Cells communicate with each via mechanical and chemical signals. This enables the formation of complex structures such as aggregates of unicellular organisms, and the robust unfolding of genetic programs during embryogenesis. Such cellular systems typically are composed of thousands to tens of thousands of cells: they are large enough to allow a statistical physics approach, yet not so large as to trivialize the role of fluctuations. In this talk I will describe an approach to understanding cellular systems based on many-body theory. This allows a systematic scheme within which to calculate cell diffusion coefficients and correlation functions. It also forms the foundation of highly optimized computer algorithms which allow tens of thousands of interacting cells to be simulated in three dimensions - thus offering the possibility of scalable simulation of real biological systems.

We study a class of mass transport models where mass is transported in a preferred direction around a one-dimensional periodic lattice and is globally conserved. The model encompasses both discrete and continuous masses and parallel and random sequential dynamics and includes models such as the Zero-range process and Asymmetric random average process as special cases. We derive a necessary and sufficient condition for the steady state to factorise, which takes a rather simple form.

Journal-ref: J. Phys. A: Math. Gen vol. 37 (2004) L275-280.

I will report on an analysis of a suggested path to self-organized criticality. Originally, this path was devised to "generate criticality" in systems displaying an absorbing-state phase transition, but closer examination of the mechanism reveals that it can be used for any continuous phase transition. The Ising model as well as the Manna model is used to demonstrate how the finite-size scaling exponents depend on the tuning of driving and dissipation rates with system size. The findings limit the explanatory power of the mechanism to non-universal critical behavior.

A language is a dynamic object exhibiting slow but sure change as it is used by its speakers. Although a number of theories exist to explain the changes that occur, in this talk I shall focus on a recent proposal grounded in evolutionary principles. By employing tools well-known from statistical physics, I shall formulate and discuss a very simple mathematical model of language change. It turns out this model is closely related to existing descriptions of genetic variation in biological populations, some properties of which I will discuss. I shall conclude with some speculation as to the utility of such an approach to more realistic linguistic systems.

Considering a lattice formulation of the two-species predator-prey model, we show, at a mean-field level, that spatial constraints and discreteness of the system generically invalidate the celebrated description a la Lotka-Volterra. In particular, we show how a mean-field approach, which takes into account the site restriction, is able to predict the second-order phase transitions between an absorbing and a fluctuating phase in the density of predators and prey that are actually (numerically) observed in dimensions D>1. Because of the volume exclusion, we show that mean-field theory correctly predicts that the phase portrait of the active state is not characterized by limit cycles, but by straight-line and, more generically, by spiralling trajectories. We will also mention some recently introduced variants of the model, where, in addition to spatial constraints, interactions of longer range (next-nearest neighbor) are incorporated. In this case, we show that a mean-field analysis predicts the existence of a first-order phase transition.

We show the results from computer simulations of various Lotka-Volterra models on a lattice with site restrictions. Many non-trivial patterns are observed for different values of the stochastic parameters in the co-existent phase. For all of the models considered it is shown that the phase transition between the active to absorbing state belongs to the universality class of directed percolation. Also, we show that introducing an effective "stirring" in the stochastic dynamics can render the models to behave essentially as predicted by the mean-field analysis.

While conventional semiconductor lasers employ electrical pumping for carrier injection, many recent efforts to improve the performance of mid-infrared antimonide-based semiconductor lasers have focused on optical pumping methods. One successful approach has been the optical pumping injection cavity (OPIC) laser, in which the quantum well active region is enclosed between distributed Bragg reflector (DBR) mirrors in order to achieve multiple passes of the pump beam to enhance absorption and thereby carrier excitation. Previously, fixed wavelength sources have been used for optical pumping of OPIC laser structures, with limited tuning available by adjusting the incident angle. By tuning the pump wavelength incident on an OPIC laser using an optical parametric oscillator, minimum threshold intensities and maximum slope efficiencies are demonstrated at the resonance of the DBR cavity surrounding the active region, further demonstrating the potential of OPIC lasers as efficient mid-infrared sources.

This talk will present research motivated by development of a suite of simulations for socio-technical systems over the past decade. I will discuss the dynamics of several systems being modeled, the mathematics of simulation, and the variety of representations and solution techniques required as the scope of the problem varies. In particular I will focus on the phase diagram of vehicular traffic, iterative solution methods to large game-theoretical problems, graph-theoretical approaches to epidemiology, and how these can be combined into an integrated application.

Carbon nanotubes are likely the strongest materials known. Experimental studies of fracture of individual nanotubes are very complicated and results generated in different experiments do not agree. We use quantum-mechanical calculations to shed light on the experimental discrencies and provide accurate estimates of fracture properties of carbon nanotubes. Stress-vs-strain curves of pristine and defected carbon nanotubes are studied to understand the effect of defects on elastic and fracture properties of carbon nanotubes. Defects include generalized Stone-Wales rotations, vacancy defects, holes and chemical functionalization of the sidewall. We are also interested in the modulation of electrical conductivity of carbon nanotubes by oxygen molecules present in air. Experiments indicate that whereas the resistance of a multi-walled carbon nanotube is not affected by exposure to air, partially-burned multi-wall carbon nanotubes can be used as gas-sensors in a reproducible manner. Quantum-mechanical calculations are used to uncover the possible mechanisms of gas-sensitivity of carbon nanotubes.

We created a "bacterial carpet" - a solid-fluid interface capable of active pumping - by attaching Serratia marcescens bacteria to PDMS or polystyrene surfaces. The cell bodies formed a densely packed monolayer, while their flagella continued to rotate, churning the nearby fluid. Motion of tracer beads close to the surface was dramatically enhanced compared to Brownian motion in the bulk fluid. The flow field contained complex, ever-changing linear patterns (rivers) and rotational patterns (whirlpools). This surface performs active mixing equivalent to diffusion with a coefficient of 2×10^-7 cm²/s. When attached to polystyrene beads, the bacteria create "auto-mobile beads" that perform greatly enhanced translation and rotation. Given the size and strength of the flow patterns near the carpets, the motion must be generated by small numbers of coordinated flagella. In a more confined geometry, large-scale self-coordination occurs, leading to self-pumping channels.

Much of the effort in semiconductor spin physics is motivated by the promise to utilize spin-polarized currents in a spin transistor. Narrow-gap semiconductors feature a high mobility and strong spin-orbit coupling, and offer opportunities for spin manipulation in transport-based spin devices. We describe and demonstrate a method to create spin-polarized ballistic electrons through spin-orbit coupling in a two-dimensional electron system in an InSb/InAlSb heterostructure. Reflection of a spin-unpolarized injected beam from a lithographic barrier leads to the creation of two fully spin-polarized side beams, in addition to an unpolarized specularly reflected beam, because the strong spin-orbit coupling in InSb quantum wells leads to different reflection angles for different spin polarizations. We present experimental results verifying the realization of the method. To further demonstrate ballistic electron transport in InSb/InAlSb heterostructures, we fabricated antidot lattices and measured their magnetotransport properties. Ballistic transport features are observed up to moderately high temperatures. The temperature dependence of the magnetoresistance peaks reveals a scattering time particular to antidot lattices, with a linear dependence on temperature in the range 0.4 K to 50 K. The very weak temperature dependence of the width of the ballistic magnetoresistance peaks indicates negligible thermal smearing for electrons in the InSb quantum well, a result of the small electron effective mass.

In order to possess a second order nonlinear optical (NLO) susceptibility, a material must be non-centrosymmetric at both the microscopic and macroscopic levels. One approach for achieving polar order of organic thin films is the fabrication of ionic self-assembled multilayers (ISAMs). In this method, alternating layers of oppositely charged species are adsorbed onto a substrate. The adsorption of each layer occurs on a timescale of minutes, is self-terminating, and yields highly homogeneous films. When the NLO-active material is a polymer with charged chromophore sidegroups, there is generally a large degree of cancellation of chromophores oriented in opposite directions. To eliminate this issue, we have developed a hybrid covalent / ionic self-assembly in which the mechanism of adsorption alternates in successive layers, yielding highly oriented chromophores. Highly stable thin films with electro-optic coefficients greater than 20 pm/V have been fabricated by this approach.