Session P1: Electron Transport in Single Molecules

Electrochemical gate-controlled conductance of single molecules

The ability to measure and control current through a single molecule is a basic requirement towards the ultimate goal of building an electronic device using single molecules. This ability also provides one with a rather unique opportunity to study charge transport, a phenomenon that plays vital roles in many chemical, electrochemical and biological processes, on a single molecule basis. To reliably measure the current, one must: 1) provide a reproducible contact between the molecule and two probing electrodes; 2) find a signature to identify that the measured conductance is due to not only the sample molecules but also a \textit{single} sample molecule; 3) provide a third gate electrode to control the current. The method that we have used to create individual molecular junctions is to bring two electrodes into and out of contact with each other in the presence of sample molecules terminated proper linkers that can bind covalently to the electrodes. The individually created molecular junctions vary in the atomic scale details of the contact configurations, and statistical analysis is used to extract the conductance of the molecular junction with the most probable configuration. When several configurations occur with comparable probabilities, the method may result in multiple conductance values. In order to control the current through a molecule, we use an electrochemical gate in which the molecular junction is immersed in an electrolyte and biased with respect to a reference. We have studied three types of molecules: electrochemically inactive molecules, electroactive molecules that undergo irreversible redox reactions, and electroactive molecules that undergo reversible redox reactions. These molecular systems exhibit rather different electrochemical gating behaviors.   

Electron Transport in Molecular Transistors

We have fabricated molecular transistors by depositing molecules between nanometer-spaced electrodes created via electromigration. Electron transport in these devices is dominated by the single-electron tunneling effect. Several examples will be discussed including (1) excitations of intramolecule vibrations in single trimetal-molecule transistors, (2) room-temperature single-electron tunneling transistors using alkanedithiols – here the transport occurs through ultrasmall Au nanoparticles spontaneously formed during thiol assembly, and (3) Kondo resonance and co-tunneling behavior in metal-porphyrin and expanded-porphyrin molecule transistors.   

Electron-vibron coupling in single molecule transistors

I discuss coupling between electron transport and vibrational degrees of freedom in single-molecule and nanotube systems. The coupling gives rise several effects such as sidebands in the differential conductance, rectification due to polaron formation,and interesting interplay with the Kondo resonance.   

Field Regulation of Single Molecule Conductivity by a Charged Atom

A new concept for a single molecule transistor is demonstrated [1]. A single chargeable atom adjacent to a molecule shifts molecular energy levels into alignment with electrode levels, thereby gating current through the molecule. Seemingly paradoxically, the silicon substrate to which the molecule is covalently attached provides 2, not 1, effective contacts to the molecule. This is achieved because the single charged silicon atom is at a substantially different potential than the remainder of the substrate. Charge localization at one dangling bond is ensured by covalently capping all other surface atoms. Dopant level control and local Fermi level control can change the charge state of that atom. The same configuration is shown to be an effective transducer to an electrical signal of a single molecule detection event. Because the charged atom induced shifting results in conductivity changes of substantial magnitude, these effects are easily observed at room temperature. [1] Paul G. Piva1,Gino A. DiLabio, Jason L. Pitters, Janik Zikovsky, Moh'd Rezeq, Stanislav Dogel, Werner A. Hofer {\&} Robert A. Wolkow, Field regulation of single-molecule conductivity by a charged surface atom, NATURE \textbf{435}, 658-661 (2005)   

Session P2: The Electronic Properties of Overdoped Cuprates: The Clean Gateway to High-Tc Superconductivity

The origin of anomalous transport in a high temperature superconductor

The metallic state of high-temperature superconductors is anomalous in that the Hall coefficient is strongly temperature dependent while the resistivity varies linearly in temperature over a wide temperature range. Although this $T$-linear resistivity gradually weakens with doping, crucially it survives until superconductivity is destroyed. Both the superconducting pairing interaction and the origin of this anomalous transport have yet to be determined, though most theoretical approaches consider them to be intrinsically linked. Through novel analysis of polar angular magnetoresistance oscillations, we have succeeded to determine the full temperature and momentum dependence of the mean free path of the charge carriers in highly doped Tl$_{2}$Ba$_{2}$CuO$_{6+\delta }$ ($T_{c}$ = 15K) up to 60K. From this, we have been able to identify the origin of the $T$-linear resistivity \textit{and }the temperature dependence of the Hall coefficient for this particular compound. Given the correlation between the appearance of the $T$-linear resistivity and the onset of superconductivity, this additional scattering is also a prime candidate for the pairing mechanism for high temperature superconductivity itself\textbf{. }   

Nodal-antinodal quasiparticle anisotropy reversal in the overdoped high-T$_{c}$ cuprates.

The cuprate superconductors can be tuned through a remarkable progression of states of matter by doping charge carriers into CuO$_{2}$ planes. The most generic feature of this tuning is a sequence from a Mott antiferromagnetic insulator, to the d-wave superconductor at intermediate doping, and eventually to an overdoped metal which is widely believed to be described by Fermi liquid theory. Of these three, the testing of Fermi liquid theory in the overdoped regime has been particularly hampered by a lack of compounds suitable for a wide range of experimental techniques. Important breakthroughs could come from the study of Tl$_{2}$Ba$_{2}$CuO$_{6+\delta }$ (Tl2201), a clean and structurally simple system with a very high T$_{c}$, whose natural doping range extends from optimal to extreme overdoping as one varies the oxygen content. Recent success in high-purity single crystal growth [1] gave us the opportunity of performing the first extensive ARPES study of the low-energy electronic structure of heavily overdoped Tl2201, which reveals a novel phenomenology: contrary to the case of under and optimally-doped cuprates, quasiparticles are sharp near ($\pi $,0), i.e. the antinodal region where the gap is maximum, and broad at ($\pi $/2, $\pi $/2), i.e. the nodal region where the gap vanishes [1,2]. This reversal of the nodal-antinodal quasiparticle anisotropy across optimal doping and its relevance to scattering, many-body, and quantum-critical phenomena in the high-Tc cuprate superconductors, is discussed. [1] D.C. Peets \textit{et al}., cond-mat/0211028 (2002); [2] M. Plat\'{e} \textit{et al}., PRL \textbf{95}, 077001 (2005).   

Scanning tunneling spectroscopy studies of Bi$_{2}$Sr$_{2}$CaCu$_{2}$O$_{8+x}$ from the strongly underdoped to strongly overdoped regime

Using atomically resolved scanning tunneling microscopy (STS), we investigate the electronic structure Bi$_{2}$Sr$_{2}$CaCu$_{2}$O$_{8+x}$ across a range of doping levels from x $\sim $ 0.1 up to as high as $\sim $0.23, with significant changes in electronic structure observed above p$\sim $0.21. New sample preparation processes [1] were used to produce heavily overdoped crystals suitable for the imaging of various forms of electronic heterogeneity. The evolution of the gap map $\Delta $(r), coherence peak height map A(r), the inelastic tunneling signatures $\omega $(r), and the quasiparticle interference LDOS modulations, as well as their interrelations across this range of doping levels, will be presented. \newline \newline Additional authors: J. Lee, M. Wang, Laboratory of Atomic and Solid State Physics, Department of Physics, Cornell University, Ithaca, NY 14853, U.S.A; K. Fujita, Department of Advanced Materials Science, University of Tokyo, Tokyo 113-0033, Japan; H. Eisaki, National Institute of Advanced Industrial Science and Technology (AIST), 1-1-1 Central 2, Umezono, Tsukuba, Ibaraki 305-8568; S. Uchida, Department of Physics, University of Tokyo, Tokyo 113-0033; and J. C. Davis, Laboratory of Atomic and Solid State Physics, Department of Physics, Cornell University. \newline \newline [1] J. Slezak, K. Fujita, J. C. Davis, in preparation (2005)   

Evolution of superconducting gap and metallic ground state in cuprates from transport

We report on fundamental characteristics of the ground state of cuprates in the limit of T=0, for both normal and superconducting states, obtained from transport measurements on high-quality single crystals of YBCO and Tl-2201, as a function of hole concentration. The superconducting gap is extracted from thermal conductivity; it is found to scale with the superconducting transition temperature throughout the overdoped regime, with a gap-to-Tc ratio of 5 [1]. The normal state is accessed by suppressing superconductivity with magnetic fields up to 60 T and is characterized by the limiting behavior of its electrical resistivity; while carrier localization is observed in YBCO at low temperature for carrier concentrations p below 0.1 hole/planar Cu, at p=0.1 and above the material remains highly metallic down to T=0 [2]. This shows that the non-superconducting state of underdoped cuprates, deep in the pseudogap phase, is remarkably similar to that of strongly overdoped cuprates, e.g. at p=0.3. We compare these results with similar measurements on other cuprates and discuss their implication for our understanding of the cuprate phase diagram. [1] In collaboration with: D.G. Hawthorn, S.Y. Li, M. Sutherland, E. Boaknin, R.W. Hill, C. Proust, F. Ronning, M. Tanatar, J. Paglione, D. Peets, R. Liang, D.A. Bonn, W.N. Hardy, and N.N. Kolesnikov. [2] In collaboration with: C. Proust, M. Sutherland, N. Doiron- Leyraud, S.Y. Li, R. Liang, D.A. Bonn, W.N. Hardy, N.E. Hussey, S. Adachi, S. Tajima, J. Levallois, and M. Narbone.   

Optical properties of Cuprates in the Normal and superconducting state.

For superconducting materials it is interesting and important to determine the kinetic energy of the conduction electrons, $<<$H$>>_{T}$, because its behavior as a function of temperature dependence, in particular at the superconducting phase transition, provides a direct and profound insight in the mechanisms by which the superconducting phase is stabilized. The intra-band optical spectral weight, W(T), is, apart from a minus sign, closely related to the kinetic energy[1]. With modern optical techniques it is possible to measure W(T) very accurately as a function of temperature. Over the past few years several teams have reported that by the superconducting phase transition affects the optical conductivity over an energy range of several electron Volts[2-8]. Some of these results were accurate enough to determine the effect of superconductivity on W(T). Here we present new optical data for a large number of underdoped and optimally doped samples of various compositions. In order to clearly distinguish the effect of the superconducting phase transition from other temperature dependencies, we use a dense sampling of temperatures (1 spectrum every Kelvin) over a broad range of temperatures and frequencies. All our data support that the change at Tc of W(T) parallel to the CuO2-planes is opposite to the trend expected from the BCS prediction. For strongly overdoped samples the observed behavior of W(T) in the normal state and in the superconducting state is qualitatively different compared to underdoped and optimally doped superconductors. \newline \newline [1] P- F. Maldague, Phys. Rev. 16, 2437 (1977). \newline [2] M. J. Holcomb, et al., Phys. Rev. Lett. 73, 2360 (1994). \newline [3] D. N. Basov et al., Science 283, 49 (1999). \newline [4] H. J. A. Molegraaf et al., Science 295 (2002) 2239 \newline [5] A. F. Santander-Syro et al., Europhys. Lett 62 (2003) 568 \newline [6] V. Boris et al., Science 304, 708 (2004). \newline [7] B. Kuzmenko et al., Phys. Rev. B 72 (2005) 144503 \newline [8] M. Ortolani et al., Phys. Rev. Lett. 94 (2005) 067002.   

Session P3: Physics Teacher Preparation at a Crisis: Innovative Programs Addressing a National Need

Improving the preparation of K-12 teachers: Contributions from physics education research

Physics education research can contribute to efforts by college and university faculty to improve the preparation of K-12 teachers to teach physics and physical science. Examples are used to demonstrate the need to help teachers deepen their understanding of basic topics and to illustrate how a research-based curriculum can assist in this process. Evidence is presented of the impact on student learning in K-12 classrooms.   

UTeach: Secondary Teacher Preparation in Science and Mathematics at the University of Texas at Austin

The UTeach Program is a joint effort of the College of Natural Sciences, the College of Education and the Austin Independent School District to recruit, prepare and support math and science teachers for the State of Texas. UTeach uses early and on-going field experiences to capture the imagination of preservice teachers and provide a foundation for more advanced pedagogical courses. With over 400 students enrolled and over 80 graduates per year, UTeach is one of the largest programs producing secondary science, mathematics, and computer science teachers in the nation. Most UTeach students are undergraduates, but around 10\% are people of many ages with strong backgrounds in mathematics or science who have decided to enter teaching. Hallmarks include: \begin{itemize} \item Four-year degree plans that enable undergraduates to obtain certification at no cost in time or money. \item Active recruitment and support including tuition reimbursement, paid internships, personal advising, and guidance by master teachers. \item Emphasis on preparing teachers who will be knowledgeable of their discipline, experienced with involving students in scientific inquiry, and practiced in employing new technologies to enhance student learning. \item A revised, streamlined professional education sequence drawing on research on learning, standards-based curricula, multiple forms of assessment, and proven strategies for achieving equity and integrating technology into math and science education. \item Program flexibility with multiple entry points (from freshman to post baccalaureate), integrated degree plans, and proficiency-based assessment, including the development of individual teaching portfolios. \end{itemize} For more information on UTeach, see \texttt{http://uteach.utexas.edu}   

Better prepared future teachers = better physics department!

A more scientifically literate society benefits physics as a profession. It is best realized by better serving all undergraduate physics students. Arguably, the most important are future K-12 teachers. In better-serving all students, the department also benefits. University of Arkansas, Fayetteville has seen a drastic change in number of majors, the number of students active in research and the number of graduates pursuing graduate work while also increasing the number of majors who decide to teach. What works to build these numbers and strengthen these resources at Arkansas will be discussed, with additional examples from other members of the growing Coalition of institutions that are seeking to improve and promote physics and physical science teacher education within physics departments. This group, the Physics Teacher Education Coalition (www.PTEC.org), is bringing together innovative ideas and practices throughout the country to help meet the critical shortage of well prepared and actively supported teachers. The program will be described and information provided for those interested in taking advantage of these efforts.   

Sustainable and Scalable Reforms in Physics Education: Research studies from Colorado PhysTEC

and STEVEN POLLOCK University of Colorado at Boulder -- While many practices developed within the physics education research community have been demonstrated as successful, they respond to calls and employ practices that echo efforts from the early part of the 20th Century.~Are we bound to the same limited success as these precursors?~We examine what it means to replicate proven reforms and to develop models for sustainable implementation of these reforms.~As part of the Colorado Physics Teacher Education Coalition, we have implemented the Tutorials in Introductory Physics, which were developed by researchers at the University of Washington. We present research on the successful implementation of these reforms at the University of Colorado and begin to answer the questions: What does it mean to replicate an educational program? and How might these educational transformations be sustained? We present empirical data on the success of reforms and the fidelity of implementation as well as theoretical frames for analyzing these data. We also present a model (the Learning Assistant program) designed for sustaining these reforms and for increasing student interest and retention in teaching.   

The AAPT/PTRA Program: Professional Development for Pre-College Physics Teachers Hosted by College and University Physics Departments

The American Association of Physics Teacher's Physics Teaching Resource Agents (AAPT/PTRA) program has a twenty year history of providing professional development for in-service pre-college physics and physical science teachers. More than 500 teachers have been prepared through NSF-funded summer institutes to provide professional development for their peers in a wide variety of venues ranging from urban, inner-city classrooms to classrooms in low population rural areas. A wide variety of inquiry-based, active engagement workshops have been developed that can assist in-service teachers at all experience and preparation levels, from new and crossover teachers to those who have taught for many years. AAPT/PTRA presenters are typically active physics teachers who share lessons learned in their classrooms on how to adopt research-based practices. College and university physics and astronomy departments interested in providing in-service professional development for pre-college teachers in their geographic areas can enter into agreement with the AAPT/PTRA program to utilize the services of these trained professional development providers. United States Department of Education Math and Science Partnership funds that are allocated to each state are an excellent source of funds that physics and astronomy departments can use to support this type of professional development for physics and physical science teachers. An example of a funded program currently in place in Texas will be presented. AAPT/PTRA program is currently funded by NSF grant {\#} ESI-0138617.   

Session P4: Keithley Award Session

Nanocalorimetry: Using Si-micromachined Devices for Thermodynamic Measurements of Thin Films and Tiny Crystals

We have used Si micromachining to fabricate membrane-based calorimeters for measuring thermodynamic properties of microgram-quantity samples over a temperature range from 1.7 to 550K in magnetic fields to 8T. Prototype scaled down devices have been made which allow precise measurements of nanogram quantities. Different types of thermometers are used for different purposes and in different temperature ranges. Current development efforts are extending the temperature range to 0.3 - 800K, and we are collaborating with the national high magnetic field lab to extend the field range to 65T in pulsed magnets. These devices are particularly useful for specific heat measurements of thin film samples (100-400 nm thick) deposited directly onto the membrane through a Si micromachined evaporation mask. They have also been used for small bulk samples attached by conducting paint or In, and for powder samples dissolved in a solvent and dropped onto devices. The measurement technique used (relaxation method) is particularly suited to high fields because thermal conductance is measured in zero field and is field independent, while the relaxation time constant does not depend on thermometer calibration. The devices have been used with little modification for thermal conductivity and thermopower measurements, and are well suited to measurements of calorimetric signals such as those occurring at phase transitions or under irreversible thermal behavior. I will discuss device fabrication and thermal analysis which allow us to precisely identify heat flow in the devices and consequent limits on the absolute accuracy, as well as possible future directions for device development. I will also briefly discuss examples of measurements on several materials of current interest: 1) amorphous Si and its alloys, 2) high precision critical temperature studies of La$_{1-x}$Sr$_{x}$MnO$_{3}$ and La$_{1-x}$Ca$_{x}$MnO$_{3}$, 3) antiferromagnetic CoO nanoparticles and thin layers, 4) Fe/Cr giant magnetoresistance multilayers.   

High-Resolution Microcalorimeter Detectors for X-ray Spectroscopy

For many decades, standard wavelength- and energy-dispersive x-ray detectors have dominated experimental physics. Recently, microcalorimeter detectors of various types that count individual photons have started to make an appearance on the experimental scene. I shall describe the development of Transition Edge Sensor (TES) detectors at NIST, with particular emphasis on their use in high-resolution x-ray spectrometry. These detectors combine the broad energy range of a SiLi or Ge detector with energy resolution approaching that of diffraction-related methods. The technologies for producing ultra-cold temperatures and for fabricating superconducting electronics have made these detectors practical to use on a daily basis. By means of careful matching of absorber and energy, detectors can be built to cover energy ranges from 1 keV to 100 keV. By extrapolating from single-pixel detectors to arrays, the possibility of large detection areas, high count rates, and even imaging is starting to look realistic. While embodying some challenging technical constraints, microcalorimeter x-ray detectors will provide attractive advantages and opportunities for physicists in a number of fields.   

Angle-Resolved High Field Low Temperature Calorimetric Measurements of Low Dimensional Materials

Quasi-two-dimensional materials exhibit a rich variety of magnetic-field-induced superconducting and magnetic states. These states are highly anisotropic with respect to magnetic field orientation; in some cases, the very existence of the state is field angle dependent. To establish the phase boundaries of these high-field, low temperature, angle-dependent states, we have fabricated miniature rotatable calorimeters for measurements of specific heat and the magnetocaloric effect at temperatures ranging from 0.1K to 20K in magnetic fields up to 20, 35 or 45 tesla. The sample orientation relative to the applied field can be continuously varied at low temperature along a single axis (with a resolution of 0.02 degrees) and at room temperature along a second axis (with a resolution of 2 degrees). The sample temperature can be programatically set and regulated to better than 0.1 percent over the entire field and temperature range, allowing field sweeps at constant temperature in addition to temperature sweeps at fixed fields. In this talk, I will discuss the design, performance, and evolution of our calorimeter and recently obtained results, including the calorimetric observation of an angle-dependent magnetically enhanced FFLO superconducting state in a heavy fermion superconductor and an angle-dependent quantum fluctuation induced ``plateau state'' at 1/3 of the saturation magnetization in a quasi 2D S = 1/2 Heisenberg antiferromagnet.   

Some non-traditional approaches to thermal and thermodynamic measurements

Three non-traditional measurement methods for measurement of thermal and thermodynamic quantities are explored. Each method is not commonly used, has some astonishing advantages, and produces outstanding accuracy with little chance for error for several reasons. One reason in common is that for each method, only one quantity is measured. The methods include noise spectroscopy for the measurement of the elastic tensor of solids (and maybe other states of matter), third-harmonic measurements of thermal conductivity and specific heat, and impulse methods for obtaining the difficult-to-acquire ``ZT'' for thermoelectrics.   

Session P5: Shedding Light on the Enigma of the Transition to Turbulence in Pipes and other Shear Flows

The Transition to and from Turbulence in a pipe

A discussion of experimental investigations of the stability of flow along a pipe will be given. The transition to turbulence is catastrophic when a well--defined amplitude of injected perturbations is exceeded. The stability threshold scales inversely proportional to the Reynolds number, $Re$, with a sharp cut off at low $Re$ values. On the other hand, the decay from the turbulent state exhibits systematic exponential behavior with diverging timescales which are indicative of critical behavior. The long transients contain spatio-temporal coherence which suggest connections with recent theoretical developments.   

Transient growth and subcritical transition in shear flows

The possibility for disturbance growth in shear flows which are linearly stable is discussed, and it is shown that a necessary condition is that the underlying linear operator is non-normal, i.e. that it is associated with non-orthogonal eigenfunctions. Since the non-linear terms are conservative it is only by utilizing linear growth mechanisms associated with the non-normal linearized operator that energy growth is possible also for subcritical finite amplitude disturbances. The non-normal effects are manifested in the possibility for large transient growth of the disturbance energy, large response to forcing and large sensitivity of the eigenvalues. The optimal transient growth and response to forcing are calculated as the norm of the matrix exponential and resolvent, respectively. The optimal disturbances are streamwise vortices and the optimal responses are streaks of high and low velocity in the streamwise direction. These flow structures are prevalent in all subcritical transitional shear flows, including pipes and channels. It is shown by direct numerical simulations that transition scenarios initiated by the optimal disturbances have low transition thresholds. The dependence of the thresholds on the Reynolds number is also presented. Finally, extensions of the transient growth concept to more complex flows are discussed and examples of its use given.   

Self-Sustaining Process and Exact Coherent Structures in Shear Flows

The Self-Sustaining Process (SSP) is a weakly nonlinear theory of a fundamental three-dimensional nonlinear process in shear flows. It is the basic mechanism that enables enhanced momentum transport and the redistribution of the mean shear energy into smaller scales and, ultimately, turbulent motions. I will briefly review the 40 years of observations of streaks and coherent structures in the near wall region of turbulent shear flows that led to the formulation of the SSP theory. A primary impact of the SSP, besides providing some level of mechanistic understanding, has been to provide a method to calculate unstable traveling wave solutions of the Navier-Stokes equations. This approach has now been successfully carried out in all canonical wall-bounded shear flows with stress as well as velocity boundary conditions. The traveling waves thus obtained show striking similarity with the observed near-wall coherent structures, earning them the name of `exact coherent structures'. Furthermore, those unstable waves have been shown to capture basic statistics of turbulent flows remarkably well, thereby providing hope for a quantitative theory of turbulence over smooth walls. The traveling waves come in many kinds: small scales, large scales and multi-scales. The asymptotics of the large scale traveling waves as the Reynolds number goes to infinity is remarkably simple and confirms the asymptotic validity of the SSP. These large scale coherent states may yield a new promising target for the control of turbulence in shear flows.   

Travelling waves in pipe flow and their relevance for transition to turbulence

The problem of understanding the nature of pressure-driven fluid flow through a circular straight pipe remains one of the oldest problems in fluid mechanics. The steady, unidirectional parabolic (laminar) flow solution named after Hagen (1839) and Poiseuille (1840) is linearly stable yet temporally and spatially disordered 3-dimensional (turbulent) solutions can easily be triggered at sufficiently large flow rates (Reynolds 1883). In contrast with Rayleigh-Benard convection where transition to turbulence proceeds along an orderly sequence of bifurcations at well-defined values of the thermal driving, the transition in a pipe is abrupt, dependent on the level of ambient disturbances in the system and, at least close to the threshold flow rate, transient. The recent discovery of travelling wave solutions (which represent saddle points in phase space) in this system has at last provided a theoretical stepping stone towards rationalizing the transition process. We will discuss the structure of these waves as well as evidence of their relevance during the transition process.   

Two scenarios for dynamics of perturbations in pipe Poiseuille flow

Two experiments on perturbations in circular pipe flows and their possible theoretical interpretations are discussed to illustrate complexity of the problem. The experimental data by A. Kaskel (1961) are discussed within the framework of spatial transient growth theory, and we argue that the phenomenon of transient growth was observed in the pipe-flow experiments. Another experiment (Eliahou et al, 1998) illustrates how weak streamwise vortices provide instability of the secondary disturbances which, in turn, amplify the steady vertical structures. The latter is consistent with the self-sustaining process scenario. These examples, and more recent DNS and experimental studies represent typical controversies that arise in the study of complex systems.   

Edge of chaos in the transition to turbulence

We study the boundary of the laminar region near the onset of turbulence. Approaching the boundary from the laminar side, the lifetime of perturbations increases, diverges when the boundary is reached, and varies chaotically for larger amplitudes. In the chaotic region, lifetimes vary sensitively with amplitude, consistent with the strange saddle picture of the turbulence proposed earlier. The trajectory on the edge between the laminar and chaotic regions is asymptotic to a single well defined state, essentially independent of the type of perturbation. The edge then becomes the stable manifold of this structure. In the case of a model shear flow, the edge states are simple or period doubled or chaotic trajectories. The case of pipe flow shows less variability and the edge state seems to remain close to a state with simple vortical structure. \\ This is joint work with T.M. Schneider (U Marburg), J.D. Skufca (Clarkson U) and J. Yorke (U Maryland).   

Session P6: Quantum Spin Dynamics in Molecular Nanomagnets

Universal Mechanism of Spin Relaxation in Solids

Conventional elastic theory ignores internal local twists and torques. Meantime, spin-lattice relaxation is inherently coupled with local elastic twists through conservation of the total angular momentum (spin + lattice). This coupling gives universal lower bound (free of fitting parameters) on the relaxation of the atomic or molecular spin in a solid [1] and on the relaxation of the electron spin in a quantum dot [2]. \newline \newline [1] E. M. Chudnovsky, D. A. Garanin, and R. Schilling, Phys. Rev. B \textbf{72}, 094426 (2005). \newline [2] C. Calero, E. M. Chudnovsky, and D. A. Garanin, Phys. Rev. Lett. \textbf{95}, 166603 (2005).   

Organization of Single Molecule Magnets on Surfaces

The field of magnetic molecular clusters showing slow relaxation of the magnetization has attracted a great interest for the spectacular quantum effects in the dynamics of the magnetization that range from resonant quantum tunneling to topological interferences. Recently these systems, known as Single Molecule Magnets (SMMs), have also been proposed as model systems for the investigation of flame propagation in flammable substances. A renewed interest in SMMs also comes from the possibility to exploit their rich and complex magnetic behavior in nano-spintronics. However, at the crystalline state these molecular materials are substantially insulating. They can however exhibit significant transport properties if the conduction occurs through one molecule connected to two metal electrodes, or through a tunneling mechanism when the SMM is grafted on a conducting surface, as occurs in scanning tunnel microscopy experiments. Molecular compounds can be organized on surfaces thanks to the self assembly technique that exploits the strong affinity of some groups for the surface, e.g. thiols for gold surfaces. However the deposition of large molecules mainly comprising relatively weak coordinative bonds is far from trivial. Several different approaches have started to be investigated. We will briefly review here the strategies developed in a collaboration between the Universities of Florence and Modena. Well isolated molecules on Au(111) surfaces have been obtained with sub-monolayer coverage and different spacers. Organization on a large scale of micrometric structures has been obtained thanks to micro-contact printing. The magnetic properties of the grafted molecules have been investigated through magneto-optical techniques and the results show a significant change in the magnetization dynamics whose origin is still object of investigations.   

First Principles Calculations and Spin Models

Single magnetic molecules are fascinating entities. The individual transition metal ions have well defined spin states associated with localized d-orbitals bonded to ligands, which mediate the effective exchange or magnetic coupling among spins. At low temperatures and magnetic fields the internal complexity of the molecule can often (but not always) be ignored, with only the total collective spin determining the ground state and first few excited states. Using Mn$_{12}$ and V$_{15}$ as prototypes, this talk will describe a more reductionist approach and describe first principles electronic structure calculations used to gain insight into the electronic and magnetic structure of the individual transition metal ions and their interactions. Various spin coupling schemes and phenomenological Hamiltonians will be presented and compared to a variety of experimental results. Many colleagues and students from a number of institutions have contributed to this work and will be acknowledged during the talk.   

Density-Functional Theory of Molecular Magnets

Molecular magnets are large (a few nanometers in diameter), well-defined, discrete molecules consisting of several transition metal ions interacting through organic and/or inorganic ligands. Among thousands of synthesized molecular magnets, there is a class of molecular magnets known as single-molecule magnets (SMMs) which have large effective magnetic moments and behave as single-domain magnetic nanoparticles in an external magnetic field. They are particularly interesting because of observed quantum tunneling of magnetization and their possible applications in magnetic recording and molecular electronics. In this talk, I will demonstrate how quantum mechanics can be used to study the properties of SMMs from a first-principles vantage point. In particular, I will present density-functional calculations of the electronic, vibrational, and magnetic properties of selected SMMs, such as the total magnetic moment, electronic energy gaps, Raman scattering spectra, exchange constants, spin excitation energetics, and magnetic anisotropy barriers. I will also discuss what types of molecular environmental changes can significantly influence the exchange interaction, magnetic anisotropy, and observed quantum tunneling in the SMMs.   

Session P7: Focus Session: Physics of Transcriptional Regulatory Networks

Programming bacterial dynamics by synthetic killer circuits

In addition to offering insight into biological ``design'' principles, de novo engineering of synthetic gene circuits may impact broad areas including computation, engineering, and medicine. However, it remains challenging to realize predictable and robust circuit performance due to noise in gene expression and cell-to-cell variation in phenotype. We address these issues by using cell-cell communication to regulate cell killing to enable precise programming of bacterial dynamics. To establish cell-cell communication, we take advantage of quorum sensing systems that many bacteria use to detect and respond to changes in the cell density. As a prototype example, we have built and characterized a population control circuit in bacterium Escherichia coli. This circuit autonomously controls cell density by regulating the death rate using a quorum sensing module. Upon activation, the circuit will lead to a stable steady state or sustained oscillations in cell density, as predicted by a simple mathematical model. Further exploiting this design strategy, we have constructed a synthetic predator-prey ecosystem, where two E. coli populations regulate each other's growth and death by engineered two-way communication. Systems such as this will enable us to explore complex ecological dynamics in a well-defined experimental framework.   

Molecules, nonlinearity, and function in regulatory networks.

Biological regulatory networks are capable of sophisticated functions, such as integrating chemical signals, storing memories of previous molecular events, and keeping time. These networks often contain feedback loops, which can promote bistability and oscillation. However, feedback alone is not enough. Strong nonlinearities in the network dynamic are also needed. It is known that many regulatory proteins form higher-order complexes and multimers. I will discuss two important sources of nonlinearity in multimerization. First, ample experimental evidence suggests that protein subunits \textit{in vivo} can degrade less rapidly when associated in complexes. For homodimers, this effect leads to a concentration dependence in the protein degradation rate. Theoretical analysis of two model gene circuits in bacteria, i.e. switch and oscillator, demonstrates that this effect can substantially enhance the function of these circuits. Second, active proteins can often be sequestered into inactive complexes. This molecular titration can lead to sharp nonlinearities, and suggests a scenario for the rapid evolution of bistable or oscillatory circuits in nature.   

Combinatorial Regulation in Yeast Transcription Networks

Yeast has evolved a complex network to regulate its transcriptional program in response to changes in environment. It is quite common that in response to an external stimulus, several transcription factors will be activated and they work in combinations to control different subsets of genes in the genome. We are interested in how the promoters of genes are designed to integrate signals from multiple transcription factors and what are the functional and evolutionary constraints. To answer how, we have developed a number of computational algorithms to systematically map the binding sites and target genes of transcription factors using sequence and gene expression data. To analyze the functional constraints, we have employed mechanistic models to study the dynamic behavior of genes regulated by multiple factors. We have also developed methods to trace the evolution of transcriptional networks via comparative analysis of multiple species.   

Gene expression dynamics during cell differentiation: Cell fates as attractors and cell fate decisions as bifurcations

During development of multicellular organisms, multipotent stem and progenitor cells undergo a series of hierarchically organized ``somatic speciation'' processes consisting of binary branching events to achieve the diversity of discretely distinct differentiated cell types in the body. Current paradigms of genetic regulation of development do not explain this discreteness, nor the time-irreversibility of differentiation. Each cell contains the same genome with the same $N (\sim $ 25,000) genes and each cell type $k$ is characterized by a distinct stable gene activation pattern, expressed as the cell state vector $S_{k}(t)$ = {\{}$x_{k1}(t) $,.. $x_{ki}(t)$,.. $x_{kN}(t)${\}}, where $x_{ki}$ is the activation state of gene $i$ in cell type $k$. Because genes are engaged in a network of mutual regulatory interactions, the movement of $S_{k}(t)$ in the $N$-dimensional state space is highly constrained and the organism can only realize a tiny fraction of all possible configurations $S_{k}$. Then, the trajectories of $S_ {k}$ reflect the diversifying developmental paths and the mature cell types are high-dimensional attractor states. Experimental results based on gene expression profile measurements during blood cell differentiation using DNA microarrays are presented that support the old idea that cell types are attractors. This basic notion is extended to treat binary fate decisions as bifurcations in the dynamics of networks circuits. Specifically, during cell fate decision, the metastable progenitor attractor is destabilized, poising the cell on a `watershed state' so that it can stochastically or in response to deterministic perturbations enter either one of two alternative fates. Overall, the model and supporting experimental data provide an overarching conceptual framework that helps explain how the specifics of gene network architecture produces discreteness and robustness of cell types, allows for both stochastic and deterministic cell fate decision and ensures directionality of organismal development.   

Network theory and prediction of regulatory switches

While the influence of the high intracellular concentration of macromolecules on cell physiology is increasingly appreciated, its impact on the function of intracellular molecular interaction networks remains poorly understood. To test the effect of molecular crowding on the function of metabolic networks, we introduce a modified form of flux balance analysis that takes into account the constraint imposed by the limit on the attainable concentration of enzymes in the crowded cytoplasm. We demonstrate and experimentally confirm that the method can successfully predict the existence of regulatory points that allow switching from high to low biomass yield pathways when changing cellular growth rate. These results demonstrate that molecular crowding represents a bound on the achievable functional states of metabolic networks, and provide a systematic approach to uncover potential regulatory points in cellular metabolism.   

Session P8: Focus Session: Jets, Shocks & Splashes

The secret of splashing: interplay of air and roughness

We studied splashing on both smooth and rough dry surfaces using high speed photography. For smooth substrates, a striking phenomenon is observed: splashing can be completely suppressed by decreasing the pressure of the surrounding gas. The threshold pressure where a splash first occurs is measured as a function of the impact velocity and found to depend on the molecular weight of the gas and the viscosity of the liquid. Both experimental scaling relations support a model in which the compressibility of the gas is responsible for creating the splash[1]. For the case of rough substrates, we systematically varied both the surface roughness and the pressure of the surrounding gas and found two distinct contributions to a splash. One is caused by air and has the same characteristics as the ``coronal'' splash observed on smooth substrates. A second, ``prompt'' splash, contribution is caused by surface roughness. We have also measured the size distribution of the droplets emitted from a splash. For a smooth surface, a broad distribution of droplet sizes is found at high gas pressures. As the gas pressure is lowered towards the splash/no-splash transition the distribution gets more and more peaked at a characteristic size. For a rough surface, the distribution is strongly correlated with the surface roughness. [1] L Xu, W W Zhang and S R Nagel, Phys. Rev. Lett. 94, 184505 (2005)   

Statistical Mechanics of a Geophysical Jet

We investigate the equal-time statistics of an equatorial jet in a two-dimensional quasi-geostrophic model of a planetary atmosphere on a rotating sphere\footnote{R. S. Lindzen, A. J. Rosenthal, and R. Farrell, J. Atmos. Sci. {\bf 40}, 1029 (1983).}. Potential vorticity is advected by the barotropic flow and at the same time relaxes towards the zonal shear flow of an underlying equatorial jet. A transition to turbulence occurs at sufficiently slow relaxation rates. Statistics accumulated by direct numerical simulation\footnote{Akio Arakawa, J. Comp. Phys. {\bf 1}, 119 (1966).} are compared to those obtained by a simple cumulant expansion. We study rigorous upper bounds on the instability size\footnote{T. G. Shepherd, J. Fluid. Mech. {\bf 196}, 291 (1988).} and discuss the limitations of the cumulant expansion.   

High-Speed X-ray Investigation of Granular Jets

When a heavy sphere is dropped onto a bed of loose, fine sand, a large, focused jet of sand shoots upward.\footnote{Thoroddsen, S. T. and Shen, A. Q. Phys. Fluids {\bf13}, 4-6 (2001).} \footnote{Lohse, D. et al. Phys. Rev. Lett. {\bf93} (2004).} Experiments at reduced air pressure reveal that the jet in fact consists of two components: a wispy, thin jet that varies little with pressure followed by a thick air-pressure-driven jet \footnote{Royer, J. et al. Nature Physics, December 2005.} . To observe the initial stages of jet formation inside the granular bed, we employed x-ray radiography using the high-intensity beams available at the Advanced Photon Source. This technique allowed us to image the motion of the sphere and the evolution of the void left behind it at frame rates up to 6600 frames per second. The x-ray movies reveal that gravity-driven collapse produces the initial, thin jet, while the compression of an air pocket trapped below the surface drives up the thick jet. We also find that the interstitial air alters the compressibility of the sand bed. In vacuum a visible compaction front precedes the ball, while at atmospheric pressure the sand flows out of the way of the ball, behaving more like an incompressible fluid.   

Measurement of Stopping Force in Low Velocity Impact Cratering

The time dependent stopping force on a ball dropped into a granular medium has been measured using an accelerometer embedded within the ball. The velocity dependence of the force shows two distinct behaviors: (1) for impacts with large (200 $\mu $m) irregularly shaped sand particles, $F(v)\propto v^{1 \mathord{\left/ {\vphantom {1 2}} \right. \kern-\nulldelimiterspace} 2}$; for impacts with 100 $\mu $m spherical glass beads, $F(v)\propto \left( {v-v_0 } \right)$. The accelerator apparatus yields reproducible, low noise data that reveals peculiar features such as a downward acceleration pulse just before the ball comes to rest.   

Giant bubble-pinchoff

Self-similarity has been the paradigmatic picture for the pinch-off of a drop. Here we will show through high-speed imaging and boundary integral simulations that the inverse problem, the pinch-off of an air bubble in water, does not obey self-similarity (of the first kind): A disk is quickly pulled through a water surface, leading to a giant, cylindrical void, which at collapse creates an upward and a downward jet. The neck radius h(tau) of the void does NOT scale with the inertial power law exponent 1/2 (i.e., does not obey ``Rayleigh-scaling''). This is due to a second length-scale, the inverse curvature of the void,which follows a power-law scaling with a different exponent. Only for infinite Froude numbers the scaling exponent 1/2 is recovered. In all cases we find the void-profile to be symmetric around the minimal void radius up to the time the airflow in the neck deforms the interface.   

Nano-liquid bridges in ambient conditions

The dynamics of nano-liquid bridges in an ambient gaseous environment is studied using molecular dynamics simulations. Under these conditions new behavior close to break-up is found, compared to the behavior in vacuum. The probability for appearance of a long-thread structure close to pinch-off, versus the appearance of a double-cone profile in vacuum, depends on the density of the ambient gas. The stochastic lubrication equation that has been introduced by Moseler and Landman [1] for the case of break-up in vacuum is modified to include an additional term representing the effect of the ambient gas. Numerical integration of the modified stochastic lubrication equation shows good agreement with the molecular dynamics simulations. [1] M. Moseler and U. Landman, Science 289, 1165(2000).   

Splashing on dry, smooth inclined surfaces

We investigate splashing of drops on dry, smooth inclined surfaces. The asymmetry of the impact leads to an azimuthal variation of the ejected rim. We show that under certain conditions only part of the rim splashes. A model for the azimuthal splash threshold is compared both with the data and with existing splash criteria.   

Levitation of Falling Spheres in Stratified Fluids

The motion of sphere's falling under the influence of gravity is a classical problem dating back to Galileo and earlier. How a falling body additionally interacts with its environment is an equally challenging problem and involves strong coupling between the body and fluid via hydrodynamic drag. We present new phenomena$^2$ concerning the motion of a sphere falling through a sharply stratified (two layer) fluid in which the falling heavy body stops and reverses its direction (bounces) before ultimately returning to descent. Shadowgraph imaging shows the physics responsible for this surprising motion is a coupling between the body and the ambient boundary layer fluid, which is endowed with a negative potential energy as it is drawn into the lower layer, forming a rising turbulent plume. The hydrodynamic coupling between the sphere and this plume temporarily arrests the motion, even causing the bead to rise back through the transition layer. We present measurements of this trapping phenomena, and report the long residence times in which the sphere is trapped within the transition layer as a function of the bottom layer fluid density field for an array of different sized spheres. $^2$ N. Abaid, D. Adalsteinsson, Akua Agyapong, and R. M. McLaughlin, ``An Internal Splash: Falling Spheres in Stratified Fluids,'' Physics of Fluids, 16, no. 5, 1567-1580, 2004.   

Sodium luminescence from laser-induced bubbles in salt water

Luminescence from collapsing laser-induced bubbles in salt water (up to 1M NaCl) has been studied. We find a new emission pulse from the 589 nm sodium line that arrives about 50 ns prior to the main blackbody luminescence pulse. This may be related to the dynamics of the compressional heating process in the bubble. We have also noticed in the salt water that the time duration of the blackbody pulse is reduced by up to 30\% from the duration in pure water, and this has been observed in several other alkali salt solutions.   

The dynamics of a flexible loop in a high-speed flow

We study the behavior of an elastic loop in a fast-flowing soap film. The loop is wetted into the film and is held fixed at a single point against the oncoming flow. We interpret this system as a 2D closed flexible body moving in a quasi-2D flow. The loop is deformed by the flow, and this coupled fluid-structure system shows bi-stability: stationary and oscillatory. In its stationary state, the loop essentially remains motionless and its wake is a von K\'{a}rm\'{a}n vortex street. In its oscillatory state, the loop sheds two vortex dipoles within each oscillation period. The frequency of oscillation of the loop is linearly proportional to the flow velocity.   

Superfluid-like shock waves in nonlinear optics

It is well-known, but often underappreciated, that condensate dynamics has analogies with nonlinear light propagation in optics. In both cases, a single, macroscopic wavefunction describes the coherent wave behavior of interest. Here, we take advantage of this correspondence and examine superfluid-like spatial shock waves by propagating coherent light through a nonlinear crystal. We report the observation of both 1D and 2D shock waves, their nonlinear behavior as a function of intensity, and double-shock wave collisions. Analytical calculations and numerical simulations show excellent agreement with the experimental results. The fine structures and features observed here match similar observations in previous shock studies using superfluids and BEC, obtained in this case in a table-top apparatus, without the need for vacuum isolation, ultracold temperatures, etc. Moreover, the inherent optical advantages of easy control of the wavefunction input and direct imaging of the output make us optimistic that the nonlinear photonic systems described here will lay the foundation for a whole series of condensate-inspired experiments, many of which would be difficult (if not impossible) to perform in the corresponding condensed matter environments.   

Session P10: Focus Session: Frontiers in Computational Chemical Physics IV

First Principles Dynamics Beyond the Born-Oppenheimer Approximation

The dynamics of molecules in excited electronic states almost invariably involves breakdown of the Born-Oppenheimer approximation, necessitating treatment of quantum mechanical effects for both electrons and nuclei. The ab initio multiple spawning (AIMS) method has been developed in order to model molecular dynamics in excited states from first principles, solving both the electronic and nuclear Schr\"{o}dinger equations ``on the fly.'' We discuss some recent developments in the AIMS methodology and applications to photodamage in DNA bases. Theoretical results are compared directly to femtosecond spectroscopy experiments. Recent attempts to couple the AIMS approach with optimization algorithms to redesign fluorescent proteins will also be discussed, if time allows.   

First principles studies of CO adsorption and oxidation on the Cu$_{2}$O(100) surface

This work is motivated by the experimental results [1] indicating that the rate of CO oxidation on Cu$_{2}$O surface is much higher than that on Cu and CuO surfaces. To gain insight into the nature of this effect we study from first principles the energetics of adsorption and oxidation of CO on Cu$_{2}$O(100). Applying the \textit{ab initio} thermodynamics approach [2] to the surface in contact with gaseous O$_{2}$, we find that the O-termination of Cu$_{2}$O(100) is preferred for all reasonable range of temperature and the O$_{2}$ pressure. We find that CO adsorbed on surface O associates with it to form CO$_{2}$ without any activation barrier. On the other hand, CO adsorbing on a surface Cu atom, it is found to slide first towards the neighboring O atom to form CO$_{2}$ as in the previous case. We analyze the local densities of electronic states and valence charge densities of the systems to rationalize the obtained results. \begin{enumerate} \item T.-J. Huang and D.-H. Tsai, Catal. Lett. 87, 173 (2003). \item K. Reuter and M. Scheffler, Phys. Rev. B \textbf{65}, 035406 (2002). \end{enumerate}   

Electronic structure and bonding properties of K and K $^{+}$ on graphite under external electric field

The effect of an external electric field on the adsorption of K and K$^{+}$ on the graphite (0001) surface, are studied by means of first- principles total-energy calculations. The results were obtained with the pseudopotentials LCAO method (SIESTA code) and the Generalized Gradient Approximation (GGA) for the exchange-correlation potential. The structural parameters, bonding properties, and electronic structure of the K and K$^{+}$-graphite system are studied in the triangular (2x2) overlayer phase as a function of the external electric field magnitude. We find an important change in the K and K$^{+}$-graphite bonding as a consequence of the charge transfer from the adatom towards the substrate induced by the electric field. However, we find that none of the investigated systems show diffusion of K or K$^{+}$ into graphite even with a strong electric field. The results are discussed in the light of the experimental observed diffusion of K into graphite, presumably induced by external electric fields.   

Carbon-based nanostructured materials for enhanced H$_{2}$ production

A key fundamental limit of the thermal splitting of bulk water is the fact that the ground state of oxygen is paramagnetic, whereas the ground state of water is diamagnetic. Here, we propose to explore a new paradigm in H$_{2}$ production: a process in which the system remains on the spin singlet potential surface throughout the reaction, by exploiting the catalytic role of defective carbon substrates. Using first principles modeling techniques, we found evidence that mono-vacancy defects in graphite and carbon nanotubes give rise to a rich chemistry, yielding many possible water dissociation pathways, some of which have activation barriers lower than half the value for the dissociation of bulk water. This reduction is caused by spin selection rules that allow the system to remain on the same spin surface throughout the reaction. These novel reactions enhance the hydrogen yield and the reaction rate. In the presence of water only, this reaction is self-limiting: when all of the defects are oxidized, the reaction is complete, and no further H$_{2}$ is produced. There are several possibilities to achieve regeneration of the active surface sites, such as photo-excitation, vibrational excitations or further reaction with other molecules. We will discuss this exploration in the context of a complete cycle of energy storage and release through the production of H$_{2}$.   

Quantum dynamics with wavepackets and density matrices: A novel computational tool with applications to biological enzymes.

A recently developed computational approach for simultaneous dynamics of electrons and nuclei is discussed. The approach is based on a synergy between quantum wavepacket dynamics and ab initio molecular dynamics. The quantum dynamics is performed using an efficient banded, sparse and Toeplitz representation for the discretized free propagator that is formally exact. Ab initio molecular dynamics is achieved by using (a) an extended Lagrangian formalism, known as atom-centered density matrix propagation, that effects an adjustment of time-scales of the electronic motion, (b) Born-Oppenheimer dynamics. The quantum dynamics and ab initio dynamics schemes are coupled through a time-dependent self consistent field-like procedure. Higher order coupling between the subsystems is inherent when the Born-Oppenheimer procedure is used as opposed to atom-centered density-matrix propagation. A fundamental computational bottleneck associated with the computation of the interaction potential between the ab initio and quantum dynamical subsystem are overcome through a novel importance sampling approach and this aspect is also discussed. Further generalization for periodic quantum dynamical treatment in extended systems is outlined.   

First-Principles pH Theory

Despite being one of the most important macroscopic measures and a long history even before the quantum mechanics, the concept of pH has rarely been mentioned in microscopic theories, nor being incorporated computationally into first-principles theory of aqueous solutions. Here, we formulate a theory for the pH dependence of solution formation energy by introducing the proton chemical potential as the microscopic counterpart of pH in atomistic solution models. Within the theory, the general acid-base chemistry can be cast in a simple pictorial representation. We adopt density-functional molecular dynamics to demonstrate the usefulness of the method by studying a number of solution systems including water, small solute molecules such as NH$_3$ and HCOOH, and more complex amino acids with several functional groups. For pure water, we calculated the auto- ionization constant to be 13.2 with a 95 \% accuracy. For other solutes, the calculated dissociation constants, i.e., the so- called pK$_{\rm a}$, are also in reasonable agreement with experiments. Our first-principles pH theory can be readily applied to broad solution chemistry problems such as redox reactions.   

Efficiency and accuracy in transition-metal chemistry: a self-consistent GGA+U approach

Transition-metal centers are the active sites for a broad variety of biological and inorganic chemical reactions. Notwithstanding this central importance, density-functional theory calculations based on local-density or generalized gradient approximations often fail qualitatively and quantitatively in describing energetics, multiplet structures, reaction barriers, and geometries around the active sites. We suggest here an alternative approach, mutuated from the Hubbard U correction to solid-state problems, that provides an excellent agreement with accurate, correlated-electron quantum chemistry calculations in paradigmatic test cases that range from the ground state of the Fe$_2$ dimer to the potential energy surfaces for the addition-elimination of molecular hydrogen on FeO$^+$. The Hubbard U is determined with a novel self-consistent procedure based on a linear-response approach.   

First principles calculation of the x-ray absorption spectra of ice and liquid water

Recent interpretations of x-ray absorption spectra (XAS) of ice and liquid water propose that the standard, tetrahedral model of the liquid should be replaced with a model where each water molecule possesses two stronger and two weaker hydrogen bonds to nearest neighbor molecules. We have investigated this issue and find no conclusive evidence to discount the standard model. Using density functional theory (DFT) calculations we find an excellent agreement with experiment for the XAS of ice I. We perform TIP4P classical molecular dynamics simulations of the liquid at 300K. Using 10 statistically uncorrelated snapshots of 32 molecules in our DFT calculations, we compute the XAS of this standard liquid model and also find a reasonable agreement with experiment. The spectral differences between liquid and solid arise from both structural disorder and the presence of dangling hydrogen bonds.   

Exploiting unitary invariance in ab initio molecular dynamics: Applications to spectral decomposition and surface reactions

The methodology of ab initio molecular dynamics, wherein finite-temperature dynamical trajectories are generated using forces computed ``on the fly'' from electronic structure calculations, has benefited significantly from its combination with maximally localized electronic orbitals. The latter exploit the unitary invariance of the total energy to generate orbitals with maximum spatial locality. These orbitals resemble the classic textbook picture of molecular orbitals and, hence, are useful tools for analyzing electronic structure. In addition, maximally localized orbitals, expanded in localized basis sets, are a key component in linear scaling methods. In this talk, it will be shown how techniques from quantum field theory can be used to reformulate ab initio molecular dynamics in such a way that maximally localized orbitals are generated automatically and dynamically as the calculation proceeds. As an application of the technique, it will be shown how IR spectra can be decomposed to reveal particular structures in aqueous solutions. A second application will focus on the addition of organic molecules to the Si(100)-2x1 surface.   

Engineering protein structure and function with computational protein design

Understanding molecular folding has important applications to understanding biology and to developing new therapeutics and new materials.~ Protein design also opens new ways to probe the determinants of folding and to facilitate the study of proteins. Such design is complicated, however, by~the conformational complexity of proteins and by the large numbers of possible sequences. Recent computational methods for identifying the properties of amino acid sequences likely to fold to a given three-dimensional structure will be presented. Several examples of structures so designed, which have been experimentally synthesized and characterized, will be presented.   

First-Principles Calculations of van't Hoff Plots for Novel Hydrogen-Storage Materials

A van't Hoff plot, log(P) vs. 1/T, provides information on the free-energy change in a reaction and is widely used to characterize hydrogen-storage materials. Recently, a new reaction of LiBH$_4$ destabilized by MgH$_2$ (yielding over 11 wt.\% of H$_2$) was proposed.\footnote{J.J. Vajo et al., J. of Phys. Chem. B 109, p.3719 (2005)} Here we investigate this reaction and its products by first-principles calculations and construct the van't Hoff plot for a direct comparison to experiment. Although it is often assumed that there is a constant slope in the van't Hoff plot for ease of interpretation, we find an important non-linearity arising from temperature-dependent vibrational entropy difference, etc. This non-linearity can be critical for an accurate comparison to experimental data, and between various reactions to determine optimal hydrogen-storage systems. Including these effects, we find agreement with recent measurements.   

Application of Generalized Sturmians to the Bound States of Two-Electron Atoms and Molecules

A variation on the method of Generalized Sturmians [J. Avery, $\underline {Hyperspherical\ Harmonics and \ GeneralizedSturmians}$, Kluwer, 2000], is applied to the calculation of the ground and excited states of two-electron atoms and molecules (etc. He, H$_2$). In the present implementation of this method, each determinant formed from a set of primitive one-electron Sturmians, is required to separately solve the Sch\"odinger equation. In the process, the screening constant of each one- electron Sturmian orbital is non-iteratively uniquely determined. The resultant generalized eigenvalue problem however has a non-positive- definite overlap matrix. The method of \lq corresponding orbitals\rq\ [H.F. King et. al. J. Chem. Phys. 47, 1936 (1967)] is used to produce a positive-definite overlap matrix. A CI calculation is then performed whereby the Hartree-Fock calculation is avoided. Results will be presented and compared with Hartree-Fock based CI calculations.   

Characterizing the potential energy landscape by its geodesic paths

We suggest that the time evolution of a condensed-matter system is related to a unique exploration path in its multidimensional potential energy surface. We show that sampling from what we call the potential energy landscape filling ensemble, we can study the potential energy landscape of a monatomic Lennard-Jones system without the complications of barrier hopping processes. The ensemble defined to include all the configurations with potential energy less than a specified value, allows us to sample the geodesic path between two randomly selected configurations. The geodesics were tentatively related to the dynamics of the system under the assumption that the geodesic path corresponds to the most efficient exploration route on its potential energy surface. The derived dynamic parameters were compared with those obtained from a molecular dynamics simulation. The agreement we found offers us a new method for relating the dynamics of a system to the topology of its static potential energy surface.   

Session P11: Focus Session: Aerosols, Clusters, Droplets: Physics and Chemistry of Nanoobjects II: Helium Nanodroplets II, Aerosols, and Miscellaneous

High resolution infrared and microwave spectra of OCS solvated in helium clusters.

In recent years, exciting progress has been made in determining the onset and following the evolution of a bulk phase property, namely superfluidity, in the microscopic size regime. Our previous microwave and infrared studies of small He$_{N}$-OCS and He$_{N}$-N$_{2}$O clusters extended up to N=8 and N=19, respectively, and the infrared spectra with CO or CO$_{2}$ as probe reached almost up to N=20. We have now been able to extend the studies on He$_{N}$-OCS to much larger N-values in both the infrared and microwave regions. The B rotational constants that were extracted from the spectra show unexpected, non-classical behavior as a function of N, the number of helium atoms. We will present the experimental techniques used and an interpretation of the observed trends in spectroscopic observables.   

Recurrences in rotational dynamics and superfluid response in doped He-HCCCN and He-N${}_2$O clusters

Recent experiments on He-N${}_2$O complexes revealed the oscillatory behavior of the rotational constant in the range of cluster sizes corresponding to the completion of the first solvation shell. We use the path-integral Monte Carlo approach to show that this phenomenon can be associated with a non-monotonic size evolution of the non-classical rotational inertia and superfluidity, the origin of which can be traced back to combined solvent layering and bosonic exchange effects. Using the dopant molecule as an experimental microscopic probe of superfluidity, we show that in small doped helium clusters superfluidity builds up in stages correlated with the filling and completion of solvation shells.   

Infrared spectra and intensities of H$_{2}$O-N$_{2}$, H$_{2}$O-O$_{2}$ and H$_{2}$O-Ar complexes in superfluid He droplets.

The infrared spectra of H$_{2}$O-N$_{2}$, H$_{2}$O-O$_{2}$ and H$_{2}$O-Ar complexes in superfluid He droplets were measured in the range of the stretching vibrational bands of water molecules. The infrared intensities of anti-symmetric stretching band of these complexes showed no significant increase with respect to that of a single H$_{2}$O molecule as opposed to the predicted intensities in previous theoretical calculations. From the analysis of the observed spectra, it was found that H$_{2}$O in H$_{2}$O-O$_{2}$ and H$_{2}$O-Ar rotates nearly freely inside the complexes, while that in H$_{2}$O-N$_{2}$ does not. The conformation of these complexes were estimated from the rotational constants obtained from the analysis.   

Spectroscopic Properties of Aerosols and their Microscopic Origin

Large molecular aggregates with sizes ranging from less than nanometers up to microns play an important role in atmospheric processes, as components of the interstellar medium, and as drug delivery systems in medicine. The vibrational dynamics of these particles can be strongly influenced by intrinsic particle properties such as size, shape, or surface area. These phenomena are discussed here for several pure and composite ice particles which consist of CO$_{2}$, N$_{2}$O, NH$_{3}$, SO$_{2}$, their isotopomers, and different carbohydrates. The aerosol are generated in collisional cooling cells, by supersonic expansions, and by rapid expansion of supercritical solutions [1]. The vibrational dynamics is studied in situ with a rapid scan Fourier transform infrared spectrometer. We demonstrate that only the combination of experiments with microscopic models leads to a comprehensive understanding of the various features observed in the infrared spectra. The corresponding molecular model (exciton model [1,2]) allows us not only to calculate spectra for large molecular aggregates, but also to derive propensity rules for the occurrence of characteristic effects in infrared spectra of particles. \newline \newline [1] R. Signorell, \textit{Mol. Phys}. \textbf{101}, 3385, (2003). \newline [2] R. Disselkamp and G. E. Ewing, \textit{J. Chem. Soc. Faraday Trans. }\textbf{86}, 2369, (1990).   

Kinetics and Products of Radical-Initiated Oxidation of Organic Particles Using Aerosol CIMS

Ambient aerosol can contain a significant fraction of organic material which may react with trace gases in the atmosphere. Recently, it has been proposed that reactions with radical species, such as OH and Cl, may constitute a substantial loss mechanism for organic particles. In particular, the radical-initiated oxidation could lead to the creation of smaller, more volatile species which remove mass from the particles. An accurate assessment of the importance of these radical reactions requires measurements of their rates of reaction as well as identification of the subsequent products. We are exploring OH- and Cl-initiated reactions using Aerosol CIMS (chemical ionization mass spectrometry) to monitor changes in the compositions of the aerosol as well as the gas phase. This technique is well-suited to the study of organic species since the mass spectra contain very little fragmentation. These experiments provide insight into the oxidative processing which may potentially alter many critical properties of organic aerosol, including hygroscopicity and their ability to act as cloud condensation nuclei.   

Gas-phase infrared spectroscopy of ionic uranyl coordination complexes

The uranyl dication (UO$_{2}^{2+})$ is the primary carrier of uranium in environmental and biological systems, yet relatively little is known about its chemical properties. Mass spectrometric studies have confirmed that the properties, and in particular the reactivity of the uranyl center is rather dependent on its coordination environment. The current study uses IRMPD spectroscopy of ionic uranyl complexes, recorded using a free-electron laser coupled to an FTICR mass spectrometer, to investigate the effects of cluster size and composition on the infrared spectra. The central observation from the current study is that the asymmetric uranyl stretch is quite sensitive to the coordination environment, showing clear, reproducible, incremental redshifts as the number and electron donating ability of the ligands is increased. These spectral trends are accurately reproduced by computed frequencies from DFT calculations. Results from complexes containing anionic ligands are also presented, and are consistent with the presence of a significant barrier to electron transfer within the complexes.   

Stable highly symmetric dopant encapsulated binary clusters

While clusters composed of rare gas atoms exhibit enhanced stabilities for high symmetry geometries, magic numbers in simple metal clusters are determined by the number of delocalized valence electrons. Altering the composition of binary clusters allows to tailor independently the cluster geometry (number of atoms) and electronic properties (number of delocalized electrons). We produce beams of binary clusters with a dual-target dual-laser vaporization source. Size and composition dependent stability fluctuations are investigated with photofragmentation and mass spectrometry, and ionization energies with threshold laser ionization spectroscopy. We recently studied clusters of noble metals doped with transition metal atoms, and of group IVa elements doped with di- and trivalent metal atoms. Evidence is presented for the existence of combined closures of shells of atoms and shells of electrons for specific binary species. Phenomenological interpretations of new electronic shell closures are compared with DFT calculations of their geometry and electronic structure.   

Factors Controlling Polymorph Formation in Nonphotochemical Laser-Induced Nucleation (NPLIN) of Aqueous Glycine Solutions

The supersaturation and polarization dependence of nonphotochemical laser-induced nucleation (NPLIN) was studied in aqueous glycine solutions at wavelengths of 532 and 1064 nm, using linearly, circularly and elliptically polarized light. We observed a narrow supersaturation window (SS=1.45-1.54) for ``polarization switching,'' i.e. different polarizations producing different polymorphs. We also observed that, within this window, a small range of ellipticities near unity could induce the nucleation of the alpha polymorph, and that this range depended on supersaturation. Similar ``polarization switching'' behavior was observed at wavelengths of 1064 and 532 nm, although the supersaturation window became narrower at lower laser intensities at both wavelengths. Order-parameter ellipsoids and triangles based on optical Kerr alignment are presented to aid the interpretation of the experimental results.   

Density Functional Studies of Magic Metal-(C$_{60}$)$_2$ Clusters

Previous experimental studies of C$_{60}$-metal clusters revealed that clusters with composition C$_{60}$Ba$_{32}$ and (C$_{60}$K$_6$)$_n$K$^+$ appeared as magic peaks in the Time-of-Flight mass spectra due to geometric and electronic stability, respectively. Recent experiments using a new heating technique have revealed a different set of magic peaks, which cannot be explained by either one of the aforementioned mechanisms. We present theoretical studies addressing the stability and bonding of these newly observed magic clusters. Molecular density functional calculations have been performed to determine the most energetically stable geometrical configurations for $M_n$(C$_{60}$)$_2$ clusters, with $M=$ K, Ba and $1 \leq n \leq 6$. The bonding mechanisms have been analyzed in some detail. The results indicate that for barium containing clusters, ionic bonding (transfer of the valence $6s$ electrons to the C$_{60}-\pi^*$ orbitals) and covalent bonding (between the barium $5d$ and C$_{60}-\pi^*$ orbitals) are the dominant mechanisms. Moreover, a metal cluster is formed between the two fullerenes (Ba-Ba bonding). For the potassium containing clusters, only ionic bonding was found. Calculations of the Gibbs free energies indicate that Ba$_3$(C$_{60}$)$_2$ and K$_4$(C$_{60}$)$_2$ are the most stable structures, in agreement with experimental results. The role of the entropy is found to be very important in determining which clusters are magic.   

Photoelectron spectroscopy as a structural probe of intermediate size clusters

We examine the utility of photoelectron spectroscopy (PES) as a structural probe of Si$_{n}^{-}$ in the n = 20 -- 26 size range by determining isomers and associated photoelectron spectra from first principles calculations. Across the entire size range, we consistently obtain good agreement between theory and experiment [Hoffmann et al., \textit{Eur. Phys. J. D} \textbf{16}, 9 (2001)]. We find that PES can almost invariably distinguish between structurally distinct isomers at a given cluster size, but that structurally similar isomers usually cannot be reliably distinguished by PES. For many, but not all, sizes the isomer giving the best match to experiment is the lowest-energy one found theoretically. Thus, combining theory with PES experiments emerges as a useful source of structural information even for intermediate size clusters.   

Nearly Free Electron Gas in a Silicon Cage

Theoretical investigations of the ground state geometries, electronic structure, spin magnetic moment and the stability of the metal encapsulated MSi$_{12}$ ( M= Sc, Ti, V, Cr, Mn, Fe, Co, Ni) clusters have been carried out within a gradient corrected density functional formalism. The ground state of most MSi$_{12}$ clusters are shown to have the lowest spin multiplicity as opposed to the high spin multiplicity of free transition metal atoms. Consequently, a proper inclusion of the spin conservation rules is needed to understand the variation of the binding energy of M to Si$_{12}$ clusters. Using such rules, CrSi$_{12}$ and FeSi$_{12}$ are found to exhibit the highest binding energy across the neutral while VSi$_{12}^{-}$ has the highest binding energy across the anionic MSi$_{12}^{- }$series. It is shown that the variations in binding energy, electron affinity and ionization potential can be rationalized within an 18-electron sum rule commonly used to understand the stability of chemical complexes and shell filling in a confined free electron gas.   

Single and Multiple Rings, and Cages in SiO$_{x}$ Clusters

Theoretical studies on the geometry, electronic structure and stability of Si$_{n}$O$_{m}$ clusters have been carried out within a gradient corrected density functional formalism. It is shown that the ground states of small Si$_{n}$O$_{n}$ clusters containing upto 4 units are single rings. The first Si-Si bond appears at Si$_{5}$O$_{5}$, and starting at this size, the elementary rings begin to assemble into multiple rings that eventually lead to cages. The ground state structures at larger sizes have a central core of pure Si atoms decorated by outer shell of SiO units. An analysis of the fragmentation patterns shows that Si$_{7}$O$_{7}$ and Si$_{10}$O$_{10}$ are particularly stable species. The results of our investigations on the Si$_{n}$O$_{n-1}$ and Si$_{n}$O$_{n+1}$ species will also be presented. In particular, we will examine possible reaction mechanisms that could lead to the formation of SiO$_{2}$ from SiO molecules in interstellar space.   

Session P12: Metal Islands and Clusters

The crossover from collective motion to periphery diffusion for small adatom-islands on Cu(111) and Ag(111)

The diffusion of two dimensional adatom islands (containing 2-100 atoms) on Cu(111) and Ag(111) has been studied, using the newly developed self-learning Kinetic Monte Carlo (SLKMC) method [Phys. Rev. B 72, 115401, 2005]. A variety of multiple and single atom processes are revealed in the simulations and the size dependence of the diffusion coefficients and effective diffusion barriers are calculated. From the tabulated frequencies of events found in the simulation, we show a crossover from diffusion due to the collective motion of the island to a regime in which the island diffuses through the periphery dominated mass transport. This crossover occurs for island sizes of 8 to 11 atoms. For islands containing 19 to 100 atoms the scaling exponent is 1.5, which is in good agreement with previous work. The diffusion of islands containing 2 to 10 atoms can be explained primarily on the basis of a linear increase of the barrier for the collective motion with the size of the island.   

Excitation of frustrated translation and nonadiabatic adatom hopping induced by inelastic tunneling

The dynamics of lateral manipulation for cobalt/Cu(111) has been investigated combining the model of vibrational heating and first-principles density functional calculations~[1]. The frustrated translational mode responsible for lateral excitation is identified as a vibrational resonance involving a concerted motion between the adatom and surface phonons. The calculated frequency shows good agreement with the onset energy for adatom hopping induced by inelastic tunneling. Simulation of the power law, compared with experiment, suggests that the atom hopping overcomes a nonadiabatic barrier due to the nonequilibrium local heating of the translational mode. \\ \\ \noindent [1] Kai Liu and Shiwu Gao, Phys. Rev. Lett. 95, 226102 (2005)   

Metal Island Coalescence Stress for Varying Surface Traction

During Volmer-Weber thin film growth, discrete metal islands grow on a substrate. When the separation between their adjacent free surfaces becomes small enough, islands coalesce and trade surface energy for elastic strain energy. While it is understood that traction between island and substrate directly influences coalescence stress, questions remain. For instance, wafer curvature measurements during growth of low traction systems indicate zero stress in the growing film. This results from a lack of mechanical coupling between film and substrate so island stress cannot be accurately determined from experimental data. We examine, via atomistic simulations, coalescence between two metal islands as a function of island size and island-substrate traction. We reveal the stress state in the coalesced structure for low traction where entire islands are able to slide along the substrate. This is compared to higher traction where islands can only slide near the point of coalescence resulting in local tensile strain. We conclude by examining the dependence upon island size of the traction above which only local sliding is observed. 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.   

Effects of Strain on Island Morphology and Size Distribution in Submonolayer Island Growth

We have carried out kinetic Monte Carlo simulations of submonolayer heteroepitaxial growth using a model in which the strain energy is approximated by a long-range $1/r^3$ interaction. For the case of irreversible growth with rapid island relaxation due to edge-diffusion we find that the island-shape changes from square to rectangular in the presence of sufficient strain. However, in this case fluctuations play an important role and the scaled island-size distribution is only weakly affected by strain. General scaling forms for the island-width and island-length distributions are also derived. Simulation results for a reversible model of Cu/Ni(001) growth, for which an interesting shape transition has been experimentally observed, are also presented and compared with experiments.   

Incidence-angle dependence of optical reflectivity difference from an ultrathin film on a solid surface

The oblique-incidence reflectivity difference (OI-RD) technique is a special form of polarization-modulated nulling ellipsometry that has been judiciously used in studies of ultrathin films and other surface-bound changes on a solid or even liquid substrate. We have recently studied the incidence angle dependence of the optical reflectivity difference signals in response to ultrathin films on transparent and opaque substrates. We find that the classical three-layer model reproduces the experimentally obtained angular dependence for a monolayer of xenon on Nb(110) and for a monolayer of protein molecules on functionalized glass. We report the findings of this recent experimental investigation and explore the enhancement of the optical response near the Brewster angle (or its equivalent for opaque substrates) in thin film detection.   

Time Resolved Study of Ordering Between Quantum Well Islands by Wide-Area k-Space Mapping

The growth of Pb quantum well islands on Si(111) was investigated in time resolved manners. The wide-area momentum space mapped in our x-ray diffraction techniques allows simultaneous observation of the lateral ordering and height distributions of Stranski-Krastanov (S-K) islands during deposition, annealing, and cycling between deposition and annealing. These islands are formed not by strain relaxation but through quantum phase separation. The ordering of these islands exhibits unusual behaviors not well described by existing strained island theories. The population of islands always decreases monotonically during all three processes. However, the island ordering was stable without coarsening with time, at least within the experimental condition. This study suggests that quantum well effect may provide a new mechanism for uniform self-assembled nanostructures.   

Sublimation of Atomic Layers from a Chromium Surface

We employ low-energy electron microscopy to study the kinetics of thermal etching, or sublimation, of Cr(001) at $\sim $1100 K. Atomic layers are removed from the surface by spontaneous nucleation and growth of two-dimensional vacancy islands, by rotation of spiral steps, and by island decay. The growth rates of vacancy islands and the rotation frequencies of double spirals are measured as a function of temperature, and the results are correlated with activation barriers of surface processes. Mass transport between the surface and bulk is shown to be unimportant.   

Iso-coverage diffusion zones observed with LEEM during annealing of Ag on Si(001)

During high temperature annealing of 3D Ag islands grown on Si(001) PEEM images reveal the emergence of a bright zone surrounding the decaying islands. Microdiffraction patterns from these bright areas display a (2x3) Ag reconstruction. The decaying Ag islands act as sources of Ag adatoms which then diffuse on the surface. The (2x3) reconstruction spreads away from the island to a distance determined by the interplay of diffusion and desorption, and by the local coverage of Ag: the outer boundary of the imaged diffusion zones constitutes an ``iso-coverage boundary.'' We describe the time and temperature behavior of these iso-coverage zones, and present a simple continuum model that describes the iso-zone size as a function of the diffusion and desorption rates.   

Imprinting Chirality into inorganic CuO Thin Films

Switzer et al. [1] have shown, that thick ($>$300nm) films of CuO grown electrochemically in the presence of chiral tartaric acid (TA) acquires a chiral orientation with respect to the growth surface. We have investigated this growth on Au(100) in the presence of chiral TA for low film thicknesses by X-ray Photoelectrons Spectroscopy and X-ray Photoelectron Diffraction (XPD). The resulting XPD patterns were analyzed by single scattering cluster calculations. XPD revealed that using chiral L(+)- or D(-)-TA in the deposition process results in a chiral CuO surface which exhibits mirror-symmetric, non-superimposable patterns with the corresponding chirality imprinted already for film thicknesses below 3nm. Whereas the XPD patterns of the CuO films deposited with the racemic DL-TA and the ``achiral'' meso-TA are completely symmetric. The selectivity of enantiomeric CuO films was demonstrated by subsequent deposition of CuO from a solution containing DL-TA onto a CuO film grown with only one of the enantiomeric forms. Additionally, films with alternating chirality were produced. [1] J. A. Switzer, H. M. Kothari, P. Poizot, S. Nakanishi, E. W. Bohannan, Nature, 2003, 425, 490   

Geometry and Electronic Structure of Alumina Supported Ag Clusters

We use DFT/GGA to model silver clusters supported on Al-terminated alpha-alumina. A variety of cluster sizes and geometry are considered. We find that the adsorption of the Ag clusters causes dramatic surface relaxations and polarization of the Ag clusters as some Ag atoms of the clusters bond with O, while others bond with Al. For comparison, we also model adsorption of a single Ag atom, which bonds at a hollow site (relative to surface oxygen). We use the electronic structure of the adsorbed clusters to interpret the bonding interaction between the Ag atoms and surface O and Al atoms. The differences in Ag-surface bonds within the cluster give rise to different electronic environments at each Ag atoms. This effect has implications of this result to reactivity.   

Adsorptive and chemical properties of supported silver clusters

Unlike bulk metal slabs, clusters of transition metals on nonmetallic supports have a finite electron reservoir. Additionally, the interaction of clusters with the support can lead to polarization in the cluster. These effects give rise to adsorption properties quite different from those of bulk metal. We characterize the adsorption of different atomic and molecular species on silver clusters supported by Al-terminated alpha-alumina using DFT/GGA calculations. A range of cluster sizes and cluster geometries is onsidered and the adsorption energetics and geometries are compared with those of bulk silver metal.   

Alkali ion scattering from sputter-induced Au nanoclusters

The neutralization of scattered 3 keV $^{23}$Na$^{+}$ ions is used to probe the confined states of sputter induced Au nanoclusters on a TiO$_{2 }$(110) substrate. The neutral fraction of Na scattered from Au nanoclusters deposited on TiO$_{2}$(110) is high compared to bulk Au, as described earlier (G.F. Liu, Z. Sroubek, J.A. Yarmoff, Phys. Rev. Lett. \textbf{92}, 216801 (2004)). Here, normal incidence 0.5 keV Ar$^{+}$ ion bombardment of a thin Au film is employed as an alternative method for the self-organized formation of Au nanoclusters. The interplay between curvature dependent sputtering and surface diffusion during Ar$^{+}$ bombardment works to develop the Au nanoclusters. The neutral fraction of the scattered Na gradually increases from 5{\%} to 50{\%} as the cluster dimensions decrease due to the Ar$^{+}$ sputtering. XPS is used to quantify the reduction of Au coverage with sputtering time. STM revels a decrease of both the rms roughness and the correlation length of Au nanoclusters with Ar$^{+}$ fluence.   

CO oxidation reaction on Au clusters supported on the TiO$_{2}$(110) surface

We have studied by means of density functional theory (DFT) plane waves calculations, the CO oxidation reaction on Au$_{8}$ clusters adsorbed on both the stoichiometric and reduced rutile TiO$_{2}$(110) surfaces. O$_{2}$ molecules bind at the interface between the TiO$_{2}$(110) surface and the gold particle, where they become partially charged through the population of the anti-bonding molecular orbital, resulting in the extension of the O-O bond length to values of peroxo-like states. When CO is co-adsorbed on the gold-particles, the Langmuir-Hinshelwood and the Elay-Rideal mechanisms are possible for the oxidation reaction. Both the catalytic cycles have been considered and the theoretical results have been compared with experimental findings obtained on size-selected Au$_{8}$ clusters deposited on thin, oxidized and reduced TiO$_{2}$ films.   

High photocatalytic activity of immobilized non-crystalline nanometric titanium oxide: key role of interface

Thinnest coatings prepared by \textit{chemical} deposition of\textit{ non-crystalline} nanomertic titanium oxide particles (2R=5.0 nm) show important photocatalytic activity, which can be higher than that of the reference sample, crystalline \textit{Degussa} P-25 TiO$_{2}$. We describe an original method of the sol particles preparation and immobilization on complex supports in the sol-gel reactor. The effect of the coating thickness on its photocatalytic activity is studied. We show that the very first layer of the deposited nanoparticles possesses both high mechanically stability and the highest efficiency. We describe main features on the nanocoatings behavior in photocatalysis (gas-phase trichloroethylene degradation) and show that their internal efficiency increases with the decrease of the deposited mass. These results suggest a new approach to the active component design, while earlier non-crystalline TiO$_{2}$ has been considered inactive.   

Ab-Initio Studies of Properties of 2D Rare-Earth Silicide Surfaces

Rare Earth overlayers on the Si(111) surface have attracted interest due to their novel properties [1,2] and the unusual reconstruction that is formed with a flat rare earth layer buried inside the Silicon [3,4,5,6]. Here we present the results of ab-initio calculations done using the CASTEP [7] code to determine the structural and electronic properties of these reconstructions. We compare these to those derived experimentally. \\ \ \\ {[}1{]} {F. P. Netzer, {\it J. Phys.: Cond. Matt.}, {\bf 7} (1995) 991-1022}\\ {[}2{]} {K. N. Tu {\it et al}, {\it App. Phys. Lett.}, {\bf 38} (1981) 626-628}\\ {[}3{]} {M. Lohmeier {\it et al}, {\it Phys. Rev. B.}, {\bf 54} (1996) 2004-2009}\\ {[}4{]} {D. J. Spence {\it et al}, {\it Phys. Rev. B.}, {\bf 61} (2000) 5707-5713}\\ {[}5{]} {D. J. Spence {\it et al}, {\it Surf. Sci.}, {\bf 512} (2002) 61-66}\\ {[}6{]} {C. Rogero {\it et al}, {\it Phys. Rev. B.}, {\bf 66} (2002) 235421-235427}\\ {[}7{]} {M. D. Segall {\it et al}, {\it J. Phys.: Cond. Matt.}, {\bf 14} (2002) 2717-2743} \\   

Session P13: Focus Session: Ultrafast and Ultrahigh Field Chemistry II: Quantum Control

Quantum Control with Nonclassical Light

Most of the experimental advances in coherent quantum control in recent years have involved ultrashort pulses and pulse shaping techniques. These pulses have been an excellent source of coherent light with precise phase relationship between the various frequency components. In several recent works we have investigated the possibility of using broadband nonclassical light, generated by down-conversion of narrow-band lasers, for coherent control. Such light, for most purposes, exhibit the properties of a broadband thermal noise, but also unique quantum correlations between spectral mode pairs at the signal and idler frequencies that are required for quantum control. We have investigated both the single-photon limit, when the light was composed of individual entangled photon-pairs, and the large signal limit, when the light is not weak but does exhibit nonclassical phase correlations. In the high-intensity limit, we have shown that coherent control of two-photon absorption can be performed with incoherent non-classical light. We showed that the signal-idler phase correlations cause the spectral quantum interference to be completely constructive for two-photon interactions that have a final state energy equal to the pump laser frequency. Consequently, even though the broadband down converted light is neither coherent nor pulsed, it induces two-photon absorption just like a coherent ultrashort pulse, and may likewise be coherently controlled by pulse-shaping techniques. We also demonstrated that pulse shaping techniques can be used in the single-photon limit, where we shape the two-photon correlation function. We demonstrate control of the quantum interference of photons at a beam-splitter, and the generation of Bell-states using polarization pulse-shaping techniques. We believe that the combination of quantum control techniques with quantum optics could add an important ingredient to the toolbox of quantum information and computing.   

Bell state generation in the presence of complicated entangling interactions

We present theoretical investigations on the entanglement generation of bipartite two- and three-level systems interacting with complicated entangling interactions and external electromagnetic fields. The theoretical method employed is based on rotating wave approximation (RWA). By using the partitioning of the complicated entangling interaction matrix, we propose a method of creating the Bell state from the initial separable state in bipartite two-level systems. In addition, by using the bipartite three-level model systems, we show how to create decoherence-free subspace when the Bell state is to be generated. The present work will be useful for realization of entanglement manipulation in the presence of complicated entangling interactions in molecular systems.   

High-Precision Coherent Control of Molecular Wave Packets

The quantum interference of two vibrational wave packets has been precisely controlled in the electronically excited state of a diatomic molecule by using a pair of fs laser pulses whose relative phase $\phi $ is locked within the attosecond time scale, and the real time evolution of that interference has been observed by another fs probe pulse. The real-time evolution shows a clear dependence on $\phi $. We have also measured a population code, which is a population ratio among the vibrational eigenstates within a WP. The population code also shows a clear dependence on $\phi $. The ordinary frequency domain interpretation based on the spectral interference of locked pulses may be useful to elucidate $\phi $ dependence of population codes, but is no longer suitable for the present real-time observation. The combination of a population code and real-time evolution is useful to obtain both phase and amplitude information stored in a WP.   

Nanolocalized Nonlinear Electron Photoemission under Coherent Control

We theoretically show that two-photon coherent control yields electron photoemission from metal nanostructures that is localized in nano-size hot spots whose positions are controllable on a nanometer scale, in agreement with recent experiments. We propose to use silver V-shapes as taylored nanoantennas for which the position of the coherently controllable photoelectron-emission hot spot can be deterministically predicted. We predict that the low-frequency, high-intensity (quasistationary) excitation of the photoemission leads to an exponentially high contrast of the coherent control. REFERENCES M. I. Stockman and P. Hewageegana, ``Nanolocalized Nonlinear Electron Photoemission under Coherent Control'', Nano Lett. 5(11), 2325-2329 (2005)   

Monitoring Molecular Dynamics using Coherent Electrons from High-Harmonic Generation

In this talk, we will discuss the first observation of intramolecular vibrational dynamics using electrons rescattered during the process of high-order harmonic generation. We excite coherent vibrations in SF$_{6}$ using impulsive Raman scattering with a short laser pulse. A second, more-intense laser pulse generates high-order harmonics of the fundamental laser, at wavelengths of $\sim $ 20-50 nm. The high-order harmonic yield is observed to oscillate, at frequencies corresponding to all the Raman-active modes of SF$_{6}$, with an asymmetric mode most visible. This is in contrast to conventional impulsive stimulated Raman spectroscopy where only the symmetric breathing mode of the molecule is easily observed. The data also show evidence of relaxation dynamics following impulsive excitation of the molecule. Our results indicate that high harmonic generation is a sensitive probe of vibrational dynamics and may yield more information simultaneously than conventional ultrafast spectroscopic techniques. Since the de Broglie wavelength of the recolliding electron is on the order of interatomic distances, i.e. $\sim $ 1.5 {\AA}, small changes in the shape of the molecule lead to large changes in the high harmonic yield. This work therefore demonstrates a new spectroscopic technique for probing ultrafast internal dynamics in molecules that uniquely combines ultrafast time resolution with atomic-scale structural information.   

Observation and control of ultrafast quantum interferences in atoms and molecules.

I will present several examples of ultrafast interferences in atoms and molecules, at the femtosecond and picosecond time scale. In a two level atom, real-time quantum state holography is performed through interferences between quantum states created by a reference pulse and a chirped pulse resulting in coherent transients. A sequence of several measurements allows one to measure the real and imaginary parts of the excited state wave function. These measurements are performed during the interaction with the ultrashort laser pulse. The extreme sensitivity of this method to the pulse shape provides a tool for electric field measurement. In a molecule, the transient interferences between two oscillating wave-packets have been observed and controlled. In a first experiment, a vibrational wave packet is created in the iodine B state. Due to anharmonicity, the wave-packet spreads and recombines in one single wave packet (revival time) or two wave-packets (half revival time). When these two wave packets cross, they transiently create a stationary wave which is observed. In a second experiment, the same situation is created by launching two wave packets in the same potential well with an ultrastable relative phase. The delay, set to 1.5 vibrational periods, is stabilized with sub 100 attosecond precision. The same transient interference pattern is observed. Moreover, the relative phase between the counterpropagating wave packets can now be controlled by scanning the interpulse delay on an optical period.   

Final resolution of the step-wise versus concerted mechanism controversy for excited-state double proton transfer in the 7-azaindole dimer in the gas phase

The excited-state double-proton transfer (ESDPT) reaction in the 7-azaindole dimer has been extensively studied by spectroscopic methods in the gas phase and in solution. Two ESDPT mechanisms, stepwise and concerted mechanisms, have been proposed so far. However, a definite conclusion has not been provided due to a lack of clear evidence. We provide final resolution of the stepwise versus concerted mechanism controversy for the ESDPT reaction in the 7-azaindole dimer by electronic spectroscopy and picosecond-time resolved spectroscopy in the gas phase (K. Sakota, C. Okabe, N. Nishi, H. Sekiya, J. Phys. Chem. A, 109, 5245 (2005)). The ESDPT reaction in the 7-azaindole dimer proceeds via the concerted mechanism. We propose a dynamic cooperative effect, where the motions of the two transferring protons couple with each other through the electronic reorganization (K. Sakota ,H. Sekiya, J. Phys. Chem. A, 109, 2718 (2005); 109,2722 (2005)).   

On the Control of Product Yields in the Photofragmentation of Deuteriumchlorid Ions (DCl$^{+})$ -- Cl + D$^{+} \quad <$ - - DCl$^{+}$ - - $>$ Cl$^{+}$ + D.

We have investigated the prospect of controlling the photofragmentation of deuterium chloride ions (DCl$^{+})$ via ultra short IR laser pulses both by experiments and by numerical solution of coupled Schr\"{o}dinger equations. The calculations provide evidence that the ratio of product ion yields Cl$^{+}$ versus D$^{+}$ can be manipulated by appropriate choice of laser pulse parameters, in particular central laser wavelength, pulse duration, intensity and chirp [1]. The analysis of time dependent populations reveals phase sensitive competition between intra- and inter-electronic state excitation. Complementary, we have performed one- and two-color fs experiments looking at the dissociation of DCl$^{+}$ ions at 800 nm [2] and in the range from 3.5$\mu $m to 7.5$\mu $m (2857cm$^{-1}$ to 1333cm$^{-1})$ [3]. In particular we show, that the ratio of product yields D$^{+}$/Cl$^{+}$ can be controlled via the chirp of the laser pulse at 4.5$\mu $m. References [1] M.V. Korolkov, K.-M. Weitzel, J. Chem. Phys. 123, 164308, (2005) [2] H.G. Breunig, A. Lauer, K.-M. Weitzel, Proceedings of the Femtochemistry VII (2005) [3] H.G. Breunig, K.-M. Weitzel, in preparation.   

Femtosecond pulse shaping in the mid infrared region using a Dazzler

We present a method to produce programmable phase- and amplitude-modulated femtosecond laser pulses in the mid infrared region (MIR: 3 -- 10 $\mu $m) by difference-frequency generation (DFG). The signal output (NIR: 1.1 -- 1.5 $\mu $m) of an optical parametric amplifier was shaped with an acousto-optic programmable dispersive filter (Dazzler), and mixed in a AgGaS$_{2}$ crystal with the idler pulse temporary stretched by passing a dispersion block to generate MIR pulses. A Dazzler provides convenient and precise way of shaping femtosecond pulses in NIR region. It is, however, not well understood how the phase and amplitude modulations are transferred from a NIR pulse to a MIR pulse via DFG process. We will discuss the analysis of the shaped NIR and MIR pulses using a frequency-resolved optical gating (FROG) and an FT-IR   

Fast Transient Electron Magnetic Resonance at 240 GHz

The zero-field splitting of the excited triplet states of organic molecules often is of the order of 1 GHz or less, and transient EPR at X-band is generally sufficient for the determination of the zero-field splitting and kinetic parameters in these type of molecules. However, information on the g-factor and g-anisotropy cannot be obtained at conventional EPR frequencies, and interpretations of the data in terms of electronic structure are mostly limited to symmetry considerations. On the other hand information of the g-anisotropy provides additional clues with respect to electronic structure, while a direct comparison with radical-ion forms of the molecules becomes possible. Experimental data of transient EPR at 240 GHz will be presented for a variety of system (fullerene-based and porphyrin-based). EPR at these very high frequencies can accurately determine the g-anisotropy and in some cases the orientation of the g-tensor with respect to the ZFS tensor. Also at these high frequencies sub nanosecond processes can be measured. Examples will be given.   

Response of Dipicolinic Acid ($C_{5}H_{5}N(COOH)_{2}$) to Ultrafast Laser Pulses

Dipicolinic acid (DPA) and its salts are common constituents of bacterial spores, including those of anthrax. It has been proposed that such spores can be detected via spectroscopic techniques which employ ultrashort laser pulses. The development of these techniques should be enhanced by a detailed understanding of the microscopic processes that transpire when a molecule is subjected to femtosecond-scale pulses of various intensities, durations, and polarizations. We have recently developed a model that can be used to perform realistic simulations of the electronic and nuclear dynamics of biological molecules (containing carbon, hydrogen, oxygen and nitrogen) when they are subjected to such pulses. The bond lengths and vibrational frequencies for a variety of test molecules are in reasonable agreement with those obtained in experiment and \textit{ab initio} calculations. Here we report results of simulations for DPA responding to femtosecond-scale laser pulses, with an analysis of the vibrational modes and electronic states which are most relevant for various choices of the laser pulse parameters.   

Session P16: Focus Session: Molecular-Scale Electronics III

Molecular sensing using point contact conductivity modulation

The electrical properties of semiconductors are sensitive to external influences, such as the adsorption of gaseous molecules. For single crystal Si surfaces, the change in conductivity induced by molecular adsorption is a very small fraction of the bulk conductivity, precluding their use as efficient sensors. Here we show that point contacts on Si surfaces in UHV environments can overcome this fundamental limitation, through the use of minority-carrier-induced conductivity modulation. Point contacts made to clean, low-doped $n-$Si(100) produce significant surface inversion layers. The inversion layer minority-carrier population is exponentially dependent upon surface charge. Slight increases in the surface charge density, from gas molecule adsorption, are detected as large increases in sample conductivity, as electrons flow in to balance the positive hole space charge. The sensitivity of this simple device structure is so high that physisorption of inert gas molecules such as He, N$_{2}$, and Ar can be detected as conductivity increases of 2 -- 100{\%}; the specific response is proportional to the molecular ionization potential. Decreasing the point contact size, from micro- to nano- to atomic-scale, increases device sensitivity because of increased minority-carrier injection ratios.   

Humidity dependence of molecular tunnel junctions with an AlOx/COOH- interface

We have studied the electron transport in planar tunneling junctions with aluminum oxide and an organic self-assembled monolayer (SAM) as the tunnel barrier. The structure of the junctions is Al/AlOx/SAM/(Au, Pb) with a junction area of $\sim $ 0.4mm$^{2}$. The organic molecules investigated include mercaptohexadecanoic acid (MHA), hexadecanoic acid (HDA), and octadecyltrichlorosilane (OTS); all of which form ordered SAMs on top of aluminum oxide. The use of a superconducting electrode (Al) enables us to determine unambiguously that these are high-quality tunnel junctions. For junctions incorporating MHA, the transport behavior is found to be strongly humidity dependent. The resistance of these junctions drops more than 50{\%} when placed in dry nitrogen and recovers when returned into the ambient. The same drop also occurs when the sample is placed into a vacuum, and backfilling the vacuum with either dry N2 or O2 has negligible effect on the resistance. For comparison, junctions with HDA show the same humidity dependence, while OTS samples do not. Since both MHA and HDA have carboxylic groups and OTS does not, the results suggest that water molecules at the AlOx/COOH- interface play the central role in the observed behavior. Inelastic tunneling spectroscopy (IETS) has also been performed to understand the role of water. This work was supported by a FSU Research Foundation PEG grant.   

Signatures of magnetism in an individual Mn12O12 molecule probed by single-electron tunneling.

We report low-temperature electron transport through individual molecular clusters, Mn12O12(O2C-R)16(H2O)4, [Mn12O12], where R is -CH3 and -CHCl2. Energy level spectroscopy with single-electron tunneling probes the ground state spin of the individual Mn12O12 molecules, and exhibits signatures of their magnetism. In particular the absence of the spin degeneracy is manifested as an energy splitting between low-lying energy manifolds of the ground state spin at zero-magnetic field, and it signifies the magnetic anisotropy of an individual Mn12O12 molecule. We also discuss the influence of this anisotropy to the electron tunneling spectrum in the presence of a magnetic field.   

ESR-STM Spectrometer for Paramagnetic Molecular Adsorbates on Surfaces

ESR-STM is a technique able to detect noise at the Larmor frequency in the tunnelling current associated with the spin dynamics of a single paramagnetic center on the surface. Several questions concerning details of this phenomenon in different magnetic fields and tunnelling currents, and for different paramagnetic centers are still debated. In this paper we describe the construction and the testing of an instrument able to detect the ESR-STM signal from organic paramagnetic molecules (DPPH and BDPA) deposed on Au(111) at different magnetic fields. First results on these molecules are presented.   

Exploring electron transport through organic monolayers using conductive tip AFM techniques

We follow an alternative approach to the study of Metal-molecule-Metal junctions that uses a combination of two atomic force microscopy (AFM) techniques. We use Nanografting to build a nanopatch of the molecules of interest and a second made of a reference molecule into a hosting self assembled monolayer (SAM) typically made of alkanethiols. After the tip is changed to a conductive one CT-AFM is used to characterized the whole system recording, at the same time, the system topography. Some of the advantages of this approach are the possibility to build and study a wide range of different M-m-M junctions and the in-situ control of the quality of the monolayers and patches. Results will be presented on saturated and unsaturated thiols self-assembled and nanografted on Au(111) surfaces. The results will be compared with those obtained by Liang and Scoles at Princeton using similar techniques.   

A Novel Automated System for Assembling Films of Nanoparticles

Due to their interesting properties, nanoparticle films have emerged as useful platforms for miniaturized chemical sensing. For nanoparticle sensors to become practical in real world applications, a reproducible method of assembly has to be implemented. This project focuses on robotic assembly techniques that deposit nanoparticle films on various substrates. We have developed a process to iteratively assemble and electronically characterize nanoparticle films using a custom robotic preparation system. The robot's design uses commercially available pneumatic and electronic actuators, valves and regulators to manage precision movements. Control of the robot is obtained by a custom Labview program which uses a TTL and GPIB interface to control relays, power supplies, and measurement circuitry. Sandia is a multiprogram laboratory operated by Sandia Corporation, a Lockheed Martin Company, for the United States Department of Energy under contract DE-AC04-94AL85000.   

Electronic Conduction in Metal/Molecule/Semiconductor Devices

In the field of molecular electronics, the contacts to the molecular elements are critical interfaces. The use of semiconductor contacts allows direct covalent bonding, provides an additional degree of freedom due to the semiconductor states, and, in certain circumstances, can minimize the effects of electrical shorting due to direct metal/substrate contacts. This talk will describe the development and electrical characterization of metal/molecule/semiconductor device structures on GaAs and Si active layers. In order to observe the conductance of the molecular species, rather than that of the semiconductor barrier, the semiconductor layers used in this study are generally highly doped. In these structures, the electronic conduction between the metal and semiconductor can be modulated by choice of molecular species. Several alkyl thiol and aromatic thiol molecules have been employed in order to determine the effects of molecular length, conjugation and intrinsic dipole moment. In certain molecules, conductance peaks or memory/switching effects have been observed. The current-voltage characteristics and conductance versus temperature both indicate that the molecular layers change the transport mechanism, generally involving a lower effective barrier height than that of a metal/semiconductor Schottky barrier. Studies on both n- and p- type substrates, including those with nanometer scale cap layers, allow the effects of the molecular and semiconductor barriers to be isolated. A basic conduction model has been developed, based on the electrostatics of the structure and thermionic-field-emission analysis of the semiconductor portion of the barrier.   

Towards a 160 kBit molecular electronic memory at 10$^{11}$ Bits/cm$^{2}$

Since its inception by Avirim and Ratner in 1974, molecular-based electronics has emerged as a promising alternative to scaled CMOS technology and its eventual integration limit. Here we present progress towards an electronically configurable, molecule-based 160,000 Bit random access memory at a Bit density approaching 10$^{11}$ Bits/cm$^{2}$. This device is based on a cross-bar architecture in which the active switching elements are bi-stable [2]-rotaxane supramolecules sandwiched between perpendicular arrays of SNAP-fabricated [1] metallic and n-Si nanowires at 34 nm pitch. Challenges in memory fabrication and testing will be discussed. [1] \textit{Science} \textbf{300}, 112 (2003); \textit{J. App. Phys.} \textbf{96}, 5921 (2004).   

Session P18: Focus Session: Carbon Nanotubes: Opto-Electronics

Electrically-Induced Infrared Emission from Carbon Nanotube Devices

The optical properties of carbon nanotubes (CNTs) are currently the focus of intense study. CNTs are direct band gap materials and their optical spectra have long been attributed to transitions between free particle bands. We show that studies of electrically-excited infrared (IR) emission from single nanotube molecules provide new insights into the electron-hole interactions in quasi-1D systems. We demonstrate strongly-enhanced electroluminescence from a partially suspended CNTFET operated under unipolar transport conditions [1]. In our devices, carriers are generated locally, when a single type of carrier is accelerated under high local electric fields at intra-molecular junctions to energies sufficient to create strongly correlated e-h pairs (excitons). This excitation mechanism contrasts with emission from radiative recombination of carriers (electrons and holes) injected from the opposite ends (source and drain) of a CNTFET operated under ambipolar transport conditions. The new excitation mechanism is about 1000 times more efficient than recombination of independently injected electrons and holes, and it results from weak electron-phonon scattering and strong electron-hole binding caused by one-dimensional confinement. We show that the light emission intensity increases exponentially with the drive current in partially suspended CNTFETs, while in 3D materials light emission is usually proportional to the product of the electron and the hole currents. The strong Coulomb interaction between electrons and holes in a 1D CNT creates bound excitons whose binding energies are more than an order of magnitude larger that those in 3D materials, preventing them from dissociating under electrical fields thus contributing little to drive current compared with that in 3D. Finally, the much higher exciton density achieved in our devices than that in typical photoluminescence experiments allows us to detect emission from higher excitation states in CNTs. \newline \newline [1] J. Chen, V. Perebeinos, M. Freitag, J. Tsang, Q. Fu, J. Liu, Ph. Avouris, Science 310, 1171 (2005).   

Exciton Formation and Electroluminescence Quenching during 1D Impact Excitation of Carbon Nanotubes Field-Effect Transistors

There are few studies addressing the influence of excitonic effects on the electro-optical response of carbon nanotube (CNT) devices. We present here near infra-red electroluminescence (EL) from unipolar single- wall carbon nanotube field effect transistors (CNFETs) at high drain-source voltages. The conditions for emission at high field reveal that a single carrier type induces EL in CNFETs through a mechanism involving 1D impact excitation. Well-resolved spectra show that the emission is assigned to the radiative recombination of the E$_{1,1}$ exciton. An emission quenching is also observed at high field and attributed to an exciton-exciton annihilation process and free carrier generation. Excitons binding energy in the order of 270 meV for 1.4 nm CNTs is inferred from the spectral features..   

Uni- and Ambipolar Light Emission From Inhomogeneous Carbon Nanotube FET's

Heterogeneities in the environment of CNTFETs can produce stationary, unipolar, electroluminescence, in addition to the normal ambipolar emission. We compare the unipolar emission with the ambipolar emission in the same device to characterize the unipolar emission process and show how the heterogeneities modify the electronic properties of the CNT. If a CNTFET is partially covered by a PMMA overlayer, changes in the IV characteristics are observed which correlate with discontinuities in the motion of the ambipolar emission at the PMMA boundary, and the generation of unipolar emission at the boundary. These PMMA induced changes show there is a step in the potential along the CNT at the boundary. Similarly, localized effects in both the ambi- and unipolar emission are observed in CNTFETs containing closed loops. The unipolar emission requires the junction between the portions of the single carbon nanotube that form the base of a loop must support the voltage drop needed to generate the light. Direct comparison of the ambipolar and unipolar emission in the same device demonstrates the efficiency of the unipolar processes.   

Carbon Nanotube p-n Junction Diodes

We describe the formation of p-n junctions along individual single-walled carbon nanotubes (SWNTs) using electrostatic doping techniques. The electrostatic doping preserves the pristine nature of CVD grown SWNTs, and when suspended, these diodes can be described by the ideal diode equation. The low background leakage currents coupled with a built-in electric field region to transport the quasi particles also makes these diodes ideal for studying the optical responses of SWNTs. We will describe several characteristics of SWNT diodes such as the quantum efficiency, origin of the quasi-particles (electrons and holes) currents, and effects due to excitons.   

Electronic Transport in Individual Carbon Nanotube P-N Junction Diodes

We have investigated electronic transport in single-walled carbon nanotube p-n junction diodes formed using gates to electrostatically dope the tube. Previous measurements [1] have shown that such diodes demonstrate nearly ideal turn-on behavior at room temperature and low biases, consistent with thermal activation over the junction barrier. We have performed measurements over a broad temperature range and have verified that the transport is by thermal activation. From the temperature dependence of the current-voltage characteristics, we can extract the nanotube band gap and the transmission coefficient through the p-n junction region. [1] J.U. Lee et al, App. Phys. Lett. \textbf{85}, 145 (2004)   

Negative differential conductance in suspended semiconducting carbon nanotubes

Suspended single-wall carbon nanotubes (SWCNTs) have been grown directly on metal electrodes with electrical contact by chemical vapor deposition. Extraordinary negative differential conductance was observed for the first time in suspended semiconducting SWCNTs. The current-voltage characteristics show an abrupt drop of conductance as the source-drain voltage is increased, followed by a constant differential conductance at higher voltage. The effect is qualitatively different from the recently reported negative differential conductance in metallic SWCNTs (\textit{Phys. Rev. Lett.} \textbf{95} 155505, 2005). We suggest that the observed negative differential conductance in semiconducting SWCNTs may be attributable to Schottky barriers between the as-grown suspended SWCNTs and the electrodes, instead of optical phonon scattering invoked in explaining the negative differential conductance in metallic SWCNTs. Our observations not only have potential applications for novel electronic devices, but also shed light on better understanding and manipulating SWCNTs transistors.   

Nanotube Film Electrodes in Electro-Optic Devices

The interface between conjugated polymers and conducting electrodes is crucial for the operation of organic electronic devices such as light emitting diodes (LEDs), electrochromics and photovoltaics. Transparent electrodes in these devices have been based mostly on indium tin oxide (ITO). There have been efforts to develop conducting polymer electrodes, and some success has been realized with films based on poly(3,4-ethylene-dioxythiophene)-poly(styrene sulfonate) (PEDOT-PSS).$^{1 }$In most cases however, the polymer conductivity is too low for such applications. Pure nanotube thin films, demonstrated to have much higher conductivities while exhibiting good transparency in the visible and near to mid IR, provide attractive alternatives.$^{2}$ Here we describe fabrication and performance of two devices: (1) an MEH-PPV polymer LED using a carbon nanotube film as the hole injecting electrode and (2) an infrared transmissive/absorptive electrochromic cell that makes use of the superior IR transmittance of the nanotube films. 1. A. A. Argun, A.Cirpan, J. R. Reynolds, Adv. Mater. 15, 1338 (2003\textbf{)}. 2. Z. Wu, \textit{et al.}, Science 305, 1273 (2004).   

Measurements of exciton binding energies in single wall nanotubes using field dependent photocurrent spectroscopy

We have used electric field dependent photocurrent measurements to distinguish between band-to-band and excitonic transitions in the excitation spectrum of a single wall nanotube capacitor. The zero field photocurrent spectrum is limited to carriers excited into continuum states that can freely diffuse from the nanotubes and into the metal contact. Application of an applied field allows for the separation of the bound exciton states via field ionization. Near the E$_{11}$ resonance, both excitonic and band-to-band transitions are resolvable with a binding energy of 109 meV. This is in reasonable agreement with recent theory for 1.3 nm diameter nanotubes$^{1}$. Near the E$_{22 }$resonance, we observe only a single field independent peak in the photocurrent spectrum indicating a fast decay of the exciton into the lower energy continuum states. Surprisingly, we are also able to resolve an exciton resonance associated with metallic nanotubes. Theory shows that in metallic nanotubes, optical transitions between the overlapping states at the Fermi energy are disallowed, giving rise to a symmetry gap$^{2}$. We measure the binding energy of the metallic exciton to be 49 meV for 1.3 nm diameter tubes. \textbf{References} (1) Perebeinos, V.; Tersoff, J.; Avouris, Ph., \textit{Phys. Rev. Lett.} \textbf{2004}, 92, 257402. (2) Spataru, C.D.; Ismail-Beigi, S.; Benedict, L.X,; Louie, S.G. \textit{Phys. Rev. Lett.} \textbf{2004}, 92, 077402.   

Optical Properties of Aligned Carbon Nanotube Mats for Photonic Applications

We studied the optical properties of the aligned carbon nanotube (16, 0), (10, 0) and (8, 4) mats for photonic device applications. We employed the ab-initio density functional calculations in the linear combination of atomic orbital formalism. We calculated the electronic structure of the carbon nanotube mats and the real and imaginary parts of the dielectric functions as functions of photon energy. The calculated dielectric functions of the aligned carbon nanotube mats show a strong anisotropy when the electric field of light is parallel or perpendicular to the tube axes. Especially, there are strong peaks in the imaginary part of the dielectric function near the absorption edges, when the electric field of light is parallel to the carbon nanotube axes. The unusual optical properties of the semiconducting carbon nanotube mats present a new opportunity for applications in new electro-optical devices in the infrared energy region. Acknowledgments: this work was funded in part by NSF (Award No. 0508245), NASA (Award No. NCC 2-1344), and ONR (Grant No: N00014-05-1-0009).   

Theory of Auger Recombination of Excitons in One-Dimensional Nanostructures

The effective Coulomb interaction is greatly enhanced in one-dimensional (1D) systems. As has been recently demonstrated for single-walled carbon nanotubes [1], this strong Coulomb interaction causes the formation of tightly bound exciton states upon optical excitation of semiconducting materials. The strength of the Coulomb interaction in 1D systems leads to a second consequence: Auger recombination of excitons, also known as exciton-exciton annihilation, can be very efficient. Here we investigate the 1D Auger process using a point-contact model for the Coulomb interaction. We show that the Auger process is essentially temperature independent, in contrast to the behavior of weakly bound excitons and free carriers in bulk semiconductors. We apply the explicit expression that we have derived to single-walled carbon nanotubes. We obtain an Auger rate of $\sim $0.6 ps$^{-1}\mu $m, comparable to the reported experimental value [2]. [1] F. Wang et al., Science \textbf{308}, 838 (2005); [2] F. Wang et al., Phys. Rev. B \textbf{70}, 241403 (2004).   

Measurement of Optical Stark Effect in Semiconducting Single-Walled Carbon Nanotubes

The optical Stark effect in quantum-confined systems, such as quantum wells, has been the subject of active interest for many years. In this paper we present the first measurement of the optical Stark effect in carbon nanotubes. In our experiment we used two-color femtosecond spectroscopy to probe the E$_{11}$ transition in the nanotubes while applying a strong optical pump beam at significantly lower photon energy (the large detuning limit). The sample was an aqueous suspension of (6,5)-enriched single-walled carbon nanotubes. An instantaneous shift in the absorption line by up to 1 meV was observed; the magnitude of the shift scaled linearly with the pump intensity. The nature of the optical Stark effect in the carbon nanotube system, with its strong excitonic transitions, will be discussed.   

Bandgap Modulation by Transverse Electric Fields in Single Wall Carbon Nanotubes

We study the variation of the electronic bandgap of semiconducting carbon nanotubes in a static electric field perpendicular to the nanotube axis. We consider three models for the transverse field profile and find that the spectrum is sensitive to the spatial variation of the transverse field. For a uniform transverse field, we show the bandgap is fixed until the field strength exceeds a critical value, in agreement with previous theoretical work. In contrast, we find no critical behavior when the applied field is localized to a region of the nanotube much smaller than its length. An arbitrarily weak field produces bound states inside the unperturbed bandgap whose binding energy vanishes as the fourth power of the applied field strength. The field strengths required to reduce the gap by a few percent are the same order of magnitude as those commonly used in scanning tunneling microscopy.   

Session P19: Focus Session: Spin Interference and Spin Hall Effect

Direct observation of the Aharonov-Casher phase

We report the direct observation of Aharonov-Casher effect, which can occur when electrons propagate in a ring structure in the presence of spin-orbit interactions and external magnetic field perpendicular to the ring plane. The transport measurements have been conducted on the series of ring structures fabricated from HgTe/HgCdTe quantum wells. We study Aharonov-Bohm type conductance oscillations as a function of Rashba spin-orbit splitting strength. We observe non-monotonic phase changes indicating that an additional phase factor modifies the electron wave function. We associate these observations with the Aharonov-Casher effect and confirm it by numerical calculations of the magneto-conductance for a multichannel ring within the Landauer-B\"uttiker formalism.   

Spin Interference Effect in a Square Loop Array including the Rashba and Dresselhaus Terms

The effect of electron wave interference to the electric conductivity ($\sigma$), including the effect of spin degree of freedom, is investigated through nanolithographically defined square (and other) loop array structures fabricated on In$_ {0.52} $Al$_{0.48}$As/In$_{0.53}$Ga$_{0.47}$As/In$_{0.52}$Al$_{0.48} $As quantum wells (QW). In this experiment, we measure $\sigma$'s of QWs as a function of magnetic field $B$ (${\bf B}$$\perp$QW plane). These samples had a gate electrode covering the entire loop array structures, where a gate voltage $V_g$ was applied between the metal gate electrode and the QW. We note that $V_g$ controls both the carrier density and the Rashba and Dresselhaus spin-orbit terms within the QWs. It turned out that the magnetoconductance $\sigma(B)$ oscillates as a function of $B$ with a period corresponding to $h/2e$, which is denoted as the AAS oscillation. We found that the amplitude of the AAS oscillation in this system also oscillated as a function of $V_g$, which is called as a ``spin interference'' effect. We investigated this effect, which is also in close relation to the ``Aharonov-Casher'' effect (electric control of the phase of the electronic wave function), in detail including both the Rashba and Dresselhaus spin-orbit terms quantitatively.   

Numrical simulation of a spin interferometer based on a single square loop with Rashba interaction

We numerically calculate the transverse conductance as a function of magnetic field in two models. One is an exact 1D model and the other is a quasi 1D square loop system which is similar to the experimental setup by Koga et. al. From the conductance curves, we employ FFT (Fast Fourier transform) and IFFT (inverse Fast Fourier transform) to extract separately the oscillatory part of conductance whose period correspond to the magnetic flux quanta (AB oscillation) and half quanta (AAS oscillation). We show that the spin precession angle $\theta$ is modulated by the Rashba interaction strength. From the curves about the amplitude of AB or AAS oscillations at B=0 versus $\theta$, we find that the node positions of $\theta$ in the exact 1D model fits well with previous theoretical calculations, but there are some deviations for the quasi 1D model.   

Ahronov-Bohm oscillations in a GaAs 2D hole system

We have grown shallow 2D hole samples in GaAs for implementing mesoscopic structures via local anodic oxidation using an AFM. In this talk we present our results of a ring device which shows clear Ahronov-Bohm oscillations. The amplitude of the resistance oscillations are about 10{\%} of the ring resistance, the strongest seen in a 2D hole system to date.   

Zero-field spin-splitting in Al$_{x}$Ga$_{1-x}$N/GaN heterostructures

We have observed the beating Shubnikov-de Haas oscillations with respect to the zero-field spin splitting of 2DEG in Al$_{x}$Ga$_{1-x}$N/GaN heterostructures. The spin-splitting energy was obtained about 9 meV from the beating SdH frequency derived by the non-linear curve fitting. A new mechanism ($\Delta _{C1}-\Delta _{C3}$ coupling) was proposed to describe the large spin splitting in wurtzite GaN, which is originated from the band folding effect and intrinsic wurtzite structure inversion asymmetry. The band-folding effect generates two conduction bands ($\Delta _{C1}$ and $\Delta _{C3})$, in which $p$-wave probability has tremendous change when $k_{z}$ approaches anti-crossing zone. The $\Delta _{C1}-\Delta _{C3}$ coupling can produce a spin-splitting energy much larger than traditional Rashba or Dresselhaus effects. This project is supported in parts by National Science Council, Core Facilities Laboratory in Kaohsiung-Pintung area, Taiwan (ROC).   

Exact Landau Levels in Two-Dimensional Electron Systems with Rashba and Dresselhaus Spin-Orbit Interactions in a Perpendicular Magnetic Field

We study a two-dimensional electron system in the presence of both Rashba and Dresselhaus spin-orbit interactions in a perpendicular magnetic field. Defining a suitable boson operator and using the unitary transformations we are able to obtain the exact Landau levels in the range of all the parameters. When the strengths of the Rashba and Dresselhaus spin-orbit interactions are equal, the Zeeman and spin-orbit splittings are independent of the Landau level index $n$. Due to the Zeeman energy, new crossing between the eigenstates $|n, k, s=1, \sigma>$ and $|n+1, k, s^\prime =-1, \sigma^\prime>$ is produced at certain magnetic field for larger Rashba spin-orbit coupling. This degeneracy leads to a resonant spin Hall conductance if it happens at the Fermi level.   

Spatial imaging of the spin Hall effect and current-induced polarization in two-dimensional electron gases

Spin-orbit coupling in semiconductors relates the spin of an electron to its momentum, and provides a pathway for electrically initializing and manipulating electron spins in zero magnetic field for applications in spintronics and spin-based quantum information processing. This coupling can be regulated with strain in bulk semiconductors and quantum confinement in semiconductor heterostructures. Using Faraday and Kerr rotation spectroscopies with temporal and spatial resolution, current-induced spin polarization\footnote{Y. K. Kato, R. C. Myers, A. C. Gossard, D. D. Awschalom, \textit{Phys. Rev. Lett.} \textbf{93}, 176601 (2004).} and the spin Hall effect\footnote{Y. K. Kato, R. C. Myers, A. C. Gossard, D. D. Awschalom, \textit{Science} \textbf{306}, 1910 (2004).} have been observed in bulk semiconductors. More recently, we have investigated the spin Hall effect and current-induced spin polarization in a two-dimensional electron gas confined in (110) AlGaAs quantum wells using Kerr rotation microscopy\footnote{V. Sih, R. C. Myers, Y. K. Kato, W. H. Lau, A. C. Gossard and D. D. Awschalom, \textit{Nature Physics} \textbf{1}, 31 (2005).}. In contrast to previous measurements, the spin Hall profile shows complex structure and the current-induced spin polarization is out-of-plane. The experiments map the strong dependence of the current-induced spin polarization to the crystal axis along which the electric field is applied, reflecting the anisotropy of the spin-orbit interaction. These results reveal opportunities for tuning a spin source using quantum confinement, strain and device engineering in non-magnetic materials.   

Spin Hall effect, spin-accumulation, and spin-currents in mesoscopic structures

Spin dependent transport effects in strongly spin-orbit coupled paramagnetic systems, such as the spin Hall Effect, have been studied extensively over the last few years. We explore how spin accumulation in a mesoscopic device could be used to observe the effect through electrical and optical means. We report calculations of spin flow in finite size samples with strong spin-orbit coupling using the non-equilibrium Green's function formalism in both the linear and the non-linear regimes. We explore different geometries and spin-orbit coupling mechanism to understand how spin relaxes near the interfaces. We will also report on the progress made in understanding the spin Hall Effect in the bulk regime and how it connects to the closely related effect of the anomalous Hall effect in ferromagnetic materials.   

The finite spin Hall effect in semicondutors

We formulate the theory of the spin Hall effect taking into account the impurity scattering effect as general as possible with the focus on the definition of the spin current. The conserved spin current (Zhang {\it et, al}. [cond- mat/0503505]) satisfying the continuity equation of spin in the bulk is considered in addition to the conventional one defined by the anti-symmetric product of the spin and velocity operators. The condition for the non-zero spin Hall current is clarified from a generic viewpoint and the following new results for explicit models are obtained: (i) spin Hall current in Rashba model is non-zero on the non-delta impurity scattering potential, and (ii) spin Hall current vanishes in cubic Rashba model on the delta impurity scattering potential.   

What is intrinsic and what is extrinsic in the spin Hall effect?

Two different forms of the spin Hall effect, intrinsic and extrinsic, have been recently proposed and observed in experiments. The intrinsic effect is caused by spin-orbit coupling in the band structure of the semiconductor and survives in the limit of zero disorder, whereas the extrinsic effect is caused by spin-orbit coupling between Bloch electrons and impurities. We treat both effects on equal footing within the framework of the exact Kubo linear response formalism. We show that the ``side-jump" term, which is usually considered part of the extrinsic spin Hall effect, is really intrinsic, because it is independent of disorder. Furthermore, it is the only non-zero {\it intrinsic} contribution to the spin-Hall effect for the linear Rashba (or Dresselhaus) spin-orbit coupling model. On the other hand, the skew scattering term is the only {\it extrinsic} contribution to the spin-Hall effect within this model. The proof based on gauge invariance holds at all orders in disorder and electron-electron interactions and to first order in spin-orbit coupling, but does not apply to more complex spin-orbit coupled bands (e.g the Luttinger model). We also study many-body effects and predict that the spin Coulomb drag will reduce the spin Hall conductivity.   

Spin Hall Effect in Doped Semiconductor Structures

We present a microscopic theory of the extrinsic spin Hall effect based on the diagrammatic perturbation theory. Side-jump (SJ) and skew-scattering (SS) contributions are explicitly taken into account to calculate the spin Hall conductivity, and we show their effects scale as $\sigma_{xy}^{SJ}/\sigma_{xy}^{SS} \sim (\hbar/\tau)/\varepsilon_F$, where $\tau$ being the transport relaxation time. Motivated by recent experimental work we apply our theory to n-doped and p-doped 3D and 2D GaAs structures, obtaining analytical formulas for the SJ and SS contributions. Moreover, the ratio of the spin Hall conductivity to longitudinal conductivity is found as $\sigma_s/\sigma_c \sim 10^{-3}-10^{-4}$, in reasonable agreement with the recent experimental results of Kato \textit{et al}. [Science 306, 1910 (2004)] in n-doped 3D GaAs system.   

Intrinsic spin Hall conductivity

In an isotropic 2D gas with general dispersion and linear-in-k spin-orbital interaction of the Rashba or Dresselhaus type in the presence of impurities we find that an intrinsic spin-Hall conductivity is finite and is of the order of the spin-orbit term squared. It vanishes only in the well-studied particular case of a quadratic dispersion.   

Imaging Stationary Flow of Spin Hall Effect-Induced Spin Densities in Mesoscopic Nanostructures

The spin Hall effect has recently attracted a lot of attention in semiconductor spintronics since it offers a novel way of all-electrical generation and manipulation of pure spin currents by employing spin-orbit (SO) couplings. To describe spin Hall transport on a scale of a few nanometers we introduce the concept of bond spin current and corresponding local flowing spin densities between the sites of the lattice model of a multiterminal SO coupled semiconductor nanostructure, and express them in terms of the nonequilibrium (Keldysh) Green functions. Our predictions for the out-of-plane polarized steady state spin densities flowing into the transverse interaction-free electrodes due to the longitudinal charge current injected into high-mobility two-dimensional electron gas (2DEG) with Rashba SO coupling crucially depend on the size of 2DEG in the units of spin precession length. In the presence of disorder, the flowing spin Hall densities remain non-zero in the bulk of the 2DEG. Moreover, we also find in-plane polarized spin densities flowing into the longitudinal leads due to the magneto-electric effect. These theoretically predicted images of mesoscopic spin Hall flow could be tested via recently advanced Kerr rotation microscopy.   

Session P20: Focus Session: Cobaltites, Nickelates and Vanadates

Intrinsic nanoscale electronic phase separation and simple percolation in La$_{1-x}$Sr$_{x}$CoO$_{3}$

The doped pervoskite cobaltite La$_{1-x}$Sr$_{x}$CoO$_{3}$ has been advanced as a model system for studying magnetoelectronic phase separation. We present here a combination of chemically sensitive high-resolution TEM, SANS, and transport data that reveal interesting new features of this phase separation. The TEM data show that the material is chemically homogenous down to nm length scales, proving that the phase separation is truly intrinsic electronic phase separation. The SANS data, which were performed at several compositions below x = 0.18 (where long-range ferromagnetism (FM) sets in), reveal that the FM clusters have a maximum size of about 2-3 nm, \textit{independent} \textit{of doping}. This demonstrates that the percolation transition that occurs at x = 0.18 is due to an increasing density of clusters with increasing x, \textit{not} an expansion of cluster size. These observations naturally explain the simple percolation observed in single crystal transport, i.e. conductivity exponents close to predicted values and a critical composition (x = 0.18) close to the expected value for the 3-D percolation limit.   

Short-range magentic correlations and dynamic orbital ordering in the thermally activated spin state of LaCoO$_3$

The cobalt perovskites La$_{1-x}$Sr$_x$CoO$_3$ show intriguing spin, lattice, and orbital properties similar to the ones observed in colossal magnetoresistive manganites. The x=0 parent compound is a non-magnetic insulator at low temperatures, but shows evidence of a spin-state transition of the cobalt ions above 50K from a low-spin to an intermediate or high-spin configuration. Using high resolution, inelastic neutron scattering, we observe a distinct low energy excitation at 0.6meV coincident with the thermally induced spin state transition observed in susceptibility measurements. The thermal activation of this excited spin state also leads to short-range, dynamic ferro- and antiferromagnetic correlations. These observations are consistent with the activation of a zero-field split intermediate spin state as well as the presence of dynamic orbital ordering of these excited states. \\ \\ Work supported by US DOE BES-DMS W-31-109-ENG-38 and NSF DMR-0454672   

Emergence of Magnetism in La$_{1-x}$Sr$_x$CoO$_3$

Orbital, spin, and charge degrees of freedom play a central role in the physics of CMR-type transition metal perovskite oxides. La$_{1-x}$Sr$_x$CoO$_3$ is a system in which a ferromagnetic, metallic state emerges when holes are doped into the parent compound, a non-magnetic, Mott insulator in the ground state. We have studied this system using elastic and inelastic neutron scattering techniques on single crystals with 0$\leq$x$\leq$0.2. With hole doping the ferromagnetic correlations between Co spins become static and isotropically distributed due to the formation of ferromagnetic droplets. The correlation length and condensation temperature of these droplets increase rapidly with metallicity due to the double exchange mechanism. Diffuse spin dynamics appear as the correlation length increases. The dynamics are broad in energy indicative of a spin wave continuum.   

Local Matrix-Cluster Interactions In La$_{1-x}$ Sr$_{x}$CoO$_{3}$.

Magneto-electronic phase separation plays an integral part in many recent advances in the understanding of correlated electron systems. We have studied the magnetically phase separated material La$_{1-x}$ Sr$_{x}$CoO$_{3}$ and the parent compound LaCoO$_{3}$, using muon spectroscopy and magnetic susceptibility measurements. The muon as a local magnetic probe is sensitive to the magnetic field distribution in LaCoO$_{3}$ in the LS state, which is a direct consequence of magnetic excitons. We believe that these excitons are interacting with the Co ions undergoing the known thermally induced spin transition. By directly comparing the results of the parent compound with La$_{1-x}$ Sr$_{x}$CoO$_{3 }$we can observe the hole-rich ferromagnetic clusters interacting with the neighboring hole poor matrix for low x. This mechanism, detected here for the first time, may play an important role in the rich electrical and magnetic properties of La$_{1-x}$ Sr$_{x}$CoO$_{3}$.   

Anisotropy-driven magnetic anomalies in Pr$_{1-x}$Sr$_{x}$CoO$_{3}$

Interest in the perovskite cobaltites has been growing steadily due to the intriguing phenomena they exhibit. It is well known that the availability of various Co ion spin states in the cobaltites provides an additional degree of freedom in comparison to the manganites. In this work we demonstrate that the cobaltites also possess another factor of considerable importance not present in the manganites -- large magnetocrystalline anisotropy. As previously reported [Mahendiran et al PRB 68 024427 (2003)] at x $>$ 0.30 Pr$_{1-x}$Sr$_{x}$CoO$_{3}$ displays an additional anomaly below the Curie temperature, where the magnetization can dramatically increase or decrease depending on applied field. We demonstrate here, using magnetometry, transport, heat capacity, and neutron diffraction, that this results from a structural phase transition from a low symmetry to higher symmetry (tetragonal) phase on reducing T. Although the Co moment is unaffected, a sharp change in the magnetocrystalline anisotropy takes place and is reflected in the hysteresis loop shape, coercivity, and remnance. The complex and puzzling behavior of the field dependence of the magnetization vs. T curves is then simply explained by the T dependent variations in hysteresis loop shape.   

Magnetic properties and phase separation in Pr$_{1-x}$Sr$_{x}$CoO$_{3}$, using $^{59}$Co NMR

Doped transition metal oxides including manganites and cobaltites have revealed a rich variety of properties that may be technologically important. The mixed valence cobaltite Pr$_{1-x}$Sr$_{x}$CoO$_{3 }$(PSCO) has a phase diagram reminiscent of La$_{1-x}$Sr$_{x}$CoO$_{3 }$(LSCO) but with a number of significant differences. For x=0.5 the system is ferromagnetic (FM) below $T_{C}$= 240 K but anomalous magnetization behavior is found close to 120 K with an associated crystal structure change from low symmetry to tetragonal with decreasing $T$. For x $<$ 0.3 no change in magnetic properties or crystal structure is found below $T_{C}$. Zero-field $^{59}$Co NMR spectra show that differences in FM character between x=0.5 and x=0.3 samples are negligibly small at temperatures in the range 3-30 K No FM line was observed for x=0.2; a narrow paramagnetic- like signal only slightly shifted from the diamagnetic $^{59}$Co spectrum is observed in high- field NMR for all three x values, providing evidence of some form of phase separation where a paramagnetic phase coexists with the FM phase. The results will be compared with the very different phase separation data previously obtained for LSCO.   

Structural and Magnetic Properties of the Kagom\'{e} Antiferromagnet YbBaCo$_{4}$O$_{7}$

The mixed-valent compound YbBaCo$_{4}$O$_{7}$ is built up of Kagom\'{e} sheets of CoO$_{4}$ tetrahedra, linked in the third dimension by a triangular layer of CoO$_{4}$ tetrahedra in an analogous fashion to that found in the known geometrically frustrated magnets such as pyrochlores and SrCr$_{9x}$Ga$_{12-9x}$O$_{19}$ (SCGO). We have undertaken a study of the structural and magnetic properties of this compound using combined high resolution powder neutron and synchrotron X-ray diffraction. YbBaCo$_{4}$O$_{7}$ undergoes a first order trigonal to orthorhombic phase transition at 175 K that breaks the trigonal symmetry of the structure. We show that this transition occurs as a response to a markedly underbonded Ba$^{2+}$ site in the high-temperature phase and does not appear to involve charge-ordering of Co$^{2+}$/Co$^{3+}$ ions in the tetrahedra. The symmetry-lowering relieves the geometric frustration of the structure, and a long-range ordered 3-D antiferromagnetic state develops below 80 K.   

Crystal Structure and Magnetic Properties of an oxygen deficient n = 2 Ruddlesden-Popper phase Sr$_{3}$Co$_{2}$O$_{5.67}$

Interest in charge, orbital, and spin state phenomena in perovskite and related cobalt oxides is a growing area of transition metal oxide physics. Recently, J. Matsuno \textit{et al.}\footnote{ J. Matsuno \textit{et al}., PRL \textbf{93}, 167202 (2004).} have found that epitaxial films of the n = 1 Ruddlesden-Popper (R-P) phase Sr$_{2}$CoO$_{4}$ are metallic ferromagnets with relatively high T$_{C} \quad \sim $ 250 K. This is particularly interesting in light of the formal oxidation state of Co, Co$^{4+}$, offering no clear source of carriers. To extend the materials chemistry and physics of the R-P series of cobaltites, we have synthesized the n = 2 R-P phase Sr$_{3}$Co$_{2}$O$_{7-\delta }$ in bulk form. The crystal structure [from neutron powder diffraction (NPD) data] of our most oxygen-deficient sample, Sr$_{3}$Co$_{2}$O$_{5.67}$ is orthorhombic \textit{Immm} with a = 3.94025(9) {\AA}, b = 3.67479(9) {\AA} and c = 20.6642(5) {\AA}. The magnetization versus temperature data show two antiferromagnetic transitions at approximately 170 K and 220 K. To further elucidate the magnetic properties of this material, we have conducted a temperature-dependent NPD study. The low temperature magnetic structure is surprisingly complex and suggestive of an incommensurate ordering wave vector. Full details and results of the NPD study will be given.   

Complex magnetic structure of YBaCo4O7

The new series of mixed-valent oxides RBaCo4O7 (R=Yb,Tb,Y) show complex structural and magnetic behavior. We have recently revealed, for the Yb analog, that a stuctural phase transition occurs in response to an extremely underbounded Ba2+ site. The symmetry lowering from orthorhombic to tetragonal, releases the frustration and allows the system to order magnetically below ~80K. Here we present our neutron diffraction study of the analog compound YBaCo4O7, that shows the same structural phase transition at high temperature and a magnetic transition at around 110K, where the system orders antiferromagnetically with propagation vector k=0. The magnetic structure, solved by global optimization algorithms, shows a non colinear Co-spins arrangement that results from the unique topology of the Co interactions. The magnetic structure is found to be strongly temperature dependent between 1.6K and 110K, which provide crucial information about the relative strengths of competing interactions.   

The effect of transition metal ions distribution on magnetic properties of Li$_{x}$(Ni$_{y}$Mn$_{y}$Co$_{1-2y})$O$_{2}$.

Li$_{x}$(Ni$_{y}$Mn$_{y}$Co$_{1-2y})$O$_{2}$ compounds have layered O(3) structure with an occupancy disorder as Ni ions migrate to the lithium layer. Ni ions provide strong antiferromagnetic (AF) exchange between the transition metal (TM) layers; therefore the degree of disorder has a pronounced effect on the magnetic properties. Ni migration is reduced when the amount of Co or Li is increased. In this work we study temperature and magnetic field dependences of magnetization and the ac susceptibility of Li$_{x}$(Ni$_{y}$Mn$_{y}$Co$_{1-2y})$O$_{2}$ with various Li and Co contents. We have shown that in LiNi$_{0.5}$Mn$_{0.5}$O$_{2}$ compound large amount of Ni on Li sites facilitates AF order within the TM layer, while interlayer Ni ions contribute to the net magnetic moment. This is consistent with the ``flower'' order of the TMs proposed from the Monte-Carlo simulations. With increasing Co content, the ``flower'' structure is destroyed and a spin glass state is observed in Co-containing compounds. This work is financially supported by the US Department of Energy, Office of FreedomCAR and Vehicle Technologies, through the BATT program at LBNL.   

Observation of Magnetic Memory Effect and Photo-induced Magnetism in Y$_{0.33}$Sr$_{0.67}$CoO$_{3-\delta}$

We prepared the Y$_{0.33}$Sr$_{0.67}$CoO$_{3-\delta }$by the conventional solid state method which sintered under the O$_{2}$ flow. The sample was finally annealed under the oxygen and nitrogen atmosphere. A DC magnetization jump was found about 200 K with a large thermal hysteresis at 0.01 T indicating a kind of magnetic memory effect. The magnetization jump comes from the inter-spin state transition on Co$^{3+}$ ion from low to intermediate spin state. The magnetic memory effect gradually disappears with the magnetic field increase and the jump temperature ($T_{J})$ shifts to low temperature. Annealed samples indicate high $T_{J}$, $T_{C}$ and the magnetization coming from the oxygen content difference. Under the irradiation of a pulsed near-infrared laser ($\lambda $ = 1050 nm), the $T_{J}$ shifts to low temperature and the magnetization below T$_{J}$ decreases. Photo-induced effect is weakened with the magnetic field. Laser irradiation may suppress spin-state transition of the part Co$^{3+}$ ions.   

Local Electronic and Spin Structure of GdBaCo2O5.5 from X-ray Absorption Spectroscopy

The family of RBaCo$_{2}$O$_{5+\delta}$ cobaltates is known to exhibit a rich variety of magnetic behavior as a function of oxygen content and temperature. We present x-ray absorption measurements on detwinned single crystals of GdBaCo$_{2}$O$_{5.5}$, where the structure is comprised of alternating rows of CoO$_{6}$ octahedra and CoO$_{5}$ pyramids. GdBaCo$_{2}$O$_{5.5}$ exhibits successive paramagnetic, ferromagnetic, and antiferromagnetic phases, and also exhibits a ``spin blockade'' effect upon doping. These unusual behaviors are believed to stem from the nearly degenerate spin states of the Co$^{3+}$ ions which can potentially vary from low (S=0), intermediate (S=1), to high (S=2) spin states. Our recent x-ray absorption measurements provide the first measurements of the local electronic and spin states. Measurements of the temperature and polarization dependence of the x-ray absorption at the oxygen K edge clearly indicate an abrupt change in the orbital populations at the metal-insulator transition at T $\sim$ 360 K. We combine our spectroscopic measurements with atomic multiplet and LSDA+U calculations to provide a first insight into the true nature of the spin state transitions which govern the unusually rich magnetic properties of the RBaCo$_{2}$O$_{5+\delta}$ cobaltates.   

High Pressure and High Resolution Magnetization of GdBaCo$_{2}$O$_{5.5}$

We present the results of two rather diverse experiments designed to reveal new features of the complex magnetic properties of GdBaCo$_{2}$O$_{5.5}$ and, in particular, lanthanide/transition metal coupling in oxide materials. First, high resolution magnetization, M (H,T), measurements on an untwinned single crystal show a small but non-zero coupling between the 3d-shell Co-based magnetic order and the 4f shell Gd paramagnetism. Second, high pressure magnetization measurements on a polycrystalline sample suggest a weakening of the ferromagnetic interplane coupling at the expense of a strengthening of the antiferromagnetic interplane coupling. In the measured pressure range, however, no unambiguous pressure-induced spin transition was observed. Finally, low temperature isothermal magnetization measurements under pressure indicate a slight weakening of the effective Co-generated molecular field, affecting the Gd paramagnetism. The totality of data contained in this work suggests that there is a small but definitive molecular field effect at the Gd sites, which is a function of the strength of transition metal magnetism ordering at higher temperatures.   

Electron-phonon coupling and low-temperature structure of NaV$_2$O$_5$

NaV$_2$O$_5$ is an extraordinary example of a structure where charge, spin and lattice degrees of freedom strongly interact. This low-dimensional compound is characterized by V atoms arranged in the form of ladders. At ambient conditions it is found to be quarter-filled with one electron distributed over one rung of the ladder. Going below T$_c$ = 34K, NaV$_2$O$_5$ undergoes a phase transition involving a reordering of the V charges, a lattice deformation and a spin pairing. In order to investigate the mechanisms driving the phase transition, parameters of electron-phonon and spin-phonon for the $\Gamma$ point phonons are determined from ab initio calculations within density functional theory. They are compared to the corresponding parameters of the isostructural CaV$_2$O$_5$, where no phase transition occurs. Moreover, ab initio results of several candidates for the low-temperature supercell of NaV$_2$O$_5$ are presented and analyzed in terms of total energies and electric field gradients.   

Quasi-one-dimensional electronic structure of $\beta^{\prime}$-Cu$_x$V$_2$O$_5$ ($x$=0.33$\sim$0.65) studied by photoemission

$\beta^{\prime}$-Cu$_x$V$_2$O$_5$ is a quasi-one-dimensional (quasi-1D) oxide that undergoes a metal to insulator transition (MIT) when $x$ decreases from 0.65 to below 0.60. It becomes a superconductor below 6K under pressure around 3GPa. In the metallic phase, the electrical resistivity along the chain direction is 30 times larger than that across the chain direction, making this material suitable for studies of quasi-1D electronic structures. We present the first photoemission spectra of $\beta^{\prime}$-Cu$_x$V$_2$O$_5$ in both metallic (x=0.60, 0.65) and insulating (x=0.33, 0.55) phases. Angle-integrated spectra show a clear indication of the MIT. Nonetheless the intensity near the Fermi energy ($E_{\rm F}$) is heavily suppressed in the metallic phase, just as in the spectra of Li$_{0.9}$Mo$_6$O$_{17}$ and certain other low dimensional oxides [1]. We observe a single band crossing $E_{\rm F}$ along the chain direction, around the $\Gamma$-point of the Brillouin zone, only in the metallic phase angle-resolved spectrum. Fermi surface intensity maps have clear 1D character and the Fermi wavevector changes according to the concentration of the dopant. [1] G.-H. Gweon, J.W. Allen, and J.D. Denlinger, Phys. Rev. B {\bf 68}, 195117 (2003).   

Session P21: Microfluidic Physics III

Multi-Point Holographic Micro-Velocimetry

We show how holographic optical trapping can be used for the multi-point measurement of fluid flow in microscopic geometries. An array of microprobes can be simultaneously trapped and used to map out the fluid flow in a microfluidic device. The optical traps are alternately turned on and off such that the probe particles are displaced by the flow of the surrounding fluid and then re-trapped. The particles' displacements are monitored by digital video microscopy and directly converted into velocity field values. The validity of the technique is demonstrated for the case of the flow around a spinning sphere and the flow at the outlet of a micro-channel.   

A Hybrid Microwave Source and Irradiator for Biological Lab On a Chip Applications.

Using a standard lithographic process, we have built a hybrid microwave irradiator for use in microwave enhanced chemistry and localized, rapid heating. The device combines a 100mW microwave source with a near field antenna to produce an entirely on-chip system for delivering microwave energy into a thin($<$100$\mu $m) layer above a substrate. The antenna utilizes a serpentine wire pattern to produce a thin layer of intense microwave electromagnetic field intensity that falls off exponentially in distance away from the substrate. The device, including RF electronics, was built on a standard 1'' by 3'' glass slide, and several antenna pixel sizes are tested for Biological Lab On a Chip Applications. This work is made possible by the NSEC NSF grant PHY-0117795.   

Simulations of Contact Line Motion in Partially Miscible Fluids

We report on extensive molecular-dynamics simulations of contact line motion in partially miscible fluids confined between two solid walls and sheared in a Couette geometry. Our results show that diffusion alone cannot remove the stress singularities at the contact line or lead to no-slip boundary conditions on the fluid velocity. Computed velocity fields show that there is a substantial drop of the fluid velocity near the contact line, which is associated with the gradient of the fluid-solid interfacial tension in the same region. However, the fluid velocity does not fall to zero at the contact line, in contrast to the case where fluids are immiscible. The nonzero velocity leads to a net advective flux across the fluid-fluid interface, which is balanced by the diffusive flux induced by the concentration gradient. The advective and diffusive fluxes across the interface are only significant in the very first layer of fluid atoms.   

Macromolecular Liquids Slip Over Solid Surfaces: Experimental Studies of the Slip Length

We present a novel method to asses the slip length and viscosity of thin films of highly viscous Newtonian liquids. We quantitatively analyze dewetting fronts of low molecular weight polystyrene melts on Octadecyl- (OTS) and Dodecyltrichlorosilane (DTS) polymer brushes [1]. Using a thin film (lubrication) model derived in the limit of large slip lengths, we can extract slip length and viscosity of films with thicknesses between 50 nm and 230 nm and temperatures above the glass transition. We find slip lengths from 100 nm up to 1 micron on OTS and between 300 nm and 10 microns on DTS covered silicon wafers. The slip length decreases with temperature. The obtained values for the viscosity are consistent with independent measurements [2]. [1] R. Fetzer, K. Jacobs, A. Muench, B. Wagner, T.P. Witelski, Phys. Rev. Lett. 95, 127801 (2005) [2] R. Fetzer, K. Jacobs, M. Rauscher (to be published)   

Source of Shear Dependent Slip at Liquid/Solid Interfaces

Slippage at liquid/solid interfaces can strongly influence transport behavior in micro- and nanoscale systems. Previous molecular dynamics (MD) studies of simple and polymeric fluids subject to planar shear at small Reynolds number have shown that the slip length increases as a power law in the shear rate for moderate to high values. The corresponding boundary condition provides a new generalization of the Navier slip law. In this talk, we examine what physical mechanism is responsible for the shear rate exponent by focusing on the collision events between the fluid particles in the first layer and the adjacent wall particles comprising a crystalline surface. By examining the interfacial frictional force as a function of the fluid sliding velocity, we recover similar behavior as inherent in the generalized slip condition and determine that the dominant frictional response stems from the repulsive part of the Lennard-Jones interaction potential. A reduced kinetic model describing the scattering of a single molecule with a given slip velocity along a crystalline surface helps explain the saturation in the frictional force at large sliding velocities. These results elucidate how different is the slip behavior at liquid/solid interfaces from that observed in rarefied gases.   

Apparent Slip at Hydrophilic Surface: Flow Profile within 1 nm from the Surface

Fluid dynamics within small channels draws great interest due to the development of microfluidic devices, yet details about flow immediately at a solid surface remain too vague.~ Here, by using fluorescence energy transfer (FRET and fluorescence quenching) approaches, we measured the flow rate of fluorescence quencher molecules within 1 nm from the quartz surface within a specially-designed microfluidic device.~ In parallel, we have simulated the flow dynamics at the surface, in order to separate cleanly the actual near-surface velocity from the confounding effects of near-surface diffusion.~~   

Slip versus Friction : Modifying the Navier condition

The modeling of fluid-solid interfaces remains one of the key challenges in fluid mechanics. The prevailing model, attributed to Navier, defines the fluid ``slip'' velocity as proportional to the wall shear and a parameter defined as the slip length. Several works have in turn proposed models for this slip length but no universal model for the slip velocity has been accepted. We present results from large scale molecular dynamics simulations of canonical flow problems, indicating, that the inadequacy of this classic model, stems from not properly accounting for the pressure field. We propose and validate a new model, based on the fundamental observation that the finite ``slip'' velocity is a result of an imbalance between fluid and solid intermolecular forces. An excess force on the fluid elements will lead to their acceleration which in turn may result in a slip velocity at the interface. We formulate the slip velocity in terms of fluid-solid friction $F_f}$ and propose a generalized boundary condition: $F_{f} = F_{s} + F_{p} = \lambda_u u_{s} + \lambda_{p} p$ where $\mbox{ p}$ denotes the pressure, and $\lambda _u $and $\lambda _p$ the viscous and static friction coefficients, for which universal constants are presented. We demonstrate that the present model can overcome difficulties encountered by the classical slip model in canonical flow configurations.   

Slip and Air-Entrainment at Water-Solid Interfaces

A number of recent studies performed with water flow past hydrophobic microchannels have reported the existence of `slip' at wall and suggested the existence of the interfacial gas layer as the underlying mechanism for the slip motion, yet the details are much disputed. We combine microscopy and advanced laser spectroscopy to directly and non-invasively detect the interfacial gas layer in flowing water past micro/nano-channels whose surface chemistry and gap spacing are varied. We observe that the dimension of the gas layer strongly depends on surface hydrophobicity and flow rates. Surprisingly, we have also observed the slip motion of water over hydrophilic surfaces with a strong dependence on liquid-loading conditions. We propose a mechanistic theory about air-entrainment that can account for our observations to elucidate the origin of the gas formation at water-solid interface and its consequence on slip motion.   

Rheology of sub-nanometer thick water films

Knowing the behavior of water in small volumes is essential for the understanding of many processes in biology, tribology, and geophysics. Water under nano-confinement plays a crucial role in biological and technological systems. Here, we report an experiment in which an atomic force microscope tip approaches a flat solid surface in purified water, while small lateral oscillations are applied to the tip. The normal and lateral forces acting on the tip are measured directly and simultaneously as a function of water thickness. We find that, for hydrophilic surfaces, oscillatory solvation forces are present in the last four adjacent water layers where the dynamic viscosity is measured to grow up orders of magnitude in respect to bulk water. The same effects are present for atomically smooth surfaces and slightly rough surfaces. Oscillatory solvation forces have been detected also when the confining flat surface was hydrophobic.   

Wetting morphologies on surfaces nanopatterned with chemical stripes

Here we investigate the wetting of simple, volatile liquids on model chemical nanopatterns created using Local Oxidation Nanolithography. This technique makes use of a biased, metallic AFM tip to locally oxidize the methyl-terminations of a self-assembled monolayer (octadecylthrichlorosilane) into carboxylic acid termination[1]. With this method we have realized parallel, 50 to 500 nm wide, wettable stripes (carboxylic) embedded into a non-wettable (methyl) surface. Several organic (polar, non-polar), volatile liquids have been condensed onto the nanopatterned surface and the resulting wetting morphologies have been studied in-situ by using an environmental AFM. Initially the liquid only condenses on the wettable stripes to form a thin liquid film. Close to saturation the liquid morphology becomes drop-like. Eventually, when more and more liquid is condensed on the stripes, the liquid drops may ``spill over'' into the non-wettable spacer so that neighboring lines merge and undergo a ``morphological wetting transition''. For all of these regimes we show that long-range forces are relevant to nanoliquid ``shape''. Results will be compared with those of Density Functional Theory.[1] R. Maoz, S. Cohen, and J. Sagiv, Adv. Mater. \textbf{11}, 55 (1999)   

Self-propelled film-boiling liquids

We report that liquids perform self-propelled motion when they are placed in contact with hot surfaces with asymmetric (ratchet-like) topology. Millimeter-sized droplets or slugs accelerate at rates up to 0.1 g and reach terminal velocities of several cm/s, sustained over distances up to a meter. The pumping effect is observed when the liquid is in the film-boiling regime, for many liquids and over a wide temperature range. We propose that liquid motion is driven by a viscous force exerted by vapor flow between the solid and the liquid. This heat-driven pumping mechanism may be of interest in cooling applications, eliminating the need for an additional power source.   

Ratcheting motion of capsules on tailored substrates

We study the motion of microcapsules on attractive surfaces. The capsules, modeled as fluid-filled elastic shells, represent polymeric microcapsules or biological cells. Certain periodic surface patterns give rise to directed capsule motion for a symmetric energy input, such as an oscillatory shear flow. We use a numerical model which integrates a lattice spring representation of the capsule shell and the substrate with a lattice Boltzmann representation for the fluid regions. We consider, as a surface pattern, a series of asymmetric ramps. The minimum shear necessary to drive a capsule ``forward'' over one ramp is less than that needed to drive the capsule ``backward'' over a ramp. We show under what conditions it is possible to move the capsule forward, in a ratcheting motion, via an imposed oscillatory flow. These patterned surfaces could be used to control capsule motion precisely, based on flow and surface properties. They coud also be used to efficiently sort capsules based on their size or material properties.   

Electrowetting for Digital Microfluidics

Droplet based chemistry promises to greatly impact biomedical research, providing new avenues for high throughput, low volume assays such as drug screening. Electrowetting on Dielectric (EWOD) is an excellent technique for manipulating microscopic drops of liquid. EWOD uses buried electrodes to locally change the surface energy between a droplet and a substrate. We present microfabricated devices for moving droplets with EWOD. One example of such a device consists of a series of 16 interdigitated electrodes, decreasing in size from 1mm to 20 microns. Each electrode is addressable by an independent, computer controlled, high voltage supply. This work made possible by a gift from Phillip Morris and the NSEC NSF grant PHY-0117795.   

Surface mediated liquid transport on nanotube

The surface mediated liquid transport on nanotubes was studied using a nanotube-based liquid transport system. Microscale liquid droplets were formed and transferred to nanotubes using the liquid transport system integrated with a nano-manipulator. If the spreading parameter $S$ is larger than a threshold value $S_{c}$, the liquid spontaneously flows out of the liquid droplet through a thin precursor film formed along the nanotube surface. The liquid transport on nanotube surfaces was studied \textit{in situ} by measuring the volume flow rate which was obtained from a direct observation of the droplet. The flow rate dependence on the size of nanotubes and surface energy were also investigated. The surface mediated liquid transport phenomenon can be exploited for the development of nanoscale liquid transport system for nanofabrication and nanoscale devices for biological and chemical applications. Reference: Kyungsuk Yum and Min-Feng Yu, Surface-mediated liquid transport through molecularly thin liquid films on nanotubes, Phys. Rev. Lett. 95, 186101 (2005)   

Session P22: Focus Session: Spin Transport in Metals

Disorder, itinerant ferromagnetism and the anomalous Hall effect in two dimensions

This talk will describe research motivated by the lack of consensus on what happens in a band ferromagnet such as iron when the itinerancy of the electrons, which carry spin information, is compromised by disorder. We address this challenging problem by performing \textit{in situ} studies of magnetotransport in a series of films having sheet resistances varying from 50 to 1,000,000 Ohms. In the weakly disordered regime of this two-dimensional system, where the quantum corrections to the conductivity have logarithmic temperature dependence, we find a surprising scaling of the longitudinal and anomalous Hall (transverse) resistances. For higher disorder the scaling breaks down and the anomalous Hall resistance $R_{xy}$ saturates at a constant value near 100 Ohms. These results imply the presence of an \textit{anomalous Hall insulating} state where the longitudinal $L_{xx}$ and transverse $L_{xy}$ conductivities approach zero with a ratio $R_{xy}~=~L_{xy}/L_{xx}^{2}$ that remains constant.   

Theory of Anomalous Hall Effect in Ferromagnets

Mechanism of the anomalous Hall effect (AHE) in ferromagnetic metals has been controversial over many decades. Karplus-Luttinger initiated the discussion by focusing on the intrinsic thermodynamic Hall current produced by the band structure with a spin-orbit interaction. Later, it was argued that instead, scattering by impurity or disorder together with the spin-orbit interaction distorts the electron motion as the skew scattering or the side jump and these extrinsic contributions dominate over the Hall current. Here, we reexamine this issue by fully taking account of both the impurity scattering and the anomalous velocity in terms of the quantum transport theory. We demonstrate that apart from the conventional nonequilibrium transport current, an equilibrium Hall current flows even in the presence of dissipation in metals. This equilibrium Hall current contains the intrinsic one which has a topological non-perturbative nature associated with degeneracy of the band dispersions in the momentum space. We also show that there appears a crossover from the extrinsic regime to the intrinsic as the electron damping rate becomes comparable to or larger than the energy scale of the spin-orbit coupling. This resolves the long standing puzzle on the mechanism and reveals a new small energy scale governing the quantum transport in multi-band systems.   

Quantum correction to the anomalous Hall conductivity of ferromagnetic metallic films

Motivated by new anomalous Hall effect (AHE) data on polycrystalline Fe films we calculate the interaction correction to the AHE for both the skew scattering and the side jump mechanisms. The correction to the longitudinal conductivity is also considered. We use a model of randomly located short range impurity potentials of arbitrary scattering strength, inducing spin-orbit scattering, and an isotropic band ferromagnet. The quantum correction is found to depend sensitively on the strengths of the scattering potential and the spin-orbit interaction. In the limit of weak scattering, known results are recovered.   

Spin Hall effects in diffusive normal metals.

We study transport in normal metals in an external magnetic field. We employ the Keldysh formalism to find transport equations in the presence of the spin-orbit interaction, interaction with magnetic impurities, and nonmagnetic impurity scattering. This system exhibits an interplay between a transverse spin imbalance (spin Hall effect) caused by the spin-orbit interaction, a Hall effect via the Lorentz force, and spin precession due to the Zeeman effect. The spin and charge accumulations are computed numerically in experimentally relevant thin film geometries. The out-of-plane spin Hall potential is suppressed when the Larmor frequency is larger than the spin-flip scattering rate. The in-plane spin Hall potential vanishes at a zero magnetic field and attains its maximum at a finite magnetic field before spin precession starts to dominate. Spin injection via ferromagnetic contacts creates a transverse charge Hall effect that decays in a finite magnetic field due to spin precession.   

Resistance and Scattering Anisotropy of Al Interfaces with Co, Fe, and Co(91)Fe(9).

The properties of normal/ferromagnetic metal interfaces, described by the interface specific resistance, AR* (A = area, R = resistance) and spin scattering anisotropy, $\gamma $, are of both fundamental interest and practical interest for optimizing current-perpendicular-to-plane (CPP) magnetoresistance (MR) and current-induced magnetization-switching (CIMS) in nanopillars. From measurements of the CPP resistances and MRs of sputtered [Al/F]x$N$ (F= Fe, Co, Co(91)Fe(9)) multilayers with $N$-bilayers, and Al/F-based exchange-biased spin-valves, we are able to estimate 2AR$^{\ast }$ and $\gamma $ for each metal pair at 4.2K. In each case, 2AR* is large and $\gamma $ is small, comparable to values of 2AR* $\sim $ 9 f$\Omega $m$^{2}$ and $\gamma \quad \sim _{ }$0.03 for Permalloy (Py)/Al interfaces [1], and each differing by an order of magnitude from the parameters for well-studied Co/Cu and Py/Cu interfaces (2AR$^{\ast }\sim $ 1 f$\Omega $m$^{2}$, $\gamma \quad \sim $ 0.8). The values of AR* with Al are too large to be explained by the resistivities of alloyed Al/F interfaces. The similarity of results for Py, Fe, Co, and Co(91)Fe(9) strongly suggests that spin dependent scattering at Al/F interfaces is determined mainly by the properties of Al. [1] N. Theodoropoulou et al., J. Appl. Phys. (In Press, 2006).   

Current-Perpendicular-to-Plane (CPP) Resistance of Cu/Al Interfaces

. The recent discovery [1] that the current-perpendicular-to-plane (CPP) interface specific resistance (area A times resistance R) of Py/Al (Py = Permalloy = Ni(84)Fe(16)) is almost 10 times larger than that of Py/Cu, led us to examine also the resistances and magnetoresistances of Py/Cu/Al/Cu/Py exchange-biased spin-valves (EBSV) and [Cu/Al]x$N$ multilayers with $N$ repeats. Using two different techniques, we estimate the interface specific resistance of sputtered Al/Cu interfaces as 2AR(Al/Cu) $\sim$ 2 f-ohm-m$^2$ at 4.2K. However, some of the data from these two techniques, as well as from the Cu/Al/Cu EBSV studies, show unusual behaviors that suggest that the Cu and Al atoms might not always stay where they are deposited. We will describe our CPP-resistance results, and plan to supplement them with x-ray and cross-sectional TEM studies. [1] N. Theodoropoulou et al., J. Appl. Phys. (In Press, 2006).   

Spin transport and spin-flip scattering in magnetic multilayer structures

The existence of spin-flip scattering at the interface between ferromagnetic (F) and nonmagnetic (N) layers of magnetoresistive F/N/F structures can significantly reduce the size of the magnetoresistance, limiting the sensitivity and increasing the power consumption of F/N/F devices such as GMR magnetic field sensors, magnetic read heads, and MRAM's~[1]. Detecting and measuring the degree of spin flip scattering in F/N/F structures can allow further optimization in such devices as well as increase the understanding of interfacial spin transport. Our nonlocal spin injection and detection experiments on mesoscopic Co-Al$_2$O$_3$-Cu-Al$_2$O$_3 $-Co spin valves provide evidence for the existence of interfacial spin-flip scattering in magnetoresistive devices~[2]. By extending the conventional picture of spin-dependent interfacial resistances (R$_{\uparrow}$, R$_{\downarrow}$) to include two additional spin-flip scattering channels (R$_{\uparrow\downarrow}$,R$_{\downarrow\uparrow}$)~[3] we have shown that the nonlocal resistance contains information about both the degree of spin polarization and the degree of spin-flip scattering at the F/N interface. The magnitudes of R$_{\uparrow\downarrow}$ and R$_{\downarrow\uparrow}$ depend on the relative orientation of the detector magnetization and the nonequilibrium magnetization in the normal metal. We have observed that the difference in spin-flip scattering between up and down channels vanishes at low temperatures, but for T$>$100K it increases nonlinearly with temperature. Further evidence for the presence of interfacial spin-flip scattering can be obtained from noise measurements, which are extremely sensitive to the microscopic transport details. \noindent [1] \textit{Spin Dependent Transport in Magnetic Nanostructures}, edited by S. Maekawa and T. Shinjo (Taylor \& Francis, New York, 2002). \noindent [2] S. Garzon, I. \v{Z}uti\'{c}, and R. A. Webb, Phys. Rev. Lett. \textbf{94}, 176601 (2005). \noindent [3] E. I. Rashba, Eur. Phys. J. B \textbf{29}, 513 (2002).   

Measurement of spin-diffusion length in sputtered Ni films

The spin-diffusion length of the electron represents a fundamental transport parameter and plays a central role in the development of spintronic devices. While several measurements of the spin-diffusion length in normal metals and in ferromagnetic alloys exist, measurements in elemental ferromagnets (Ni, Fe, Co) have been scarce. We present a novel sample geometry using giant magnetoresistance (GMR) for the measurement of the spin-diffusion length in elemental ferromagnets with weak scattering asymmetry. We report the first measurement of the spin-diffusion length of Ni using an exchanged-biased Permalloy-based spin-valve, Py/Ni/Cu/Py, where the Ni layer acts as a “GMR spoiler” layer when its thickness becomes greater than the spin-diffusion length in Ni.   

Magneto-transport in Fe$_{3}$O$_{4}$/Nb:SrTiO$_{3}$ schottky junction diode

Among the half metallic ferromagnets, Fe$_{3}$O$_{4}$ is of particular interest because of its half metallicity, high curie temperature and a charge ordering transition at 120K (popularly known as Verwey transition (T$_{V}))$. These materials are also expected to show 100{\%} spin polarization. In view of these fascinating properties, we studied temperature dependent transport, magnetic, structural and interface characteristics of epitaxial schottky junctions between Fe$_{3}$O$_{4}$ and Nb:SrTiO$_{3}$ (with different Nb concentrations). Epitaxial thin films of Fe$_{3}$O$_{4}$ were grown on Nb:SrTiO$_{3}$ substrates by PLD technique. The films show epitaxial growth along (100)-axis direction. We also performed HR-TEM and EELS study to ensure a better quality interface. In the temperature range above T$_{V}$, 300K-130K, the I-V characteristic shifts towards higher forward bias voltage upon lowering temperature. On further decreasing temperature (below T$_{V})$, the trend is reversed. Junction parameters such as the Schottky barrier height (\textit{$\phi $}$_{B})$ and ideality factor (\textit{$\eta $}) are extracted using thermionic emission theory at all temperatures. These parameters show interesting and systematic trends above and below T$_{V}$. From the magnetic field dependence of non linear I-V characteristics data, a spin polarization of $\sim $80{\%} is estimated for the magnetite electrode at T$_{V}$.   

Spin filtering of hot electrons in ferromagnetic layers

We present a spin dependent transport experiment where spin polarized electrons, injected from vacuum, are spin-filtered when entering into a thin ferromagnetic layer. The role of the interface between this spin-filter layer and the ``current collector'' is analysed. In a first geometry, the ``current collector'' is a semiconductor and the transmitted current is measured through a Fe/Oxide/GaAs diode. In that case, the measured electrons are selected at the interface, at an energy higher than the Schottky barrier. In a second geometry, the spin filter is a self standing thin layer Au/Co/Au and the ``current collector'' is a faraday cup. In that case, only electrons that overcome the vacuum energy level of the gold are measured. In both cases, the spin dependent part of the collected current is measured when the ferromagnetic layer is reversed. We have found an energy domain where both the collected current and its spin dependence increase over orders of magnitude. The role of the interface in the spin-dependent transport is discussed and a spin dependent transport model is presented.   

Mesoscopic Conductance Fluctuations in Cobalt Nanoparticles

We~present measurements of mesoscopic conductance fluctuations in Cobalt particles of diameter 200nm. Samples are made by e-beam lithography and shadow metal deposition. Co particles are not single domain; domain walls are nucleated at the contacts between Co and Cu-reservoirs. We obtain the dependence of~peaks in differential resistance with the applied voltage and the magnetic field during the magnetization reversal process at 0.03K temperature. The conductance fluctuations with the magnetic field are caused by a mechanism different from the usual Aharonov-ohm effect. In particular, domain walls are found to generate significant mesoscopic fluctuations. We obtain that electron transfer across the domain wall is associated with a phase change of about 5$\pi $. We explain how this phase-shift arises from a not perfectly parallel spin-transport across domain walls. The dephasing time is very short, $\tau _\phi \sim ps$. Fast dephasing is correlated with the strong magnetocrystalline anisotropy in Co. This work was performed in part at the Georgia-Tech electron microscopy facility. We thank P. Brouwer for valuable discussions. This research is supported by the David and Lucile Packard Foundation grant 2000-13874 and Nanoscience/Nanoengineering Research Program at Georgia-Tech.   

Spin transport through atomic scale chromium Coulomb islands

Electrical current through metallic islands coupled via tunnel barriers to external leads is governed by the Coulomb repulsion and can be brought down to single electron transport. The spin-degeneracy of the electrons can be lifted by choosing both the leads and the islands to be magnetic. The combination of spin-splitting and Coulomb blockade creates a device geometry capable of resonant tunneling of a single spin-direction only. Maximum effect can be obtained by minimization of the size of the Coulomb islands in order to suppress spin-relaxation. We report on our efforts to make a spin-resonant tunneling device using atomic size clusters of chromium atoms, submerged in an alumina-barrier in a conventional magnetic tunnel junction set-up. The $300x300\;\mu\rm{m}^2$ size magnetic tunnel junction consists of a cobalt bottom electrode, an aluminum-oxide tunnel barrier, a delta-doping layer of chromium in the range of $1-6\;\AA$, an alumina tunnel barrier, and a permalloy top-electrode. Transport measurements reveal Coulomb blockade behavior.   

Orbital nature of spin magnetic moment

In view of the application of spintronics, it is important to understand the physical observable of the spin, i.e. its magnetic moment. We show by constructing a wave-packet in the upper bands of the Dirac equation that the spin magnetic moment (Bohr magneton) is a direct result of the self-rotation in the wave-packet. In this sense, an non-relativistic electron is really a rotating charged object, confirming the original speculation on the physical nature of the electron spin. In a Bloch band of a crystal, a wave-packet can acquire additional self-rotating orbital angular momentum, giving rise to a change of the spin magnetic moment and causing a modification of the g-factor from 2.   

Session P23: Quantum Spin Chains II

Thermal transport of dimerized and frustrated spin-1/2 chains

We present a numerical study of thermal transport in dimerized and frustrated spin-1/2 chains at finite temperatures. Since these models are nonintegrable, the thermal Drude weight scales to zero in the thermodynamic limit. The conductivity at finite frequencies, however, is non-zero and we discuss the scaling with system size as well as the extrapolation to the zero-frequency limit. Results for three cases are presented. First, the dimerized chain is studied in the limit of weakly coupled dimers. In this case, interactions of the elementary triplet excitations are weak, which should allow for an analytical description based on a bond-boson operator representation. Second, we compare the thermal conductivity of the frustrated chain in the massless and the massive regime of this model. Finally, we extract the zero-frequency thermal conductivity of the isotropic two-leg spin ladder and discuss implications for the interpretation of recent experiments for La$_5$Ca$_9$Cu$_{24}$O$_{41}$.   

A systematic method for constructing a spin-singlet basis for quantum antiferromagnets

We present a new method for constructing a complete orthonormal basis for the singlet states of quantum spin-1/2 lattice systems. Our approach can be used for any dimension and an arbitrary lattice symmetry. In this talk the main group- and graph-theoretical steps are explained. The general theory is then applied to the one-dimensional quantum antiferromagnet. Exploiting the symmetries of closed rings, we can drastically reduce the number of basis states for the different eigenstates of the Hamiltonian (in the singlet sector). The method allows to calculate in an efficient way expectation values of any operator.   

Crossed Spin-1/2 Heisenberg Chains as a Quantum Impurity Problem

Using equivalencies between different models we reduce the model of two spin-1/2 Heisenberg chains crossed at one point to the model of free fermions. The spin-spin correlation function is calculated by summing the perturbation series in the interchain interaction. The result reveals a power law decay with a nonuniversal exponent.   

Comparison of Magnetic Field-Modified Electronic Excitations in Ni(II) Compounds

NTDN (Ni[tn]$_2$[NO$_2$]$_2$) can be considered a paramagnetic analog material to the Haldane compounds NENP and NENB (Ni[en]$_2 $NO$_2$ClO$_4$ and Ni[en]$_2$NO$_2$BF$_4$; where en = C$_2$N$_2 $H$_{8}$ and tn = C$_2$N$_3$H$_10$). Except for the different bonding of one NO$_2$ group and the absence or presence of spin chains, NTDN and the Haldane compounds have nearly identical electronic coordination around the Ni$^{2+}$ ions. Here, we report and compare the magnetic field ($H$)-dependent polarized optical transmittance of the three materials in the range 9,000 to 22,000 cm$^{-1}$. The $H$ dependence is manifest in the varying intensities of certain electronic absorptions with applied field. Although all three materials possess similar $H$- sensitive excitations, the details of the $H$ dependence differ with the magnetic ground states. In NTDN, the intensity changes commence at $H$ = 0 and saturate at $\approx$ 10 T, whereas in the Haldane compounds, the onset of changes is at the gap- closing critical field, $H_C$, above which the intensity is linearly modified with field. The mechanism of the $H$- dependence is yet to be clarified and probably depends on the nature of the electronic excitation. Intensity variations with applied field are observed in both Ni$^{2+}$-to-NO$_2^-$ charge transfer transitions and Ni$^{2+}$ $d-d$ spin forbidden excitations.   

Magnetic Chains Assembled with Atomic Precision

We report the first study of small magnetic chains constructed and studied \textit{in-situ} with atomic precision. By positioning Mn atoms one at a time with an STM, we are able to follow the low energy excitation spectrum of the chain as it changes dramatically with length. The low energy spectra of chains built from an odd number of atoms display a spin-flip excitation similar to that of a single Mn atom. In even-length chains the spin flip excitation is absent and is replaced by an excitation at higher energies that splits into three distinct modes in a magnetic field. We interpret these results as direct evidence of the antiferromagnetic coupling of the individual atomic spins. Quantitative comparison of our results with the Heisenberg model allows us to directly obtain the coupling strength J$\sim $6meV between atomic spins and suggests that the total spin of the individual Mn atoms on the surface is 5/2.   

Resonant Quantum Tunneling of spin chains in 3D-magnetically ordered CoTAC

CoTAC is a well know molecular spin chain system which orders in an antiferromagnetic canted state at $4.15\ K$. We show that, below $300\ mK$, the dynamics of the magnetization in the $c$-direction are governed by resonant quantum tunneling of the magnetization. This conclusion is based on a number of experimental observations: the temperature independence of the relaxation of the magnetization, speeding up of the relaxation at a well defined magnetic field value ($1025\ Oe$) and the increase of the magnetization each time this field is crossed during a succession of minor loops. The key to understanding this behavior could be the absence of resonance in zero field. We propose a mechanism to describe this behavior using a simple model of domain wall nucleation which explains many of the unusual experimental observations.   

Extended Valence-Bond Solid Picture for Quasi-One-Dimensional Quantum Magnets

Ground-state gapped phases of quasi-one-dimensional quantum magnets are given an comprehensive interpretation based on an extended valence-bond solid (VBS) picture. We introduce composite spins with the enlarged spin magnitude in the ground state and regard the system as an effective single spin chain that consists of composite spins. The relevance of the composite spins in the ground state is revealed by the effects of dimerization in the spin-spin couplings. In order to characterize the gapped phases, we inspect the configuration of valence bonds by calculating the appropriate order parameters by the quantum Monte Carlo method with the continuous-time loop algorithm. The so-called short-range resonating-valence bond solid state is identified to be an extended VBS state. Dimensional crossover to the two-dimensional systems is also discussed.   

Luttinger sum rules in the 2-channel spin 1 Kondo chain

We present numerical results, using the DMRG method, for the zero-temperature phase diagram of the one-dimensional 2-channel Kondo lattice model, coupled to $S=1$ localized spins. Unlike the previously studied 1-channel $S=1/2$ case, the $S=1$ localized spins in the 2-channel model permit a gapped Haldane state, which stabilizes the weak-coupling (``small'' Fermi surface) paramagnetic phase. We focus on the paramagnetic region approaching half-filling, where the interesting possibility has been recently raised of a non-trivial crossover region where the Fermi wavenumber varies continuously.   

Coupling of Spin and Charge Degrees of Freedom in a Hydrodynamic Two-Fluid Approach

We use a hydrodynamic approach to study correlated quantum one- dimensional systems. One can derive an effective hydrodynamics in the limit of smooth densities and slow velocities. In the limit when gradients of density and velocity can be neglected, one obtains an integrable system. We discuss the origin of this integrability and, as an application, we calculate some non trivial correlation functions. The Hydrodynamic description is a useful tool to study the dynamics of a system. We introduce a two-fluid hydrodynamic description of one-dimensional spin 1/2 fermions with contact interactions. It is known that linearized hydrodynamics (bosonization) exhibits spin-charge separation. In the full {\it non-linear} theory, the spin and charge degrees of freedom are coupled to each other, therefore spin waves carry charge as well. We discuss the dynamics of such system.   

Coupled one-dimensional magnetic systems: the truncated conformal spectrum approach

We develop a combined analytical/numerical technique in order to understand coupled, possibly strongly, one dimensional magnetic systems. The approach trades on exact knowledge of the underlying one dimensional subsystem and the concomitant ability to compute exactly matrix elements of operators coupling the subsystems together. With these matrix elements in hand, the fully coupled system is first truncated and then diagonalized numerically. In this way we obtain both the spectrum and correlation functions of the system. The truncation can be improved upon through a renormalization group procedure. As a test case we consider coupled quantum Ising chains.   

Random antiferromagnetic spin chains with enlarged symmetry

We present the asympotically exact solution of some random antiferromagnetic spin chains with enlarged symmetry groups. Using a generalization of the strong disorder real space renormalization group method, we considered both the isotropic SU(N) and the anisotropic SU(4) chains with totally antisymmetric irreducible representations. In the first case, the system is governed by a universal infinite-randomness fixed point (IRFP), with activated dynamical scaling between energy $\left(\Omega\right)$ and length $\left(L\right)$ scales $\Omega\sim\exp\left(-L^{\psi}\right)$ (with the tunneling exponent $\psi=1/N$), and average correlation function decaying as a power law with exponent $\eta=4/N$. All thermodynamic quantities are universal with exponents depending only on $N$. In the second case, relevant for systems with SU(2)$_{spin}\otimes$SU(2)$_{orbital}$ symmetry, we determined the full phase diagram as a function of the mean anistropy and its variance. All stable fixed points are of the IRFP variety with activated dynamical scaling. The tunneling exponents span a wide range of values and can even be larger than the SU(2) value, $\psi=1/2$.   

Thermodynamics and magnetization process of the frustrated ferromagnetic spin-1/2 chain

We report a study of the thermodynamics and magnetization curve of a $J_1$--$J_2$ spin-1/2 chain with a ferromagnetic nearest neighbor coupling $J_1$ and an antiferromagnetic next-nearest neighbor interaction $J_2$. This model has recently been suggested to describe the properties of different compounds such as LiCuVO$_4$ and Rb$_2$Cu$_2$Mo$_3$O$_{12}$. We present results for both the specific heat and the magnetic susceptibility for the whole parameter range of $J_2/4<|J_1|<0$ obtained by exact diagonalization of up to $N=24$ sites. The specific heat exhibits a two-peak structure for $J_1<0$, originating from, first, the proximity to a ferromagnetic ground state and, second, antiferromagnetic fluctuations at higher energies. Furthermore, the magnetization process at zero temperature is analyzed by means of the density matrix renormalization group technique. Particular emphasis is given to the presence (or rather absence) of magnetization plateaus and a comparison with other theoretical results for the ground-state phase diagram.   

The Magnetic phase diagram of the spin-chain system Ca$_{2+x}$Y$_{2-x}$Cu$_5$O$_{10-\delta}$ : Oxygen hole-doping

Recently, K. Kudo et al.\footnote{K. Kudo, S. Kurogi, and Y.Koike, Physical Review B \textbf{71}, 104413 (2005)} studied the magnetic ground state in the edge-sharing CuO$_2$ chains in the spin- chain system Ca$_{2+x}$Y$_{2-x}$Cu$_5$O$_{10-\delta}$. In that study, the antiferromagnetic transition temperature decreases with increasing \emph{x} and disappears around \emph{x}=1.4 followed by the appearance of a spin-glass phase at \emph{x}=1.5. We propose that the oxygen content should be included in the hole doping effect by $p=1/5(x-2\delta)$ in the spin-chain system. We present x-ray diffraction, magnetic susceptibility, specific heat and iodometric titration measurements\footnote{This work is supported by the Robert A. Welch Foundation grant No.F-1191 and the National Science Foundation grant No. DMR-0210383} which indicate that an oxygen deficiency shifts the magnetic features toward higher \emph{x}. For example, for $x=1$ samples, the single crystals of Ref.1 are equivalent to our oxygen deficient polycrystalline sample with $\delta \approx 0.5$. Such a composition has an only slightly suppressed N\'eel temperature, while for nearly fully oxygenated $x=1$ samples, the antiferromagnetic transition is completely suppressed.\footnote{M. D. Chabot, and J. T. Markert, Physical Review Letters \textbf{86}, 163 (2001)}   

Low temperature magnetization and the excitation spectrum of antiferromagnetic quantum Heisenberg rings

We have performed quantum Monte Carlo calculations to obtain the low temperature magnetization and differential susceptibility for finite, antiferromagnetic Heisenberg rings of intrinsic spins $s = 1/2, 1,3/2,2,5/2,3,7/2$. From these data we have determined the level-crossing fields as well as the dependence of the minimal excitation energies on the total spin quantum number $S$. For large intrinsic spins ($s \geq 3/2$) we find that the data exhibit scaling behavior, approaching the classical limit proportional to $s^{-1.05}$. Since this limit is approached so slowly, even $s=7/2$ spins are distinctly non-classical. We have also found for large $s$ that as the number of spins $N$ increases, the energy gap between the ground state and the first excited state approaches zero proportional to $1/N^\alpha$, where $\alpha \approx 0.76$ for $s=3/2$ and $\alpha \approx 0.84$ for $s=5/2$. Finally, we demonstrate the usefulness of our results by examining the Fe$_{12}$ molecular ring, leading to a new, more accurate estimate of the exchange constant for this system than has been obtained heretofore.   

Magnetic Properties of a Coordination Polymer [Mn$_{3}$[(OH)$_{2}$Na$_{2}$(3-cnba)$_{6}]$_{n}$

Magnetic properties of [Mn$_{3}$(OH)$_{2}$Na$_{2}$(3-cnba)$_{6}$]$_{n}$ (3-Hcnba = 3-cyanobenzoic acid), a newly discovered three-dimensional coordination polymer, were investigated using magnetic susceptibility $M(T)$/$H$ and isothermal magnetization $M(H)$. The crystal structure of [Mn$_{3}$(OH)$_{2}$Na$_{2}$(3-cnba)$_{6}$]$_{n}$ is triclinic with a space group $P$-1. The lattice parameters are $a$ = 6.663 {\AA}, $b$ = 12.971 {\AA}, $c$ = 14.161 {\AA}, \textit{$\alpha $} = 70.13\r{ }, \textit{$\beta $} = 88.43\r{ }, and \textit{$\gamma $} = 76.47\r{ }. The results of $M(T)$/$H$ on powder samples show that the effective moment \textit{$\mu $}$_{eff}$ of Mn$^{2+}$ is 5.88 \textit{$\mu $}$_{B}$ at temperatures above 100 K, close to the expected value for a free Mn$^{2+}$ ion. Below 3 K, [Mn$_{3}$(OH)$_{2}$Na$_{2}$(3-cnba)$_{6}$]$_{n}$ orders antiferromagneticly. A sudden slope change in $M(H)$ measured at is observed at a very small critical field of H$_{c} \quad \approx $ 20 G, suggesting a metamagnetic transition. Above 20 kG, $M(H)$ starts to saturate, reaching a value equivalent to 1.7 \textit{$\mu $}$_{B}$ per Mn$^{2+}$ ion. The magnetic behavior of the complex is interpreted in terms of an effective ferrimagnetic Mn(II) chains in which spin moments are linked by interactions in an AF-F-AF (F = ferromagnetic and AF = antiferromagnetic) sequence in the triangular magnetic repeating unit.   

Session P26: Focus Session: Protein Dynamics in Folding and Function

NMR Studies of Enzyme Structure and Mechanism

At least three NMR methodologies pioneered by Al Redfield, have greatly benefited enzymology: (1) the suppression of strong water signals without pre-saturation; (2) sequence specific NH/ND exchange; and (3) dynamic studies of mobile loops of proteins. Water suppression has enabled us to identify unusually short, strong H-bonds at the active sites of five enzymes (three isomerases and two esterases), and to measure their lengths from both the chemical shifts and D/H fractionation factors of the deshielded protons involved (J. Mol. Struct. 615, 163 (2002)). Backbone NH exchange studies were used to detect regions of an NTP pyrophosphohydrolase in which NH groups became selectively protected against exchange on Mg(2+) binding, and further protected on product (NMP) binding, thus locating binding sites as well as conformationally linked remote sites (Biochemistry 42, 10140 (2003)). Dynamic studies were used to elucidate the frequency of motion of a flexible loop of GDP-mannose hydrolase (66,000/sec) containing the catalytic base His-124, from exchange broadening of the side chain NH signals of His-124 in the free enzyme. The binding of Mg(2+) and GDP-mannose lock His-124 in position to deprotonate the entering water and complete the reaction.   

Interference between relaxation and parameters for protein structure determination

The effect of cross-correlated relaxation on scalar and dipolar coupling measurements is analyzed. We compare one-bond proton-carbon scalar and dipolar couplings of protein methine and methylene sites obtained by monitoring proton and carbon magnetization. Apparent J-coupling constants of the same pair of nuclei vary depending on the type of magnetization involved. The discrepancies are of different magnitude for methine and methylene moieties. Dynamic frequency shifts are partially responsible for the observed differences. More importantly, the largest observed variations can be explained by processes of magnetization transfer originated by cross-correlated relaxation. These later effects are not cancelled when obtaining residual dipolar couplings.   

Multiple Quantum Relaxation Probes of Protein Dynamics on Multiple Timescales

Several effects may lead to significant differences between the relaxation rates of zero-quantum coherences (ZQC) and double-quantum coherences (DQC) (collectively known as multiple-quantum coherences) generated between a pair of spin 1/2 nuclei in solution. These include the interference between the anisotropic chemical shifts of the two nuclei participating in formation of the ZQC or DQC, the individual dipolar interactions of each of the two nuclei with the same proton, and the slow modulation of the isotropic chemical shifts of the two nuclei due to conformational exchange. Motional events that occur on a timescale much faster than the rotational correlation time (picosecond-nanosecond) influence the first two effects, while the third results from processes that occur on a far slower timescale (microsecond-millisecond). An analysis of the differential relaxation of ZQC and DQC is thus informative about dynamics on the fast as well as the slow timescales. We present here a set of NMR experiments that measure the differential relaxation of ZQC and DQC involving several backbone and sidechain nuclei in proteins. These measurements provide significant insight into the complex dynamic modes that exist in the protein backbone and sidechains. A detailed understanding of these dynamic modes may provide clues into the role of dynamics in modulating protein function.   

A Plastic Explosive-Degrading Enzyme

The enzyme nitroreductase catalyzes reduction of high explosives such as TNT and RDX. Although a well-resolved $^{1}$H$^{15}$N-HSQC is obtained at 37 $^{\circ}$C, the HSQC at 4 $^{\circ}$C is concentrated between 7.5 and 8.5 ppm and is comprised of sharp overlapped peaks. Thus, it appears that the protein denatures upon cooling. However, the non-covalently-bound FMN cofactor is not released at the lower temperature. Similarly, ultra-violet CD spectroscopy shows that the protein retains essentially full secondary structural content at 4 $^{\circ}$C. Thus, it appears that nitroreductase exists as an ensemble of rapidly interconverting loose structures at lower temperature, only adopting a single long-lived structure above 20 $^{\circ}$C. Both saturation transfer from water and solvent proton exchange measurements, demonstrate that resonances of the poorly-dispersed spectrum represent protons closer to water, and in faster exchange with it. Thus we propose that the single well-defined structure is favored entropically, by release of water molecules that solvate the protein at 4 $^{\circ}$C. We propose that the loosely structured state plays a role in accommodating binding of diverse substrates.   

Protein-Protein Interactions during Bacterial Chemotaxis using Methyl TROSY Nuclear Magnetic Resonance.

During bacterial chemotaxis, the histidine autokinase CheA interacts with the chemotaxis receptors with the help of the coupling protein CheW. The CheA-CheW interaction is typical of many macromolecular complexes where protein-protein interactions play an important role. In this case a relatively small protein, CheW (18 kDalton), becomes part of a much larger complex. Here we describe a new method to map the residues at a protein-protein interface for macromolecular complexes of molecular weight greater than 100 kDalton. The method exploits the C13 methyl TROSY methodology developed in Lewis Kay's laboratory. The essence of the Kay approach is that a portion of the intensity of HMQC spectra of individual -(13)CH3 resonances in an otherwise deuterated macromolecule have much reduced dipole-dipole relaxation and remain sharp and relatively easy to detect , even in macromolecules of molecular mass 100 kD or greater. The reduction in dipolar interactions is lost if a given methyl group comes in close contact with other protons such as those supplied by the interface of a protonated interaction partner. Comparing the -(13)CH3 resonances of a protein of interest in the presence of a protonated versus deuterated interaction partner allows the methyls at the interface can be identified. The application of the approach for establishing points of contact between CheA and CheW will be discussed.   

Structural Basis for Specific Membrane Targeting by the HIV-1 Gag Protein.

In HIV-1 infected cells, newly synthesized retroviral Gag polyproteins are directed to specific cellular membranes where they assemble and bud to form immature virions. Membrane binding is mediated by Gag's matrix (MA) domain, a 132-residue polypeptide containing an N-terminal myristyl group that can adopt sequestered and exposed conformations. Membane specificity was recently shown to be regulated by phosphatidylinositol-(4,5)-bisphosphate (PI(4,5)P2), a cellular factor abundant in the inner leaflet of the plasma membrane (PM). We now show that phosphoinositides, including soluble analogs of PI(4,5)P2 with truncated lipids, bind HIV-1 MA and trigger myristate exposure. The phosphoinositol moiety and one of the fatty acid tails binds to a cleft on the surface of the protein. The other fatty acid chain of PI(4,5)P2 and the exposed myristyl group of MA bracket a conserved basic surface patch implicated in membrane binding. Our findings indicate that PI(4,5)P2 acts as both a trigger of the myristyl switch and as a membrane anchor, and suggest a structure-based mechanism for the specific targeting HIV-1 Gag to PI(4,5)P2-enriched membranes.   

Vinculin Tail Dimerization and Paxillin Binding

Vinculin is a highly conserved cytoskeletal protein that is essential for regulation of cell morphology and migration, and is a critical component of both cell-cell and cell-matrix complexes. The tail domain of vinculin (Vt) was crystallized as a homodimer and is believed to bind F-actin as a dimer. We have characterized Vt dimerization by Nuclear Magnetic Resonance (NMR) Spectroscopy and identified the dimer interface in solution by chemical shift perturbation. The Vt dimer interface in solution is similar to the crystallographic dimer interface. Interestingly, the Vt dimer interface determined by NMR partially overlaps the paxillin binding region previously defined coarsely by deletion mutagenesis and gel-blot assays. To further characterize the paxillin binding site in Vt and probe relationship between paxillin binding and dimerization, we conducted chemical shift perturbations experiments using a paxillin derived peptide, LD2. Our NMR experiments have confirmed that the paxillin binding site and the Vt dimerization site partially overlap, and we have further characterized both of these two binding interfaces. Information derived from these studies was used to identify mutations in Vt that selectively perturb paxillin binding and Vt self-association. These mutants are currently being characterized for their utility in structural and biological analyses to elucidate the role of paxillin binding and Vt dimerization in vinculin function.   

Protein dynamics and allostery

Cooperativity and allostery are phenomena of universal importance in biological systems. According to the classical ``mechanical'' view allosteric interactions are mediated by a series of discrete changes in bonding interactions that alter the protein conformation. Nevertheless, proteins may have adopted additional mechanisms for energetically linking distant sites, thereby allowing a signal to be propagated over long distances. We have identified a cooperative biological system wherein allosteric interactions appear to be mediated exclusively by transmitted changes in protein motions. Changes in the structure, fast and slow protein motions and the redistribution of the native-state ensemble along the cooperative reaction coordinate have been characterized.   

Protein Dynamics in an RNA Binding Protein

Using $^{15}$N NMR relaxation measurements, analyzed with the Lipari-Szabo formalism, we have found that the human U1A RNA binding protein has ps-ns motions in those loops that make contact with RNA. Specific mutations can alter the extent and pattern of motions, and those proteins inevitably lose RNA binding affinity. Proteins with enhanced mobility of loops and termini presumably lose affinity due to increased conformational sampling by those parts of the protein that interact directly with RNA. There is an entropic penalty associated with locking down those elements upon RNA binding, in addition to a loss of binding efficiency caused by the increased number of conformations adopted by the protein. However, in addition to local conformational heterogeneity, analysis of molecular dynamics trajectories by Reorientational Eigenmode Dynamics reveals that loops of the wild type protein undergo correlated motions that link distal sites across the binding surface. Mutations that disrupt correlated motions result in weaker RNA binding, implying that there is a network of interactions across the surface of the protein. (KBH was a Postdoctoral Fellow with Al Redfield from 1985-1990). This work was supported by the NIH (to KBH) and NSF (SAS).   

Axial Rotation of Lipids in Membranes

This study was motivated by Mary Roberts and Al Redfield, who proposed that the observed 10 ns decay time in their phosphorous NMR measurements of DPPC in bilayers originated from axial rotation of the lipid. Analyses of correlation functions and average first passage times from 50 ns molecular dynamics of DPPC bilayers strongly support their interpretation. The rotational anisotropy of a lipid in a bilayer is close to 1.0, in contrast to the 2.5 expected for a hydrodynamic cylinder with lipid dimensions. This implies that axial rotation is dominated by headgroup, not tail, interactions.   

High Resolution Field Cycling NMR in Biopolymers in Solution: Current and Potential Applications.

I have been exploring the feasibility and utility of performing high resolution relaxation and cross-relaxation (NOE) in an unmodified shared commercial NMR instrument (PNAS 101:17066-17071) in collaboration with Mary Roberts of Boston College and Elan Eisenmesser of Brandeis. We can move a sample from 11.7 T to low (fringe) field 40- 80 cm above the commercial probe, and back, in 0.3 to 0.5 sec. I am making the system move the sample more gently, to avoid protein denaturation, using a stepping-motor timing-belt linear-motor. I willreview our initial papers on a DNA octamer, and phospholipidvesicles using phosphorus NMR. We emphasize phosphorus,in part, because relaxation studies over a wide range can make behavior of this biologically important species more readily interpretable than at high field alone. Then I will discuss a range of biochemical experiments, not yet done, using the full capability of our commercial instrument for multidimensional preparation and detection before and after field-cycling, and utilizing other nuclear species (H, C, and N). These involve especially use of electron-paramagnetic species and/or studies of small molecules in fast binding exchange with larger ones, and/or increased hetero-NOE effects at low field, to get new information on dynamics and structural preferences. Predictably we have found that, once we start to work on a specific biopolymer problem, we think of more things to do than we expected at first.   

Session P27: Electronic Structure II

Analytic structure of Bloch functions for linear molecular chains

In this talk I will discuss Hamiltonians of the form $H=-{\bf \nabla}^2+v(x,y,z)$, with $v (x,y,z)$ periodic along the $z$ direction, $v(x,y,z+b)=v(x,y,z)$. The wavefunctions of $H$ are the well known Bloch functions $\psi_{n,\lambda}(x,y,z)$, with the fundamental property $\psi_{n,\lambda}(x,y,z+b)=\lambda \psi_{n,\lambda}(x,y,z)$ and $\partial_z\psi_{n, \lambda}(x,y,z+b)=\lambda \partial_z\psi_{n,\lambda}(x,y,z)$. I will give the generic analytic structure (i.e. the Riemann surface) of $\psi_{n,\lambda}(x,y,z)$ and their corresponding energy, $E_n (\lambda)$, as functions of $\lambda$. I will also discuss several applications, like a compact expression of the Green's function or the asymptotic behavior of the density matrix and other correlation functions for insulating molecular chains.   

First-principles calculation of Hubbard parameter: Constrained local density functional approach with Maximally localized Wannier function

We present a new ab initio method for calculating effective onsite Coulomb interactions of itinerant and strongly correlated electron systems. The method is based on constrained local density functional theory formulated in terms of maximally localized Wannier functions. This scheme can be implemented with any basis, and thus allows us to perform the constrained calculation with plane-wave-based electronic-structure codes. We apply the developed method to the evaluation of the onsite interaction of 3d transition-metal series. The results are discussed using a heuristic formula for screened Coulomb interactions. This work was supported by NAREGI Nanoscience Project, Ministry of Education, Culture, Sports, Science and Technology, Japan.   

Electronic structure of Cu$_{2-x}$S and related compounds

Chalcosite Cu$_2$S and digenite Cu$_{1.8}$S are possibly interesting semiconductors for photovoltaic applications. Their electronic structure is poorly understood because their crystal structure is complex. If consists of a close-packed lattice of S with mobile Cu occupying various types of interstitial sites with a statistical distribution depending on temperature. As a starting point for understanding these materials, we investigated the simpler antifluorite structure. Both local density approximation (LDA) and self-consistent quasiparticle GW calculations with the full-potential linearized muffin-tin orbital method give a semimetallic band structure with the Fermi level pinned at a degenerate Cu-d band state at $\Gamma$. A random distortion of the Cu atoms from the perfect antifluorite positions inside each S cage is found to break the degeneracy of the $d$ state at $\Gamma$ and thus opens up a small gap of about 0.1 eV in LDA. The experimental evidence for a semiconducting gap of about 1 eV is critically examined. To gain further insight into the Cu d and s-band shifts beyond LDA, we considered other Cu compounds such as Cu$_2$O and CuBr. We compare their LDA and GW band structures and determined the effective masses and Kohn-Luttinger Hamiltonian parameters for CuBr.   

The Applicability of Different Quantum Mechanical Methods to Transition Metal Oxides

Numerous quantum mechanical methods and basis sets have been applied extensively to organic molecules. However, the performance of these is not well understood for transition metal oxides. We employed different methods along with several basis sets for optimizing the geometries in the gas phase and calculating the IR spectra as well as thermodynamic properties including Gibbs free energy and enthalpy of linear, trigonal and tetrahedral metal oxides. The MCSCF and DFT methods generally give the most accurate results for organic and inorganic molecules. Surprisingly, our studies showed that the results obtained for iron(III) oxides at the GVB and MP2 levels gave more accurate results than the MCSCF and hybrid methods. Similarly, for aluminum and chromium oxides, the calculations with MP2 and PBE yielded thermodynamic properties, which are closer to experimental values.   

Zeroth-moment dielectric sum rule applied to electron damping

The first and inverse-first frequency moments of the dielectric function, epsilon(q,omega), are given by the f-sum rule and Kramers-Kronig transformation of the static dielectric function. Model expressions for these quantities are plentiful. The square of the zeroth moment must be less than the product of the above two, by Cauchy-Schwartz. (It equals that product in single-plasmon-pole models). In this work, we present simple ways to estimate the zeroth moment as a function of q. This facilitates an improved model for epsilon(q,omega) that requires minimal computation and exhibits realistic behavior without use of a pole model. We apply this to calculating the electron self-energy, particularly lifetime damping effects in insulators near the band gap.   

Relativistic real-space multiple scattering calculations of EELS

We present an extension of the real space multiple scattering code FEFF8 for {\it ab initio}, relativistic calculations of electron energy loss spectra (EELS), which is applicable both to periodic and non-periodic systems. The approach explains the observed relativistic shifts in the magic angle. \footnote{B. Jouffrey, P. Schattschneider and C. Hebert, Ultramicroscopy {\bf102}, 61 (2004).} In addition, the method can account for experimental parameters such as collection and convergence angles of the microscope and sample orientation. We also discuss relativistic effects on inelastic electron scattering including the density correction to the stopping power. Our results are compared with other approaches and with experiment.   

Prediction of Born-Oppenheimer Interatomic Forces Using Orbital-Free Density Functional Theory with Approximate Kinetic Energy Functionals

Rapid calculation of Born-Oppenheimer forces is essential for driving the so called quantum region of a multi-scale molecular dynamics (MD) simulation. The orbital-free (OF) DFT approach is appealing but has proven difficult to implement because of the challenge of constructing reliable orbital-free approximations to the kinetic energy functional. To be maximally useful for multi-scale simulations, an OF-KE functional must be local (i.e. one-point). In the face of these difficulties, we demonstrate that there is a way forward. By requiring only that the approximate functional deliver high-quality forces, by exploiting the ``conjointness'' hypothesis of Lee, Lee, and Parr, by enforcing a basic positivity constraint, and by parameterizing to a carefully selected, small set of molecules we are able to generate a OF-KE functional that does a good job of describing various H$_q$Si$_m$O$_n$ clusters as well as CO and H$_2$O (providing encouraging evidence of transferability).   

Fitting of Molecular Densities by Compact, Atom-Centered Expansion

Use of an orbital-free (OF) version of DFT requires both a suitable approximate Kohn-Sham kinetic energy functional and a systematic but simple model of the system density. We report useful approximations to the KS density via a very compact expansion in atom-centered functions. Spherically averaged, isolated-atom densities are used as basis functions to expand spherically symmetric atom-centered contributions. A simplified expansion in real spherical harmonics is then added to the spherically symmetric contributions. Although drastically simplified, such representations of the density nevertheless result in impressively small mean square deviations relative to the target KS density. The fitted density can then be combined with an approximate OF-KE functional we have developed \footnote{V.V. Karasiev, S.B. Trickey, and F.E. Harris, J. Comp. Aided Mat. Des. (2005) (accepted).} to generate energy surfaces. These energy surfaces have shapes similar to those arising from true KS densities, and are therefore suitable for calculation of forces to drive molecular dynamics simulations.   

Magnetic Field Effects upon Exchange-Correlation in the Hooke's Atom

Extending Density Functional Theory (DFT) to coulombic systems in a non-vanishing magnetic field in a computationally feasible way is highly desirable. Even though the current DFT (CDFT) formalism is long-established, there still are no generally applicable, reliable $E_{xc}, {\mathbf A}_{xc}$ functionals analogous with the LDA. Progress can be made by comparison study on a solvable correlated system. Hooke's atom is well-known in ordinary DFT because its Schr\"{o}dinger equation can be solved exactly for some coupling strengths and numerically with high accuracy for the rest. Hence exact Kohn-Sham quantities are readily available. Using our extensions (exact and numerical) to non-zero B-field, we examined the effects on exchange- correlation holes and energies and considered possible ways to include the essential ones in $E_{xc}, {\mathbf A}_{xc} $. In our tests, the CDFT vorticity variable, $\nu$, turns out to be a computationally difficult quantity which may not be appropriate in practice to describe external B field effects on $E_{xc}, {\mathbf A}_{xc}$.   

The problems in the density functional theory with the total spin and space symmetry and the invariant properties of the electron density.

The problems in the density functional theory (DFT) arising when it is applied to the spin and space multiplets are discussed. It is rigorously proved that the electron density of an arbitrary $N$-electron system does not depend upon the value of the total spin $S$ of the state and preserves the same analytical form for all states with the definite $S$. It is also proved that the diagonal element of the full density matrix is invariant respecting all operation of the group symmetry of the state, i.e, it is a group invariant. From these results follows that the problems in DFT with the total spin and degenerated states cannot be solved within the framework of density matrix formalism.   

Dehydrogenation in catalyst activated MgH$_{2}$

Dehydrogenation in catalyst activated magnesium hydride (MgH$_{2})$ has been investigated using \textit{ab initio} Molecular Dynamics (MD) simulation and Nudged Elastic Band (NEB) method. Our calculation explains why small amount of Nb$_{2}$O$_{5}$ catalyst can substantially improve the thermodynamics and kinetics of MgH$_{2}$. We show that Nb$_{2}$O$_{5}$ promotes the creation of Mg vacancies and that the hydrogen desorption from the vicinity of Mg vacancies occurs in molecular form and is exothermic. The activation energy barrier for H$_{2}$ desorption in vacancy containing magnesium hydride (1.02 eV) is much lower than that in the pure magnesium hydride (3.30 eV). Therefore, the effective catalyst for dehydrogenation in MgH$_{2}$ will be one that can easily facilitate MgO formation.   

A method for biased surface electronic structure: a planewave non-repeated slab approach

We have developed a new formalism for calculating electronic structures in a symmetric/asymmetric slab model [1]. The method can treat not only surfaces exposed to vacuum but also biased surfaces. To solve the Kohn-Sham equation, we adopt a conventional Kohn-Sham solver in a repeated slab model. On the other hand, for the Poisson equation, we solve it in a whole space along surface normal direction. Owing to this treatment we can easily obtain work functions of the surface and we can calculate polarized surfaces without dipole correction. By introducing an effective screening medium and imposing appropriate boundary conditions to the Poisson equation, we can calculate a surface that is placed in front of electrode. In this model it is possible to apply a bias voltage to the surface by changing the Fermi energy of the surface. Thus we can calculate electronic and geometric structures of the biased surface. This model corresponds to the experimental setup for the scanning tunneling microscopy or back-gate field effect transistor. The important advantage of the method is that we can easily implement it in a conventional first-principles calculation method. References: [1] M. Otani and O. Sugino, submitted to PRB.   

Many-Body Electronic Structure of Curium metal

We report computer-based simulations for the many-body electronic structure of Curium metal. Cm belongs to the actinide series and has a half-filled shell with seven $5f$ electrons. As a function of pressure, curium exhibits five different crystallographic phases. At low temperatures all phases demonstrate either antiferromagnetic or ferrimagnetic ordering. In this study we perform LDA+DMFT calculations for the antiferromagnetic state of high-pressure fcc modification of Curium metal.   

Session P29: Biomolecular Structure and Functions

Interactions between model bacterial membranes and synthetic antimicrobials.

Antimicrobial peptides comprise a key component of innate immunity for a wide range of multicellular organisms. It has been shown that natural antimicrobial peptides and their analogs can permeate bacterial membranes selectively. There are a number of proposed models for this action, but the detailed molecular mechanism of the induced membrane permeation remains unclear. We investigate interactions between model bacterial membranes and a prototypical family of phenylene ethynylene-based antimicrobials with controllable hydrophilic and hydrophobic volume fractions, controllable charge placement. Preliminary results from synchrotron small angle x-ray scattering (SAXS) results will be presented.   

Observation of membrane fusion between individual virus particles and supported lipid bilayers

A portion of the host cell membrane is incorporated into newly produced, enveloped virus particles during an active infection. Fusion of that viral membrane with the membrane of targeted host cells is generally accepted to be a key step for the infection of normal cells as a virus spreads among a normal cell population. For the best studied enveloped viruses, viral proteins catalyze the membrane fusion reaction during a low pH step along the cellular endocytotic pathway. To gain a better understanding of the molecular mechanisms underlying viral membrane fusion, we have constructed an \textit{in vitro} fluorescence assay to allow high resolution, real time measurements of Sindbis viral fusion to supported lipid bilayers. Single particle tracking is used to observe individual virus particles. The mixing of a fluorescent dye incorporated into the viral membrane with the supported bilayer reports fusion. We present results regarding the effects of different lipid blends as well as different buffer conditions on membrane fusion for Sindbis virus. We compare the fusion of virus produced in mammalian cells to that from insect cells. .   

Simulating Domain Formation and Fusion in Lipid Bilayers

The lipid dynamics is the source of the variety of membrane structures and dynamics found in celss. Many of the interesting phenomena of biomembranes involve time scales of at least microseconds, which have been beyond simulations until recently. Coarse-grained models of lipid molecules have been developed to reach these long time scales and maintain the essential physical character of the molecules. Using these models in molecular dynamics simulations, time scales in the $\mu$s to ms range are treatable. As examples, simulations of domain formation and membrane fusion will be presented. In mixed lipid systems, the formation of domains is now understood to be an active component in biological processes. Our simulations reveal the dynamics of the lipid molecules that form domains in binary systems. In particular, the correlation between the two monolayers of the bilayer is dependent on the molecular structure of the lipid molecules. Membrane fusion is a fundamental process of cellular transport and infection processes. Understanding the basic principles governing membrane fusion has many important consequences. The coarse-grained molecular dynamics simulations show how lipid molecular structure influences the fusion process.   

Lipid Coupling in Asymmetric Supported Lipid Bilayers Revealed by Fluorescence Correlation Spectroscopy

In biological systems, phospholipids asymmetry in two leaflets is a key feature of cell membranes for membrane biogenesis, intracellular fusion and signal transduction. Detailed information of the interactions and dynamics of the asymmetric membranes is paramount for design of applications. Here we use fluorescence correlation spectroscopy (FCS) to measure the coupling between 1, 2-dilauroyl-\textit{sn}-glycero-3-phosphocholine (DLPC) and 1, 2-dipalmitoyl-\textit{sn}-glycero-3-phosphocholine (DPPC) in asymmetric planar-supported bilayers (PSLBs), at temperatures where DLPC is in the fluid phase but DPPC is in the gel phase. Asymmetric PSLBs were prepared by placing dilute fluorescent-labeled 1, 2-dimeristoyl-\textit{sn}-glycero-3-phosphoethanolamine (DMPE) in DLPC leaflet as the probe for measuring lateral diffusion within the host leaflet environment. By constructing asymmetric bilayers where DLPC is alternatively in the top and in the bottom leaflet, we compare lipid coupling between the two leaflets with frictional interaction between the leaflets and the nanometer-thick water layer that separates the bottom leaflet from the solid support.   

Dissipative Particle Dynamics Simulation of Structure Properties of Lipid Micelles

We chose dissipative particle dynamics (DPD) simulation to study the micelle structure properties. The self-assembly lipid is modeled by a flexible chain with head and tail particles. By changing interaction parameters between the solution particles and the lipid particles, three types of solution are considered: water, oil and water-oil mixture. It is found that the tail/head chains forming the micelle core have a very disorder distribution. The relation between the core radius with the aggregation number as well as the number of oil particles inside is obtained based on the simulation results and the molecular packing parameter. The mean density of particles in the core is higher than that in the simulation box. For micelles without oils inside, the density depends on the number of head particles in a lipid molecule, and the dependence becomes weaker as more oil particles are captured. At the core surface, head particles form clusters with water particles incorporated between the clusters. Comparisons with results from other simulation method such as MD show that DPD achieves high performance in simulating micelle formation and structure properties.   

Anomalously Slow Domain Growth in Membranes with Asymmetric Transbilayer Lipid Distribution

The effect of asymmetry in transbilayer lipid distribution on the phase separation of self-assembled multicomponent fluid vesicles is investigated numerically via dissipative particle dynamics. We show that this asymmetry induces a spontaneous curvature wich alters significantly the morphology and dynamics of the lipid mixture. In particular, at intermediate tension, domain growth is found to be anomalously slow dynamics. In contrast, in the limiting cases of low and high tensions, the dynamics proceed toward full phase separation.   

Molecular simulation studies of tail-length effects in mixed-lipid bilayers

Because lipid lateral diffusion is slow on the time-scale accessible to atomistic molecular dynamics (MD) simulations, equilibrium clustering and segregation in mixed-lipid bilayers are impractical to study through conventional computational approaches. A hybrid MD-Monte Carlo method employing lipid mutation moves within the semi-grand canonical ensemble method has been implemented for mixtures of lipids of differing tail lengths. For DLPC:DPPC mixtures, equilibration is demonstrated during simulation runs nearly two orders of magnitude shorter than estimates for standard MD. Statistical measures of lateral association in bilayer slabs and of partitioning among membrane micro-environments (flat bilayers, inner and outer leaflets of curved bilayers, and bilayer edge) will be presented.   

Stabilization of Model Membrane Systems by Disaccharides. Quasielastic Neutron Scattering Experiments and Atomistic Simulations

Trehalose, a disaccharide of glucose, is often used for the stabilization of cell membranes in the absence of water. This work studies the effects of trehalose on model membrane systems as they undergo a melting transition using a combination of experimental methods and atomistic molecular simulations. Quasielastic neutron scattering experiments on selectively deuterated samples provide the incoherent dynamic structure over a wide time range. Elastic scans probing the lipid tail dynamics display clear evidence of a main melting transition that is significantly lowered in the presence of trehalose. Lipid headgroup mobility is considerably restricted at high temperatures and directly associated with the dynamics of the sugar in the mixture. Molecular simulations provide a detailed overview of the dynamics and their spatial and time dependence. The combined simulation and experimental methodology offers a unique, molecular view of the physics of systems commonly employed in cryopreservation and lyophilization processes.   

Protein crystals on phase-separating model membranes

We study the interplay between the crystallization of proteins tethered to membranes and separation within the membranes of giant unilamellar vesicles (GUVs) composed of DOPC, sphingomyelin (SM), and cholesterol. These model membranes phase separate into coexisting liquid domains below a miscibility transition temperature. This phase separation captures some aspects of the formation of lipid rafts in cell membranes and demonstrates the influence of membrane composition on raft formation. Real cell membranes have a much more complicated structure. There are additional physical constraints present in cell membranes, such as their attachment to the cytoskeleton and the presence of membrane bound proteins. The self-association of membrane proteins can influence the membrane phase behavior. We begin to investigate these effects on model tethered protein- loaded membranes by incorporating a small amount of biotin-X- DPPE into our GUVs. The biotinylated lipid partitions into a cholesterol-poor phase; thus, streptavidin binds preferentially to one of the membrane phases. As streptavidin assembles to form crystalline domains, it restricts the membrane mobility. We examine the effect of this protein association on lipid phase separation, as well as the effect of the lipid phase separation on the crystallization of the tethered proteins.   

Model systems to investigate the effect of cholesterol on the transfection efficiency of lipoplexes

Motivated by its important role in lipid-mediated gene delivery, we have studied the effect of cholesterol on membrane fusion. While recent work in our group has identified the membrane charge density as a critical parameter for transfection efficiency (TE) of lamellar, DOPC containing cationic lipid-DNA (CL-DNA) complexes [1-3], this model cannot fully explain the effect of cholesterol, suggesting that a different mechanism is responsible for the observed enhancement of TE. A model system using negatively charged giant vesicles has been developed to mimic the interaction of the cell membrane with CL-DNA complexes containing cholesterol. Differences in fusogenic properties have been observed as a function of the amount of cholesterol present in the CL-DNA complexes, and a fluorescence resonance energy transfer based assay was employed to quantify this effect. X-ray diffraction confirms that the lamellar structure seen with CL- DNA complexes is retained with the addition of cholesterol. Funding provided by NIH GM-59288 and NSF DMR-0503347. [1] A.J. Lin et al, \textit{Biophys. J.}, 2003, V84:3307-3316. [2] K. Ewert et al, \textit{J. Med. Chem.}, 2002, V45:5023-5029. [3] A. Ahmad et al., \textit{J. Gene Med., }2005, V7:739-748.   

Had a drink last night? How alcohol interacts with biological membranes

We have performed extensive 100 ns molecular dynamics simulations to study the effect of methanol and ethanol on two different lipid bilayer systems (POPC and DPPC) in the fluid phase at 323 K [1,2]. We studied both structural changes induced by the alcohols and the dynamics of the system. It turned out that ethanol was able to penetrate the membranes whereas methanol was not able to do so. In particular, ethanol prefers to be accommodated in the vicinity of the lipid headgroup region. We also determined the dependence of lipid chain ordering on ethanol concentration and quite surprisingly found that to be non-monotonous. We explain that in terms of modified surface tension [2]. Finally, we determined lifetime of hydrogen bonds to be about 1 ns and found that be in excellent agreement with NMR results. \newline \newline [1] B.W. Lee, et al, Fluid Phase Equilibria 225, 63-68 (2004) \newline [2] M. Patra et al, Biophys. J., in press 2005   

Solid domain rafts in lipid vesicles and scars

The free energy of a crystalline domain coexisting with a liquid phase on a spherical vesicle may be approximated by an elastic or stretching energy and a line tension term. The stretching energy generally grows as the area of the domain, while the line tension term grows with its perimeter. We show that if the crystalline domain contains defect arrays consisting of finite length grain boundaries of dislocations (scars) the stretching energy grows linearly with a characteristic length of the crystalline domain. We show that this result is critical to understand the existence of solid domains in lipid-bilayers in the strongly segregated two phase region even for small relative area coverages. The domains evolve from caps to stripes that become thinner as the line tension is decreased. We also discuss the implications of the results for other experimental.   

Efficient calculation of mechanical properties of multicomponent phospholipid bilayers with Monte Carlo simulations

We present a systematic study of the mechanical properties of multicomponent phospholipid bilayers. Two sets of systems are considered. The first consists of a mixture of DioleoylPhosphatidylethanolamine (DOPE) and DioleoylPhosphatidylcholine (DOPC) phospholipids. These two molecules have different head groups but the same chain length. The second system consists of a mixture of DilauroylPhosphatidylcholine (DLPC) and DistearoylPhosphatidylcholine (DSPC); these molecules have different chain lengths but the same head group. We use atomistic and coarse grain models, coupled to advanced Monte Carlo simulation techniques, to examine the structure and mechanical properties of the bilayers. Our results for pure systems are in quantitative agreement with experiment. Experimental data for mixed bilayers are not available, but our results indicate that their mechanical behavior is highly non-linear, a finding that we can interpret in terms of the composition and the resulting structure of the mixtures.   

Roughness effect on vesicle adhesion characterised by a novel micropipette-based technique

Numerous biological processes have to go through a cell adhesion process, which make the fundamental study of the adhesion of cells on solid substrate a key research topic in cellular biophysics. We will present our work on the adhesion of a single vesicle on a substrate. A vesicle is held at the end of a micropipette mounted on a micromanipulator and put into contact with a surface. We developed a novel technique to directly measure adhesion using the spring-constant of an L-shaped micropipette when pulling the vesicle from the substrate. The substrate is made of a micropatterned polymer film coated with a thin layer of gold to promote adhesion with the vesicle. The effect of the surface roughness can therefore be carefully characterized.   

Structure of Cholesterol Helical Ribbons, Self-Assembling Biological Springs.

Helical ribbons with characteristic pitch angles form spontaneously in a variety of quaternary surfactant-lipid-sterol-water solutions. These helical ribbons form in a variety of axial lengths, widths and radii. Surprisingly, however, they all have pitch angles of either 11 or 54\r{ }. Our X-ray diffraction studies of individual ribbons confirm that the remarkable stability of each of the two pitch angles is related to a crystalline nature of the ribbons. The small size (of 100 x 10 x 0.1 $\mu $m$^{3})$ and the significant curvature of the ribbons produce weak and broad Bragg peaks. Therefore, novel methods are used to analyze these data. The structure of these ribbons is similar to that of cholesterol monohydrate. Interestingly, there is an evidence for a superlattice structure, resembling that found in thick films of cholesterol grown at the air-water interface.   

Session P30: Focus Session: Organic Interfaces

Properties of Organic Molecules at Metal Surfaces

The adsorption and self–assembly of organic molecules at surfaces has recently been investigated extensively, both because of the fundamental interest and for prospective applications in nanoelectronics and nanophotonics [1, 2]. Molecule–molecule and molecule–substrate interactions can be tuned by the appropriate choice of substrate material and symmetry. Upon molecular adsorption, surfaces typically do not behave as static templates, but often rearrange dramatically to accommodate different molecular species [3, 4]. This presentation reviews recent experimental work using Scanning Tunneling Microscopy, which is providing new insight into fundamental properties such as molecular diffusion [5, 6] and self–assembly via surface templating [7] and hydrogen bonding driven by co-adsorption [8]. \newline \newline [1] F. Rosei et al., \textit{Prog. Surf. Science} \textbf{71}, 95 (2003). \newline [2] F. Rosei, \textit{J. Phys. Condens. Matter} \textbf{16}, S1373 (2004). \newline [3] F. Rosei et al., \textit{Science} \textbf{296}, 328 (2002). \newline [4] R. Otero, F. Rosei, et al., \textit{Nanoletters} \textbf{4}, 75 (2004). \newline [5] M. Schunack, T.R. Linderoth, F. Rosei, et al., \textit{Phys. Rev. Lett.} \textbf{88}, 156102 (2002). \newline [6] J.A. Miwa, S. Weigelt, H. Gersen, F. Besenbacher, F. Rosei, T.R. Linderoth, submitted. \newline [7] R. Otero, Y. Naitoh, F. Rosei et al., \textit{Angew. Chem.} \textbf{43}, 4092 (2004). \newline [8] K.G. Nath, O. Ivasenko, J.A. Miwa, H. Dang, J.D. Wuest, A. Nanci, D.F. Perepichka, F. Rosei, submitted.   

Direct measurements of contact resistances in asymmetric pentacene thin-film transistors with polyaniline and gold electrodes

We have fabricated asymmetric pentacene thin-film transistors with one gold electrode and one polyaniline (PANI) electrode connected to the same pentacene channel. Surface potential measurements reveal large potential drops at the gold/pentacene contact, but not at the PANI/pentacene contact during operation. We observe, however, some potential drop along the PANI electrode outside the scan window due to the bulk resistance of PANI. To minimize the potential drop across the PANI electrode, we have fabricated asymmetric devices with one exposed gold electrode, and one PANI-coated gold electrode. The surface potential profiles of these asymmetric devices with the PANI-coated gold electrode as the source electrode and the exposed gold electrode as the drain electrode, and vice versa, reveal no additional potential drop across the PANI-coated gold electrode. Quantification of the contact resistance indicates that the PANI/pentacene contact is significantly less resistive than the gold/pentacene contact whether the PANI-coated gold electrode is used as the source, or the drain electrode. Additionally, charge injection from the exposed gold electrode and the PANI-coated electrode appears to be more difficult than charge extraction from these electrodes.   

Nm-resolution studies of Au/molecular-film/GaAs junctions using ballistic electron emission microscopy (BEEM)

BEEM was used to image and quantify lateral homogeneity and energy band alignments at molecule/electrode interfaces in Au/dC-X/GaAs structures, where dC-X are dicarboxylic ligands with X= H, OCH$_{3}$, CF$_{3}$, CN, or CH$_{3}$ [1]. Transport through such junctions was proposed to be dominated by ``pinholes'' in the dC-X film, with the Au/GaAs Schottky barrier height (SBH) at pinholes modified by the surrounding molecular film dipole [1]. BEEM images of dC-CH$_{3}$ with V$_{tip }< \quad \sim $1.38 eV indeed revealed isolated 20-40 nm sized ``pinholes'' with measured local SBHs ranging from 0.90 -- 1.0 eV, consistent with the model [1]. However, between the pinholes we also observed a new conduction channel for V$_{tip }> \quad \sim $1.38 eV, possibly due to transport through the LUMO of the dC-CH$_{3}$ film itself. BEEM measurements for the other --X groups also showed non-uniform, film-dependent SBH, but for those films the pinholes were too dense or the films too transparent to resolve isolated pinholes. All dC-X films were stable under the BEEM hot-electron flux. \newline \newline [1] H. Haick \textit{et al}., Adv. Mater. \textbf{16}, 2145 (2004).   

A Possible Mechanism For Photoinduced Effects In Molecule-Based Magnets

A mechanism based on charge transfer processes between ligand and metal, for photoinduced effects on magnetic order that are observed in manganese-tetracyanoethylene molecule-based magnet is proposed. In order to support the mechanism, Monte Carlo calculations for a double exchange model with antiferromagnetic interaction between nearest neighbor manganese core spins, J$_{AF}$ on two dimensional metal-ligand lattice are performed. Depending on strength of J$_{AF }$ and the number of electrons in the system, total average magnetization and average angles of core spins are calculated.   

The electronic structure and polymerization of a self-assembled monolayer

Irradiation-induced modifications of electronic structure in the monomolecular insulator [1,1';4',1''-terphenyl]-4,4''-dimethanethiol (TPDMT) films have been investigated by photoemission and inverse photoemission. A dominant effect is cross-linking of the TPDMT film, which result in a quasipolymeric material with a smaller gap between the highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) but improved the dielectric properties. The cross-linking processes are also accompanied by changes in molecular orientation. The photoemission intensities of organic molecular layers generally obey the Debye-Waller temperature dependence but not always. With the example of a monomolecular film formed from [1,1';4',1''-terphenyl]-4,4''-dimethanethiol, we show that pronounced deviations from Debye-Waller temperature behavior are possible and are likely caused by temperature dependent changes in molecular orientation.   

A first-principles study of $\pi$-stacking in charged oligothiophenes in the presence of counterions

We investigate the structural and electronic properties of charged oligothiophenes and of their $\pi$-stacking interactions with extensive density-functional theory (PBE, B3LYP) and post-Hartree-Fock (MP2) calculations. We pay particular attention to the role of counterions and of the solvation medium in tuning the balance between electrostatic repulsion and chemical bonding, including explicitly hexafluorophosphate counterions, and exploring the role of polarizability and surface tension for different solvents. Our calculations show that Coulomb's repulsion of the charged oligothiophenes is rapidly screened by solvation, counterions, or both, leading to stable $\pi$-dimerized systems (or higher stackings) where binding is driven by $\pi$-bond hybridization. Furthermore, we studied the charge-transfer properties of the counterion-oligomer system, as well as site preferences for counterions, highlighting the relevance of a proper treatment of correlations and self-interaction in describing the electronic-structure of these systems.   

Orientation of Fluorophenols on Si(111)

Oriented adsorption of switchable organic molecules at surfaces is an important prerequisite for single molecular electronics [1, 2]. As model systems we select polar fluorophenols with tailored dipole moments and investigate their adsorption on the Si(111)7$\times $7 surface by near edge x-ray absorption fine structure spectroscopy (NEXAFS). A strong polarization dependence of the $\pi $* transitions is observed in fluorinated phenols, while phenol itself is isotropic. A quantitative model is developed to convert polarization-dependent NEXAFS data into orientational information. The model includes three angular degrees of freedom, two of them fixed and the other with a Gaussian distribution. Such a situation is encountered in a variety of self-assembled monolayers (SAMs) with tailored end groups [3]. [1] T. A. Jung, R. R. Schlittler, J. K. Gimzewski, Nature \textbf{386}, 696, (1997) [2] A. J. Mayne, M. Lastapis, G. Baffou, L. Soukiassian, G. Comtet, L. Hellener and G. Dujardin, Phys. Rev. B \textbf{69}, 045409 (2004) [3] Y.Y. Luk, N. L. Abbott, J. N. Crain and F. J. Himpsel, J. Chem. Phys. \textbf{120,} 10792 (2004)   

Session P31: Nanotubes: Theory and Experiment

Building and Deploying Community Nanotechnology Software Tools on nanoHUB.org -- Non-Equilibrium Green's Function Simulations of the Impact of Atomic Defects on the Performance of Carbon Nanotube Transistors.

The Network for Computational Nanotechnology (NCN) is a multi-university, NSF-funded initiative with a mission to lead in research, education, and outreach deploying a unique web-based infrastructure (http://nanoHUB.org) to serve the nation's National Nanotechnology Initiative. Around 30 research codes/community tools are available and all the NCN services are free of charge. One such community tool is the CNTFET simulator based on NEGF techniques and the Finite-Element-Method (FEM) to treat three-dimensional (3D) electrostatics. We are able to simulate electronic transport in experimentally demonstrated 3D CNT devices with atomistic potential and charge resolution. Currently, we are investigating the effects of atomistic defects in the CNT devices such as vacancies and charged impurities.   

Separation of single-walled carbon nanotubes into metallic and semiconducting groups: a simple and large-scale method

Separation of a large number of single-walled carbon nanotubes (SWNTs) into groups each with specifically metallic and semiconducting properties is an extremely important task for technology application. Even though effective methods (1, 2) have been devised, they suffer from drawbacks such as either the yield is low (3) or expense is high (4). In this work, we study the problem from a theoretical approach, we notice that based on the first principles calculations the binding strengths of methylamine to the semiconducting [13, 0] SWNT are only 36$\sim$61\% of that to the metallic [7, 7] SWNT, which suggests that the amines is much more attractive toward the pure metallic than the semiconducting SWNTs. Therefore starting from as-prepared SWNTs and with the assistance of amines, we achieved SWNTs with enriched metallic properties over semiconducting in a convenient and large-scale manner. References: (1) D. Chattopadhyay, L. Galeska, F. Papadimitrakopoulos, Journal of the American Chemical Society 125, 3370 (MAR 19, 2003). (2) H. P. Li et al., Journal of the American Chemical Society 126, 1014 (FEB 4, 2004). (3) R. Krupke, F. Hennrich, H. von Lohneysen, M. Kappes, SCIENCE 301, 344 (JUL 18, 2003). (4) M. Zheng et al., Science 302, 1545 (NOV 28, 2003).   

Polaron superconductivity model for Li-doped nanotube-zeolite composite

We propose a polaron superconductivity model for Li-doped nanotube-zeolite composite, in which the 4 Angstrom carbon nanotubes are embedded in the zeolite matrix, with a nanotube-nanotube wall separation of less than 10 Angstroms. The small separation implies inevitable nanotube-nanotube coupling, leading to a 3D anisotropic superconductor. Here we calculate the mean-field superconducting transition temperature based on the fact that each adsorbed Li atom, situated in the middle of the nanotube, donates an electron to the nanotube so as to form an ion-electron system. In addition, the Li ion is trapped in a shallow well formed by the (5,0) nanotube, with a periodicity of 4.3 Angstroms. So the Li vibration resembles that of an optical phonon. We have evaluated both the electron-electron interaction and the electron-phonon interactions in the presence of the screening effect, and solved the Eliashberg-Gorkov equation to obtain the superconducting transition temperature.   

Theoretical Model for a Carbon Nanotube-Based Magnetometer at Non-Zero Temperatures

We present a complete description of electronic current in metallic single-walled carbon nanotubes under the influence of axially oriented magnetic fields at nonzero temperatures. We include in our model [1] the diameter of the carbon nanotube, the temperature and length of the nanotube. We find that the current in a zigzag carbon nanotube that is metallic at zero magnetic field is strongly modulated by varying the magnitude of an axially oriented magnetic field. We use this property, to propose a design of a carbon nanotube based directional magnetometer that could be designed to sense magnetic fields from 1 T to 8 T and at temperatures from 0 K up to 100 K. [1] Vladimir Dobrokhotov and Christopher Berven, ``Electronic Transport Properties of Metallic CNTs in an Axial Magnetic Field at nonzero Temperatures: A Model of an Ultra-small Digital Magnetometer,'' accepted for publication 11-2005 Physica E   

Effect of Short-Range Electron Correlation in Nanotubes

We study the effect of short-range electron correlation in the zigzag and armchair carbon nanotubes (CNT). We derived the dynamic pair interaction potential between two electrons in the tubule incorporating Hubbard type local field factor. The dispersion of plasma modes at different values of angular momentum, and chirality angle and single-particle excitations are derived as well. We also evaluate the self-energy part due to the interaction of an electron with acoustic mode.   

Electron Standing Waves in Semiconducting Carbon Nanotubes: Spatially-Resolved Scanning Tunneling Spectroscopy

Electronic modulation patterns were observed, from the gap states of semiconducting single-wall carbon nanotubes, using spatially-resolved scanning tunneling spectroscopy. Some modulations show single peaks, with the period twice of the lattice constants, while others show double peaks.. Both modulations are localized within a few nano-meters, enclosed by exponential decay functions. The modulation patterns are well understood in terms of the squared wavefunctions, derived from the simple quantum mechanical potential well models. Our model can be applied to the bound states of metallic carbon nanotubes as well.   

Electronic structure of carbon nanotubes adsorbed on Si(001) vicinal surfaces.

We have investigated adsorption of carbon nanotubes on Si(001) vicinal surfaces using Density Functional Theory total energy and electronic structure calculations. Energetically favorable adsorption orientations and positions of carbon nanotubes were searched by total energy calculations, and detailed atomic structure of carbon nanotubes adsorbed at most probable adsorption sites have been obtained by full structure relaxation. Adsorption energy shows both direction and site dependence since the carbon nanotube form covalent bonds with the clean vicinal surface. Charge transfer between the nanotube and the surface happens mainly at the interface, which results in a quasi one-dimensional electron channel. Electronic states of carbon nanotube and silicon surface are highly rehybridized and mixed. A metallic carbon nanotube could behave as small gap semiconductor depending on adsorption site. Insertion of $sp^3$ bonded carbon atoms in $sp^2$ bond network introduces energy gap in electronic structure of the nanotube near Fermi level. This energy gap could be eliminated if there are surface states available for charge transfer. But such surface states are not available for particular nanotube directions, and non-metallic electronic structure appears. Dangling bond states of silicon surface, which appears as energy bands near the Fermi level, attracts electrons from the nanotube provided that these states are localized near the nanotube. Termination of surface dangling bonds in the vicinity of adsorbed nanotube could help tune the electronic properties of adsorbed nanotube.   

The II-VI nanostructure zoo

We present predictions of a completely new family of nanotubes, related nanostructures and fullerene-like cages formed from the II-VI semiconductor mercury telluride. Our predictions are supported by first-principles calculations on the structures. The structures are remarkable in several ways: They are all more stable than the planar form of HgTe from which they are formed; they are radically altered from the tetrahedral bulk forms, and a strong interaction with the electronic structure results in a semimetal-semiconductor transformation; and for the larger armchair tubes isomerisation leads to striking structures formed from heavy modification of the tube, accompanied by large changes in the bandgap. For the nanotubes two simple rules for preferred coordination of Hg and the Hg-Te-Hg bond angles explain the structural stability of the nanotubes, and the formation of their exotic isomers. The cage structures are based upon the Archimedean and Platonic solids, where key requirements in terms of numbers of vertices, number of triangular faces and their connectivity determine the viable subset of structures.   

Nonlinear ac conductivity of disordered Luttinger liquids

We consider low energy charge transport in one-dimensional electron systems with short range interactions under the influence of a random potential. At zero temperature, the linear ac conductivity vanishes like $\sim \omega^2 (\ln (1/\omega))^2$. Much less is known about the \emph{non-linear conductivity}. At zero temperature and frequency, charge transport is only possible by tunneling of charge carriers, which can be described by instanton formation. The nonlinear dc conductivity is characterized by $I \sim \exp( - \sqrt{E_0 / E}) $ provided the system is coupled to a dissipative bath [1]. Combining RG and instanton methods, we calculate the nonlinear ac conductivity and discuss the crossover between the nonanalytic field dependence of the electric current at zero frequency and the linear ac conductivity at small electric fields and finite frequency [2].\\[0.5cm] [1] S.~Malinin, T.~Nattermann, and B.~Rosenow, Phys. Rev. B {\bf 70}, 235120 (2004).\\[0cm] [2] B.~Rosenow and T.~Nattermann, accepted for publication in Phys. Rev. B; preprint cond-mat/0408042.   

Renormalization of a single impurity potential of arbitrary strength in a Tomonaga-Luttinger liquid

We study the renormalization flow of a single impurity potential of arbitrary strength in a Tomonaga-Luttinger liquid (TL). It is known that an impurity potential in TL is effectively renormalized by electron-electron interaction, with different manners in weak and strong potential limits for spin dependent models $K_{s}\neq1$. This fact strongly suggests that the fixed points of an impurity potential should shift as varying potential strength. In order to determine the scaling fixed points at arbitrary potential strength, we extend boundary bosonization scheme to the problem of arbitrary potential strength, and calculate the local density of states (LDOS) as a function of temperature and distance from the impurity. The impurity scaling flow is determined from the ratio between LDOS at the boundary and in the bulk. For $K_{s}=1 $, the phase boundary is given by $K_{\rho}=1$ irrespective of the potential. For $K_{s}\neq 1$, we find that the fixed points shift from $K_{\rho}\sim 2-K_{s}$ to $K_{\rho}=1/K_{s}$ as increasing the potential strength from $0$ to $\infty$. We also discuss how the scaling behavior appears in transport experiments.   

Adsorbed monolayers on suspended single-walled carbon nanotubes

A monolayer of adsorbates on a single-walled carbon nanotube presents the possibility of extending earlier studies of two-dimensional monolayer systems on graphite to the quasi-one-dimensional regime, by effectively imposing cylindrical boundary conditions. The monolayer can be detected either via its effect on the nanotube's conductance or by using the nanotube itself as a vibrating microbalance. Many adsorbates are known to affect the conductance, through a variety of mechanisms. Amongst these are O$_{2}$ and the noble gases Xe and Kr, whose phases and ordering on 2D graphite are well known. Our experiments so far have indicated that the presence of an O$_{2}$ layer on a nanotube close to liquid nitrogen temperatures can be detected using a threshold shift. We are now fabricating individual suspended nanotube devices with the initial aim of studying cylindrical commensurability effects on the phases of noble gases using the microbalance technique.   

Superplastic single-walled carbon nanotubes

Theoretical prediction on the maximum achievable tensile strain of a single-wall carbon nanotube (SWCNT) is less than 20{\%}, but experiments indicate a much lower attainable strain of less than $\sim $6{\%}. Here we report that, at temperatures of above 2000\r{ }C, SWCNTs deform superplastically, with a tensile elongation to failure nearly 280{\%}, and a diameter reduction of fifteen times. With this remarkable dimension change, the electronic property changes correspondingly from a metal with a pseudogap to a semiconducting state with a tunable gap up to 2 eV. Such superplastic deformation originates from plastic deformation mechanism dominated by the nucleation and motion of the kinks as well as atom diffusion in SWCNTs at high temperatures. Variable range hopping conduction is observed in the localized state due to scattering by point defects and kinks in the quasi-one-dimensional system.   

Unoccupied electronic states of Multiwall Carbon Nanotubes Arrays

We have grown multiwall carbon nanotube (MWCNT) arrays by CVD both from pyrolysis of Fe-Phthalocyanine and decomposition of Acetylene on Fe covered SiO$_{2 }$/Si(111) substrates. The characteristic diameter of the tubes is 50 nm for both type of samples. Even though the films show good alignment in the bulk, they do present some disorder of the tubes at the top of the films. Inverse photoemission spectra from these samples are similar to those obtained from HOPG. The main differences are in: the non existence of what has been recognized as an image charge state on graphite and some additional intensity very close to the Fermi level ($\varepsilon _{F})$. A similar intensity has been measured previously by photoemission in a symmetrical position with respect to $\varepsilon _{F}$. This increased metallic character could, at this point, be interpreted as defects from the closure of the tubes or as a manifestation of van Hove oscillations in the unoccupied density of states.   

Electrostatics of Straight and Wiggly Nanotubes in External Electric Field

Distribution of charge induced in a straight nanotube (NT) by external electric field parallel to the NT axis is found as a function of the NT length and radius. As the voltage drop along the NT exceeds the gap, positive and negative charge regions emerge at the NT ends. These regions are separated by a neutral strip at the NT center. External field is unscreened within the neutral strip, while it is strongly suppressed outside the strip. For a NT of a wiggly shape, the induced charge distribution represents alternating positively and negatively charged regions separated by neutral strips.   

Session P32: Spin Glasses

Effects of high magnetic fields on the spin-glass states in disordered manganites

Magnetization and magnetoresistance were measured in single crystals of random alloys RE$_{1-x}$AE$_{x}$MnO$_{3}$ (RE and AE denote the rare-earth and alkaline-earth ions at the perovskite A-site) in pulsed high magnetic fields up to 50 T with a long time duration ($\sim $10 ms) and up to 140 T with a short time duration ($\sim \mu $s). The crystals exhibit the spin glass behaviors at low temperatures in zero field. In high magnetic fields, Sm$_{1-x}$(Ba$_{1-y}$Sr$_{y})_{x}$MnO$_{3}$ showed prominent metamagnetic transitions, whereas RE$_{1-x}$Ba$_{x}$MnO$_{3}$ (RE=Sm, Eu, Gd) showed a smooth magnetization saturation with just kinks in the derivative of the magnetization. Moreover, in the metamagnetic phase transitions, peculiar time dependence and the pre-history dependence were found in the hysteresis of the magnetization. It was also found that the magnetization is accompanied with a colossal magnetoresistance. These behaviors were interpreted in terms of the developments of the clusters and the orbital orders by magnetic fields, which are dependent on the average A-site ionic radius and the randomness.   

What is new for spin-glass in a quasi-2D system

In conventional spin glasses, magnetic interaction is not strongly anisotropic and the entire spin system is believed to be frozen below the spin-glass transition temperature. Along {\em any} direction, spin correlations are highly disordered. In La$_2$Cu$_{0.94}$Li$_{0.06}$O$_4$, for which the in-plane exchange interaction dominates the interplane one, only a fraction of spins with antiferromagnetic correlations extending to neighboring planes become spin-glass. The remaining spins with only in-plane antiferromagnetic correlations remain spin-liquid at low temperature. Spin correlations are highly disordered only along the interlayer direction, but highly ordered in-plane. Such a novel partial spin freezing out of a two-dimensional spin-liquid observed in this cold neutron scattering study is likely due to a delicate balance between disorder and quantum fluctuations in the quasi-two dimensional $S$=1/2 Heisenberg system.   

Spin Glasses at the Bond Percolation Threshold

Low energy excitations for the Edwards-Anderson model on hyper- cubic lattices at the bond percolation threshold $p_c$ are investigated. At $T=0$, $p_c$ separates paramagnetic and spin glass phases. At the ``edge'' of the ordered state, these excitations are characterized by a distinct scaling exponent. This exponent allows to determine the shape of the phase boundary, $T_c(p)\sim(p-p_{c})^\phi$, for $p\to p_c^+$, which is experimentally measurable in $d=3$. At $p_c$, very large spin glass systems can be studied with an {\it exact} reduction algorithm\footnote{Europhys. Lett. {\bf 67}, 453 (2004)} to produce accurate scaling behavior. For more information, see http://www.physics.emory.edu/faculty/boettcher/   

Spin-glass correlations in classical dipoles

We present analytic mean-field and high-temperature expansion results on thermodynamics of classical Ising dipoles in the strongly diluted regime. These display a broad distrubution of couplings resulting in strong enhancement of the spin-glass transition temperature from the naive estimate. We comment on reliability of these results and their relationship to other approaches.   

Computing Barriers in Spin Glasses

The energy barriers $E_B$ between low-lying states in spin glasses are expected to scale as a power of the system size: $E_B \sim N^{\psi/d}$ for $N$ spins in a $d$-dimensional spin glass. Whether the barrier exponent $\psi$ is equal to the stiffness exponent $\theta$ (where the cost of minimal large scale excitations scales as $N^{\theta/d}$) is an unsolved question in general. In an attempt to solve this question with some rigor in large theoretical spin glass samples, numerical simulations for barriers in spin glasses on a hierarchical lattice have been carried out, using an exact algorithm for computing the barrier to the monotone growth of connected domains. The resulting $\psi$ is sensitive to the distribution of weights on the bonds between spins. These distributions give different weights to bonds that appear at different stages of the hierarchical generation of the lattice. Results for $\psi$ and $\theta$ will be presented for various lattices, including variations of Cayley trees and lattices that satisfy Migdal-Kadanoff approximations, and distributions that plausibly emulate finite-dimensional spin glasses.   

The End of Aging in a Spin Glass

Aging phenomena in complex systems has been used as an important tool to investigate the physics of complexity. In particular aging effects in spin glasses, measured using the Thermoremenant Magnetization (TRM) decays, have been instrumental as a probe of complex equilibrium and non-equilibrium dynamics. Current theoretical and experimental analysis suggest that the TRM decay of spin glasses is mainly composed of two terms; The ``stationary'' term which does not depend on the sample history and dominates the short time decay ($<$1s) and a long time aging term which depends on the samples history. We report finding that aging found in spin glass materials, has a finite lifetime and that after aging has ended there is a third component of the magnetization decay. This decay is independent of the waiting time, logarithmic in nature and part of the same mechanism that produces aging. Finally we find that the logarithmic decay implies a maximum aging time (MAT) that is very strongly dependent on temperature and ranges from short times near the spin glass transition temperature to many times the current best estimates of the age of the universe for low temperature.   

Memory and aging effect in hierarchical spin orderings of stage-2 CoCl$_{2}$ graphite intercalation compound

Stage-2 CoCl$_{2}$ graphite intercalation compound undergoes two magnetic phase transitions at $T_{cl}$ (= 7.0 K) and $T_{cu} $ (= 8.9 K). The aging dynamics of this compound is studied near $T_{cl}$ and $T_{cu}$. The intermediate state between $T_ {cl}$ and $T_{cu}$ is characterized by a spin glass phase extending over ferromagnetic islands. A genuine thermoremnant magnetization (TRM) measurement indicates that the memory of the specific spin configurations imprinted at temperatures between $T_{cl}$ and $T_{cu}$ during the field-cooled (FC) aging protocol can be recalled when the system is re-heated at a constant heating rate. The zero-field cooled (ZFC) and TRM magnetization is examined in a series of heating and reheating process. The magnetization shows both characteristic memory and rejuvenation effects. The time $(t)$ dependence of the relaxation rate $S_{ZFC}(t)=(1/H)$d$M_{ZFC}(t)$/d$\ln t$ after the ZFC aging protocol with a wait time $t_{w}$, exhibits two peaks at characteristic times $t_{cr1}$ and $t_{cr2}$ between $T_{cl}$ and $T_{cu}$. An aging process is revealed as the strong $t_{w}$ dependence of $t_{cr2}$. The observed aging and memory effect is discussed in terms of the droplet model.   

Bose glass vs. Mott glass in site-diluted S=1 Heisenberg antiferromagnets

Making use of large-scale quantum Monte Carlo simulations, we investigate the ground-state phase diagram of the square-lattice S=1 Heisenberg antiferromagnet with strong single-ion anisotropy and in presence of site-dilution of the magnetic lattice. Mapping the spins onto Holstein-Primakoff bosons, the single-ion anisotropy is seen to play the role of a repulsive on-site potential for the bosons. The clean limit of the model shows an anisotropy-driven quantum phase transition from an XY ordered (superfluid) phase to a quantum disordered (Mott insulating) phase. A similar transition is also driven by the application of a uniform field on the disordered state. Adding site dilution to the model, the non-trivial interplay between quantum fluctuations and lattice randomness gives rise to a novel quantum-disordered Mott-glass phase in zero field, with a gapless spectrum and yet a vanishing uniform susceptibility. Upon applying a field, such phase is turned into a Bose glass, with gapless spectrum and finite susceptibility. The above picture is directly relevant for experiments on doped quasi-low-dimensional Ni compounds, such as the recently investigated NiCl$_2$-4SC(NH$_2$)$_2$ (V.S. Zapf et al., condmat/0505562).   

Impurity effects on frustrated ferro- and ferrimagnets in one dimension

We have investigated impurity effects on magnetization for frustrated one-dimensional ferro- and ferrimagnets. Using the density-matrix renormalization group method and the exact diagonalization method, we confirmed that the magnetization decreases significantly by doping non-magnetic impurities. In a special case, the magnetization can vanish due to a single impurity in finite chains. Introducing the picture of magnetic domain inversion, we numerically investigated the impurity-density dependence of magnetization. In particular, we show that the magnetization substantially decreases down to less than 60{\%} from that of the corresponding pure system by doping an infinitesimal density of impurities. We also formulate conditions for the materials which may show this anomalous impurity effect.   

Frustrating interactions in oxides induced by non-magnetic impurities

An antiferromagnetic host material doped with non-magnetic impurities, such as Zn-doped La$_2$CuO$_4$, is generally believed to represent an excellent model case of the site-dilution of a magnetic substance. We demonstrate that there exist a significant qualitative correction to such a picture: an impurity can induce substantial frustrating interactions between spins that are nearest neighbors of the impurity site. Not only this effect explains discrepancies between experimental data and the site-dilution theory, but it could also lead to some important modification of the behavior of doped antiferromagnets close to the percolation. We study the 2D, $S=1/2$ copper-oxide plane with Zn impurities starting from the microscopic three-band Hubbard model. We show that, for a wide range of the model parameters, the substantial superexchange interactions between the next- and next-next-nearest neighbor Cu spins around the impurity site can be generated via the virtual transitions through the oxygen orbitals. Surprisingly, the interaction across the impurity $J''_{Zn}$ is greater than the next-nearest neighbor interaction $J'_{Zn}$ due to a partial cancellation of the super- and the cyclic exchanges for the latter. This study is completed by the $T$-matrix calculation of the staggered magnetization $M(x)$ as a function of Zn doping $x$. The predicted range of $J'_{Zn}$ and $J''_{Zn}$ agrees with the values needed to explain experimental deviation of $M(x)$ from the results of the site-dilution theories.   

Spin Polarization Measurments of Co$_{1-x }$-- Pt$_{x}$ alloys by Point Contact Andreev Reflection Spectroscopy

Recently Kaiser \textit{et al.,}$^{1}$ compared the spin polarization measured by spin resolved tunneling spectroscopy (Tedrow-Meservey) and the magnetic moment of Co$_{1-x }$-- Pt $_{x}$ alloys. We have measured the transport spin polarization, P$_{c}$ and magnetic properties of the same series of samples using Point Contact Andreev Reflection Spectroscopy (PCAR). All films with x varying from 0 to 100{\%} and a thickness of $\sim $1000 {\AA} were grown on Si substrates covered with $\sim $250 {\AA} of SiO$_{2}$ by magnetron sputtering. We will present a correlation between spin polarization and magnetization for this series of magnetic alloys and compare our results with the ones obtained in Ref. [1]. 1. C. Kaiser \textit{et al.}, PRL \textbf{94} 247203 (2005).   

Charge degrees of freedom in frustrated lattices

We explore systematically the charge degrees of freedom in frustrated lattices. A model of spinless fermions on a checkerboard lattice with nearest-neighbor hopping t and Coulomb repulsion V is used at half and quarter fillings. Quantum fluctuations reduce the classical macroscopic degeneracy. For the strongly correlated limit V >> t, an added electron decays into two quasiparticles with fractional charge. We study the classical correlations and, by means of quantum field theory as well as axact diagonalisation, we also investigate the possibility of a confined or deconfined phase as well as the statistics of these quasiparticles.   

Probing the Almeida-Thouless line away from the mean-field model

In order to test the existence of a spin-glass phase in a field at finite temperatures, results of Monte Carlo simulations of the one-dimensional long-range Ising spin glass with power-law interactions in the presence of a (random) field are presented. By tuning the exponent of the power-law interactions, we are able to scan the full range of possible behaviors from the infinite-range (Sherrington-Kirkpatrick) model to the short-range model. A finite-size scaling analysis of the correlation length indicates that there is no transition in a field with non-mean field critical behavior at zero field. This suggests that there is no Almeida-Thouless line for short-range Ising spin glasses away from the mean-field regime.   

Session P33: Focus Session: Friction, Fracture and Deformation III

Atomistic Simulation of Size Effects in Bending a Single Crystal

We perform atomistic Monte Carlo simulations of bending a Lennard-Jones single crystal in two dimensions. In examining initial yield, we find an apparent ``reverse'' size effect. However, when strain rate effects are taken into account, we demonstrate that the size effect disappears. Once geometrically necessary dislocations coalesce to form grain boundaries, we observe a size effect of the usual kind, e.g. smaller samples support a higher scaled bending moment than larger samples. We compare simulation results with recent experiments on bending of highly annealed nanowires [B. Wu et al, Nature Matls 4, 525, 2005.] Finally, we observe a topological instability in the evolution of a grain boundary intersecting a free surface under compressive stress. The grain boundary buckles and nucleates a protruding grain, suggesting a novel mechanism for the formation of a hillock on a compressed metal surface.   

Analytical calculation of energy barrier for dislocation nucleation from a crack tip

In a ductile material a crack subjected to a subcritical applied load may respond by emitting dislocations via thermal activation. Computer simulations show the activation energy to be strongly dependent on the applied stress. To understand this result we use conformal mapping techniques to analyze the interaction of a straight screw dislocation with a parallel crack in a strip geometry. The energy barrier for dislocation escape from the crack tip is calculated explicitly and it is found to be a sensitive function of the applied stress, in qualitative agreement with simulation results. Scaling properties of the activation energy are also determined. This analytical result permits us to formulate hypotheses regarding the factors controlling the observed strain rate. To test such hypotheses we finally calculate the strain rate as a function of temperature and applied stress and compare our results with observations.   

XRay Scattering in a Deformed Crystal by a Phase Field Method

We demonstrate the use of a phase field method for dislocated crystals, developed by one of us, for computing the scattering of Xrays. The model addresses deformation on a single slip plane by dislocations of a single burgers vector interacting with a set of point obstacles. The obstacles are introduced in two modes; one randomly on the slip plane, and the second in straight ``walls.'' The obstacles simulate blocking interactions by dislocations on different slip planes, and the ``walls'' represent the intersection of a secondary slip plane with the primary plane being simulated. In the small angle case, the scattering source is the local dilatation induced by the dislocations on the slip plane, and in the Bragg case, the scattering source is the change in local lattice constant. The small angle results show scattering with oscillations attributable to the width of the ``walls.'' In the Bragg case, the Laue spots are broadened by the dislocations, and the results directly confirm the picture of dipolar wall scattering introduced many years ago by H. Mughrabi.   

Atomistic Dislocation Dynamics in Phase Field Crystals: Long Time Scale Properties

The fundamental dislocation processes of glide, climb, and annihilation are studied on diffusive time scales within the framework of a continuum field theory, the Phase Field Crystals (PFC) model. Glide and climb are examined for single edge dislocations subjected to shear and compressive strain, respectively, in a two dimensional hexagonal lattice. It is shown that the natural features of these processes are reproduced without any explicit consideration of elasticity theory or ad hoc construction of microscopic Peierls potentials. Particular attention is paid to the Peierls barrier for dislocation glide/climb and the ensuing dynamic behavior as functions of strain rate, temperature, and dislocation density. It is shown that the dynamics are accurately described by simple viscous motion equations for an overdamped point mass, where the dislocation mobility is the only adjustable parameter. The critical distance for the annihilation of two edge dislocations as a function of separation angle is also presented.   

Studies of the Dislocation Glass

We report the large scale simulations of 2D dislocation systems with overdamped dynamics. 40,000-1,000,000 dislocations were studied with a combination of coarse graining, Fast Fourier Transform and stochastic methods. Both glide and climb processes were considered, as well as the local rotation of crystal axes. Simulations were performed at zero and finite temperatures, with and without dislocation annihilation. When climb processes were included, the system exhibited the formation of dislocation cells/patterns even in equilibrium, without the application of shear. This is in close correspondence with recent experiments on GaAs by P. Rudolph et al. (2005). The distribution function of cell sizes can exhibit a fractal dimension. At long times the system shows glassy dynamics. In particular, aging was observed through the waiting time dependence of the correlations and the effective diffusion. In certain parameter ranges the formation of cells leads to an initial exponential decay of correlations. This is followed by the growth of cells, generating a power law temporal decay in the long time domain. Data for both time domains and for all waiting times can be collapsed onto a single master curve when a t/tw scaling is applied.   

Nonlinear acoustic effects from dislocation-based hysteretic kinking solids under stresses

We argue that proposed recently mechanism explaining inelastic hysteresis in non-linear elastic systems indeed can be explained by means of formation of dislocation-based incipient kink bands (IKB). Using acoustic waves we investigated possible dislocation related mechanisms responsible for nonlinear dynamic response of IKB solids. In this work, for the first time we observed IKB formation and reversibility directly using acoustic coupling technique (ACT) measuring ultrasonic waves attenuation as a function of stress and acoustic emission signatures during compression test of nanolaminated layered ternary carbide (MAX phases) samples. We confirm here that the dynamic behavior of these non-linear elastic systems is due to the interaction of dislocations with the stress waves.   

Simulations of Nano-indentation and Shear Banding in Amorphous Solids

Molecular dynamics simulations of a number of amorphous systems reveal the structural changes that accompany plastic localization. We have simulated both two-dimensional and three-dimensional systems in nanoindentation\footnote{Y. Shi and M.L. Falk, ``Structural transformation and localization during simulated nanoindentation of a non-crystalline metal film,'' Applied Physics Letters, Vol. 86, pp. 011914 (2005).}, uniaxial tension\footnote{Y. Shi and M.L. Falk, ``Strain localization and percolation of stable structure in amorphous solids,'' Physical Review Letters, Vol. 95, pp. 095502 (2005).} and compression in plane strain\footnote{Y. Shi and M.L. Falk, ``Does metallic glass have a backbone? The role of percolating short range order in strength and failure,'' Scripta Materialia, Vol. 54, pp. 381 (2005).}. The degree of strain localization depends sensitively on the quench rate during sample preparation, with localization only arising in more gradually quenched samples. Careful analysis of the strain rate dependence of the localization allows us to extrapolate to the low strain rate limit. This analysis reveals a transition from localized flow to homogeneous flow at a critical value of the potential energy per atom prior to testing. This transition occurs in both two- and three- dimensional systems. The transition appears to be associated with the k-core percolation of short range order (SRO) in the two-dimensional system$^{2}$. We have used a generalization of the Frank-Kasper criterion to identify SRO in the three-dimensional systems. Only in certain systems does this method predict a percolation transition corresponding to the transition in mechanical behavior. We discuss the non-uniqueness of this measure of SRO, and consider whether a more rigorous definition could be derived which applies to systems far from the hard-sphere limit.   

A statistical model of plastic deformation in disordered media

Plastic deformation at the macroscopic scale is assumed to stem from series of successive localized plastic events. A random elastic limit is associated to each site of a discrete mesh. Using a quasi-static driving, one site at a time undergoes plastic shear. The local plastic threshold is then renewed. The localized slip induces long range elastic interactions of quadrupolar symmetry. These additional internal stresses are then used to determine the next weakest site. The model gives rise to a macroscopic plastic flow, corresponding to a genuine depinning transition. We obtain an asymptotic macroscopic yield stress. The transient regime can be associated to a hardening phenomenon of pure statistical origin. Beyond the average plastic behavior we observe stress fluctuations following a universal distribution (only dependent on the system size $L$). Shear deformation presents at all scales spatial and temporal fluctuations of universal character. We observe shear band-like structures which persist only during a finite time $\tau \propto L^z$ and which present a clear anisotropic character with a system size dependent width $w\propto L^\zeta$.   

Scaling laws in fracture of metallic glasses

Brittle metallic glasses themselves can be seen as a model system to study the mechanical properties of metallic based glassy materials. We report a brittle Mg-based bulk metallic glass which approaches the ideal brittle behavior. However, a dimple-like structure is observed at the fracture surface by high resolution scanning electron microscopy, indicating some type of `ductile' fracture mechanism in this very brittle glass. We also show a clear scaling correlation between the fracture toughness and plastic process zone size for various glasses. The results indicate that the fracture in brittle metallic glassy materials might also proceed through the local softening mechanism but at different length scales. The full text of this work has been published under the title \textit{Fracture of Brittle Metallic Glasses: Brittleness or Plasticity} by the authors in Physical Review Letters 94, 125510 (2005).   

Mapping Elasticity at the Nanoscale

In the last few years Atomic Force Acoustic Microscopy has been developed to investigate the elastic response of materials at the nanoscale $^{[1],[2]}$. We have extended this technique to the real-time mapping of nanomechanical properties of material surfaces. This mapping allows us to investigate the local variation of elastic properties with nanometer resolution and to reduce the uncertainties that arise from single measurements. Quantitative measurements are acquired by first performing an accurate calibration of the elastic properties of the Atomic Force Microscope’s probes with respect to single crystal reference materials. A wide variety of surfaces with different mechanical properties have been investigated to illustrate the applicability of this technique. \\ $^{[1]}$ U. Rabe \emph{et al.}, Surf. Interface Anal. $\bf{33} $, 65 (2002)\\ $^{[2]}$ D.C. Hurley \emph{et al.}, J. Appl. Phys. $\bf{94}$, 2347 (2003)   

Session P35: Nanostructure Fabrication, Quantum Point Contacts, and Single Electron Transistors

Atomic-Scale Modeling of Shape Stability-Regimes and Stacking in InAs/GaAs Quantum Dot Nanostructures

From a thermodynamic point of view, quantum dot (QD) growth is governed by the balance between energy gain due to strain relief and energy cost due to formation of QD side facets and edges. Both contributions are accounted for by an interatomic potential of the Abell-Tersoff type that we developed recently. We relax realistic InAs/GaAs QD nanostructures using this interatomic potential and compare the resulting total energies. To investigate the experimentally observed shape sequence of `hut'-like QD's dominated by \{317\} facets and `dome'-like QD's dominated by \{101\} facets, we compare the energy of a homogenous InAs film and differently sized InAs QD's with either shapes. We identify three regimes: For coverages below about 1.9 monolayers InAs the film is most stable, followed by small `hut'-like QD's and larger `dome'-like QD's. This is in line with the experimentally deduced critical coverage for the 2D to 3D growth transition, and the shapes of small and larger QD's. We can also explain the growth correlation in QD stacks: Our calculated potential-energy surfaces of free-standing QD's in different lateral positions above overgrown QD's show an energy gain of about 20 meV per In atom for the experimentally observed vertical QD alignment.   

Raman Spectroscopy of InAs/GaAs Quantum Dots Patterned by Nano-indentation

Patterns of InAs/GaAs quantum dots (QDs) grown by the combination of nanoindentation technique and molecular beam epitaxy were studied. The resulting QDs tend to preferentially nucleate on indented areas rather than other regions. We studied the strain on the indentations, regions surrounding the indents, and non-indented areas. The QD LO mode for the patterned areas shifted by 8 cm-1 when compared to the non-patterned area. The biaxial strain in the indented areas producing this shift is four times larger than that in non-indented areas, explaining the QD preference within these areas. This larger strain suggests that QDs on the indentations can be formed by depositing a smaller InAs amount than that required to form QDs on non-indented areas, thus obtaining QDs only on the pattern.   

Self-Assembled Unstrained InGaAs Quantum Dashes

We describe a technique for MBE-based fabrication of unstrained quantum dashes with Al$_{x}$In$_{y}$Ga$_{1-x-y}$As alloys lattice-matched to InP substrates. Templates for lattice-matched quantum dash growth are obtained by combining molecular beam epitaxy with \textit{in situ} etching by arsenic bromide. A seed layer of self-assembled InAs quantum dashes is converted into nanotrench templates through overgrowth followed by strain-enhanced etching. We have explored limitations on the accessible range of alloy compositions imposed by the etch process and found that strain-induced etching is limited to compounds with low Al content. Nanotrench templates can be filled with lattice-matched alloys of varied compositions to define barriers and quantum wires that could lead to optoelectronic devices in a spectral range around 1.5 $\mu$m. Here we also present Atomic Force Microscopy and Photoluminescence data obtained from self assembled unstrained In$_{0.53}$Ga$_{0.47}$As Quantum Dashes.   

Microanalysis of quantum dots with type II band alignments

We will discuss the structural characterization of a system consisting of undoped self-assembled InSb quantum dots having a type II band alignment with the surrounding In$_{0.53}$Ga$_{0.47}$As matrix. This differs from systems using conventional type-I quantum dots that must be doped and that rely on intersubband transitions for infrared photoresponse. Type II dots grown in a superlattice structure combine the advantages of quantum dots (3-dimensional confinement) with the tunability and photovoltaic operation of the type II superlattice. We grew a high surface density of InSb quantum dots with a narrow distribution of sizes and shapes and free of dislocations within the body of the dots. The dots are relaxed due to an array of misfit dislocations confined at the basal dot/matrix interface. This makes burying the dots with InGaAs not feasible without generating dislocations due to the large dot/matrix lattice mismatch. We are experimenting with strain-compensating or graded strain overlayers to lower the lattice mismatch.   

Feasibility Study of Directed Self-Assembly of Semiconductor Quantum Dots

Strain mismatched semiconductors are used to form Self-Assembled Quantum Dots (SAQDs). An important step in developing SAQD technology is to control randomness and disorder in SAQD arrays. There is usually both spatial and size disorder. Here, it is proposed to use spatially varying heating as a method of to direct self-assembly and create more ordered SAQD arrays or to control placement of single dots or dot clusters. The feasibility of this approach is demonstrated using a 2D computational model of Ge dots grown in Si based on finite element analysis of surface diffusion and linear elasticity.   

Controlling the self-assembly of Ge quantum dots grown by pulsed laser deposition

Growth dynamics and morphology of self-assembled Ge quantum dots (QD) on Si(100)-(2x1) by nanosecond pulsed laser deposition are studied by in situ reflection high-energy electron diffraction (RHEED) and post deposition atomic force microscopy (AFM). The effects of the laser fluence and substrate temperature on the QD formation are investigated. The QD density increased dramatically (from 3$\times $10$^{7}$ cm$^{-2}$ to 6.3$\times $10$^{8}$ cm$^{-2})$, while the average lateral size decreased (from 362 nm to 107 nm) when the laser fluence was increased from 23 J/cm$^{2 }$to 70 J/cm$^{2}$. Their shape also changed from large huts, observed at 23 J/cm$^{2}$, to domes observed at the highest fluence. At 150$^{\circ}$ C, misaligned QDs formed resulting in diffused RHEED pattern. At 400$^{\circ}$ C and 500$^{\circ}$ C, transmission RHEED patterns were observed indicating the growth of oriented hut and dome QDs. Around 600$^{\circ}$ C, the QDs were formed on top of textured surfaces.   

Defect engineering in periodic gradient-index optical thin films

For thin film deposition with obliquely incident vapour flux, ballistic shadowing limits growth to nucleation sites, forming a porous columnar microstructure. Combined with advanced substrate rotation in a technique known as glancing angle deposition (GLAD), precisely controlled nanoscale architectures are formed. \textit{In situ} variation of the angle of incidence provides dynamic control of the resulting film porosity, allowing the design of continuously varying periodic refractive index profiles to produce thin film interference filters. Intentional nanostructural defects can be introduced, such as uniaxial and biaxial constant index layers or index profile discontinuities, creating defect modes in the filter optical stopbands. Structural and optical characterizations of these periodic structures were performed, with the goal of understanding the relationship between the spectral properties of the film and the engineered nanostructure, demonstrating the high degree of control obtainable over the resulting filter properties using the GLAD process.   

Strontium titanate transformation to highly conductive nanolayers

Developing fabrication methods for electronically active nanostructures is an important challenge of modern science and technology. Fabrication efforts for crystalline materials have been focused on state-of-the-art epitaxial growth techniques. These techniques are based on deposition of precisely controlled combinations of various materials on a heated substrate. We report a method that does not require deposition and transforms a nanoscale layer of a complex crystalline compound into a new material using low energy Ion Beam Preferential Etching (IBPE). We demonstrate this method by transforming a widely used insulator model system, SrTiO3, into a transparent conductor. Most significantly, the resistivity decreases with decreasing temperature as 2.5 power of T and eventually falls below that of room temperature copper. These transport measurements imply a crystal quality in the conduction channel comparable to that obtained with the highest quality growth techniques. The universality of low energy IBPE implies wide potential applicability to fabrication of other nanolayers. David W. Reagor, Vladimir.Y. Butko, Nature Materials, v.4, 593, August 2005.   

Dependency of quantum pumping on transmission mode and dot size

We have used e-beam lithography to fabricate sub-micron metal gates on a two dimensional electron gas with mean free path on the order of several micrometers. Negative biases were applied to the metal gates to confine electrons in a small area ($\sim \mu $m$^{2})$ forming a so-called quantum dot. Two quantum point contacts (QPCs) served the entrance and exit of electrons in the dot are located in line. Quantum charge pumping phenomena of the open dot in the absence of an external bias was observed using two independent ac voltages with the same frequency, 1$\sim $80MHz, but a phase difference between them. Similar pumping results were reported by Marcus et al. earlier. However, due to the differences in the geometrical arrangements, the behaviors are somehow different including that our pumping current is one order more magnitude bigger and does not increase linearly with frequency for the entire measuring range. Moreover, we found that the pumping current seems increase with decreasing transmission mode numbers of the two QPCs. When the mode number goes to zero and the open dot transforms to closed dot, the pumping current vanishes. The results and measurements of the dependences of quantum charge pumping on transmission mode and dot size will be presented and discussed.   

Local Density of States of a Quantum Point Contact Near Pinchoff

Over the last decade, there has been great interest in how electrons flow through a quantum point contact (QPC) as it is just opened up, before a fully transmitting 1D conduction channel is available. Remarkably, there does not seem to be a smooth transition from tunneling to ballistic transport. Instead, a shoulder appears in the conductance versus channel width, at a conductance of roughly 0.7 times that of an open spin-degenerate channel. Experiments have built a consensus that this so-called ``0.7 structure'' is related to electron spin and electron-electron interaction, but the detailed description remains controversial. To study this system, we have made devices where one of the two QPC gates is actually a tunnel barrier to a third lead, fabricated on a GaAs/AlGaAs heterostructure. With this third lead, it is possible to probe the density of states in the QPC channel from the side as the QPC opens from pinchoff through the first channel. We acknowledge support from the ONR Young Investigator Program, Award No. N00014-01-1-0569 and a Research Corporation Research Innovation Award, No. RI1260.   

Single electron transistors in GaN/AlGaN heterostructures.

We study transport properties of two single-electron transistors (SETs) in a GaN/AlGaN heterostructure. The first SET accidentally formed in a quantum point contact near pinchoff. Its small size produces large energy scales: a charging energy of 7.5 meV, and well-resolved excited states. The second, intentionally-fabricated SET is much larger. Hundreds of consecutive, uniformly-spaced Coulomb oscillations yield a charging energy of 0.85 meV. Excited states are not resolvable in Coulomb diamonds, and Coulomb blockade peak height remains constant with increasing temperature, indicating that transport is through multiple quantum levels even at the 450 mK base electron temperature of our measurements.   

Current Biased Real Time Charge Detection in a Single Electron Transistor

Measurements of charge fluctuations in an AlGaAs/GaAs single electron transistor (SET) are presented. The SET consists of a lateral quantum dot created by confining a two-dimensional electron gas using nanometer-size surface electrodes. The charge on the quantum dot is detected by changes in conductance of a nearby quantum point contact. We discuss noise and bandwidth characteristics of our charge detection method, which uses commercially available voltage amplifiers. Our real time charge detection capabilities are used to investigate charge dynamics on the SET in a magnetic field parallel to the two-dimensional electron gas. This work is supported by the U.S. Army Research Office under Contract No. W911NF-05-1-0062, by the National Science Foundation under Grant No. DMR-0353209, and in part by the NSEC Program of the National Science Foundation under Award No. PHY-0117795.   

Real Time Electron Hopping Phenomena in a Single-Electron Transistor

Utilizing a current-biased quantum-point-contact charge sensor, we observe electrons hopping on and off a AlGaAs/GaAs single-electron transistor (SET) in real time. An electron tunnels between the extended states in the leads and the lowest-energy state localized in the lateral quantum dot created by nanometer-size surface electrodes. We observe changes in the tunneling rates, caused by the spin splitting in a magnetic field B applied parallel to the 2DEG. We have also observed single-electron photo-ionization of the SET by application of microwave radiation. This work is supported by the ARO (W911NF-05-1-0062), the NSF (DMR-0353209) and in part by the NSEC Program of the NSF (PHY-0117795).   

Shot noise and strong feedback effects in nanoelectromechanical systems

Quantum nanoelectromechanical systems have attracted much attention recently, offering potential for applications as well as insight into fundamental physics. Using a quantum noise approach, we study theoretically a nanomechanical oscillator coupled to a superconducting single-electron transistor (SSET). Incoherent Cooper pair tunneling processes in the SSET can lead to a negative damping instability, where the oscillator's amplitude increases as it absorbs energy from the SSET \footnote{Clerk, Bennett, NJP {\bf 7}, 238 (2005).}. Here, we focus on the current noise of the SSET in the negative damping regime, in which the growing amplitude of the oscillator becomes large enough that the motion of the oscillator and the dynamics of the SSET depend strongly on each other. We describe the inherent non-linearity of this regime using effective, energy-dependent damping and temperature, and discuss characteristic timescales for dynamics in the system. The current noise is of particular interest because it can be directly observed, and current experiments are probing this regime \footnote{K. Schwab {\it et al.} (in preparation).}.   

Finite size effects in the decay of metastable states in one-dimensional resonant tunneling structures

We study the current switching in a double-barrier resonant tunneling structure in the regime where the current-voltage characteristic exhibits intrinsic bistability, so that in a certain range of bias two different steady states of current are possible. Near the upper boundary $V_{th}$ of the bistable region the upper current state is metastable, and because of the shot noise it eventually decays to the stable lower current state. We find the time of this switching process in strip-shaped devices, with the width small compared to the length. The mean switching time $\tau$ increases exponentially as the bias $V$ is tuned inside the bistable region from its boundary value $V_{th}$. The one-dimensional geometry of the problem enables us to obtain analytically exact expressions for the exponential factor and to calculate the prefactor of $\tau$ for an arbitrary length of the strip. Furthermore, we evaluate the mean time of switching in ring-shaped devices, with the widths small compared to their diameters.   

Session P36: Focus Session: Plasmon Resonances in Nanostructures

Photoluminescence from a gold nanotip as an example of tabletop Unruh-Hawking radiation.

Conversion of zero-point quantum fluctuations into real thermal photons which may occur in a curved space-time is the main mechanism behind the Hawking radiation and the Unruh effect [1]. Up to date no experimental verification of these effects and the related dynamical Casimir effect has been reported. Here we argue that the recently observed infrared photoluminescence from a gold nanotip, which is mediated by surface plasmons (SP) propagating over a curved metal tip surface [2], constitutes an example of such zero-point to real photon conversion. Since SP wavelength may be very short, a surface plasmon wave packet propagating along a curved metal surface with radius of curvature $R\sim1$ micrometer may be considered as a classical particle (this would correspond to the ray optics approximation). The centripetal acceleration of such particle may be as large as $a\sim c^2/R\sim 10^{22}g$. According to ref. [1], such particle perceives vacuum as a bath of thermal radiation with temperature $T=\hbar a/2\pi kc\sim 1000K$. Nonlinear optical mixing of SPs with the thermal quanta from this bath looks like infrared photoluminescence in the laboratory reference frame. This work was supported in part by NSF grants ECS-0304046, CCF- 0508213 and ECS-0508275. References [1] W.G. Unruh, Phys.Rev.D 14, 870 (1976). [2] M.R. Beversluis, A. Bouhelier, and L. Novotny, Phys.Rev.B 68, 115433 (2003).   

Surface plasmon dielectric waveguides

We demonstrate that surface plasmon polaritons can be guided by nanometer scale dielectric waveguides on top of a gold film. In a test experiment plasmons were coupled to a curved 3 micrometer radius dielectric stripe, which was 200 nm wide and 138 nm thick using a parabolic surface coupler. This experiment demonstrates that using surface plasmon polaritons the scale of optoelectronic devices based on dielectric waveguides can be shrunk by at least an order of magnitude.   

Synthesis and Optical Properties of Star-shaped Gold Nanoparticles

Here we describe the synthesis, structure, and optical properties of ca. 100 nm star-shaped gold nanoparticles. Seed mediated, surfactant directed synthesis yields nanoparticle solutions sufficiently monodisperse that extinction spectra reveal plasmon bands representative of their structure. Single particle spectroscopy measurements demonstrate that these nanoparticles exhibit multispectral, multidirectional polarized scattering. Through correlated structural characterization by electron microscopy, each scattering component can be assigned to the different points on the star-shaped structure. The plasmon resonances were also found to be extremely sensitive to the local dielectric encironment, yielding sensitivities as high as 1.41 eV photon energy shift per refractive index unit. These properties suggest that the star-shaped gold nanoparticles may be highly valuable for certain biosensing and microscopic imaging paradigms.   

Nanorice: a new hybrid nanostructure

The plasmon hybridization method [1] is applied to nanorice, a new metallic nanostructure which combines the properties of two popular tunable plasmonic nanoparticle geometries: nanorods and nanoshells. The particle consists of a prolate spheroidal dielectric core and a thin metallic shell, bearing a remarkable resemblance to a rice grain. The nanorice particle shows far greater geometric tunability of the optical resonance, larger local field intensity enhancements and far greater sensitivity as a surface plasmon resonance (SPR) nanosensor than any previously reported dielectric-metal nanostructure. The tunability of the nanorice particle arises from the interaction of primitive plasmons associated with the inner and outer surfaces of the shell. The results from plasmon hybridization are compared to FDTD simulations. \newline \newline [1] E. Prodan and P. Nordlander, J. Chem. Phys. 120(2004)5444-5454   

Photonic Crystal Effects in Surface Enhanced Raman Scattering from Nanocluster/Nanoshell Arrays

We study the local optical properties of one-dimensional solid nanosphere dimer arrays with large array spacings, using the generalized Mie theory. We have obtained a large Raman cross section enhancement with magnitude of $10^{11}$ purely by electromagnetic effects, which is higher if compared with that of an isolated nanosphere dimer and in the literature. A coupled dipole approximation is used to understand this enhancement and the plasmon resonance shift relative to the isolated dimer. We have also studied the nanoshell dimer array and found even higher enhancement with magnitude of $10^{13}$. Our studies show that the nanoshell arrays with proper spacings have clear advantages in single molecule surface enhanced Raman spectroscopy (SMSERS).   

Strongly anisotropic optical composites

We study the macroscopic electromagnetic properties of nano-structured meta-materials formed by plasmonic nanowires embedded in a dielectric host. We show that nanowires have a significant effect on the effective dielectric constant of the system even in the case when their concentration is below 15{\%}. The effect of dielectric properties of nanowires as well as the effect of inclusion concentration, shape and local configuration disorder on effective dielectric constant is explored via numerical simulations. Further, we develop an analytical description of the effective dielectric properties of nanowire composites and study the limit of its validity. We demonstrate that it is possible to use plasmonic nanowire composites to construct strongly anisotropic low-loss optical materials. Proposed applications include polarizers, reflectors, high-energy-density nano-waveguides, and the recently discovered non-magnetic low-loss left-handed media.   

Metal Nanoparticle Enhanced Fluorescence -- Role of Particle Plasmon Resonance

We report on a systematic investigation of the enhancement of fluorescence by proximity to Ag nanoparticles whose size, shape and spacing are varied systematically using electron beam lithography. Our measurements indicate that enhancement of both absorption and radiative decay takes place. We compare our observations with expectactions based upon coupling to particle plasmons.   

Role of Cylindrical Surface Plasmons in Enhanced Optical Transmission

We investigate the role of cylindrical surface plasmons in enhancing the optical transmission from nanoarrays of dielectric coaxial cylinders embedded in a metal film. Finite difference time domain (FDTD) simulations identify transmission peaks at long wavelengths as being associated with the fields produced by the individual coaxial cylinders, and these peaks move out to increasingly long wavelengths as the dielectric ring becomes narrower. An analysis of cylindrical surface plasmon dispersion relations show that these peaks are due to resonances from surface plasmons propagating on the cylindrical metal-dielectric interfaces whose wave functions increasingly overlap as the ring narrows. The counterintuitive behavior of the wavelength of the peak is a direct consequence of the negative dielectric constant of the metal film and would not occur for a perfectly conducting or dielectric film. This resonant surface plasmon mechanism closely accounts for the dependence of the position of the simulated transmission peaks on ring geometry and the length of the coaxial cylinders.   

Plasmonic properties of non-concentric nanoshells

The plasmon hybridization method [1]is applied to nanoeggs, i.e., nanoshells with a non-concentric (offset) core. In contrast to concentric nanoshells, the particle exhibits a multitude of dipole active plasmon resonances. These resonances are formed by hybridization of the multipolar plasmon resonances associated with the inner and outer surfaces of the metallic shell. The reduced symmetry introduced by the offset of the core causes a significant admixture of dipolar components in all plasmon modes. The hybridization is shown to depend strongly on the asymmetry of the particle. The results compare very well with results from FDTD simulations. The non-concentric nanoshell particles are shown to provide large electric field enhancements on open-ended surfaces. \newline \newline [1] E. Prodan and P. Nordlander, J. Chem. Phys. 120(2004)5444-5454   

Plasmonic properties of the metallic nanosphere/thin wire system.

The plasmon hybridization method [1] is applied to a metallic nanosphere positioned near an infinitely long metallic wire. The plasmon resonances of the sphere are found to be shifted and to depend on the polarization of the incident light. In the limit of a thin wire, a virtual state consisting of propagating low energy wire plasmons is induced. The state is similar in nature to the virtual thin film state recently predicted and observed for a nanosphere near a thin metallic film [2]. \newline \newline [1] E. Prodan and P. Nordlander, J. Chem. Phys. 120(2004)5444-5454. \newline [2] F. Le, N. Z. Lwin, J.M. Steele, M. Kall, N.J. Halas, and P. Nordlander, Nano Lett. 5(2005)2009-2013.   

Optical Response of Metal Nanoparticle Chains

The excitation of surface plasmon on metal nanoparticles is interesting to many researchers because of its variety of applications. By arranging nanoparticles in different ways, many interesting properties can be observed. For metal nanoparticle chains, there is a red (blue) shift on the plasmon resonant frequency for longitudinal (transverse) excitation. We present the results on this splitting of plasmon resonant frequency for Ag nanoparticle chains with diameters around 10nm, calculated by the multiple scattering theory (MST) and the ways to understand the results using simple models. MST calculations are performed on the extinction of finite silver nanosphere chains embedded in glass matrix. The transmission and reflection of an infinite 2D arrays of silver nanospheres are also calculated to understand the interaction between nanoparticle chains. The results are in agreement with recent experiments. The splitting of plasmon-resonance modes associated with different polarizations of the incident light is further understood by employing simple models. Results on the effect of order and disorder in nanoparticle chains are also presented.   

Emergence of collective plasmon excitation in a confined one-dimensional electron gas

We present a theoretical study of the electronic excitation in a confined one-dimensional electron gas~[1], which is utilized to model atomic chains created in recent experiments. The length dependence of the excitation spectra is obtained from the linear response theory within the random phase approximation and time-dependent density functional theory. As the length of the chains increases, the dipole excitation spectra shows a transition from electron-hole pair excitations to collective plasmon excitation. The trend of the length-dependent plasmon resonance is predicted, and the nature of the plasmon resonance is also elaborated. \\ \\ \noindent [1] Shiwu Gao and Zhe Yuan, Phys. Rev. B 72, 121406(R) (2005).   

Electron energy-loss spectroscopy study of surface plasmons in Au nano-particles.

We have studied the surface plasmon excitations of $\sim $10 nm Au nano-particles with various shapes (such as sphere, rod, and triangle) by electron energy-loss spectroscopy (EELS) using a 0.2 nm electron probe in a scanning transmission electron microscope (STEM). EELS spectra were investigated as a function of distance from the probe to the nanoparticle surface, i.e., the impact parameter, and four surface plasmon peaks at 10, 15, 24, and 33 eV could be identified in the loss spectra in the 10 to 40 eV range where the dielectric constant of Au is positive (and still less than one). These high energy surface excitations are anomalous and in sharp contrast to the well-known surface plasmon of Au at 2.4 eV in the visible spectral range where the dielectric constant is negative. Spectral imaging studies also conclusively show that these high energy surface excitations are indeed localized at the surface of the nano-particles.   

Plasmons in the metallic nanoparticle-film system as a tunable impurity problem

We show that the plasmon resonances of a metallic nanoparticle interacting with a metallic film is an electromagnetic analog of the spinless Anderson-Fano model [1].~ The three characteristic regimes of this model are realized here, where the energy of the nanoparticle plasmon resonance lies above, within, or below the energy band of surface plasmon states. The latter regime is experimentally observed and identified. Our approach [1] is generalized to describe a nanoshell on a metallic film and to account for the screening effects caused dielectric backgrounds. These three interaction regimes are controlled by film thickness and the aspect ratio of the nanoshell. The results are compared with Finite-Difference Time-Domain (FDTD) simulations using realistic dielectric functions. \newline \newline [1] ~F. Le, N. Z. Lwin, J.M. Steele, M. Kall, N.J. Halas, and P. Nordlander, Nano Lett. 5(2005)2009-2013.   

\textbf{Surface plasmon interference spectroscopy of metal films}

Circular nanoslits manufactured by focused ion beam in silver films were used to excite surface plasmon polaritons and to generate plasmon interference patterns. Changes of the plasmon interference periods at changing the excitation wavelength were imaged by a near-field scanning optical microscope and scaled by the known nanostructure dimensions allowing precise plasmon wavelength measurements. The plasmon dispersion curves for our film thickness were calculated in different approximations and a proper fitting function for the experimental data was chosen. This allowed to retrieve the frequency dependence of the dielectric function of our silver film, which is different from usually cited Johnson-Christy and Palik data but falls in the range of values reported in literature. The results of fitting indicate to the important role of losses, which can not be neglected in definition of the real part of the dielectric constant even in the Drude region. Our technique is a useful tool for the local characterization of the dielectric function sensitive to the structure of metal films potential for photonics applications.   

Session P38: Superconductivity-Optical Spectroscopy of Cuprates

Optical Study of Optimally Doped and Overdoped YBCO

Thin films of the optimally-doped and overdoped high temperature superconductor (YBa$_{2}$Cu$_{3}$O$_{7-\delta }$ and Y$_{0.7}$Ca$_{0.3}$Ba$_{2}$Cu$_{3}$O$_{7-\delta }$ with T$_{c }$= 90 K and 79 K, respectively) have been investigated by optical spectroscopy in the ab-plane. In the normal state, with increasing the carrier concentration in the CuO$_{2}$ planes, spectral weight is lost in the high-frequency charge-transfer band and transferred to lower frequencies. With increased doping, the free-carrier (Drude-like component) plasma frequency increases, consistent with a charge density increase. However, the superfluid density decreases in this regime (overdoped region) and a substantial normal-fluid component still exists in the low frequency part of the optical conductivity well below $T_{c}$.   

The optical conductivity of Ortho II YBa$_2$Cu$_3$O$_{6.5}$.

The a-axis optical properties of the ortho II phase of YBCO (every other chain filled, $T_c$= 59 K) were derived from reflectance data over a wide frequency and temperature range. Above 200 K the spectra are dominated by a broad background of scattering that extends to 1 eV. Below 200 K, in the normal state, a shoulder in the reflectance signals the onset of scattering at 400 cm$^{-1}$. Below the superconducting transition temperature the superconducting condensate appears. Its spectral weight is consistent, to within experimental error, with the FGT sum rule and with independent measurements of Gd ESR. We also compare our data with magnetic neutron scattering on samples from the same source that show a strong resonance at 31 meV. Extrapolating the optical conductivity to zero frequency yields the dc resistivity of Ortho II which is in good agreement with four-probe measurements. We find that the scattering rates can be modeled as the combined effect of the neutron resonance, a bosonic background and a density of states with a pseudogap.   

Optical conductivity of Bi$_2$Sr$_2$CuO$_6$ in the optimally and overdoped regimes

Bi$_2$Sr$_2$CuO$_6$ has a much lower critical temperature ($T_c^{Max} \sim 20$ K) than other single layer cuprate superconductors, making this compound very useful to investigate normal state properties of cuprates. In addition, it is possible to oxygenate this material to a relatively high degree of overdoping ($T_c < 5$ K). We measured, from room temperature to 5 K, the optical conductivity of the same Bi$_2$Sr$_2$CuO$_6$ film thermally treated to be in the optimally and overdoped regimes. The in-plane resistivity in the optimally doped regime depends lineraly on the temperature. It develops a positive curvature in the overdoped regime as expected from its more Fermi liquid like behavior. Surprisingly, despite its linear resistivity, the frequency dependent scattering rate in the optimally doped sample has a quadratic behavior. We will discuss our results in terms of the possible scenarios for the normal state conductivity and infer its effects on the superconducting state.   

Doping and temperature dependent optical properties of Bi$_2$Sr$_2$CaCu$_2$O$_{8+\delta}$

We report on the ab-plane reflectance of underdoped (UD), optimally doped (OPT), and overdoped (OD) Bi$_{2}$Sr$_{2}$CaCu$_{2}$O$_{8+{\delta}}$ (Bi-2212) samples [$T_{c}$ = 69 K(UD), 96 K (OPT), 82 K (OD), 80 K (OD: annealed from the OPT sample), 65 K (OD) and 60 K (OD)]. We analyzed the measured reflectance data and previous data of two underdoped Bi-2212 samples ($T_c$ = 67 K (UD) and 82 K (UD)) to extract the doping dependent optical constants. Bi-2212 is one of the most important cuprate systems widely studied by ARPES and tunnelling. We also calculate the doping dependent dc resistivity from extrapolation of the optical conductivity and the doping dependent superfluid density and the optical self- energy by using the extended Drude model. With these quantities in hand, we will discuss some current issues: the kinetic energy change at $T_c$, the role of the magnetic resonance mode, and a possible quantum critical point.   

Polarization dependence of charge-transfer excitations in La$_2$CuO$_4$

We have carried out an extensive resonant inelastic x-ray scattering (RIXS) study of La$_2$CuO$_4$ at the Cu K-edge. Multiple charge-transfer excitations have been identified using the incident photon energy dependence of the cross section and studied carefully with polarizations E//c and E //ab. An analysis of the incident photon energy dependence, the polarization dependence, as well as the K-edge absorption spectra, indicates that the RIXS spectra reveal rich physics about the K-edge absorption process and momentum-dependent charge-transfer excitations in cuprates.   

Evidence of an anisotropic charge-excitation gap in stripe-ordered \boldmath La$_{2-x}$Ba$_x$CuO$_4$ with $x=1/8$ \unboldmath

The {\it ab}-plane optical properties of a cleaved single crystal of La$_{2-x}$Ba$_x$CuO$_4$ for $x=1/8$ ($T_c \simeq 2.4$~K) have been measured over a wide frequency and temperature range. The low-frequency conductivity is Drude-like and shows a metallic response with decreasing temperature. However, below $\simeq 60$~K, corresponding to the onset of charge-stripe order, there is a rapid loss of spectral weight below about 40~meV, resulting in a major reduction in the number of free carriers. This suggests a partial gapping of the Fermi surface. Surprisingly, the sample is still metallic and becomes a superconductor at low temperature. This material is a striking example of how charge and spin stripe order, metallic behavior and superconductivity can coexist.   

Anisotropic Drude response in Mg(B$_{1-x}$C$_{x})_{2}$

There exists an unsolved issue on MgB$_{2}$ that a plasma edge estimated from optical spectra so far is conspicuously inconsistent with a band calculation. It is mysterious since experimental band dispersions by ARPES are marvelously coincident with the band calculation. We report on a- and c-axis optical responses in Mg(B$_{1-x}$C$_{x})_{2 }$ using small single crystals and a FTIR spectrometer combined with an optical microscope. It was verified that the Drude responses show the anisotropy between a- and c-axis. The observed plasma frequencies $\omega_{p'}$ are considerably small compared to a theoretical value ($\sim $7eV), whereas the bare ones $\omega _p $, estimated from a sum rule of optical conductivity $\int {d\omega \sigma (\omega )=(\pi /2)\omega _p^2 } =(\pi /2)\varepsilon _\infty \omega _p^{'2} $, are fairly coincident with it. We also discuss an effect of carbon-substitution on carrier concentration and scattering rate in a multiband system. This work was supported by the New Energy and Industrial TechnologyDevelopment Organization(NEDO) through ISTEC as the Collaborative Research and Development of Fundamental Technologies for Superconductivity Applications.   

Mode coupling effect in infrared spectra of Tl$_2$Ba$_2$Ca$_{n-1}$Cu$_n$O$_x$ (n=1, 2, 3)

We performed in-plane optical reflectance measurements on Tl$_2$Ba$_2$Ca$_{n-1}$Cu$_n$O$_x$ with n=1, 2, 3. The single crystals were successfully grown by flux method with maximum T$_c$=90 K, 109 K, and 119 K for n=1, 2, 3, respectively. For all three phases near optimal doping, the reflectance roughly has a linear-frequency dependence in the normal state, but displays a pronounced knee structure followed by a dip-like feature at higher frequency below T$_c$. Such characteristic features were commonly ascribed to the coupling of electrons with a bosonic mode. Very remarkably, we found that the energy levels of those features scale with T$_{c}$ for the three phases. The results suggest against a phonon origin for the bosonic mode. We also investigated the spectral evolution with doping for Tl-2201 crystals in the overdoped side. We found that the mode coupling effect weakens with doping and disappears in the heavily overdoped sample. Meanwhile, the optical scattering rate evolves from a linear- $\omega$ dependence to a shape with upward curvature in the normal state. Both the temperature and frequency dependence of the scattering rate can be described by a power law relation. Compared with ARPES results, we suggest that the overall decrease of the scattering rate may mainly originate from the increase of the quasiparticle life time near the ($\pi$,0) region in the Fermi surface.   

A Complete Charge Model For The High T$_{c}$ Superconductivity

Focusing on the recent transport, far-infrared (far-IR), and neutron scattering data, we establish a model for the high temperature superconductivity (HTS) built on the two-dimensional (2D) square electronic lattices in the CuO$_{2}$ planes and their three-dimensional ordering. We demonstrate that our model captures all the essential physics that bring about a coherent picture of the HTS and offers the key to the mechanism for the HTS.   

Temperature Dependent of the Optical Spectral Weight in Correlated Metal Nd$_{1-x}$TiO$_{3 }$(x=0.095)

We investigated the infrared reflectance of Nd$_{1-x}$TiO$_{3}$, which is a hole-doped transition-metal-oxide system. In the metallic sample with $x$=0.095 (hole concentration=3$x$=0.285), the partial optical spectral weight, $\omega (\Omega ,T)=\int_0^\Omega {\sigma _1 (\omega ,T)d\omega } $, turns out to be a linear function of $T^{2 }$at different cutoff frequencies. Recent optical studies of LSCO [1] and BSCCO [2] also found that the optical spectral weight varies quadratically with temperature, i.e. $\omega (\Omega ,T)\simeq \omega _0 -B(\Omega )T^2$, in both superconductors and nonsuperconducting metals. The coefficient $B(\Omega )$ was considered as a ``thermal response'' of the carriers. In our study, for Nd$_{1-x}$TiO$_{3 }(x$=0.095), $B(\Omega )$ exhibits distinct features which we compare to both cuprates and conventional metals. [1] M. Ortolani \textit{et al}., Phys. Rev. Lett. \textbf{94}, 067002 (2005). [2] H. J. A. Molegraaf \textit{et al}., Science \textbf{295}, 2239 (2002).   

Phonon Shifts at the Superconducting Transition in \textit{$\kappa $}-(BEDT-TTF)$_{2}$-Cu(NCS)$_{2}$

We have measured the phonon spectra above and below the superconducting transition in \textit{$\kappa $}-(BEDT-TTF)$_{2}$-Cu(NCS)$_{2}$. This organic superconductor has a transition temperature near 10K. These measurements were carried out using inelastic x-ray scattering at sector 3ID at the Advanced Photon Source. The incident x-ray energy was 21.657 keV and the resolution of the spectrometer was about 2.0 meV. We have observed significant phonon shifts at several energies in these spectra at several points in the Brillouin zone. Here we discuss the behavior of the phonons and possible reasons for the shifts.   

Session P39: Focus Session: Superconductivity-Theory and Computation (Mainly First Principles)

Effect of Spin Fluctuations on Electron-Phonon Superconductivity

Most of our intuition regarding conventional superconductivity is derived from the McMillan formula. What are often believed to be generic properties of the Eliashberg equations, in reality hold only in the regime where the McMillan formula is applicable. I will show how common beliefs, such as a monotonic relation between the reduced gap and the coupling constant, or an exponential behavior of the NMR relaxation with the gap as the activation parameter, fail for more complicated spectral functions. Most interestingly, I will demonstrate that our conventional wisdom totally fails us in the rapidly developing field of superconductors near a magnetic instability. In particular, I will derive (and test against numerical solutions of the Eliashberg equations) an analogue of the McMillan formula, fully accounting for the pair-breaking effect of spin fluctuations, and will show that these {\it increase} the phonon isotope effect, sometimes by as much as a factor of two. This is counterintuitive and opposite to the effect of the high-energy Coulomb interactions (the so-called Coulomb pseudopotential). I will also discuss the possibility of observing this effect in specific materials, such as MgCNi$_3$. This work has been done in collaboration with Oleg Dolgov (MPI Stuttgart) and Alexander Golubov (U. Twente).   

Properties from spin-phonon coupling in high-T$_C$ superconductors: HgBa$_2$CuO$_4$ and La$_{(2-x)}$Sr$_x$CuO$_4$

The mechanism of spin-phonon coupling (SPC) in high-T$_C$ copper oxides is explored from band calculations on LSCO and HBCO systems. The LMTO band calculations, based on the local density approximation, are made for cells containing frozen phonon displacements and/or spin waves within the CuO plane. The virtual crystal approximation is used for studies of hole doped systems. The main result is that phonons are favorable for spin waves and vice-versa, and that pseudogaps appear naturally in the band structures of striped materials with strong SPC. The wave length of the spin-phonon modulation is related to doping, and the mutual enhancement of SPC is strongest when the non-doped system is close to an anti-ferro magnetic ground state. The calculated band results are used for modelling of different properties, such as isotope effects, phonon softening, shear dependences and T-variations. The results are discussed and compared with experiment. It is speculated that perpendicular SPC, with different behavior along x- and y-directions, can produce double gap structures. A moderate correction to LDA, which stabilizes the AFM state for the undoped material, will enhance the coupling constant for spin fluctuations $\lambda_{sf}$ for doped cases. These results suggest that properties of high-T$_C$ superconductors should depend both on phonons and magnetic fluctuations.   

Superconductivity of Li, Al and K under pressure

C Franchini$^{\S}$, N. N. Lathiotakis$^{\dag}$, A. Floris$^{\dag\S}$, A. Sanna$^{\S}$, M. A. L. Marques$^{\dag}$, M.~L{\"u}ders$^{\ddag}$, S. Massidda$^{\S}$, E. K. U. Gross$^{\dag}$, A. Continenza$^{*}$.\\ $^{*}$ CASTI - INFM and Dip. Fis., Univ. di L'Aquila, I-67010 Coppito (L'Aquila) Italy; $^{\S}$ SLACS INFM and Dip. Fis., Univ. di Cagliari, I-09042 Monserrato (Ca), Italy; $^{\dag}$ Institut f{\"u}r Theoretische Physik, Freie Universit{\"a}t Berlin, Arnimallee 14, D-14195 Berlin, Germany; $^{\ddag}$ Daresbury Lab., Warrington WA4 4AD, United Kingdom.\\ Extreme pressure strongly affects the superconducting properties of ``simple'' metals, like Li, K and Al. Using the new ab-initio method of density functional theory of the superconducting state, we report investigations on the superconducting properties of dense Li, K and Al. Our results show an unprecedented agreement with experiments, assess the predictive power of the method over a wide range of densities and electron-phonon couplings, and provide predictions for K, where no experiments exist so far. For fcc K we predict a superconducting phase transition at 18 GPa, with a maximum critical temperature of about 2 K at 23 GPa, the pressure where the crossover between the fcc and the K $III$ structure experimentally occurs. We studied the effect of pressure on the electronic and vibrational properties of alkali, showing a progressive phonon softening near the K point of the Brillouin zone and a concomitant enhancement of the electron-phonon coupling constant $\lambda$.   

Superconductivity of Alkali Metals under High Pressure

We calculated the superconductivity properties of alkali metals under high pressure using the results of band theory and the rigid-muffin theory of Gaspari and Gyorffy. Our results suggest that at high pressures Lithium, Potassium, Rubidium and Cesium would be superconductors with transition temperatures approaching 10-20 K. Our calculations also show that Sodium would not be a superconductor under high pressure even if compressed to less than half of its equilibrium volume. We found that the compression of the lattice strengthens the electron-phonon coupling through a delicately balanced increase of both the electronic and phononic components of this coupling. This increase of the electron-phonon coupling in Li is due to an enhancement of the s-p channel of the interaction, while in the heavier elements the p-d channel is the dominant component.   

Superconductivity and Lattice Instability in Compressed Lithium from Fermi Surface Hot Spots

Lithium, a simple metal not superconducting above 5mK at ambient pressure, becomes a 20 K superconductor at 50 GPa. This high T$_c$ is shown to arise from critical (formally divergent) electron-phonon coupling to the transverse phonon branch along intersections of Kohn anomaly surfaces with the Fermi surface. First principles linear response calculations of the phonon spectrum and spectral function $\alpha^2 F(\omega)$ reveal (harmonic) instability already at 25 GPa. Our results imply that the fcc phase is anharmonically stabilized in the 25-38 GPa range.   

DFT Study of the Single-Band Layered TMO LiNbO$_2$

We establish using first principles methods that LiNbO$_2$ is a realization of a triangular lattice ``single band'' system. The bandwidth (less than 2 eV) suggests the interesting possibility of correlation effects that should be kept in mind. We present a tight-binding model for the valence band of LiNbO$_2$, composed primarily of Nb d$_{z^2}$ states, finding that intralayer second neighbor hopping $t_2 ~ 100$ meV is dominant over the significantly smaller first neighbor interactions $t_1 ~ 70$ meV. The nearest neighbor coupling is strongly modified by oxygen displacements, and the electron-phonon coupling may provide the coupling mechanism for superconductivity in Li-deficient samples ($T_c \approx 5$K). We will present the Nb-centered Wannier function, which provides insight into this unusual electronic structure. Calculations of the Born effective charges for the metal ions are also found to have anisotropy that reflects the layered nature of the electronic bonding. Their deviation from formal charge values indicates important covalent character, which is also evident in the Wannier function.   

Electron-phonon Interaction in Graphite-Intercalation Compounds

After the discovery of superconductivity with a Tc of 11.5 K in Ca-intercalated graphite (CaC6 )[1], the interest in graphite-intercalation compounds has been revived. Different pairing mechanisms, based on excitons or electron-phonon interactions, have been put forward [2]. In this contribution we first analyze, using the NMTO[3] method, the electronic structure of CaC6. We then propose a simple model, based on pure graphite, to explain superconductivity in this class of compounds. Implications on the design of new materials with similar superconducting properties are also discussed. \newline [1] T.E. Weller {\em et al.}, Nature Physics {\bf 1}, 39 (2005). [2] G. Csany {\em et al.}, Nature Physics {\bf 1}, 42 (2005); I. I. Mazin, cond-mat/0504127; M. Calandra and F. Mauri, . [3] O.K. Andersen and T. Saha-Dasgupta, Phys. Rev. B 62, R16219 and O.K. Andersen, T. Saha-Dasgupta and S. Ezhov, Bull. Mat. Sci. 26, 19 (2003).   

Gap anisotropy in density functional theory of the superconducting state

The discovery of superconductivity in MgB$_{2}$ ($T_{c}=39.5$K), with the clear presence of two gaps, has renewed the interest not only in electron-phonon mediated superconductivity, but also on the problem of anisotropic superconductivity. Here we use the recently introduced density functional theory of the superconducting state, that allows calculations of material-specific properties without the use of any adjustable parameters. The method, extended to ${\bf k},{\bf k'}$ resolved matrix elements of phonon-mediated and coulomb interactions, allows for a fully ${\bf k}$-resolved gap structure. Within this approach, we obtain the critical temperature and the two gaps of MgB$_{2}$ in good agreement with experiment. We will report on the existence of two different gaps also in Pb, and show that this is related to the different strength of the electron-phonon coupling associated with the two bands crossing the Fermi level. The calculated anisotropy is in good agreement with experiment. The same aproach is used for Nb3Sn, where recent experiments (Guritanu et al., Phys. Rev. B 70, 184526 (2004)) point to a possible two-gap behaviour. Our calculations show how our formalism is able to capture, in absence of any ad-hoc model, the features of multi-gap superconductors.   

Band structure trend in cuprates and correlation with T$c\ max$

Parameters in model Hamiltonians are derived from LDA band structures for cuprate high T$_c$ superconductors. The materials and structural dependences of these are discussed. The most essential material dependent parameter is the range of the intralayer hopping. Furthermore, the range of this hopping correlates with T$_c$, i.e. materials with larger hopping ranges have higher maximum T$_c$'s.   

First-Principles Construction of the Zhang-Rice singlet: Role of the apical oxygen in the mobility of the doped hole.

The Zhang-Rice singlet (ZRS) has been well accepted as the most relevant low-energy states in high $T_c$ cuprates. Based on a novel Wannier state analysis [1] of the LDA+$U$ electronic structure, a realistic ZRS is constructed from properly orthogonalized local Cu $d_{x^2-y^2}$ and symmetric combination of O-p states ($p^{(s)}$), leading to a realistic derivation of low-energy effective t-t'- t''-J Hamiltonian. Interestingly, symmetrized apical oxygen $p_z$ orbital with the Cu $d_{z^2}$ symmetry is found to be close to the ZRS in energy (~0.7 eV) and thus significantly facilitates the hopping to the second and third nearest neighbors. [1] W.-G. Yin, D. Volja, and W. Ku, cond-mat/0509075.   

Band structure and Fermi surface of Ba$_{2}$Ca$_{3}$Cu$_{4}$O$_{8}$F$_{2}$

Recently Y. Chen \textit{et} \textit{al}. have measured the Fermi Surface (FS) of the fluorinated four CuO$_{2}$ layer superconductor, Ba$_{2}$Ca$_{3}$Cu$_{4}$O$_{8}$F$_{2}$, using angular resolved photoemission spectroscopy (ARPES). Surprisingly, they found only two large ($\pi $,$\pi $) centered hole FS sheets, while four would have been expected. In order to investigate the reason for this we have performed first-principles electronic band structure calculations for this compound. As expected four antibonding copper-oxygen bands cross the Fermi level of which the two have dominantly orbital character on the two inner-layers and the other two have most orbital character on the outer-layers. The splitting between these bands is, however,much smaller than the splitting between the two measured FS sheets. The fluorine were claimed to replace all apical oxygens, however, by partly substituting apical oxygen as well as oxygen in the four CuO$_2$ layers by fluorine, good agrement with the experimental FS could be obtained.   

The role of the Fermi surface sampling in first-principles calculations of electron-phonon coupling

A quantitative understanding of the electron-phonon interaction is crucial to the understanding of conventional and possibly high-T$_{\rm c}$ superconductivity, as well as to the study of transport and spectroscopic (such as optical and photoemission) properties of bulk and nanoscale systems. Despite the enormous interest in calculating electron-phonon interaction from first principles, present methods carry severe practical limitations. We present here a comparative study of several existing methods for computing this quantity. We show that, independent of the approximation adopted, a common computational bottleneck is that the Fermi surface must be sampled with extremely high accuracy, leading to prohibitively expensive calculations for complex systems. We also reformulate the problem of evaluating phonon linewidths in terms of self-consistent linear response theory, and demonstrate our approach through application to magnesium diboride.   

Session P40: Focus Session: Pathways to Practical Quantum Computing I

Solid state technologies for quantum computers

Impressive progress is now being made in realizing the rudiments of quantum computers in solid state devices. I will discuss two of them. First, single electron quantum dots now offer a higly coherent spin state for use as a qubit. Decoherence effects, arising from hyperfine interactions and the spin-orbit interaction, are well on their way to being understood and controlled. Second, Josephson junction devices, in many forms, are showing promise as qubits. The dynamics of these electric circuits can be designed to exhibit a wide variety of quantum effects; good two-level systems can be produced by careful design, and careful schemes for decoupling from the environment. Coupling to harmonic modes offer a wide variety of ``quantum optic'' realizations in the microwave regime.   

CNOT logic for Josephson phase qubits

Josephson junctions have demonstrated enormous potential as qubits for scalable quantum computing architectures. Here we study the speed and fidelity of four controlled-NOT gate implementations designed for capacitively coupled phase qubits. One gate applies to qubits fixed permanently in resonance, two require varying the dc current bias, and the fourth applies to permanently detuned qubits. Realistic simulations suggest that these implementations can be demonstrated with good fidelity using existing superconducting circuits.   

Violation of Bell's Inequality using Josephson Phase Qubits

Recent improvements of the measurement visibility and coherence times in Josephson Phase Qubits have enabled first tests of two- qubit quantum gates and examination of quantum phenomena using these devices. Here, we present an experiment in which we attempt to violate Bell’s Inequality, which would be further proof that the system at hand behaves in a truly quantum mechanical way. The violation of Bell’s Inequality is the primary argument against the possible existence of a hidden- variable-theory as an alternative to quantum mechanics. This experiment illustrates the use of coherent control over capacitatively coupled qubits with always-on coupling, including the establishment of the system in eigenstates of the coupling, e.g. the singlet state. Single qubit rotations combined with a simultaneous, fast, high-visibility readout allow for state- tomography on the system.   

Experimental State Tomography using Superconducting Quantum Bits

The superconducting approach to building a scalable quantum computer has enjoyed tremendous successes in the past several years with coherence times now sufficiently long to implement quantum gates on a system with coupled qubits. In order to quantify the performance or fidelity of the gates, quantum state tomography is required. Successful state tomography relies on high measurement fidelities and the ability to perform arbitrary rotations in the transverse plane of the Bloch sphere. Here, we have made significant progress towards overcoming these challenges and present, for the first time, experimental data on single and two-qubit state tomography.   

Towards single shot read-out in circuit quantum electrodynamics (QED)

In recent experiments we have demonstrated the resonant coherent coupling of individual photons to a single qubit implemented as a Cooper pair box in a high quality superconducting cavity [1]. In the non-resonant case, the dispersive coupling between the qubit and the cavity field is used to perform quantum non-demolition (QND) measurements of the qubit state [2]. Using this read-out technique we have performed high visibility measurements of Rabi oscillations and Ramsey fringes [3]. Here we present a detailed experimental and theoretical analysis of the cavity response for continuous and pulsed measurements in a wide range of cavity drive amplitudes. We also discuss an optimal read-out strategy for qubits in a continuous QND measurement and aim at demonstrating single shot read-out in the circuit QED architecture [4].\\ \\ $[1]$ A. Wallraff et al. Nature (London) 431, 162 (2004)\\ $[2]$ D. I. Schuster et al. Phys. Rev. Lett. 94, 123602 (2005)\\ $[3]$ A. Wallraff et al. Phys. Rev. Lett. 95, 060501 (2005)\\ $[4]$ A. Blais et al. Phys. Rev. A 69, 062320 (2004)\\   

Coherent control in circuit QED

Superconducting charge qubits fabricated inside a transmission line resonator have been used to successfully demonstrate strong interaction of an artificial atom with a single photon [1]. This architecture has also been used to show high-visibility and long coherence time (T$_{1}\sim $7 $\mu $s, T$_{2}\sim $ 500 ns) Rabi oscillations [2] and in the detailed study of measurement-induced dephasing [3]. Here we will discuss protocols to realize one and two-qubit logical gates in circuit QED. These are based on resonant and off-resonant irradiation of the transmission line resonator. First experimental results towards the realization of these gates will be presented. Supported by NSA and ARDA under ARO Contract No. W911NF-05-1-0365 and the NSF under Grants No. ITR-0325580 and No. DMR-0342157. [1] A. Wallraff \textit{et al.}, Nature 431, 162 (2004). [2] A. Wallraff \textit{et al.}, Phys. Rev. Lett., 95, 060501 (2005). [3] D. Schuster \textit{et al.}, Phys. Rev. Lett., 94, 123602 (2005).   

Superconducting SET backaction on the Cooper-pair box

We report on measurements of the backaction of a superconducting single electron transistor (SSET) measuring a Cooper-pair box qubit. During the weak, continuous measurement made by the SSET, the charge noise acts on the Cooper-pair box. The quantum nature of that noise is able to dephase, relax and even excite the qubit. This noise depends strongly on the operating point of the SSET. We operate the SSET near the double Josephson quasiparticle (DJQP) feature, where the backaction of the SSET is well understood (A. Clerk, et al., Phys. Rev. Lett. 89, 176804 (2002)), and where there are no quasiparticle poisoning effects. Measurements of the relaxation time of the Cooper- pair box reveal the symmetric component of the quantum noise and measurements of the steady-state polarization reveal the anti-symmetric component. Both measurements vary as expected with SSET operating point and confirm this model of SSET backaction.   

Information Flow in the Readout of a Superconducting Quantum Bit

Quantum computation requires efficient and well controlled coupling between qubits. Superconducting qubits can be strongly coupled using passive electrical circuit elements, but one of the major remaining challenges is to eliminate uncontrolled coupling to parasitic degrees of freedom. I will present experimental results on charge qubits integrated with a novel readout device -- the Josephson bifurcation amplifier (JBA). New experiments using the improved readout fidelity and speed of the JBA quantify parasitic losses and shed light on their mechanism.   

Mach-Zehnder-type Interferometry in a Strongly Driven Persistent-Current Qubit

We have demonstrated Mach-Zehnder-type interferometry with a niobium superconducting persistent-current qubit. The qubit’s ground and first-excited states exhibit an anti-crossing. Driving the qubit with a large-amplitude harmonic excitation sweeps it through this anti-crossing two times per period. The induced Landau-Zener (LZ) transitions act as coherent beamsplitters, and the accumulated phase between LZ transitions varies with the driving amplitude. We have observed quantum interference fringes as a function of the driving amplitude for 1 to 20 photon excitations. We present and discuss these results.   

Double Quantum Dot Molecule Coupled with Single-Electron Transistors for Quantum Computation Applications

We describe the fabrication of a series-coupled double quantum dot (DQD) with side-coupled single-electron transistors (SETs). The DQD are intended to work as qubits, and the SETs perform the quantum spin measurements. The device was fabricated on a GaAs/AlGaAs heterostructure using a one-step, two-angle, evaporation of aluminum. Our design is compatible with modern semiconductor techniques, and if proven successful, can readily be scaled into larger integrated qubit systems with spin manipulation and measurement circuitry. Our preliminary experimental results indicate that both the QDs and SETs have single-electron tunneling behaviors with good reproducibility. We will report on progress towards the in-situ detection of the spin and charge of a single electron trapped in the semiconductor quantum dots.   

Charge fluctuation induced dephasing of exchange coupled spin qubits

Exchange coupled {\it spin} qubits in semiconductor nanostructures are shown to be vulnerable to dephasing caused by {\it charge noise} invariably present in the semiconductor environment. This decoherence of exchange gate by environmental charge fluctuations arises from the fundamental Coulombic nature of the Heisenberg coupling, and presents a serious challenge to the scalability of the widely studied exchange gate solid state spin quantum computer architectures. We explore the properties of the resulting exchange gate errors, and estimate dephasing times for coupled spin qubits in a wide range (from 1 nanosecond up to more than 1 microsecond) depending on the exchange coupling strength and its sensitivity to charge fluctuations in a particular nanostructure.   

Solid-state quantum teleportation between nanomechanical modes

We study a quantum teleportation scheme between two nanomechanical modes without local interaction. The nanomechanical modes are connected by and linearly coupled to the continuous variable modes of a superconducting circuit made of transmission line and Josephson junctions. The phase sensitive measurement during the teleportation can be conducted by a superconducting single electron transistor operated as an rf mixer. Using a Wigner function approach, we calculate the fidelity of transferring coherent state under finite temperature and non-unit detector efficiency. We show that a fidelity above the classical limit of $1/2$ can be achieved for a large range of parameters.   

Spin transport and quasi 2D architectures for donor-based quantum computing

The original Kane quantum computer architecture is based on a single line of $^{31}$P atoms spaced a few tens of nm apart in an isotopically pure $^{28}$Si lattice with electrodes above and between donor atoms. This architecture suffers from major technical issues including strong spatial oscillations in the nearest neighbour donor electron exchange coupling strengths at the scale of a single lattice site and an inability to limit the effect of a given electrode to its nearest donor or donor pair. Through the introduction of a new donor electron spin transport mechanism, a 2D donor electron spin quantum computer architecture is proposed. This new architecture addresses the exchange coupling and cross-talk issues, as well as a host of other physical barriers to implementation.   

Session P41: Focus Session: Dielectric, Ferroelectric, and Piezoelectric Oxides III

Luminescence and Raman based real time imaging of ferroelectric domain walls

Novel methods for real-time imaging of ferroelectric domain walls open up new possibilities for advancing physical understanding of domain wall structure, properties, and defect interactions. Instead of destructive chemical etching and subsequent optical or electron microscopy, domain walls are imaged nondestructively in real-time by photoluminescence microscopy using dilute doping by rare earth ions as “designer defects” whose luminescence is affected by the domain walls. Uisng a combination of high spatial and spectral site-selectvity in laser confocal and near field opptical microscopy, domain structure changes on a 100 nm length scale can be observed with a temporal resolution of 5ms, as demonstrated in LiNbO$_3$ and LiTaO$_3$. Imaging using Raman spectroscopy (that does not require rare earth doping) will also be described. These new imaging methods reveal that domain wall widths and structures are velocity-dependent, and they provide active, real time feedback needed for precise laser-writing of ferroelectric domain patterns.   

Study of Ferroelectric Domains in a Phase Separated Multiferroic Mixture by Variable Temperature Electrostatic Force Microscopy

We present a variable temperature Electrostatic Force Microscopy (VTEFM) study on a mixed multiferroic crystal. The sample was synthesized by the floating zone method. It was cut and polished with the surface normal to the growth direction. The chemical phase separation is clearly seen by polarized optical microscopy. The transition temperature is about 25K and 900K for the two different phases. The VTEFM images taken at 77 K reveal the ferroelectric domains, with typical sizes in the order of micrometers.   

A new fundamental limit of ferroelectric devices and its domain dynamics in ultrathin ferroelectric BaTiO$_{3}$ films

Phenomena in ultrathin ferroelectric (FE) films, such as the critical thickness and the domain structures, have attracted much interest for a few years. We fabricated fully-strained SrRuO$_{3}$/BaTiO$_{3} $/SrRuO$_{3}$ capacitors, whose BaTiO$_{3}$ layer thicknesses were between 5 and 30 nm, using the laser molecular beam epitaxy. We found that rapid decay of net polarization occurs due to large depolarization field [1]. Using the Monte-Carlo simulations, this decay can be explained by the domain formation dynamics, governed by the domain nucleation process. We found a universal relation between the decay exponent and nucleation energy barrier, regardless of film thickness and temperature. This universal relation will provide a fundamental thickness limit for practical FE devices, set by net polarization decay. [1] D. J. Kim et al. Phys. Rev. Lett, in press.   

Raman Studies of Ferrolectric Domain Walls in Lithium Tantalate and Niobate

The local structure of ferroelectric domain walls and its dependence on intrinsic defects and dopants is of great interest both from a basic science and a application point of view. For instance, in the ferroelectrics LiNbO$_3$ and LiTaO$_3$ that are widely used in nonlinear and electro-optical devices, the stability, shape, switching fields and smallest achievable domain size are determined by the defect concentration. Using confocal Raman spectroscopy we investigated the perturbation of the phonon modes across a domain wall as a function of sample stoichiometry (i.e.: the number of intrinsic defects. For all samples, we find that in the spectral vicinity of the E(TO$_8$) and E(TO$_9$) the Raman intensity is enhanced in the domain wall region. In order to elucidate the origin of this enhancement, we investigate the directional dispersion of the observed change and perform measurements under variation of pump and probe light polarization and sample orientation. On the basis of these results, we will discuss structural models of the domain wall.   

Time-Resolved Observations of Soft Phonon Modes in Strained BaTiO$_{3}$/Si Heterostructures

Ferroelectric thin films such as BaTiO$_{3}$, grown on Si(100) substrates, have enormous potential for applications ranging from non-volatile random access memories to electro-optic gates for quantum information processing architectures. Optical techniques provide powerful means for obtaining time-resolved information about the ferroelectric soft mode in these materials. Using a two-color pump-probe arrangement, we observe THz-frequency soft modes in strained BaTiO$_{3}$/Si heterostructures grown by oxide-molecular beam epitaxy.   

Distribution function of random electric fields in disordered ferroelectrics thin films

We present the calculation of first moment $E_0$ and variance $\Delta E$ of distribution function of random fields in a ferroelectric of finite size. This defines completely the distribution function in gaussian limit. Specific calculations have been performed for the case of slab-shaped ferroelectric thin film. We have shown that $E_0$ and $\Delta E$ can be expressed through the integrals from first and second degree of Green's function of such confined geometry ferroelectric in $k$ - space. To obtain the Green's function, we solve the differential equation minimizing Landau free energy of a ferroelectric with respect to the boundary conditions on its surfaces. We show, that the distribution function of random fields in the finite-size ferroelectric differs from that of the unbounded bulk material. For example, both $E_0$ and $\Delta E$ depends on film thickness $L$. Knowledge of this distribution function permits to calculate the observable physical properties of ferroelectric thin films made from ferroelectric relaxors. Our method of calculation of $E_0(L)$ and $\Delta E(L)$ can be easily generalized for ferroelectric of arbitrary shape.   

Interface Induced Ferroelectric Phase Transformation in SrTiO3.

The transport properties across bicrystal interfaces in SrTiO3 are quantified with 4-pt probe, Hall measurements, scanning impedance microscopy and scanning tunneling microscopy. The properties are related to the structure determined by transmission electron microscopy, energy loss spectroscopy and first principles calculations. An anomaly in the temperature dependence of the transport properties arises from the charge trapped at the interface, which induces dipole ordering adjacent to the boundary. This represents the first observation of interface induced ferroelectricity in SrTiO3.   

Nanoscale structural dynamics of ferroelectric thin films

The emerging capability to visualize dynamical phenomena at small scales in both distance and time simultaneously has important implications in understanding ferroelectric materials. We have used time-resolved synchrotron x-ray microdiffraction to probe polarization switching and piezoelectric response at the sub-nanosecond time scale and the sub-micrometer spatial scale in lead zirconium titanate thin films. Based on time resolved maps of the polarization and piezoelectric distortion, the polarization switching domain wall velocity can be measured directly. The magnitude of this velocity and its scaling with electric fields suggest that significant improvements in switching speed can be made in optimized thin film structures.   

External and internal magnetic-field effects on ferroelectricity in orthorhombic rare-earth manganites

We report the dielectric and magnetic properties of the perovskite (Eu,Y)MnO$_3$ crystal {\it without} the presence of the $4f$ magnetic moments of the rare earth ions. The subject compound, (Eu,Y)MnO$_3$, was controlled the average ionic radius of the $A$ site so that it was the same as that of TbMnO$_3$ in which the intriguing magnetoelectric effect has been recently discovered. The (Eu,Y)MnO$_3$ crystal was found to have two distinct ferroelectric phases with polarization along the $a$ ($P_a$, $T$$\le$23K) and $c$ ($P_c$, 23K$\le$$T$$\le$25K) axes in the orthorhombic $Pbnm$ setting in a zero magnetic field. In addition, we have demonstrated a magnetic-field-induced switching between these ferroelectric phases: $P_a$ changed to $P_c$ by the application of magnetic fields parallel to the $a$ axis ($H_a$). In analogy to the case of $P_c$ in TbMnO$_3$, this result is possibly interpreted as follows. In the case of (Eu,Y)MnO$_3$, Mn $3d$ spins rotate in the $ab$ plane and $P_a$ would emerge in a zero field. In the $H_a$, the field will force the spins to rotate in the $bc$ plane, in which $P_c$ would be stabilized. Magnetization measurements supported this interpretation: We confirmed the change of the spin rotation axis of the helix from the $c$ axis to the $a$ axis induced by application of the $H_a$ because there is no $4f$ moments acting as an internal magnetic field and interacting with the $3d$ spins. Results obtained with other rare-earth manganites such as (Gd,Tb)MnO$_3$ and (Eu,Ho)MnO$_3$ will be presented.   

Dielectric and vibrational properties of crystalline and amorphous high-k lanthanum aluminate

High-k oxides are the focus of intense research for their applications in MOS and FLASH devices. A material currently in focus is LaAlO$_3$, with a dielectric constant of ~23-24 in the crystal phase, and similar values of around 20-22 in the amorphous phase (although values as low as 15 have also been reported). We have studied LaAlO$_3$ in both phases [1] to identify possible reasons for this apparent conservation of the dielectric properties upon amorphization. Amorphous samples were generated by melt-and-quench using a combination of pair potentials and ab initio dynamics. The linear response density-functional perturbation-theory approach was used to study dynamical response and phonons. We indeed find a large dielectric constant (~24) in the amorphous: the rationale is that the expected reduction of the anomalous effective charges is compensated by the appearance of new low-frequency (weakly) IR-active modes, whose character is a mixture of La translations (IR in the crystal) and Al-O octahedra rotations (Raman in the crystal). A similar behavior is expected in any rare earth aluminate exhibiting a similar perovskite-related structure (e.g. scandates). \\ 1) P. Delugas, V. Fiorentini, and A. Filippetti, Phys. Rev. B 71, 134302 (2005), and to be published.   

Composition dependence of the diffuse scattering in relaxor (1-$x$)Pb(Mg$_{1/3}$Nb$_{2/3}$)O$_{3}$-$x$PbTiO$_{3}$ ($0\leq x\leq0.40$)

We have studied composition dependence of diffuse scattering in the relaxor system (1-$x$)Pb(Mg$_{1/3}$Nb$_{2/3}$)O$_{3}$-$x$PbTiO$_{3}$ (PMN-$x$PT) with $x=0$, 10, 20, 30, and 40\% by neutron diffraction. The addition of ferroelectric PbTiO$_{3}$ (PT) modifies the ``butterfly" and ``ellipsoidal" diffuse scattering patterns observed in pure PMN ($x=0$), which are associated with the presence of randomly oriented, polar nanoregions (PNR). The spatial correlation length $\xi$ derived from the width of the diffuse scattering increases from 12.6~\AA\, for PMN ($x=0$) to 350~\AA\, for PMN-20\%PT, corresponding to an enlargement of the PNR. The integrated diffuse scattering intensity, which is proportional to $\chi''$, grows and reaches a maximum at $x = 20$\%. Beyond $x =$30\%\,PT, a concentration very close to the morphotropic phase boundary (MPB), no diffuse scattering is observed below $T_{C}$, and well-defined critical behavior is observed. By contrast, the diffuse scattering for $x \leq 20$\% persists to low temperatures, where the system retains an average cubic structure ($T_{C}=0$). We can simulate the wave vector dependence of the diffuse scattering by assuming that it arises from the condensation of a soft transverse-optic (TO) phonon.   

Multiresonance, multifrequency spectroscopy of rare-earth and transition ions in ferroelectrics.

The usual scheme of many methods for material investigation includes an emitter of electromagnetic waves and detectors for the wave registration. Typical output of one-frequency methods is an image with a space resolution of about the used wavelength. The methods are very successful for the study of lattice structures, their transformations at phase transitions, domains etc. Another approach uses sweeping of the frequency or another external parameter: electric or magnetic field, pressure etc. Typical result of a measurement is a spectrum or a dependence of measured characteristic on the sweeping parameter. The spectra do not contain direct evidence about the space structure of a lattice or defects. However, they contain very important information about the energetic characteristics of interactions of lattice ions, intrinsic and extrinsic defects. We present results of multifrequency research of defects in oxide crystals involving optical spectroscopy, microwave and radiofrequency spectroscopy: electron paramagnetic resonance, and electron nuclear double resonance.   

Femtosecond Spectroscopy of LuMnO$_{3}$

Hexagonal LuMnO$_{3}$ manganite is a ferroelectric and strongly frustrated antiferromagnetic crystal. Strong coupling between lattice, electronic, and magnetic degrees of freedom makes it a promising electronic material. We have used femtosecond pump-probe spectroscopy to study the interaction of electron excitations with lattice vibrations in real time. Optical excitation of a Mn $d_{(x^2-y^2),(xy)} \to d_{(3z^2-r^2)} $transition served as the primary excitation step. With both pump and probe beam polarization perpendicular to the c axis, the probe reflectivity shows a sharp drop due to saturation of the transition, recovering on a timescale of 1 ps. We also observed displacive excitation of a coherent optical phonon vibration at 3.6 THz, which is assigned to an A1 symmetry mode involving Lu ion motion along the c axis. This mode was excited in longitudinal (LO) and transverse mode (TO) geometries. While the LO-TO frequency splitting is small ($<$0.1 THz), a remarkable phase reversal of the reflectivity curve was observed. This is attributed to a large linear electro-optic effect (Pockels effect), induced by the THz electric field associated with the LO mode.   

Session P42: Focus Session: Planetary Materials III

Dissociation of CaIrO$_3$-type MgSiO$_3$ in the gas giants

CaIrO$_3$-type MgSiO$_3$ is the planet-forming silicate stable at pressures and temperatures (PTs) beyond those of Earth's core-mantle boundary. We have found using first principles quasiharmonic free energy computations that this mineral dissociates into MgO and SiO$_2$ at PTs expected to occur in the cores of the gas giants ($>\sim$10 Mbar, 10,000 K). This transformation should be important also for modeling the internal structure of two recently discovered terrestrial exoplanets: a dense Saturn orbiting HD149026b and a super Earth orbiting GJ876d. We propose a low pressure route experiment to confirm this dissociation.   

Thermal Electrons and Thermal Conductivity in Oxide Minerals at P and T Relevant to Terrestrial Exoplanets

The recent discovery of an extrasolar planet, with 7.5 times the mass of the Earth, has prompted investigation of a new range of parameter space, 3 times higher in temperature T and 10 times higher in pressure P than the Earth's mantle. We estimate thermal conductivity k(T) of minerals under these extreme conditions. The radiative portion of k(T) is large above the mid-lower post-perovskite mantle, where T reaches 5000-6000K. At T higher than 5000 K, free electron carriers are thermally activated with the population n(T) increasing as exp(-E*/2kT), where E* is the band gap energy of around 5 eV. Free carriers damp electromagnetic waves at frequencies below the plasma frequency, estimated to be close to 1 eV, shutting down radiative heat transport. Although thermal holes have low mobility, we find that thermal electrons are quite mobile, with small effective masses and weak scattering. Therefore, they become dominant carriers of heat. We predict electrical resistivity as low as 1000 micro-ohm cm.   

Elasticity of MgO Under Direct Pressure Measurement: Insights on Current Pressure Scales.

Recent high pressure studies indicated that the inaccuracy and inconsistency of the pressure scales used for pressure determination in different studies might be an importance source that gives rise to the apparent discrepancy in the derived phase equilibrium and physical properties for mantle minerals. In this study, P and S wave velocities and unit cell parameters (density) of MgO are measured simultaneously up to 11 GPa 1073K using combined ultrasonic interferometry and in-situ X-ray diffraction techniques, from which the elastic bulk and shear moduli as well as their and temperature pressure derivatives are obtained independent of pressure. These properties are subsequently used to calculate the primary pressures at the observed strains for comparison with those derived from previous proposed MgO pressure scales. Additionally, a comparison of the primary pressure obtained from MgO with those inferred from the enclosed internal pressure calibrant (NaCl) gives an opportunity to evaluate the Decker NaCl scale as well. Our results suggest that current pressure scales may bear larger uncertainties than originally claimed.   

A high PT scale based on density functional calculations of MgO

In situ crystallography based on diamond anvil cells have recently been extended to the multi-Mbar regime. Temperatures in these experiments have crossed the 2,000 K mark. Yet, current high PT standards of calibration produce too large uncertainties to the point of inhibiting clear conclusions regarding the importance of certain phenomena for planetary processes at these high PTs, e.g., the post-perovskite transition in Earth’s mantle. We propose a calibration based on thermal equations of state (EoS) of MgO obtained from LDA quasiharmonic (QHA) calculations. These EoSs agree very well with several calibrations at relatively low PTs. This gives further support to our predictions made within the range of validity of the QHA.   

Phase stability and elasticity of CaSiO$_3$ perovskite at high pressure and high temperature from Ab inito molecular dynamic calculations

We report the dynamics of the structure and elastic properties of CaSiO3 perovskite from \textit{ab initio }molecular dynamics (AIMD) calculations at high pressure (P up to 130 GPa) and high temperature (T up to 5000K). Our calculations indicate three separate stability fields: metrically orthorhombic, tetragonal and cubic, with the tetragonal phase dominating the pressure and temperature region between room temperature and 4000K. The cubic phase is not entirely stabilized even at temperatures of the Earth's lower mantle. Calculated X-ray diffraction patterns indicate small super- lattice reflections that could result from the octahedral rotations throughout the P-T region investigated. The calculated elastic constants and velocities are independent of temperature at constant volume. Referenced to room pressure and 2000K, we find: Gr\^{u}neisen parameter is $\gamma $(V) = $\gamma $0(V/V0)q with $\gamma $0 = 1.53 and q = 1.02(5), and the Anderson Gr\^{u}neisen parameter is given by ($\alpha $/ $\alpha $0) = (V/V0)$\delta $T in which $\alpha $0 = 2.89 x 10-5 K-1 and $\delta $T = 4.09(5). Using the third order Birch Murnaghan equation of state to fit our data, we have for ambient P and T, K$_{0}$ = 236.6(8) GPa, K$_{0}$' = 3.99(3), and V$_{0}$ = 729.0(6) {\AA}3.   

Computational study of the pressure behavior of post-perovskite phases

The recent discovery of the post-perovskite phase transition (CaIrO$_3$ structure) in MgSiO$_3$ has lead to theoretical and experimental investigations of silicates, germanates and oxides that could take this structure. We have employed density functional-theory to explore a series of new compounds with the post-perovskite structure under pressure. We analyze the effects of the Si substitution by tetravalent cations on the perovskite-to-post-perovskite transition and on the crystal structure of post-perovskite. Cations Ti$^{4+}$ and Zr$^{4+}$ prefer the post-perovskite structure. We also explore the sesquioxides Al$_2$O$_3$ and Rh$_2$O$_3$ and compare their structural evolution with the one of MgSiO$_3$. For Rh$_2$O$_3$ we observe an enhancement of the ionic character of the type II structure with pressure.   

Thermodynamics of Mg$_2$SiO$_4$ liquid from first principles molecular dynamics simulations

As the main medium through which planetary differentiation occurs, silicate liquids have a central role in the study of the Earth. We determine the structural and thermodynamic properties of Mg$_2$SiO$_4$ liquid using first principles molecular dynamics in the framework of density functional theory and the local density approximation (LDA). Calculations, performed in the canonical ensemble with a Nose thermostat, span a range of pressures (0 - 150 GPa) and temperatures (3000 - 6000 K). Simulations are performed over 3000 time steps (femto seconds), yielding the total energy and average pressure from which the equation of state, heat capacity and Gruneisen parameter are determined. Preliminary results show that, in addition to the increase in Si coordination with pressure, about one third of the O atoms are not bound to Si at low pressure, a fraction which vanishes with increased Si coordination.   

Dissociation of Ringwoodite investigated by first principles

The dissociation of Ringwoodite, Mg$_2$SiO$_4$ gamma-spinel, into MgO and MgSiO$_3$ perovskite is believed to be associated with the 660-km discontinuity in Earth's mantle. Details of this transition are important to clarify its effect on mantle convection: it is believed to inhibit flow across the ``660'' discontinuity. We have investigated the phase boundary using quasiharmonic free energy computations within the LDA and GGA. Once more the GGA transition pressure, P$_{tr}$, is higher and in much better agreement with the limited experiments available. The higher GGA P$_{tr}$ can be rationalized by close inspection of the relationship between GGA and LDA functional forms. Our predictions of density, bulk modulus, and bulk velocity jumps across the transition are consistent with seismic observations.   

Structure and freezing of MgSiO$_3$ liquid in Earth's interior

Silicate liquids are primary agents of mass and heat transport, yet little is known of their physical properties or structure over most of the mantle pressure regime. We have applied density functional theory within the local density approximation to the study of silicate liquids via Born-Oppenheimer first principles molecular dynamics. The simulations are performed in the NVT ensemble with a Nose thermostat. We find that over the pressure regime of Earth's mantle the mean Si-O coordination number increases nearly linearly with compression from four-fold to six-fold. The Gr\"uneisen parameter of the liquid increases markedly on compression, in contrast to the behavior of mantle crystalline phases, and in accord with expectations based on the pressure-induced change in structure of the liquid. The density contrast between liquid and crystal decreases nearly five-fold over the mantle pressure regime and is 4 \% at the core-mantle boundary. The melting curve, obtained via integration of the Claussius-Clapeyron equation yields a melting temperature of $5400 \pm 600$ K at the core mantle boundary. Our results support the notion of buoyantly stable silicate melts at the core-mantle boundary.   

A laboratory method for modeling synthesis of coesite in the earth's surface by combining local mechanical collision with shear stress and high static pressure.

A laboratory method of combining the high-energy mechanical ball milling and high static pressure has been suggested for modeling synthesis of coesite in the earth's surface. A window of milling time, a mechanical collision-induced intermediate phase of $\alpha $-quartz and its condition of easily crystallizing into coesite induced by high static pressure 3.0 GPa, 923 K, $<$ 1.0 min have been discovered. The condition has a much shorter synthesizing time and lower synthesizing critical pressure than that obtained before. The Raman spectrum for the coesite synthesized by the present method has the biggest number of peaks, and have covered over the information of those natural and synthesized coesite reported before. Here We clarify the implications of the coesite synthesized by this method in geo-science, and suggest another possible formation mechanism of coesite in the earth's surface, which is different from the hypothesis of plate subduction-exhumation in the earth that was based on the coesite formation condition of high static pressure in laboratory.   

Novel Perovskite Compounds Synthesized under Elevated High Pressure.

High pressure synthesis is very powerful to stabilize new compounds with perovskite-like structure, as it has been very well established in the studies of Earth mantle. This has been widely demonstrated in the research of transition metal oxides. Here we introduce some new transition metal compounds that have been recently synthesized under elevated high pressure.   

Structure of MgO(MgSiO$_{3})_{n}$ in Earth's Lower Mantle: ab initio calculations

Ruddlesden-Popper (RP) compounds are composed of alternating perovskite-type and rocksalt-type structural elements. MgSiO$_{3}$ and MgO are found as separate phases in Earth's lower mantle. Both structural elements occur also in the hypothetical RP-series MgO(MgSiO$_{3})_{n}$. It is interesting to explore the high pressure-high temperature stability of such RP-structures. Using the augmented plane wave implementation of Density Functional Theory we investigate the structural stability at lower mantle conditions of the member with $n$ = 1 e.g. Mg$_{2}$SiO$_{4}$. The goal of the present calculations is to test the stability of this Ruddlesden-Popper phase relative to $\gamma$-(Mg,Fe)$_{2}$SiO$_{4}$ and the assemblage MgSiO$_{3}$-perovskite + MgO magnesiow\"{u}stite. We will present our results of this study.   

Heat capacity measurements of sub-milligram quantities of mantle minerals

Knowledge of heat capacities and standard entropies of mantle minerals is necessary for thermodynamic modeling of high P-T equilibria. However many of these materials can only be prepared in milligram quantities in a multianvil apparatus or in microgram quantities in a diamond anvil cell. This eliminates traditional adiabatic calorimetry techniques for Cp measurements. Our microcalorimeters have been used to successfully measure thin films, multilayers, and magnetic single crystals. Using these ``calorimeters on a chip'', we are measuring the heat capacity of the Fe$_{2}$SiO$_{4}$ olivine and spinel polymorphs from 2 K to room temperature. This will provide a direct measurement of the entropy of the olivine-spinel transition and will uncover possible magnetic phase transitions at low temperature in the spinel phase. We would like to thank the DOE for funding this research.   

Carbon under extreme conditions: phase boundaries from first-principles theory

We present predictions of diamond and BC8 melting lines and their phase boundary in the solid phase, as obtained from first principles calculations. Maxima are found in both melting lines, with a triple point located at $\sim 850~\rm GPa$ and $\sim 7400~\rm K$. Our results show that hot, compressed diamond is a semiconductor which undergoes metalization upon melting. On the contrary, in the stability range of BC8, an insulator to metal transition is likely to occur in the solid phase. Close to the diamond/ and BC8/liquid boundaries, molten carbon is a low-coordinated metal retaining some covalent character in its bonding up to extreme pressures. Our data provide constraints to the carbon equation of state, which is of critical importance to devise models of, e.g., Neptune, Uranus and white dwarf stars, as well as of extra-solar carbon planets. This work was performed under the auspices of the U.S. Dept. of Energy at the University of California/Lawrence Livermore National Laboratory under contract no. W-7405-Eng-48.   

Synthesis of New Cubic C$_{3}$N$_{4}$ Phase under High Pressure and High Temperature

Synchrotron-based X-ray diffraction studies were carried out on a graphite-like C$_{3}$N$_{4}$ ($g$- C$_{3}$N$_{4})$ phase subjected to high pressures up to 38 GPa and high temperatures of up to 3000 K using the laser-heated diamond-anvil cell. Laser-heating the sample to 1800 K at pressure between 20 and 38 GPa, a new set of diffraction pattern appeared, showing positively that a high-pressure phase was formed. Upon decompression of the post-lasered sample to 1 atmospheric pressure, the X-ray diffraction peaks of high-pressure phase were replaced completely by a new pattern, thus demonstrating a new metastable phase was formed retrogressively. X-ray diffraction data on the recovered phase show that it is a new cubic phase that does not match to any high-pressure phases in C$_{3}$N$_{4}$ predicted theoretically.   

Session P43: Novel Phases in Quantum Gases

Stability of Bosonic atomic and molecular condensates near a Feshbach Resonance

Fermions near a Feshbach resonance exhibit a smooth crossover between a Bose-Einstein condensed state of molecules and a BCS superfluid of Cooper pairs. We study the analogous problem in Bosons, where there is a possibility of a phase transition between a molecular condensate (MC) and an atomic condensate (AC). We show that on the molecular side of the resonance at low densities, a MC-AC continuous transition is precluded by the AC state having a negative compressibility [cond-mat/0507460]. Instead, there is a mechanical collapse to a liquid-like state, analogous to a first order phase transition. We predict that sufficiently high densities (beyond those currently achieved in experiments) will push the system beyond its tricritical point and allow a continuous MC-AC phase transition.   

Phase diagram of Bose-Fermi mixtures in one-dimensional optical lattices

The ground state phase diagram of the one-dimensional Bose-Fermi Hubbard model is studied in the canonical ensemble using a quantum Monte Carlo method. We focus on the case where both species have half filling in order to maximize the pairing correlations between the bosons and the fermions. In case of equal hopping we distinguish between phase separation, a Luttinger liquid phase and a phase characterized by strong singlet pairing between the species. True long-range charge density waves exist with unequal hopping strengths.   

Luttinger's theorem and Phase Transitions in Bose-Fermi mixtures

A mixture of bosonic and fermionic atoms with a Feshbach resonance between the two can exhibit a range of phases as the energy of a fermionic bound state is varied. In each uniform phase a generalized statement of Luttinger's theorem can be made regarding the two Fermi surfaces, one associated with the atomic fermion and one with the bound-state molecule. The various phases can then be characterized by their different Luttinger constraints, which also depend on the presence or absence of a bosonic condensate. Interesting parallels can be drawn between this system and two others: the transition to the fractionalized Fermi liquid in Kondo lattice models, and fermion-pair condensation in the presence of mismatched Fermi surfaces.   

Spontaneous Symmetry Breaking and Defect Formation in a Quenched Ferromagnetic Spinor Bose-Einstein Condensate

We observe spontaneous symmetry breaking in a spinor Bose condensate of $^{87}$Rb that is quenched across a quantum phase transition to a ferromagnetic state. Using high spatial resolution maps of the vector magnetization of the condensate, we directly observe the spontaneous formation of inhomogeneous ferromagnetic regions separated by un-magnetized defects. The growth of these ferromagnetic regions are due to a dynamical instability, which determines their typical size and the time for their formation in accord with our observations.   

Thermodynamic properties of Bose-Hubbard model.

We perform Monte Carlo simulations of bosons in a three-dimensional optical lattice. We present accurate data for the ground state phase diagram and for the finite-temperature thermodynamic properties, including specific heat and entropy. Our data form a basis for an accurate thermometry of the system.   

Existence of Roton Excitations in Bose Einstein Condensates: Signature of Proximity to a Mott Insulating Phase

Within the last decade, artificially engineered Bose Einstein Condensation has been achieved in atomic systems. Bose Einstein Condensates are superfluids just like bosonic Helium is and all interacting bosonic fluids are expected to be at low enough temperatures. One difference between the two systems is that superfluid Helium exhibits roton excitations while Bose Einstein Condensates have never been observed to have such excitations. The reason for the roton minimum in Helium is its proximity to a solid phase. The roton minimum is a consequence of enhanced density fluctuations at the reciprocal lattice vector of the stillborn solid. Bose Einstein Condensates in atomic traps are not near a solid phase and therefore do not exhibit roton minimum. We conclude that if Bose Einstein Condensates in an optical lattice are tuned near a transition to a Mott insulating phase, a roton minimum will develop at a reciprocal lattice vector of the lattice. Equivalently, a peak in the structure factor will appear at such a wavevector. The smallness of the roton gap or the largeness of the structure factor peak are experimental signatures of the proximity to the Mott transition.   

Supersolid Bosons on Frustrated Optical Lattices

We consider an ultra-cold Bose gas on a triangular optical lattice subject to nearest neighbor repulsion, and determine the phase diagram using quantum Monte Carlo simulations. Already in the hard-core limit the system is found to exhibit an extended supersolid phase emerging from an order-by-disorder effect as a novel way of a quantum system to avoid classical frustration. We analyze the nature of the supersolid phase and its stability in competition with phase-separation, which we find to occurs in other regions of parameter space. Possible experimental realizations of our scenario and extensions to other lattice geometries are discussed as well as the connection to the physics of frustated quantum antiferromagnets.   

Frustrated two-dimensional XY models with cold atoms in optical lattices

We consider a system of cold bosonic atoms in a rotating optical lattice at finite temperature. We show that such system exhibits a non-trivial dependence of the condensation temperature and the superfluid order parameter on the vortex density due to commensuration effects of the vortex and optical lattices. We identify several vortex filling/lattice geometry combinations for which the vortex ordering pattern exhibits subtle order-by- disorder effects due to an interplay between multiple degeneracy of frustrated vortex configurations and thermal fluctuations.   

Winding Numbers in Rotating Bose Gases

The exact ground states of zero-temperature rotating Bose gases confined in quasi-two-dimensional harmonic traps are investigated numerically, for small numbers of alkali atoms. As the rotation frequency increases, the interacting Bose gas undergoes a series of transitions from one quantum Hall state to another. By tracking the change in ground state energy with an applied phase twist, we are able to calculate the winding (Chern) number characterizing the topological nature of the various bosonic quantum Hall states.   

Parafermionic states in rotating Bose-Einstein condensates

Rotating Bose-Einstein condensates in a trap are the place of a very rich physics. It has been predicted that, under appropriate conditions, they will behave like two dimensional electron systems in the fractional quantum Hall effect regime. In addition to the usual fractions, more exotic phases have also been predicted at filling factor $\nu=k/2$. These parafermionic states are described by the Read-Rezayi (RR) wave functions. We study how the system size and interaction act on the overlap between the true ground state and corresponding RR state. The quasihole excitations of the RR states are known to obey non-Abelian statistics. We numerically evaluate the degeneracy of these states and show it is in agreement with a formula given by E. Ardonne. We compute overlap between low-energy true eigenstates and quasihole ground states, and discuss in which cases such description is valid.   

Quantum magnetism with multicomponent polar molecules in an optical lattice

We consider dipolar molecules in an optical lattice prepared as a mixture of states with angular momentum $\ell=0$ and $\ell=1$. The $1/r^3$ interaction between molecules for this system is produced by exchanging a quantum of angular momentum between two molecules. We show that Mott states of such systems have a large variety of non-trivial spin orderings including SDW state at a wavevector that can be controlled by changing parameters of the system. As the Mott insulating phase is melted, we also show that an interesting winding in the phase of the order parameter can occur. Finally, we consider ways of detecting such phases experimentally.   

Biaxial nematic phase of two dimensional disordered rotor models and spin-one bosons in optical lattices

We show that the ground state of disordered rotor models with quadrupolar interactions can exhibit biaxial nematic ordering in the disorder-averaged sense. We present a mean-field analysis of the model and demonstrate that the biaxial phase is stable against small quantum fluctuations. We point out the possibility of experimental realization of such rotor models using ultracold spin-one Bose atoms in a spin-dependent and disordered optical lattice in the limit of a large number of atoms per site and also suggest an imaging experiment to detect the biaxial nematicity in such systems.   

Session P44: Organic Conductors

Interlayer magnetoresistance as a probe of the quantum coherence of electronic excitations in layered metals

Angle-dependent magnetoresistance oscillations (AMRO) have been used as a powerful tool to map out Fermi surfaces in layered metals, such as organic metals, strontium ruthenate, and an over-doped cuprate. We derive a general formula for AMRO in systems with anisotropic interlayer hopping, anistropic in-plane scattering and an anisotropic 2$d$ Fermi surface. We discuss the ability of AMRO to discriminate between coherent transport when there is a 3$d$ Fermi surface and weakly incoherent transport, where there is hopping between 2$d$ Fermi surfaces that are only defined in each layer. We illustrate these ideas by comparison with experimental measurements of AMRO in thallium cuprate [1]. \\ $[$1$]$ N. E. Hussey {\it et al.}, Nature {\bf 425}, 814 (2003).   

Interlayer Aharonov-Bohm interference in tilted magnetic fields in quasi-one-dimensional organic conductors

Different types of angular magnetoresistance oscillations in quasi-one-dimensional organic conductors, such as $\rm(TMTSF)_2X$, are explained in terms of Aharonov-Bohm interference in interlayer electron tunneling. A two-parameter pattern of oscillations for generic orientations of a magnetic field is visualized and related to the experimental data. \\ Reference: cond-mat/0509039   

Unified Theory of Magic Angles and Interference Commensurate Oscillations.

We suggest the unification theory of angular magnetoresistance oscillations in low-dimensional metals with open sheets of Fermi surfaces. It is based on an idea that effective space dimensionality of electron spectrum and electron wave functions is changed at some special directions of a magnetic field. These 1D -$>$ 2D dimensional crossovers are shown to be due to interference effects, which occur when electrons move in the extended Brillouin zone in a magnetic field. Our quantum mechanical approach allows to derive an equation which describes analytically both Magic Angles and Interference Commensurate oscillations in resistivity component, perpendicular to conducting layers, and reveals their common physical origin. We compare our results with experimental data obtained on (TMTSF)$_{2}$ClO$_{4}$ and (TMTSF)$_{2}$PF$_{6}$.   

Lebed Magic Angles in (TMTSF)$_{2}$X $_{ }$Probed by Torque, Transport and NMR

We've investigated the Lebed effect [1] in the quasi-1D molecular organic conductor (TMTSF)$_{2}$X for magnetic fields in the $b^{\prime }-c^{\ast }$-plane, via angle-dependent torque, magnetoresistance, and NMR relaxation rate. In torque versus field angle measurements for X = ClO$_{4}$ at 0.1 K, we observed distinct field induced spin density wave (FISDW) transitions but, to within our experimental accuracy of 3$\times $10$^{-11}$ Nm, we found no evidential anomalies at the Lebed magic angles. We compare this result with earlier reports of torque measurements in X = ClO$_{4}$ [2] and $^{77}$Se NMR relaxation rate measurements in X = PF$_{6}$ [3]. In fixed angles $T-$ (NMR) and $B-$sweeps (torque and magnetoresistance) in the vicinity of magic angles, no change in the FISDW position was observed. These measurements suggest that magic Lebed orientations have no effect on the metal-FISDW transition. [1] A. G. Lebed, JETP Lett. \textbf{43}, 174 (1986). [2] M. J. Naughton \textit{et al}., Phys. Rev. Lett. \textbf{67}, 3712 (1991). [3] W. Wu et al., Phys. Rev. Lett. 94, 097004 (2005)..   

Thermoelectric power and Nernst effect studies in the metallic and field-induced spin density wave states in (TMTSF)$_{2}$ClO$_{4}$

We have measured the angular dependence of thermoelectirc power (TEP) and Nernst effect of (TMTSF)$_{2}$ClO$_{4}$. At low temperatures and in the metallic state, Nernst effect shows giant resonant signals around the Lebed magic angles, while TEP is small without noticiable angular dependence. This behavior is very similar to what was observed in (TMTSF)$_{2}$PF$_{6}$ in the metallic state [Wu et al., Phys. Rev. Lett. \textbf{91} 56601(2003)]. By entering the field-induced spin density wave (FISDW) state, both TEP and Nernst signal show complicated behaviors reflecting the FISDW subphase transitions. Remarkably, the resonant Nernst effect still persists in the FISDW state and with even lager amplitude. By increasing the perpendicular field above $\sim $ 6.5 T, both TEP and Nernst effect becomes small again at all angles. Our Nernst effect results are inconsistent with some proposed models for the metallic state of (TMTSF)$_{2}$PF$_{6}$, which may suggest this phenomenon is beyond the Fermi liquid description.   

Coexistence of spin density wave and triplet superconductivity in quasi-one-dimensional Bechgaard salts

The interplay between magnetic order and superconductivity is a very important problem in condensed matter physics. In the quasi-one-dimensional (quasi-1D) organic conductor (TMTSF)$_2$PF$_6$, an antiferromagnetic state charaterized by a spin density wave (SDW) order neighbors a triplet superconducting (TSC) state on the pressure-temperature phase diagram. Experiments [1,2] suggest a coexisting region of SDW and TSC orders in the vicinity of the phase boundary. We consider a tight-binding quasi-1D electron system, and construct the Ginzburg-Landau (GL) free energy with two order parameters. In the absence of a magnetic field, the rotational symmetry of this system is broken due to spin anisotropy and spin-orbit coupling. Thus, the GL free energy has a similar form as the ordinary $\phi_1$-$\phi_2$ model, except additional gradient terms. We calculate the GL coefficients microscopically and obtain a phase diagram in zero magnetic field. This phase diagram shows a coexistence region for SDW and TSC. \\ Reference:\\ $[1]$ T. Vuletic {\it et al.}, Eur. Phys. J. B {\bf 25}, 319 (2002).\\ $[2]$ I. J. Lee {\it et al.}, Phys. Rev. Lett. {\bf 94}, 197001 (2005).   

Anomalous temperature dependence of the single-particle spectrum in the organic conductor TTF-TCNQ

The angle-resolved photoemission spectrum of the quasi-one-dimensional organic-conductor TTF-TCNQ exhibits an unusual temperature dependence in the sense that a transfer of spectral weight over an energy range of $\approx 1$\textit{eV} takes place as the temperature decreases below 260$K$. In order to investigate the origin of this behavior, we have performed Dynamical Density-Matrix-Renomalization-Group (DDMRG) calculations at zero temperature and Quantum Monte Carlo (QMC) calculations at finite temperatures for the single-particle spectral weight of the doped one-dimensional (1D) Hubbard model. We present DDMRG and QMC results for a range of the model parameters of the 1D Hubbard model and make comparisons with the photoemission data. In addition, we present zero-temperature DDMRG results on the doped 1D Hubbard-Holstein model in order to explore how the electron-phonon coupling influences the single-particle spectrum in 1D correlated conductors.   

Multiple spin sites in an organic conductor without magnetic ions.

The anisotropic low dimensional organic conductors are attractive because of the variety of ground states with unusual and exotic electronic properties. One of them is tau-[P-($S$,$S)$-DMEDT-TTF]$_{2}$(AuBr$_{2})_{1+y}$ [where y$\sim $0.75 and P-($S$,$S)$-DMEDT-TTF stands for pyrazino-($S$,$S)$-dimethyl-ethylenedithio-tetrathiafulvane], which has tetragonal crystal structure with unit cell dimensions \textbf{a}=\textbf{b}=7.3546 {\AA} and \textbf{c}=67.977 {\AA}$^{1}$. Even though there are no magnetic ions in the compound, transport measurements show magnetic ordering at low temperature and in magnetic fields. To investigate the origin of the magnetic behavior, we are conducting an ESR study. We do observe multiple resonances which indicate the existence of the multiple spins although the system does not contain magnetic ions. Moreover, the in-plane angular dependent ESR measurements reveal 4 fold symmetry. Both in-plane and out of plane ESR signal show evidence of antiferromagnetic behavior below 12 K. It is possible that the ion stoichiometric charge transfer (1+y) is the origin of the magnetic effects. Further analysis will be presented.   

Investigating the Charge Ordering Pattern in the Organic Spin Ladder (DT-TTF)$_2$Cu(mnt)$_2$

Quantum spin ladders have attracted considerable interest as intermediaries between one-dimensional chains and two- dimensional square lattices. In order to elucidate the charge ordering pattern in a model organic spin ladder, we measured the temperature-dependent infrared spectra of the organic spin- ladder candidate (DT-TTF)$_2$Cu(mnt)$_2$. We interpret the results within the context of recent theoretical predictions of various charge ordering patterns in spin ladders [1]. Comparision with spectra of the isostructural Au analogue and neutral DT-TTF compound aid in this analysis. [1] R. T. Clay, S. Mazumdar, Phys. Rev. Lett. \textbf{94}, 207206 (2005)   

Magnetic field effects on the coexisting Bond-Charge-Density waves in the quasi-one-dimensional quarter-filled bands

Magnetic field effects on spin-Peierls systems have been of interest for a long time. The theoretical phase diagram consists of three different regions containing the homogenous dimerized and undimerized phases, and a magnetic phase consisting of a soliton lattice or an incommensurate phase. We have investigated numerically spin excitations and magnetic field effects on the bond-charge-density wave (BCDW) that appears below the spin-Peierls transition in the quarter-filled band organic charge transfer solids (CTS), with the specific goal of determining whether the simplest phase diagram, obained within the spin model, applies also to the quarter-filled band where both charge and spin degrees of freedom exist. We also discuss recent experiments in quarter-filled band CTS within the context of our theory.   

Temperature dependent competition between charge-ordering and spin-Peierls transition in (TMTTF)$_2$X: the role of quantum phonons

The (TMTTF)$_2$X salts are quasi-one-dimensional materials with complex phase diagrams that feature a large number of ordered states including superconductivity. The ground states of these materials are often spin-Peierls (SP) states. However, at intermediate temperatures (100 K) a transition to a charge ordered state is also present, which may compete with the ground state SP ordering. We investigate numerically models for these materials that include three components: electronic interactions, bond-coupled phonons, and Holstein-type phonons coupled to the local charge density. These three components have different energy scales are hence expected to dominate at different temperatures. We explicitely include finite phonon frequency in our calculations using quantum Monte Carlo methods. We present charge, spin, and bond susceptibilities as a function of temperature and discuss recent experiments on these materials.   

Discrete breather energy thresholds in Discrete Nonlinear Schrodinger (DNLS) systems Discrete breather energy thresholds in Discrete Nonlinear Schrodinger (DNLS) systems

The DNLS equation has been used successfully to model physical systems as varied as the Holstein polaron, the Davydov soliton, local modes of small molecules and, more recently, optical wave guide arrays and Bose-Einstein condensates trapped in optical lattices. In addition, the DNLS also governs the slow modulations of plane waves in Klein-Gordon systems (network of oscillators). In one dimension and for a cubic nonlinearity, the DNLS is known to support discrete breather solutions - time-periodic, spatially localized excitations - with arbitrary low energy (or norm). In contrast, for higher nonlinearities or in higher dimensions, an energy (norm) threshold is known to exist, below which discrete breathers cannot be found. Using two different approaches to treat the DNLS equation--namely, an exponential ansatz and the so-called ``single nonlinear impurity'' approximation-- we derive analytical expressions for these energy thresholds and compare them to the exact numerical solutions.   

Bipolaron phase diagram of the 1D adiabatic Holstein-Hubbard model in the strong coupling limit

We derive the phase diagram for bipolarons in the one-dimensional adiabatic Holstein-Hubbard model in the strong coupling (small hopping) limit. We show the existence of a threshold value for the Coulomb interaction beyond which the ground state of the system consists of two free polarons (infinitely far apart from each other). This result, obtained by means of an exact perturbative expansion, cannot be reproduced by the usual exponential ansatz for the wave function.   

First-Principles Study of Electronic Structure in $\alpha $-(BEDT-TTF)$_{2}$I$_{3}$ at Ambient Pressure and under Uniaxial Strains.

We calculate the electronic structure of $\alpha $-(BEDT-TTF)$_{2}$I$_{3}$ at 8K and room temperature at ambient pressure and under the uniaxial strains along the a- and a-axes within the density functional theory. We discover \textit{anisotropic Dirac cone dispersion} near the chemical potential. We also extract the orthogonal tight binding parameters to analyze physical properties. An investigation of the electronic structure near the chemical potential clarify that effects of the uniaxial strain along the a-axis is different from that along the b-axis. The Dirac cone dispersion yields the linear density of states to give $T^{2}$ dependence of the carrier density upto about 100K. It may explain the experimental findings not only qualitatively but also quantitatively.   

Power-law Current-Voltage Characteristics of Charge-Ordered Organic Crystals ${\theta}$-(BEDT-TTF)$_2$MZn(SCN)$_4$ (M=Cs, Rb)

We have measured the current-voltage characteristics of charge- ordered organic crystals ${\theta}$-(BEDT-TTF)$_2$MZn(SCN)$_4$ (M=Cs, Rb) in a low current range down to $10^{-13}$ A. The current-voltage characteristics follow the power law $I{\propto} V^a$ with a large exponent (e.g., $a=8.4$ at 0.3 K for M=Cs) over a wide range of currents. The power-law characteristics are attributed to electric field-induced unbinding of electron- hole pairs which are thermally excited in the background of the two-dimensional charge order. From analysis of crossover electric fields from ohmic to the power-law characteristics, we obtain strong evidence that the electron-electron Coulomb interaction is significantly long-ranged, i.e., the screening length is greater than 10 molecule sites. A novel magnetoresistance effect, possibly due to the Pauli exclusion principle, is also presented.   

Session P45: Structural and Ferroelectric Phase Transitions

Structural symmetry of Cd$_2$Re$_2$O$_7$ from nonlinear optics

Cd$_2$Re$_2$O$_7$, a superconducting metallic pyrochlore, undergoes a second-order structural phase transition at 200~K from a cubic to tetragonal lattice. Landau theory predicts that any second-order cubic-to-tetragonal phase transition must also possess an order parameter associated with broken inversion symmetry. By observing optical second-harmonic generation, we provide a direct demonstration that the 200~K transition in Cd$_2$Re$_2$O$_7$ involves broken inversion symmetry. Moreover, we have used the polarization dependence of SHG to refine the crystal structure. We find that the low-temperature crystal symmetry is that of the $F\overline{4}2m$ space group.   

X-ray absorption spectroscopy on the phase transition of Cd$_{2}$Re$_{2}$O$_{7}$

Pyrochlore Cd$_{2}$Re$_{2}$O$_{7}$ has received considerable attention and investigated extensively because of its manifold phase transitions accompanied by profound physical properties. In this study, the relations among the electronic structure, phase transition, and some physical properties of Cd$_{2}$Re$_{2}$O$_{7}$ single crystals have been investigated. We conducted the X-ray diffraction as well as Re $L_{2,3}$-edge extended X-ray absorption fine structure (EXAFS) at different temperatures to investigate the local structure of Cd$_{2}$Re$_{2}$O$_{7}$ and temperature-dependent O $K$-edge X-ray absorption near edge structure (XANES) to probe its electronic structure during the phase transition. We also performed the band structure calculations to understand the DOS near the Fermi level of Cd$_{2}$Re$_{2}$O$_{7}$ at different temperatures.   

Pressure effects on a low carrier density metal's infrared response

We have performed pressure and temperature dependent infrared studies of semi-metal bismuth. Along with other semi-metals such as graphite and antimony, bismuth is a material of much current interest due to the interesting properties resulting from an extremely low carrier density, small effective mass, and long mean free path. A novel optical setup has been developed to measure low temperature infrared responses at pressures of up to 17kbar. We observe massive changes in bismuth's optical and infrared conductivity as a function of temperature and pressure. We discuss the origin of these changes both in terms of simple band structure effects, as well as in terms of the correlation effects that are expected in this very low-carrier density metal.   

Ferroelectric phase transition in the incommensurate phase of the K$_{2}$SeO$_{4}$ crystal

It is shown that in the incommensurate (IC) phase of the K$_{2}$SeO$_{4}$ crystal a phase transition takes place to the ferroelectric IC phase, prior to the lock-in transition to the triple-period commensurate phase. Such a behavior is due to the Lorentz microscopic electric field, which is induced by the IC domains on approaching the lock-in transition temperature. The proposed behavior of the K$_{2}$SeO$_{4}$ crystal explains the second-harmonic generation, observed in the IC phase, near the lock-in transition. It explains also the observed drop in the crystal's elastic constant c55 near the lock-in transition, which corresponds to a significant decrease of the z-polarized sound velocity in the x-direction. Continuous increasing of the polarization in the ferroelectric IC phase manifests itself as a continuous decreasing of the sound velocity. In the same model one can explain also the observation of an overdamped Raman scattering in the z(xz)y geometry in the low temperature range of the IC phase existence. Dielectric properties of the ferroelectric crystals with artificially fabricated domains are discussed.   

Normal and superconducting state properties of the (Pr$_{1-x}$Nd$_x$)Os$_4$Sb$_{12}$ system

Our previous experimental studies of the (Pr$_{1-x}$Nd$_x$)Os$_4$Sb$_{12}$ system revealed that: (1) superconductivity and ferromagnetism from both end compounds were suppressed almost monotonically toward $x = 0.55$, (2) retention of the antiferroquadrupolar order phase in (Pr$_{1-x}$Nd$_x$)Os$_4$Sb$_{12}$ to higher values of $x$ than in Pr(Os$_{1-x}$Ru$_{x}$)$_4$Sb$_{12}$, (3) two possible CEF energy level schemes in NdOs$_4$Sb$_{12}$, in which the ground state is either the $\Gamma_{6}$ doublet or $\Gamma_{8}^{(2)}$ quartet. Recent ultrasonic measurements in the NdOs$_4$Sb$_{12}$ sample ($x=1$) revealed softening of the C$_{44}$ mode, which indicated that the CEF ground state in this compound is more likely to be the $\Gamma_{8}^{(2)}$ quartet. The lattice parameter in (Pr$_{1-x}$Nd$_x$)Os$_4$Sb$_{12}$ seems to increase slightly from PrOs$_4$Sb$_{12}$ ($x = 0$) toward NdOs$_4$Sb$_{12}$ ($x = 1$). The T-x and H-x phase diagrams related to superconductivity, ferromagnetism, antiferroquadrupolar order, and the CEF energy level scheme for the (Pr$_{1-x}$Nd$_x$)Os$_4$Sb$_{12}$ system for $0 \le x \le 1$ will be discussed.   

Elastic properties of ferromagnetic heavy fermion system SmOs$_4$Sb$_{12}$

We report the elastic constants of the heavy fermion system SmOs$_4$Sb$_{12}$ by means of an ultrasonic measurement. A steep decrease associated with the ferromagnetic transition was observed at around 2 K in elastic constants C11, (C11-C12)/2 and C44. Furthermore a characteristic increase, possibly due to the ``rattling-motion'' was observed around 15 K in the elastic constants. The variation of the onset temperature and a degree of the increase as a function of the ultrasonic frequency is reasonably reproduced in terms of the Debye-type dispersion. The obtained parameters describing the rattling-motion such as a relaxation time, an activation energy and a mean square displacement will be discussed as compared with those of an isostructural compound PrOs$_4$Sb$_{12}$.   

High-Energy X-Ray Study of Short Range Order and Phase Transformations in Ti-V

Phase transformations, especially precipitation processes, are a key factor in alloy design. Understanding these processes in the framework of statistical thermodynamics requires knowledge about the atomic interaction potentials between the alloy constituents. Experimentally, these parameters can be accessed via the diffuse x-ray scattering caused by the configurational short range order and lattice distortions. We employ a bulk sensitive high energy technique to study both phenomena simultaneously in situ, probing macroscopic single crystals in transmission geometry. The data recorded by a 2D detector reveal Bragg reflections from the precipitates superimposed on the diffuse scattering of the matrix. We present a detailed study of bcc Ti-V, a typical titanium $\beta$-alloy. The diffuse scattering is mainly due to lattice distortions induced by the atomic size mismatch. Depending on the annealing temperature, growth and dissolution of hcp $\alpha$-Ti precipitates and minute fractions of TiC are observed. HRTEM experiments have been conducted to complement our results.   

Reversible Magnetostriction with Temperature in Single Crystal Tb$_{5}$Si$_{2.2}$Ge$_{1.8}$

The magnetostriction changes that accompany the phase transitions of single crystal Tb$_{5}$(Si$_{2.2}$Ge$_{1.8})$ have been investigated at temperatures between 20 K and 150 K, by measurements of the reversible component of the magnetostriction along the crystallographic ``a'' axis. Over this temperature range the shape and slope of the magnetostriction curves change, which are indicative of changes in the magnetic state, crystal structure and magnetic anisotropy. Results show a phase transition that occurs near 106 K (onset-completion range 100 - 116 K). The abrupt nature of the strain transition, its unusual hysteresis, and its temperature dependence appear to indicate a first order phase transition which can be activated by applied magnetic field or temperature. Magnetostriction measurements at temperatures below the transition region show a magnetostriction of small overall magnitude but with a high, temperature dependent anisotropy. Funded by USDoE-Office of Basic Energy Sciences   

Magnetic phase transition and spin dynamics in Li(Ni1-xFex)PO4

Elastic and inelastic neutron scattering techniques were used to study the magnetic phase transition and spin dynamics in pure and Fe substituted LiNiPO4 single crystals. Pure LiNiPO4 undergoes a first-order magnetic phase transition from a long- range ordered incommensurate phase to an antiferromagnetic ground state at TN = 20.8 K. With the substitution of Fe for Ni, the magnetic phase transition changes from first-order to second-order, and moreover, the long-range ordered incommensurate phase of pure LiNiPO4 between 20.8 K to 21.5 K was suppressed in the LiNi0.75Fe0.15PO4 sample. Inelastic neutron scattering revealed a ~2 meV energy gap and an anomalous “soft mode” in the spin wave dispersion curve along the [010] direction for pure LiNiPO4. For LiNi0.8Fe0.2PO4, however, the energy gap was reduced to 0.9 meV and the anomaly along the [010] direction reduced. The spin-wave dispersion curves were simulated using a Heisenberg Hamiltonian with Dzyaloshinski-Moriya interactions.   

Phonon Anomalies in the Martensitic Phase of Ni$_{2}$MnGa

Ni$_{2}$MnGa is a cubic ferromagnetic shape memory alloy exhibiting phonon anomalies as precursors to the Martensitic phase transformation. By cooling in a small magnetic field (1.5T) through its transformation temperature it is possible to obtain a single domain of the tetragonal Martensite phase. One can, therefore, measure the phonon dispersion curves in the Martensite phase and compare them to the high temperature cubic phase. The TA branch propagating along the [qq0], with polarization along [q-q0], shows an anomaly near the wavevector of the charge density wave (CDW) peak at q=0.425, which differs from q=0.33 observed in the cubic phase. Most interesting is a new acoustic-like branch emerging from the CDW peak. This will be discussed in relationship to phasons and amplitude modes observed in other incommensurate systems.   

Medium-range Structure of La$_{2-x}$Sr$_x$CuO$_4$ (0.0 $<$ $x$ $<$ 0.3) by Pulsed Neutron PDF Analysis

We studied the local and medium-range atomic structure in high- temperature superconductor (HTSC) La$_{2-x}$Sr$_x$CuO$_4$ (0.0 $<$ $x$ $<$ 0.3) by the pulsed neutron pair-density function (PDF) analyses. The measurement was made with the NPDF of LANSCE, LANL. Because of the high $Q$-resolution of the NPDF the PDF was determined up to 20 $nm$. The measured PDF was compared with that calculated for the average structure determined by the Rietveld analysis of the same data. We found that the measured PDF deviates from the calculated PDF in two different ways. For $x$ = 0.16 - 0.3 deviations were seen up to about 3 $nm$, strongly related to the oxygen in the CuO$_2$ plane displaced along the $c$-axis. These deviations reflect local fluctuations between the orthorhombic and tetragonal phases. A more interesting deviations were seen for $x$ = 0.04 - 0.16 up to 2 $nm$ for all compositions. The lengthscale of this local fluctuation corresponds to the in-plane coherence length, suggesting that the structure that supports HTSC may be different from the average structure, and the size of the local domains is not limited by the dopant concentration. Further implications are discussed.   

Inelastic neutron scattering study of YBaFe$_2$O$_5$

YBaFe$_2$O$_5$ belongs to a new class of oxides with the chemical formula $R$Ba$M_2$O$_5$ ($R=$~rare-earth, $M=$~transition metal), based on the perovskite structure with a doubled unit cell pyramids of five-coordinated $M$-sites. The $M$-site is mixed valent in the stoichiometric formula unit (with an average valence of +2.5). Therefore, charge and orbital ordering phenomena can exist on the $M$-site and be studied without introducing disorder. The charge ordered phase of YBaFe$_2$O$_5$ is unusual, since it does not satisfy the Anderson criterion (i.e. it is not the lowest energy electrostatic arrangement of charges), but rather orders into alternating chains of 2+/3+. This indicates that other interactions, such as electron-phonon coupling, are necessary to arrive at the chain structure. Here, we present the results of an inelastic neutron scattering study of polycrystalline YBaFe$_2$O$_5$. We find the spectrum of phonon and magnetic excitations are clearly modified at the charge- and magnetic ordering temperatures: T$_{CO}$ = 308~K and T$_N$ = 430~K, respectively.   

Session P46: FQHE

Measuring Exchange Interactions by Tunneling Deep Into the Quantum Hall Liquid

We present measurements of the tunneling density of states of a two dimensional electron gas (2DEG) in GaAs at energies up to 10 meV above and below the Fermi energy. Using time domain capacitance spectroscopy (TDCS), we determine the current-voltage (IV) characteristics for tunneling perpendicularly between a gated 2DEG and a 3D electron continuum separated by a thin tunneling barrier. In TDCS, sharp pulses are applied to the sample while measuring displacement currents from electrons entering or leaving the 2DEG, allowing tunneling IV measurements without direct electrical contact to the 2DEG. We observe changes in the Landau level structure far from the Fermi surface as we fill and empty individual Landau levels by varying the electron density and magnetic field. This provides a unique measurement of the exchange enhanced spin splitting of empty and filled Landau levels.   

Thermal Dephasing in the Laughlin Quasiparticle Interferometer

We report experiments on thermal dephasing of the Aharonov-Bohm oscillations in the novel Laughlin quasiparticle (LQP) interferometer, [1] where quasiparticles of the 1/3 FQH fluid execute a closed path around an island of the 2/5 fluid. In the $10.2 \leq T \leq 141$ mK temperature range, qualitatively, the experimental results follow a thermal dephasing dependence expected for an electron interferometer, and show clear distinction from the activated behavior observed in resonant tunneling and Coulomb blockade devices, both in the chiral Luttinger liquid ($\chi$LL) and the Fermi liquid regimes. The data fit very well the $\chi$LL dependence predicted for a $g=1/3$ two point-contact LQP interferometer. [2] The fit yields a value of the chiral edge excitation velocity, $u=1.4\times 10^4$ m/s obtained for the first time for a continuous FQH edge excitation spectrum. The small deviation from the zero-bias theory seen below 20 mK indicates yet unrecognized source of experimental decoherence, not included in theory. \newline \noindent [1] F. E. Camino et al., Phys. Rev. B $\bf 72$, 075342 (2005). \newline \noindent [2] C. de C. Chamon et al., Phys. Rev. B $\bf 55$, 2331 (1997).   

Flux Period Scaling in the Laughlin Quasiparticle Interferometer

Aharonov-Bohm superperiod was rececently reported for electron interferometer devices in the quantum Hall regime, where electron paths circle a 2D electron island. The electron island main confinement is produced by etch trenches, into which front gate metal is deposited. We determine experimentally the A-B period $\Delta_B$ at several front gate voltages $V$ for electrons ($f =1$) and Laughlin quasiparticles (2/5 embedded in 1/3). For moderate $|V| \leq 300 $ mV, on each QH plateau, we find linear dependence of $\Delta_B$ on $V$. For $f=1$, the electron A-B path area $S$ can be found from $\Delta_B$ using flux quantization condition $\Delta_\Phi =S\Delta_B =h/e$ for the flux period $\Delta_\Phi$. The A-B area enclosed by the $f=1/3$ edge channel (the 2/5 island area) is not known independently if the FQH flux period is not known a priory. The front gate voltage dependence of $\Delta_B$ provides such independent determination of the 2/5 island area. The directly measured values of $\Delta_B$ and its slope $d\Delta_B /dV$ can be combined to derive the voltage $V (1e)$ attracting a unit charge to the area of the A-B path, assuming $S$ is known. For a many-electron ($\sim$2000) 2D disc of radius $r$, the product $rV(1e)$ should be approximately constant, independent of the QH filling or the area. Thus the $f=2/5$ island area can be determined directly with a $\sim$10\% accuracy, which is quite sufficient to distinguish the physically reasonable possibilities of the flux periods $5h/e$, $5h/2e$, $1h/e$, and $h/2e$.   

Extracting fractional statistics from superperiodic Aharonov-Bohm oscillations

We consider a quantum Hall interferometer in which the quasiparticles of a fractional quantum Hall (FQH) liquid with filling factor $\nu_1=1/3$ propagate around a large ring of radius $r_1$, which is encircles an island with a smaller radius $r_2$ occupied by FQH liquid with filling factor $\nu_2=2/5$. We study the conductance oscillations that result from the incompressibility of the FQH liquid occupying the island and the constructive interference condition for the quasiparticles encircling the outer ring. Since the constructive interference condition depends on both the magnetic flux enclosed by the encircling path and the statistical phase gained by the encircling quasiparticle due to the presence of quasiparticles in the island, such conductance oscillations can be used to detect signatures of fractional statistics. We find that oscillatory period depends on both radii, $r_1$ and $r_2$. We discuss the relation between our results and the recent experiments by F.E.Camino, W. Zhou and V.J. Goldman in the context of our model.   

Resonant tunneling in fractional Hall effect

We study theoretically the possible transitions of a fractional quantum Hall island surrounded by another fractional quantum Hall state, induced by either the variation of the magnetic field or a backgate voltage, and find a rich set of possibilities in addition to the one considered previously[1],The elementary transitions correspond to the addition or removal of a composite fermion from the edge or the interior of the island; combinations of elementary transitions may occur simultaneously due to electrostatic constraints. Relevance to a recent experiment[2] is considered, which measures the resonant tunneling of composite fermions through their quasi-bound states around such a 2/5 island surrounded by the 1/3 sea. It is shown that the results are consistent with the notion of fractional braiding statistics, but can be explained on the basis of fractional charge alone. We also perform calculations based on microscopic composite fermion wavefunctions of finite systems to test the theoretical considerations. [1]J.K.Jain, S.A.Kivelson, and D.J.Thouless, Phys.Rev.Lett.\textbf{71}, 3003(1993). [2]F.E.Camino, W.Zhou, and V.J.Goldman, Phys.Rev.B \textbf {72}, 075342(2005).   

Electronic Mach-Zehnder interferometer as a tool to probe fractional statistics

We study transport through an electronic Mach-Zehnder interferometer recently devised at the Weizmann Institute. We show that this device can be used to probe statistics of quasiparticles in the fractional quantum Hall regime. We calculate the tunneling current through the interferometer as the function of the Aharonov-Bohm flux and voltage bias, and demonstrate that its flux-dependent component is strongly sensitive to the statistics of tunneling quasiparticles.   

Pauli-like principle for Abelian and non-Abelian FQHE quasiparticles

A general formulation of condensed matter physics describes the Hamiltonian as $H_0 + H_1$, where $H_0$ is a positive ``topological'' Hamiltonian with a highly-degenerate zero-energy ground state (extensive $T=0$ entropy) and $H_1$ is the ``physical'' Hamiltonian that splits this huge multiplet. Usually, $H_0$ is a non-interacting Hamiltonian, with zero modes that form a simple Fock space spanned by Wannier orbitals of low-energy electron bands, or a Landau level, etc. Systems of Laughlin FQHE quasiholes are described by a more general $H_0$ that removes low-relative-angular momentum two-particle states from the zero-mode spectrum, and the non-Abelian Moore-Read and Read-Rezayi quasihole systems involve removal of $n > 2$ particle states. The latter are candidates systems for topological quantum computation. The zero-modes count has been previously obtained by counting the number of linearly-independent polynomials of various types. I give a simpler Pauli-principle-like formulation that transparently gives the counting rules, and allows the creation of subsets of lowest-Landau-level Slater determinant states from which the zero-mode states can be constructed. This aids numerical diagonalization of $P_0H_1P_0$, where $P_0$ is the projection into the zero-modes space of $H_0$, for exact-diagonalization simulations of the manipulations of non-Abelian quasiparticles proposed for topological quantum computations.   

Quantitative study of the non-Abelian statistics of quasiholes and quasiparticles in the nu=5/2 paired Hall state

We analyze quantitatively various properties of a collection of quasihole and quasiparticle excitations of the paired composite fermion state, described by a Pfaffian wave function proposed by Moore and Read (Nucl.Phys.B 360, 362, 1991), which are relevant to the validity of the notion of non-abelian braiding statistics. Working in the spherical geometry, we study the coupling of two quasiholes as a function of their distance by evaluating both the density profile and the interaction energy of a quasihole pair. Further, we perform a numerical study to check whether the $2^{n-1}$ independent states of $2n$ quasiholes are almost degenerate, i.e.\ the coupling between these states is exponentially suppressed as a function of their separation, which will be crucial for any practical realization of non-Abelian statistics. We also compare the exact diagonalization spectra of the Coulomb interaction in the second Landau level and the model three-body contact interaction for which the Pfaffian state and its quasihole variants are known to be exact. Based on the connection between Halperin's 331 state and the Pfaffian state (Greiter, Wen, and Wilczek, PRL 66, 3205, 1991) we construct a class of quasiparticle wave functions, which we study with respect to both the three- body contact interaction and the Coulomb interaction to test their accuracy.   

Probing Non-Abelian Statistics in the $\nu=5/2$ Fractional Quantum Hall State

We analyse an interferometric experiment to detect non-Abelian quasiparticle statistics -- one of the hallmark characteristics of the Moore-Read state expected to describe the observed FQHE plateau at $\nu= 5/2$. The implications for using this state for constructing a topologically protected qubit as has been recently proposed by Das Sarma, Freedman and Nayak are also addressed.   

Composite fermions and conformal field theory

We show how Quantum Hall quasiparticle wave functions similar to those in the composite fermion theory are related to correlators of certain nonlocal operators in a conformal field theory. Charge and statistics are determined using both analytical and numerical methods.   

An Exact Solution for the Half-filled Lowest Landau Level

We present an exact solution for the interacting electron gas in the half-filled lowest Landau level on a thin torus. The low energy sector consists of non-interacting, one-dimensional, neutral fermions (dipoles). The ground state, which is homogeneous, is the Fermi sea obtained by filling the negative energy states and the excited states are the gapless neutral excitations out of this one-dimensional sea. We identify this ground state as a version of the Rezayi-Read state, and find that it develops continuously, as the circumference grows, into the Rezayi-Read state that is believed to describe the observed metallic phase in the two-dimensional system. This suggests a Luttinger liquid description of the half-filled Landau level.   

One-Dimensional Theory of the Quantum Hall System

We consider the lowest Landau level on a torus as a function of its circumference $L_1$. When $L_1\rightarrow 0$, the ground state at general rational filling fraction is a crystal with a gap---a Tao-Thouless state. For filling fractions $\nu=p/(2pm+1)$, these states are the limits of Laughlin's or Jain's wave functions describing the gapped quantum Hall states when $L_1\rightarrow \infty$. For the half-filled Landau level, there is a transition to a Fermi sea of non-interacting neutral dipoles, or rather to a Luttinger liquid modification thereof, at $L_1\sim5$ magnetic lengths. This state is a version of the Rezayi-Read state, and develops continuously into the state that is believed to describe the observed metallic phase as $L_1\rightarrow \infty$. Furthermore, the effective Landau level structure that emerges within the lowest Landau level follows from the magnetic symmetries.   

Competing liquid and solid orders at $\nu=1/5$

The lowest Landau level states at very low filling factors are accurately understood as topological quantum crystals of composite fermions. At higher fillings (but still in the lowest Landau level), on the other hand, the system forms an incompressible composite- fermion liquid. However at $\nu=1/5$, both descriptions fail to give an accurate account to the true ground state. Our numerical calculations show that for small systems the crystal has lower energy than the liquid, and only for $N\ge 10$ does the liquid become the ground state. We find that a linear combination of the CF liquid and the CF crystal wave functions provides an excellent account of the actual state for small systems. These results indicate that the $1/5$ fractional Hall state is highly susceptible to the formation of composite fermion crystallites in it. We will discuss the relevance of these results to experiment, and also the possibility of inducing a liquid-solid transition at $1/5$ by tuning the interaction.   

Electric field effects in the Hall conductivity

We study the Hall conductivity as a topological invariant under the influence of an intense electric field. We consider a model of a 2DEG in a two-dimensional lattice in the presence of an applied in-plain electric field and perpendicular magnetic field. The Hall conductivity is determined from quasiclassical calculations. In the presence of an electric field the longitudinal quasi-momentum is quantized leading to the appearance of a ‘‘magnetic Stark ladder’’, in which the bands of the Hofstadter butterfly are replaced by a series of quasi discreet levels. We show that the transverse conductivity of this levels is an integer topological invariant independent of the intensity of the electric field thus leading to an integer Hall conductivity.   

Measurement of Spin Excitations in the Fractional Quantum Hall Regime of 1/2$<\nu<$1

We use inelastic light scattering methods to investigate quasiparticle excitations of the fractional quantum Hall liquid in the filling factor range 1/2$<\nu<$1. The long wavelength spin wave mode at the Zeeman energy shows intriguing behavior. The mode is observed at the filling factors nu=1, 2/3 and 3/5 of the quantized Hall states but its intensity collapses for filling factors away from these states. In the filling factor range 1/2$<\nu<$2/3 spin excitations are observed below the Zeeman energy. These modes are interpreted as spin flips where the composite fermion Landau level quantum number and spin orientation change. The spectral lineshapes of spin flip excitations suggest a spin polarization transition between $\nu$=3/5 and $\nu$=2/3 [1]. [1] Irene Dujovne et al, PRL 95, 056808 (2005)