The crystallization of a polymer during cooling from the melt state is a complicated process, even when carried out under quiescent conditions. As a first approximation, crystallization can be viewed on the macroscopic level as occurring in two separate stages. During the primary stage, crystal nucleation and growth lead to the formation of semicrystalline superstructures (often, but not necessarily, spherulitic in nature), which grow until they impinge on each other. The secondary stage is then usually associated with the further increase in crystallinity within the primary superstructures. Secondary crystallization must be viewed as an inevitable consequence of thermodynamic and kinetic hindrances to the formation of equilibrium extended chain crystals. While profuse kinetic and morphological studies can be found in the literature for the primary crystallization stage, systematic attention has only been given recently to the secondary stage.

In this presentation, we combine results of differential scanning calorimetry, atomic force microscopy, small angle X-ray scattering and creep studies to propose a description of the secondary crystallization process and discuss a possible mechanism for the physical aging of semicrystalline polymers above their glass transition temperature.

Two mechanisms of secondary crystallization are envisioned conceptually: secondary crystal formation at low temperatures and lamellar thickening at high temperature. The crystallization time dependence of the melting behavior of secondary crystals is shown to be universal and strongly correlated with the temporal evolution of the glass transition temperature. These and related experimental observations are then rationalized in a qualitative model that considers 1) the thermal stability of crystals formed at different temperatures, 2) the influence of secondary crystallization on conformational constraints in the remaining amorphous fraction and 3) the magnitude of a scaled crystallization temperature defined by θ = (T_{α c}-T)/(T_{α c}-T_g), where T_{α c} and T_g are characteristic temperatures for the crystal and liquid phases.

We investigate linearity (in tropical algebra) of the updating operators applied to some interface growth models and how this grants subadditivity hidden behind. This ensures the presence of a law of large numbers under Eulerian scaling. Building upon previous work by T. Seppalainen, this finally produces a Hamilton-Jacobi equation at the macroscopic level via a Hopf-Lax formula identified at the microscopic level as tropical linearity.

Using this flexible framework we study the large space and time scale behavior of a totally asymmetric, nearest-neighbor exclusion process in one dimension with random jump rates attached to the particles. When slow particles are sufficiently rare the system has a phase transition. At low densities there are no equilibrium distributions, and on the hydrodynamic scale the initial profile is transported rigidly. We elaborate this situation further by finding the correct order of the correction from the hydrodynamic limit, together with distributional bounds averaged over the disorder.

Concepts of nonequilibrium statistical physics have already been employed to mimic phenomena which rely on human behavior, e.g. the emergence of collective organization in social systems. Recently, variants of the (Ising-like) voter model have been used to study quantitatively collective phenomena, such as opinion formation, or creation of consensus. However, voter-like models have mainly been studied on regular lattices justified in physical situations, but not in the context of social sciences. In socio-cultural situations, the interaction patterns between individuals find a better characterization as complex networks with connections changing in time.

We propose a new, exactly solvable model in which the voter dynamics takes place on an adaptive disordered network. The network consists of the usual agents (spins) with two possible opinions (up and down states), but in a novel dynamical approach: at each time step, the states of the agents modify the connections between them; the connections, in turn, determine the new states of the agents. In this manner, the structure of the network is quenched, as it is correlated and evolves with the dynamics of the agents. We obtain the time evolution and the final state of this sytem for arbitrary initial conditions. On short time scales the network approaches exponentially one of its completely ordered absorbing states, or some metastable states: a disordered neutral state or an active state with finite magnetization. On long time scales only the absorbing states persist.

The tremendous interest in using the spin degree of freedom in electronic devices has led to an extensive endeavor to investigate the intrinsic spin polarization of various magnetic materials. The spin polarization has been shown to be a critical factor in the performance of many spin based devices and it has been predicted that materials of high polarization are of utmost importance. The use of superconducting spectroscopy in a planar junction configuration allows for a direct electrical measurement of spin polarization which has direct technological relevance. Planar junction superconducting spectroscopy has been successfully applied to various materials including the dilute magnetic semiconductor GaMnAs, the concentrated magnetic semiconductor EuS, and had also been used to study the dependence of measured spin polarization on the effect of barrier thickness in tunnel junction based devices.

For the past 10 billion years, the typical mass of star-forming galaxies has been decreasing, seemingly in direct conflict with the prevailing model of cosmological structure formation. Using analytic arguments and reviewing recent observations, I will demonstrate that the solution to this mystery is likely to lie in the formation of supermassive black holes, which exert strong feedback on their environments as they pass through an active phase, known as a quasar. Next, I will present the results of one of the largest cosmological smooth particle hydrodynamic simulations every carried out, which includes this feedback process and can be used to make detailed observational comparisons. In particular, our modeling places us in a unique position to interpret joint measurements of the distributions of quasars, galaxies, and small-scale distortions in the microwave background. Finally, I will summarize other aspects of my ongoing research, focusing on studies that constrain the properties of the first stars.

Neutrinos are enigmatic particles which have provided us with one of the clearest proofs for physics beyond the Standard Model of particle physics, so far. At the same time they are valuable probes of astro-physics and fundamental symmetries. The more we learn about the intrinsic properties of this particle the better we will be able to exploit its potential as a messenger of possibly exotic phenomena which otherwise would remain undetected. I will give a brief review of the current status of neutrino physics and introduce the open questions in the field. The emphasis in the second part of my talk will be on a tentative road map for addressing these questions with a focus on oscillation physics.

Proteins are large macromolecules consisting of many noncovalent interactions that determine their three-dimensional structure and stability. Manipulating protein stability and/or function is desired in protein engineering and the pharmaceutical industry. The importance of conformational flexibility to function is well known. For example, enzymes must be flexible enough to mediate a reaction pathway, yet rigid enough to achieve molecular recognition. Consequently, conformational flexibility is a critical link between structure, stability and function. A difficult challenge in biophysical modeling is to develop methods that accurately predict protein flexibility and stability under given thermodynamic and solvent conditions in computing times fast enough for high throughput computational biology applications. After a brief introduction on key features of protein chemistry, I will discuss a promising new paradigm that combines constraint theory with free energy decomposition schemes. Merging these two concepts leads to a Distance Constraint Model (DCM) that is computationally tractable using fast graph-rigidity algorithms. After an overview of the theoretical approach, computational results are presented on the folding/unfolding transitions in the beta hairpin turn and in proteins using mean field approximations, and in the alpha-helix using exact transfer matrix methods.

Hosted by Alexey Onufriev.

I will present recent work on inelastic electron tunneling spectroscopy (IETS) in single-molecule transistors, self-assembled molecular monolayer cross-wire junctions, and multilayer molecular cross-wire junctions coordinated with nickel ions. IETS is an unique technique that provides in situ vibrational information for molecular electronic devices. The IET spectrum represents a molecular signature for a molecular device, simultaneously proving that the molecule of interest is present and giving insight into how the charge carriers interact with the molecule. By electrostatically and chemically tuning the molecular electronic level of devices near resonance, we observed significant modification of their vibrational features. We attribute this behavior to the onset of high-order quasi-elastic processes near the vibrational excitation thresholds of a molecular device close to electronic resonance.

Hosted by Giti Khodaparast.

Noise is ubiquitous in nature, and is becoming more important as devices are engineered down to nano scales. Examples arise in nano-mechanical oscillators and Josephson junctions, to name a few. Noise induced escape from potentials is one important dynamical aspect which is found in many areas of physics and biology where switching behavior between states is observed. In this talk, noise-induced escape from a metastable state of a dynamical system is studied close to a saddle-node bifurcation point, but in the region where the system remains underdamped. We find the energy of escape scales as a power of the distance to the bifurcation point. Moreover, we make a prediction of two types of scaling and the corresponding critical exponents. Both theory and numerical simulations will show how the scaling depends on the interaction between noise and the underlying deterministic dynamics. This work was done in collaboration with Mark Dykman, Michael Shapiro, and Oleg Kagan.

Research supported by the Office of Naval Research, Army Research Office and National Science Foundation.

Hosted by Royce Zia.

In recent years the spin of electrons has attracted significant interest due to possible applications in the novel field of spintronics. Here one seeks to obtain devices that use not only the charge of the electrons but also their spin degree of freedom in order to achieve new functionalities. In my talk I will discuss how spin-orbit coupling forms a common theme underlying many spintronics concepts because spin-orbit coupling makes the spin respond to its spatial ("orbital") environment. Thus it gives us a "control knob" with which we can steer the purely quantum-mechanical spin degree of freedom.

Often spin-orbit coupling in semiconductors can be interpreted as an effective magnetic field. Like an external magnetic field, it results in an oscillatory motion of spin noneigenstates which is the well-known spin precession. The orbital dynamics of electrons in the presence of spin-orbit coupling is likewise characterized by an oscillatory component. In the context of relativistic quantum mechanics described by the Dirac equation, the oscillatory motion has long been known as Zitterbewegung [1]. I will show that many remarkable analogies exist between the Zitterbewegung of relativistic quantum mechanics and the oscillatory orbital dynamics of electrons in semiconductors [2]. Moreover, the oscillatory orbital motion is closely related to the precession of the spin degree of freedom.

[1] E. Schroedinger, Sitzungsber. Preuss. Akad. Wiss., Phys. Math. Kl. 24, 418 (1930)

[2] R. Winkler, U. Zuelicke, and J. Bolte, cond-mat/0609005

Hosted by Jean Heremans.

The way DNA bends determines, to a large extent, how it recognizes and binds other biological molecules, and also how it can pack in cellular compartments. Traditionally, DNA was considered fairly rigid on the relevant length scales ( less than 500 Angstroms). Recent experiments have sparked an intense debate in the field by showing that DNA may, in fact, be much more flexible than previously thought. More over, the text book picture of DNA as an elastic rod obeying Landau's theory of elasticity appears to break down. The structural origins of this newly discovered flexibility of the DNA are still unknown. Atomistic molecular dynamics simulations we performed on SYS-X provide clues into the possible origins of this phenomenon. I will briefly discuss the methods we use, and then focus on our findings and their implications for the mechanism of DNA flexibility.

Hosted by Michel Pleimling.

We take any finite set $A$ and latt it alphabet. Its elements are called letters. Any finite sequence of letters is called a word. Our configuration space is $A^Z$, that is the set of bi-infinite sequences, all the terms of which are letters. $M$ denotes the set of translation-invariant measures on $A^Z$. A class of linear operators on $M$ has been studied for seveeal decades under the name of "cellular automata". We study "substitution operators", a class of operators from $M$ to $M$, which are generally non-linear, but have a natural interpretation of substituting one word by another. We concentrate at special cases, where there are only two letters called minus and plus and a small sample operators. We show that the processes generated by certain superpositions of these operators show phenomena relevant to statistical physics: ergodicity, non-ergodicity, phase transitions.

Hosted by Royce Zia.