Brithen, M. Haider, A. Smith, N. Sandler and P. Ordejon. Submitted to PRL. [2] Neugebauer et al. Phys. Rev. Lett. 80, 3097 (1998) [3] Feuillet et al. Appl. Phys. Lett. 70, 24 (1997) [4] D. Sanchez-Portal, P. Ordejon, E. Artacho, and J.~M. Soler, Int. Journ. of Quant. Chem. 65, 453 (1999).   

First Principles Study of Dihydride Chains on H-Terminated Si(100)-2$\times$1 Surface

STM observation of H-terminated Si(100)-2$\times$1 surface often shows the existence of a single dihydride-chain structure parallel to a step. That chain is located near the S$_{\rm B}$ step, but away from it more than one Si-dimer's distance. In order to discuss the mechanism of the formation of such a structure, we have performed first principles calculations. As a result, we have found that the ``rebonded'' step edges of the clean Si surface turns into one dihydride chain and a ``non-rebonded'' step edge at the time of hydrogen termination. We have also found that the chain adjacent to the step is energetically less stable than in a distant position from the step. We will show these results and discuss how the dihydride chain changes its position beyond a monohydride-Si-dimer row. This study was performed through Special Coordination Funds for Promoting Science and Technology of the Ministry of Education, Culture, Sports, Science and Technology of the Japanese Government.   

Application of a direct method to solve the structure of ($\surd $3x$\surd $3)Sb/Au (110) from surface x-ray diffraction measurements

Lack of phase information in typical x-ray diffraction measurements makes it very difficult to recover the atomic-scale structure of a crystal by direct inversion of the measured amplitudes. We have developed a direct method for surface x-ray diffraction (SXRD) where the aim is to recover the part of the surface structure that is different from the truncated bulk. The iterative algorithm we have developed employs prior knowledge of the truncated bulk structure and alternately satisfies constraints in real and reciprocal space. Here we report on an application of the method to determine the unknown structure of ($\surd $3x$\surd $3)Sb/Au (110) from experimental data. In this application, the direct method is adapted to deal with the presence of four symmetry-related domains.   

STM study of azobenzene self-assembly at clean metal surfaces

Azobenzene derivatives form a unique class of photoactive molecules that have potential for nanoscale optical applications. We have examined the self-assembly behavior of azobenzene molecules on the Au(111) and NiAl(110) surfaces using a variable temperature UHV STM. We observe a variety of low-dimensional molecular configurations, some of which can be manipulated with the tip of the STM. These structures are highly temperature dependent and typically require cryogenic operation for stable imaging.   

Density-functional study of adsorption of isocyanides on the gold (111) surface.

Density functional theory (DFT) is used to study how the isocyanides HNC and CH$_{3}$NC attach to the gold (111) surface. Slab calculations are performed for monolayers with 1 molecule per 3 gold atoms coverage. For both molecules a weak binding of about 0.2 eV is found at the top site. No binding is found at other sites.   

Synchrotron X-ray Specular Reflectivity Measurements of Dotriacontane Films Adsorbed on a Ag(111) Surface

Alkane films adsorbed on metal surfaces are of interest as model lubricants. To investigate the structure of both solid and liquid films of intermediate-length alkanes, we have conducted a series of x-ray specular reflectivity measurements as a function of temperature on dotriacontane ($n$-C$_{32}$H$_{66}$ or C32) films vapor-deposited on a single-crystal Ag(111) surface in UHV. The initial C32 coverage was sufficient to observe coexistence of a multilayer film with preferentially oriented bulk C32 particles. After heating the samples to remove the bulk particles, we obtained specular reflectivity curves at room temperature consistent with one complete C32 layer followed by partial second and third layers of progressively smaller occupancy. In each layer, the molecules are oriented with their long axis parallel to the surface. We find no evidence of a perpendicular monolayer structure as observed for C32 films deposited from solution.$^{2}$ From heating studies, we determine the melting and desorption temperatures of the first and second C32 layers.$^{ 2}$H. Mo \textit{et al}., Chem. Phys. Lett. \textbf{377}, 99 (2003).   

Atomic Force Microscopy Measurements of Topography and Friction in Dotriacontane Films Adsorbed on SiO$_2$

We are continuing Atomic Force Microscopy (AFM) studies of the structure, morphology, and lateral frictional force of dotriacontane ($n$-C$_{32}$H$_{66}$ or C32) films deposited from a heptane solution onto SiO$_{2}$-coated Si(100) wafers. Step heights observed in the topographic images are consistent with layer thicknesses inferred from synchrotron x-ray$^{2}$ reflectivity scans; i.e., the first one or two layers of the film grow with the long molecular axis parallel to the surface followed by a partial layer containing perpendicular molecules. Our AFM results extend our x-ray studies in several ways: besides single perpendicular layers, we observe islands consisting of multiple perpendicular layers. We are also able to determine the spatial extent of the perpendicular layers and the location with respect to them of preferentially oriented bulk particles that nucleate at higher coverages. The frictional forces measured correlate with topographic features: we measure the same force on top of the perpendicular layers as on top of the bulk particles. This force is smaller than that on the parallel layers and the bare SiO$_{2 }$surface. $^{2}$H. Mo \textit{et al}., Chem. Phys. Lett. \textbf{377}, 99 (2003).   

Furan Decomposition Mechanism on Vicinal Pd(111) Studied by STM and DFT

We have used scanning tunneling microscopy to investigate the behavior of furan (C$_{4}$H$_{4}$O) adsorbed on stepped Pd(111) at 199 and 225 K, as well as aspects of its decomposition after heating to a maximum temperature of 415 K. Studies conducted on two substrates with relatively narrow and wide terraces show that furan preferentially adsorbs at step edge sites on both surfaces, while evidence of molecular diffusion is seen only on the narrower vicinal planes. After heating to 288 K, 300 K, and 415 K, evidence of reaction can be observed in occupied-states STM images. Our observations support a furan decomposition mechanism wherein the heterocycle preferentially adsorbs and reacts at upper step edge sites. Ab initio calculations based on Hohenberg-Kohn density functional theory (DFT) have been performed for several high-symmetry adsorption sites of furan on flat Pd(111) and show distortions of both the furan and the top layer of Pd upon adsorption.   

Evidence of surface reconstruction during inorganic crystal nucleation under Langmuir monolayers

When a carboxylic acid Langmuir monolayer is spread on aqueous solutions of lead and carbonate ions, an inorganic film nucleates at the organic surface. \textit{In situ} synchrotron x-ray diffraction studies show that the structure is that of hydrocerrusite (2Pb(CO$_{3})\cdot $Pb(OH)$_{2})$, oriented with the hexagonal basal planes parallel to the water surface. In addition, there are peaks from a $\surd $7 $\times \quad \surd $7 superstructure of the hydrocerrusite surface lattice. The organic monolayer unit cell contracts so as to form an epitaxial match with the superstructure. The Bragg rods of the supercell reflections reveal that the surface layer is $\sim $40 {\AA} (5-6 layers) thick, with vertical layer spacing close to that of hydrocerussite. Surface reconstruction phenomena have been observed previously only on very clean surfaces under ultrahigh vacuum   

Session A42: Artificially Structured or Self-Assembled Magnetic Materials - I

Large scale and high density regular array of magnetic quantum dots by nanosphere lithography

Magnetic quantum dots are single domain magnetic particles which have attracted intense study recently for their fundamental and technological properties. Large scale preparation of monodiapersed and similar shaped magnetic quantum dots remains a challenge at high area density. X-ray lithography is expansive to set up and have limited lateral resolution for nanometer sized magnetic quantum dots. Electron beam lithography technique is also high cost and inefficient tool. This has prompted a number of alternative methods based on self-organized structure such as self-assembled diblock copolymer (Science, Vol. 290, 2000). In this paper, we report the preparation of high density patterned magnetic dots of varying size and shape by nanosphere lithography. Although nanosphere lithography has been widely used to produce mesoscopic scale dot arrays, we find that one can not directly scale them down without modification because of the conformal deformation of polystyrene spheres we have employed. We have used ion beam modification technique to controllably reopen the pore structure, alloying size and shape selective deposition of regular array of magnetic nanoparticles between 20-50 nm in size. The magnetic and electrical characterization of the large scale array has been carried out by SQUID, VSM and magnetic force microscopy and the result will be reported as a function of structural, material and processing parameters.   

Fabrication and structural characterization of ordered magnetic nanodot arrays over large area

Self-assembly of nanopores in anodized alumina is of much interest as a controlled fabrication method of magnetic nanostructures for fundamental studies and potential magnetic recording applications. Up to 10 micron thick Al films are e-beam evaporated on N-type Si substrate for porous alumina mask fabrication. By controlling anodization conditions, hexagonally ordered pores with 8-125 nm diameter and 20-160 nm periodicity are formed over $\sim $1 cm$^{2}$ area. SEM and AFM characterization shows that the pores are distributed within $\sim $10{\%} standard deviation from the mean value. Fe magnetic nanodot arrays are fabricated by subsequent e-beam evaporation of Fe and mask lift-off. The smallest dot array fabricated this way is 44 nm, which corresponds to 0.4 Tbit/in$^{2}$ density. The nanodot periodicity is confirmed by small angle neutron scattering measurements. For nanoscale exchange bias studies, Fe/FeF$_{2}$ bilayer nanodot array are prepared using low angle Ar ion etching instead of the lift-off.   

Studies of Lithographically Defined Geometrical Frustrated Magnetic Networks

We used electron beam lithography to pattern Permalloy and MnAs thin films into honeycomb, triangular, and Kagome lattices of separated ferromagnetic nanodots. In these lattices, the dot size is around 100nm with varying separation. Micromagnetic simulation suggests that these lattices can experience magnetic frustration because of the incompatibility of the lattice symmetry and the dipole interaction among the ferromagnetic dots. These lattices have been studied by MFM, both under external field and in the remnant state. This research is supported by a grant from Army Research Office Grant (DAAD19-03-1-0236) and MRSEC at Penn State University.   

Structural and magnetic characterization of Co/NiFe dot arrays using soft x-ray resonant magnetic scattering

The field dependence of Co/NiFe dot arrays was studied in a layer-selective way, and structural and magnetic characterizations were conducted. Using x-ray resonant magnetic scattering (measured at the Ni/Co $L_3$ absorption edges), element-specific hysteresis loops were obtained from the NiFe and Co layers of patterned dot arrays. One dot-array sample was grown with an oxide layer between the NiFe and Co layers, and another was grown without it. The two arrays show different field dependence of NiFe magnetization. These results were compared with the hysteresis loops from single-layered NiFe and Co dot arrays. For structural and magnetic characterization of the dot arrays, soft x-ray resonant magnetic scattering measurements in reflectivity mode were performed with circularly polarized x-rays. The work at Argonne is supported by the U.S. DOE, Office of Basic Energy Sciences, under Contract No. W-31-109-ENG-38.   

Tailoring size effects on the exchange bias of ferromagnetic-antiferromagnetic nanodots

The dependences of the exchange bias effects on the antiferromagnetic (AFM) and ferromagnetic (FM) layers thicknesses have been investigated in continuous films and sub-100 nm dots composed of NiFe-IrMn bilayers. The nanostructures were prepared by sputtering the materials on prepatterned Si square dots. At room temperature, HE in continuous films decreases as the AFM layer thickness (tIrMn) increases, whereas HE in the nanodots remains rather constant. As a result, it is possible to either enhance or reduce HE in the nanostructures, with respect to continuous films, by varying tIrMn. Such a behavior is not observed when varying the FM layer thickness. An enhancement of the coercivity and a reduction of the blocking temperature in the dots are also observed. These effects are ascribed to the 3D confinement of the AFM spin structure and the concomitant enhanced thermal activation effects in the nanostructures. The influence of the dot size and temperature, together with atomistic simulation results, will also be presented.   

Spin waves in perpendicularly magnetized nanoscale permalloy dots

Square arrays of permalloy dots, 100 nm, 200 nm, 500 nm, and 1000 nm in diameter and 40 nm thickness, were created using electron-beam lithography. Ferromagnetic resonance measurements were made at 9.7 GHz with magnetic field was applied out of plane. In addition to the usual uniform mode, a manifold of lower field spin wave (SW) modes were observed for each sample. The spacing between the SW modes exhibits an inverse relationship with the dot diameter. Micromagnetic simulations performed using the open source program OOMMF, are in agreement with experimental results. The SW mode geometry is consistent with ``drum head'' type modes described by Bessel functions as described recently in other dot structures [1]. Supported by US DOE FG02-86ER45281 (MU). \begin{enumerate} \item G. N. Kakazei, et. al., APL, 85, 443 (2004). \end{enumerate}   

Ferromagnetic resonance force microscopy investigations of micron size permalloy dots

Low temperature ferromagnetic resonance force microscopy has been used to investigate the dynamic magnetic properties of permalloy disc arrays. Two dimensional scans of the sample at various external magnetic fields reveal the influence of the tip field and the array structure on the FMRFM-signal. Due to the high force sensitivity and the large signal to noise ratio of the ferromagnetic resonance force microscope individual dots with a 1.5\,$\mu m$ diameter and a center to center distance of 1.8\,$\mu m$ are readily resolved.   

Magnetism, NMR spectra, and optical transformations in Nafion with paramagnetic nano-structure

The bulk magnetization and the $^{19}$F NMR spectra of the ionomer Nafion as-received and doped with Mn$^{2+}$, Co$^{2+}$, Fe$^{2+}$, and Fe$^{3+ }$paramagnetic ions have been studied, with and without treatment in 1H-1,2,4-triazole. As-received Nafion is diamagnetic at 300 K but below 10 K it shows a small paramagnetic ``tail'' indicating the presence of a small amount of paramagnetic centers. Nafion doped with Mn, Co, and Fe ions shows clear paramagnetic behavior at 300 K, which depends on the type and amount of doping ion and treatment in triazole. $^{19}$F NMR spectra at 13 kHz magic angle spinning show differential increases in linewidth and spinning sidebands intensities. The observed changes in magnetism are determined by the spin state of the paramagnetic ions, while NMR reflects dipole-dipole and Fermi contact interactions between them and nearby fluorines of Nafion. Doping of Nafion with Fe and Co and treatment in triazole can strongly affect its color. The most pronounced effect is the temperature- induced reversible transformation of colorless Nafion to violet (Fe doped) and the reverse phenomenon (Co doped) observed simultaneously with the magnetic transformation just below room temperature. All observed phenomena are discussed in terms of the paramagnetic ions in the nano- structure of the Nafion matrix.   

Nanoparticles studied by magnetic speckles

Magnetic nanoparticles self assemblies are promising for advanced permanent magnetic applications [1]. The recent development of Soft X-Ray Resonant Magnetic Scattering (SXRMS) provides a very good tool to study magnetic order and reversal processes in such nanostructures. The chemical selectivity and the polarization sensitivity allows to probe the magnetic configuration, as shown by recent studies on superparamagnetic Co particle assemblies [2]. Moreover, by using coherent light and 2D detection one can obtain remarkable speckle patterns that are related to the local magnetic distribution [3,4]. In this work, we present spectroscopy measurments in circular polarization as well as SXRMS measurement performed in transmission geometry (small angle scattering) on Co and Fe3O4 nanoparticles assemblies. We recorded magnetic speckles in linear polarization at specific points tempertaure and magetic field. By studying the cross-correlation between the speckles patterns, we can measure the statistical evolution of the microscopic magnetic distribution through the superparamagnetic transition. [1] H. Zeng et al. Nature \textbf{240}, 395 (2002) [2] J.B. Kortright \textit{et al.}, Phys. Rev. B \textbf{70}, (2004) in press. [3] K.Chesnel \textit{et al}., Phys. Rev. B \textbf{66}, 172404 (2002) [4] M. S. Pierce \textit{et al.}, Phys. Rev. Lett. \textbf{90}, 175502 (2003).   

Quantitative studies of the vortex state in sub-100 nm magnetic nanodots.

Magnetism at nanoscale, when the size of the structures is comparable to or smaller than the ferromagnetic domain size, offers a great potential for new physics. Specifics of magnetic reversal in such structures are important for the high-density magnetic memory. Sub-100 nm magnetic dots are fabricated using self-assembled nanopores in anodized alumina [1]. Magnetization measurements performed using SQUID magnetometer indicate transition from a vortex to a single domain state for the Fe dots. This transition is studied as a function of the dot size and magnetic field. Monte Carlo and micromagnetic simulations confirm the experimental observations. Virgin curves measured at various temperatures indicate thermally activated vortex annihilation and nucleation. Quantitative analysis of the polarized neutron reflectometry in small angle geometry yields the vortex core size of $\sim $14 nm, in a good agreement with the 13 nm obtained from the simulations. Work supported by AFOSR. [1] Kai Liu et al., Applied Physics Letters \textbf{81}, 4434 (2002).   

Frustrated Vortices

Patterning soft magnetic materials into ring structures can give rise to a vortex state of the magnetization. When two rings interact strongly, i.e., through direct contact, the vortex states should have opposite chiralities. Thus, for three interacting rings there is an obvious frustration between the magnetic states. We have fabricated isolated and contiguous arrays of permalloy rings, with diameters of 1--4~$\mu$m, widths of 0.2--1.8~$\mu$m, and thickness of 15~nm. Their field dependent magnetization was investigated with magnetic force microscopy and magneto-optical Kerr effect measurements, accompanied by micromagnetic simulations. Generally, in isolated rings the magnetization reverses via nucleation and annihilation of a vortex state. However, in the case of three interconnected rings the magnetization reversal depends on the direction of the applied field. With the field along the midpoints of two of the rings the magnetization changes gradually, while for the field tangential to the connection between two of the rings vortices develop with opposite chiralities and the third ring remains in an onion state.   

Exchange Biased Vortices

Soft ferromagnetic discs with submicrometer diameter can reverse their magnetization via nucleation and annihilation of a vortex state. We prepared 400 nm diameter discs of Ni$_{80}$Fe$_{20}$/IrMn bilayers, where the exchange coupling between the ferromagnetic Ni$_{80}$Fe$_{20}$ and the antiferromagnetic IrMn modifies the magnetization reversal. Annealing of the Ni$_{80}$Fe$_{20}$/IrMn discs in an applied field establishes an exchange bias along the direction of the magnetic field during annealing. Magneto-optic Kerr effect measurements with the field applied along the exchange bias direction reveal a typical vortex hysteresis loop, which is now shifted with respect to zero field. Magnetic force microscopy in applied fields confirm that the reversal is via a vortex state. When the applied field is rotated with respect to the exchange bias direction the nucleation and annihilation fields reduce slightly, until at a critical angle of $\approx 80^{\circ}$, beyond which no vortex nucleation is observed. Micromagnetic simulations show that beyond the critical angle the magnetization reversal occurs via rotation of a C-state.   

Theoretical analysis of the transmission Phase Shift of a Quantum Dot in presence of Kondo correlations

We study the effects of Kondo correlations on the transmission phase shift of a quantum dot coupled to two leads. Experimental determination of the phase shift made by embedding a quantum dot in one of the arms of an Aharonov-Bohm interferometer leads to a value of the phase which differs from the well-known theoretical predictions. We propose here a theoretical interpretation of these results based on Bethe Ansatz calculations combined with the scattering theory. Quantitative agreement is obtained with experimental results both in the unitary limit and the weak Kondo coupling regimes.   

Kondo Effect in an Exactly Solvable Double Dot System with Singlet Groundstate

We analyze transport through a double dot system connected to two external leads. Each dot is treated as possessing a single active level. With symmetric couplings to the leads, only a single effective lead is available to interact with the double dot. We model the system through a generalization of the Anderson model which is exactly solvable via the Bethe ansatz provided certain, not particularly restrictive, constraints are placed upon the dot parameters. From this analysis we see that the zero temperature dot system at its particle-hole symmetric point favors singlet formation. Using exact solvability, we further determine how this double dot singlet evolves under increasing temperature and magnetic field together with changes of the gate voltage moving the dot-lead system away from particle-hole symmetry. We are able to analyze both the resulting transport signatures of this evolution and the effective energy scale governing the changes.   

Session A43: Focus Session: Spin Transfer Effect I

Time-Domain Measurements of Nanomagnet Dynamics Driven by Spin-Polarized Current

The transfer of spin angular momentum from a spin-polarized current to a nanomagnet exerts torque, and can cause the magnet's moment either to reverse its direction or to enter a state of steady precession. Exploiting a new nanoscale spin valve design, we make first time-resolved measurements of these dynamics. These measurements are made in Py(4 nm)/Cu(8 nm)/Py(4 nm)/IrMn(8 nm) nanopillar spin valves in which exchange bias is used to create a non-zero equilibrium angle between the magnetic moments of the free and fixed permalloy (Py) nanomagnets. In the regime of steady-state precession, the current-driven dynamics exhibit a high degree of coherence, as evidenced by long dephasing times ($\sim $10$^{2}$ ns). Measurements of the onset of the persistent precession in response to a current step demonstrate a fast ($\sim $ 1 ns) response of the nanomagnet to variations of the current. In the switching regime, time-resolved measurements demonstrate that spin-transfer-driven magnetization reversal in our samples is accomplished via a process of coherent precession. We also make time-resolved measurements of magnetic relaxation of the free Py nanomagnet excited by a short current pulse. These measurements, made as a function of spin-polarized current bias, demonstrate that the effective Gilbert damping parameter can be tuned by the spin-transfer torque. The value of the damping parameter in the limit of no current bias significantly exceeds the damping of an extended 4 nm thick Py film -- which is attributable to substantial spin pumping in the nanopillar structure. Our results demonstrate that coherent nanomagnet dynamics can be generated by spin-transfer torques in properly designed magnetic nanostructure devices and directly measured in both the time and frequency domains. This opens a wide range of opportunities for new types of fundamental studies of nanomagnet dynamics and for novel technological applications in the areas of high frequency communications and signal processing.   

Current-driven excitations in symmetric magnetic nanopillars

An electrical current was shown to induce spin waves and reversal of magnetization in a ferromagnet. A typical experiment on current-driven excitation of a ferromagnet involves two single-domain thin film magnets separated by a nonmagnetic spacer. One magnet is ‘hard’ and used to polarize the current while the spacer is thin enough for the polarized current to get through and excite the second ‘free’ magnet. The free layer is generally thin compared to the hard one thus marking an intrinsic asymmetry of the phenomenon, i.e., for initially parallel magnetizations of the two magnets the current-driven excitation occurs only when electrons flow from the free magnet to the fixed one. In the present work we study experimentally the current-driven excitations in symmetric Co/Cu/Co nanopillars. In contrast to all the previous observations where current of only one polarity is capable of exciting a multilayer system saturated by an externally applied magnetic field, we observe that both polarities of the applied current trigger excitations in a symmetric multilayer [Phys. Rev. Lett. 93, 36602 (2004)]. This may indicate that in symmetric structures the current propels high-frequency magnetic oscillations in all magnetic layers. We argue, however, that only one layer is excited in our multilayers but, interestingly, currents of opposite polarities excite different layers. This hypothesis is supported by modeling the spin accumulation in symmetric magnetic multilayers.   

Spin-transfer torque driven de-pinning of a domain-wall in a magnetic nano-wire

We present a theoretical study of the dynamics of a magnetic domain wall trapped in a potential well, driven by current-induced spin-transfer torque, in the presence of an external magnetic field. We show the existence of two regimes and two different mechanisms for the de-pinning of the domain wall using a one dimensional model. This model reveals that in small magnetic fields the critical current for de-pinning of the domain wall is completely independent of the pinning potential and weakly dependent on the magnetic field. Whereas, above some critical field, the critical current density is sensitive to the pinning potential and, moreover, strongly decreases with increasing field. Analytical expressions of the field dependence of the critical de-pinning current are derived for each regime, in the zero damping limit. The influence of adiabatic and non-adiabatic spin-transfer torques is also discussed. These theoretical predictions are compared to experimental results and to micromagnetic simulations.   

Domain wall motion driven by an electric current

We have recently proposed [1] that an electric current in a ferromagnetic film generates two mutually orthogonal spin torques, $\mbox{\boldmath $\tau$}_1 = b_J {\bf M}\times ({\bf M} \times \frac{\partial \bf M}{\partial x})$ and $\mbox{\boldmath $\tau$}_2 = c_J {\bf M}\times \frac{\partial \bf M}{\partial x} $ where ${\bf M}$ is the magnetization vector and the constants $b_J$ and $c_J$ are proportional to the current density. By including these two spin torques in the Landau-Lifshitz-Gilbert equation; we have simulated the domain motion in a number of experimentally accessible geometries. We have found that the current-driven domain wall motion displays many unique features compared to that driven by an external field. One particular example is to predict the critical current as a function of the applied magnetic field in a ``constriction'' geometry where the domain wall is originally trapped before applying an electric current. The calculated critical current densities are compared to the existing experimental data. [1] S. Zhang and Z. Li, Phys. Rev. Lett. {\bf 93}, 127204 (2004).   

Current-Driven Domain Wall Motion in Permalloy Nanowires

The current-driven motion of magnetic domain walls (DW) in permalloy (Ni$_{81}$Fe$_{19})$ nanowires is discussed. The nanowires were fabricated by electron-beam lithography from permalloy films 10 to 40 nm thick. The wire lengths and widths were varied from 2 to 10 $\mu $m, and from 70 to 300 nm, respectively. DWs are injected into the nanowires using magnetic fields, generated either by a large electromagnet or locally by using micron-wide gold wires fabricated above and transverse to the permalloy nanowires. The injected DWs are trapped at triangularly-shaped notches fabricated along one or both edges of the nanowires. Current-driven DW motion is probed using anisotropic magnetoresistance measurements and magnetic force microscopy (MFM). DWs can be moved in the absence of any external magnetic field by current pulses, varying in length from nano- to micro- seconds. Current densities of the order of 10$^{8}$ A/cm$^{2 }$are needed. MFM images show unambiguously that DWs can be moved from one notch to another, in either direction along the nanowire, depending on the current pulse polarity, intensity and duration. The dependence of the critical current density required to move the DWs between notches on the nanowire width and notch shape and size will be discussed.   

Mobility of field and current-driven domain walls in magnetic nanowires

Recent experiments [1] have demonstrated domain wall displacements in magnetic nanowires resulting from the injection of a spin-polarized current across the wall. These observations have been attributed to spin-momentum transfer. We present direct measurements of domain wall velocities in focused ion beam etched Permalloy nanowires using high spatial resolution scanning Kerr polarimetry with high-bandwidth detection ($<$2 ns risetime). The present experiments provide instantaneous measures of domain wall velocity during the entire course of wall propagation, in contrast to the average velocities determined in previous displacement studies [2]. We will report on field-driven domain wall velocity profiles and mobilities in magnetic nanowires, and the influence of dc spin currents on these dynamic quantities. The relation to various models will be discussed. Supported by NSF-DMR-0404252 and the R. A. Welch Foundation.\newline \newline[1] M. Tsoi, et. al, Appl. Phys. Lett. 83, 2617 (2003)\newline [2] A. Yamaguchi, et. al, Phys. Rev. Lett. 92, 077205 (2004)   

SEMPA measurements of trapped domain walls in thin film nanoconstrictions.

We used scanning electron microscopy with polarization analysis (SEMPA) to image the magnetic nanostructure of domain walls trapped in patterned NiFe thin film nanoconstrictions. Currents were applied to the structures to induce spin torque driven motion of the domain wall in the nanoconstriction. Various film geometries were investigated in order to understand how the size and shape of the constriction affects the magnetic nanostructure of the domain wall. The structures were fabricated using electron beam lithography. Constriction widths varied from 40 nm to 200 nm. The non-invasive nature of SEMPA allowed successful imaging of the unperturbed, remanent state of the trapped domain walls. In 10 nm thick NiFe films, all of the observed trapped walls (within the constriction) were of the transverse type, and the domain wall widths were strongly dependent on both the width of the constriction (approximately equal to the width) as well as the shape of the constriction. Because SEMPA directly measures the magnetization direction, the image data allows meaningful quantitative comparisons to micromagnetic calculations. Simulations with inserted domain walls show good agreement with the behavior of the domain walls observed by SEMPA. Detailed comparisons will be presented. *Work supported in part by the Office of Naval Research   

Nanometer Scale Observation of Current-Induced Narrow Domain Wall Depinning in Perpendicular Spin Valves

Until now, current driven domain wall (DW) motion in magnetic wires has been experimentally studied for in-plane magnetized films. Since the DW width is large ($\sim $100 nm), only the adiabatic limit in which the current polarization follows the magnetization direction has been studied. Also, this wide DW masks any local variation in the pinning potential, thus making it difficult to probe the depinning process on a nanometer scale. Here, we report the first quantitative study of the depinning of a 1D narrow DW under a current. We use a 12nm wide Bloch DW in wires based on spin valves with perpendicular magnetic anisotropy. High sensitive electrical measurements allow us to observe current-induced DW motion between pinned sites separated by 10 nm. In spite of the strong pinning potential and narrow DW, a low critical current density of the order of 1x 10$^{7}$ A/cm$^{2}$ is found. The study of the depinning process emphasizes the crucial role thermal fluctuations and the pinning potential play in current induced DW motion process.   

Interaction of Spin-Transfer Oscillators With AC Currents and Fields

We have shown previously that a DC current flowing through a 40 nm point contact made to a spin valve structure induces high frequency, large-angle coherent magnetic precession.~ The precession frequency ranges from 5-40+ GHz, and is a strong function of the field magnitude and direction, and the current.~ Injection of an additional AC current into the device produces frequency modulation for low injected frequencies, and "injection locking" for injection frequencies near the resonant frequency.~ We will present a detailed analysis of the injection locking process, highlighting the similarities and differences with conventional oscillators, and using the injection locking as a means of determining the precession amplitude at the device.~ In addition, we will show results on the effect of AC magnetic fields on the spin-transfer resonance.~ Comparison with models to describe the effects on the magnetization trajectory will be made.   

Current-Induced High Frequency Excitations in Py-based Nanopillars

To study how high frequency excitations induced by high current densities in ferromagnetic/non-magnetic/ferromagnetic (F/N/F) nanopillars vary with applied magnetic field H and current I, we have assembled a system containing 40 GHz picoprobes, a 12 GHz spectrum analyzer, and a 40 GHz Microwave-generator-based system that can be used as a spectrum analyzer up to 40 GHz. Our first measurements, on Permalloy (Py = Ni(84)Fe(16))-based nanopillars of the form Cu(80nm)/Py(30nm)/Cu(10nm)/Py(6nm)/Cu(5nm)/Au(200nm), yielded peaks at frequencies in the range 1 to 2 GHz. We will describe how the frequencies and heights of these peaks vary with H and I.   

Bipolar High Field Excitations in Co/Cu/Co Nanopillar Junctions

Spin transfer has been studied in Co/Cu/Co pillar devices (PD) in large fields applied perpendicular to the layers and as a function of magnetic layer thickness. Sub-100 nm size junctions have been fabricated by means of a nano-stencil mask process in combination with an in-situ wedge growth mechanism. The junctions consist of a thick`fixed' Co layer and thin (0.5 to 3 nm) `free' Co layer. At high current densities excitations, which lead to a decrease in junction resistance, are observed for both polarities of the current [1]. Our results suggest that current-induced excitation of the magnetization can lead to a lower resistance state than that of a state of static parallel alignment of the layers. Intrinsic asymmetries of bilayer junctions in conjunction with lead asymmetries cause a strong asymmetry in the longitudinal spin accumulation (LSA). Recently it has been found that at high current densities such asymmetries in the LSA can cause non-uniform spin- wave excitations even in PDs with only a single ferromagnetic layer [2]. Here we compare the thickness dependence of these additional excitations in single layer junction with that of the free layer thickness dependence of the bilayer junctions. \\[4pt] [1] B. \"{O}zyilmaz et al. cond-mat/0407210.\\[0pt] [2] PRL,93, 176604 (2004).   

High Speed Spin-Transfer Switching Behavior of Low Critical Current Spin Valve Nanopillars

For spin transfer writing to be effective for MRAM, the integration of a magnetic device with a scaled CMOS transistor in a memory cell requires that $I_{c}$ for switching a thin, thermally stable element on ns time scales be $<<$ 1 mA. Since $I_{c}$ scales with the volume of the magnetic element and the square of its saturation magnetization $M_{S}$, the use of very small free layers with low $M_{S}$ can result in low $I_{c}$'s. The challenge is obtaining a large enough magnetic anisotropy to ensure thermal stability at $\sim $100 C. We have fabricated 40x120 nm elliptical Py/Cu/Py nanopillar spin valves exhibiting free layer coercive fields in accord with 3-D micromagnetic modeling. For a 4.5 nm thick free layer device, currents necessary for 100{\%} switching go from 0.6 mA for a 10 ns pulse, where thermal activation aids switching, to 2 mA for a 1 ns pulse, where there is insufficient time for thermal fluctuations and $I_{c}$ is set by the current required to transfer enough spin into the free layer to force its reversal. We will discuss the switching mechanisms of these devices in the ns regime, and our progress towards achieving fully stable devices with low $I_{c}$'s.