Variable Step Random Walks and Self-Similar Distributions

We study a scenario under which variable step random walks give anomalous statistics. We begin by analyzing the Martingale Central Limit Theorem to find a sufficient condition for the limit distribution to be non-Gaussian. We study the case when the scaling index∼ζ is∼1₂. For corresponding continuous time processes, it is shown that the probability density function W(x;t) satisfies the Fokker–Planck equation. Possible forms for the diffusion coefficient are given, and related to W(x,t). Finally, we show how a time-series can be used to distinguish between these variable diffusion processes and Lévy dynamics.     

Exact propagator of the Fokker-Planck equation with a space-dependent diffusion coefficient and a time-dependent mean-reverting force

We have investigated the algebraic structure of the Fokker-Planck equation with a variable diffusion coefficient and a time-dependent mean-reverting force. Such a model could be useful to study the general problem of a Brownian walker with a space-dependent diffusion coefficient. We also show that this model is related to the Fokker-Planck equation with a constant diffusion coefficient and a time-dependent anharmonic potential of the form V(x, t) = ?a(t)x ² + b ln x, which has been widely applied to model different physical and biological phenomena, e.g. the study of neuron models and stochastic resonance in monostable nonlinear oscillators. Using the Lie algebraic approach we have derived the exact diffusion propagators for the Fokker-Planck equations associated with different boundary conditions, namely (i) the case of a single absorbing barrier, and (ii) the case of two absorbing barriers. These exact diffusion propagators enable us to study the time evolution of the corresponding stochastic systems. Received 23 October 2001 and Received in final form 24 December 2001     

The localization transition of the two-dimensional Lorentz model

We investigate the dynamics of a single tracer particle performing Brownian motion in a two-dimensional course of randomly distributed hard obstacles. At a certain critical obstacle density, the motion of the tracer becomes anomalous over many decades in time, which is rationalized in terms of an underlying percolation transition of the void space. In the vicinity of this critical density the dynamics follows the anomalous one up to a crossover time scale where the motion becomes either diffusive or localized. We analyze the scaling behavior of the time-dependent diffusion coefficient D(t) including corrections to scaling. Away from the critical density, D(t) exhibits universal hydrodynamic long-time tails both in the diffusive as well as in the localized phase.     

The influence of surface forces on shear-induced tracer diffusion in mono and bidisperse suspensions

The shear-induced self-diffusivity of tracer particles of radius a _t = λa in a suspension of particles having a radius, a , is calculated by Stokesian dynamics for different values of the size ratio, λ , both in 2 and 3 dimensions in the binary-collision regime. The self-diffusion is found to decrease strongly when the size ratio becomes quite different from unity. On the other hand, for the same average distance of contact between two spheres, the presence of a soft force always increases greatly the diffusion compared to the effect of a hard shell which is used to model the roughness. This is particularly true for tracer particles smaller than the bath particles, where the shear-induced diffusion can be increased by many order of magnitudes in the presence of a soft force. For suspensions of monodisperse particles we show that, for low volume fraction, the diffusion coefficient is much smaller than the one predicted by the binary collision model, due to the existence of a layered structure. On the contrary at higher volume fraction, many-body collisions strongly enhance the diffusion and it appears that the value of the diffusion is quite sensitive to the presence of clusters of particles which are themselves determined by the range of interparticle forces.     

Structural and dynamic properties of polyoxyethylene sorbitan monooleate micelle in water dispersion studied by pulsed field gradient NMR

The diffusion phenomenon of a nonionic surfactant, polyoxyethylene sorbitan monooleate (POE-SMO), micelle in aqueous solution was investigated by pulsed field gradient nuclear magnetic resonance (PFG NMR) with a high gradient strength of 17.4 T/m at the diffusion timet _d varied from 3 to 300 ms. This high gradient strength allowed us to measure the slow self-diffusion coefficient of POE-SMO micelle, and the short diffusion time below 10 ms showed the restricted diffusion of the micelle. At the shortt _d the self-diffusion of the micelle was restricted and the restricted sizes were 1.8, 1.5, and 0.8 μm for the POE-SMO concentration of 100, 200 and 300 mM, respectively, and 0.6 μm for the POE-SMO only. The possible reason of this restriction was assumed to be the formation of a spatial network or a micellar clustering. Furthermore, a proton exchange between water molecule and surfactant OH group on the micelle surface was proposed. With respect to this proposal, the residence time of the proton at the micelle surface and the thickness of the surface were investigated from proton self-diffusion coefficients by PFG NMR.     

Analysis of perturbed angular correlation spectra of metal ions bound to proteins with rotational correlation times in the intermediate region

The formulation by Dattagupta of the strong-collision model, describing the effect on the perturbation function,G ₂(t) by the isotropic tumbling of an electric field gradient, is generalized to electric field gradients with no axial symmetry. The effect on the perturbation function by strong collisions is compared to the effect of rotational diffusion in the adiabatic limit. The comparison is carried out for decays with an intermediate state of spin 5/2 and for non-axially symmetric electric field gradients. It shows that the strong-collision model can be used for interpretation of PAC spectra of molecules with correlation times between the adiabatic and the fast relaxation limits. The strongcollision model is then used to determine the rotational diffusion of the cadmium substituted copper, zinc superoxide dismutase at 3°C and 25°C from^111mCd TDPAC spectra. For these analyses, the model is incorporated into a conventional least-squares fitting routine.     

On Island Formation in a Locally Perturbed Tokamak Equilibrium

To control the plasma transport at the edge of a tokamak the outer flux surfaces can be artificially destroyed by applying a resonant helical magnetic field, as it is demonstrated at Pulsator [1],[2], [3], Tore Supra [4],[5] and proposed for TEXTOR-94 [6] in the concept of “ergodic divertors”. As a measure of the efficiency of the perturbation field e.g. the level of the field line diffusion coefficient DFL the width Δ_i of the magnetic islands and the related Chirikov parameter are of importance [7],[8],[9],[10]. For the planned Dynamic Ergodic Divertor (DED) at TEXTOR-94 where the perturbation coils are located at the high field side the standard expression for Δ_i using the Fourier components of the magnetic field perturbation [7] leads to results significantly different from field line tracing calculations [11]. The standard expression is commonly used in terms of the perturbation magnetic field δB [5],[7],[8],[9],[12],[13]. But when replacing the Fourier components of the perturbation vector potential by those of the magnetic field finite aspect ratio effects have been neglected so far. For present tokamaks with ? = r/R ? 0.3 this can lead to an error in the field line diffusion of one to two orders of magnitude. In this paper it is shown that taking into account the finite aspect ratio at this point leads to correct results compared to the highly precise field line tracing calculations by the Gourdon code. The island width then is recognized to depend significantly on the poloidal position of the perturbation field. This is in contrast to the standard expression. Also the role of the choice of the magnetic coordinate system is considered.     

Incorporation of Effects of Diffusion into Advection-Mediated Dispersion in Porous Media

The distribution of solute arrival times, W(t;x), at position x in disordered porous media does not generally follow Gaussian statistics. A previous publication determined W(t;x) in the absence of diffusion from a synthesis of critical path, percolation scaling, and cluster statistics of percolation. In that publication, W(t;x) as obtained from theory, was compared with simulations in the particular case of advective solute transport through a two-dimensional model porous medium at the percolation threshold for various lengths x. The simulations also did not include the effects of diffusion. Our prediction was apparently verified. In the current work we present numerical results related to moments of W(x;t), the spatial solute distribution at arbitrary time, and extend the theory to consider effects of molecular diffusion in an asymptotic sense for large Peclet numbers, P_e. However, results for the scaling of the dispersion coefficient in the range 1<P_e<100 agree with those of other authors, while results for the dispersivity as a function of spatial scale also appear to explain experiment.     

Simulation of molecular orientational dynamics in a model polar fluid

A molecular dynamics simulation of a Stockmayer fluid with μ* = 1·0, ρ* = 0·7 and T* = 1·13 (±0·03) is reported. In addition to evaluations of a number of static properties, orientational time correlation functions C_l (t) = <P(cos δθ(t))> were calculated for l = 1 through 4 ; P_l is a Legendre polynomial and δθ(t) is the angle of reorientation of the dipole in time t. These time correlation functions are characteristic of nearly free rotation and agree well with curves calculated from a perturbation theory for the memory functions that utilizes the simulated value of the mean square torque. The angular velocity autocorrelation function for this fluid was also simulated and compared with perturbation theory. Agreement is not good, primarily because of the presence of a pronounced long time tail in the simulated function. The relationship between these results and those of other simulations and theories is discussed.     

Decay Law of Moving Unstable Particle

Quantum relativistic decay law of moving unstable particle is analytically calculated in the model case of the Breit–Wigner mass distribution. It turns out that Einstein time dilation of the moving particle decay holds approximately at times when the decay is exponential. The related correction is calculated analytically. Being very small at these times it is practically unobservable. It is shown that Einstein dilation fails for large times t when decay is not exponential. An unstable system of the kind of K ₀-meson (which is the superposition of K _s and K _I) is also considered. In this case, the violation of Einstein dilation is shown to be appreciable at all times under some condition     

Renormalization group analysis of polymer cyclization with non-equilibrium initial conditions

We develop a renormalization group approach for cyclizing polymers for the case when chain ends are initially close together (ring initial conditions). We analyze the behavior at times much shorter than the longest polymer relaxation time. In agreement with our previous work (Europhys. Lett. 73, 621 (2006)) we find that the leading time dependence of the reaction rate k(t) for ring initial conditions and equilibrium initial conditions are related, namely k _ring(t) ∝ t ^-δ and k _eq(t) ∝ t ^1-δ for times less than the longest polymer relaxation time. Here δ is an effective exponent which approaches δ = 5/4 for very long Rouse chains. Our present analysis also suggests a “sub-leading” term proportional to (ln t)/t which should be particularly significant for smaller values of the renormalized reaction rate and early times. For Zimm dynamics, our RG analysis indicates that the leading time dependence for the reaction rate is k(t) ∼ 1/t for very long chains. The leading term is again consistent with the expected relation between ring and equilibrium initial conditions. We also find a logarithmic correction term which we “exponentiate” to a logarithmic form with a Landau pole. The presence of the logarithm is particularly important for smaller chains and, in the Zimm case, large values of the reaction rate.     

Derivation of effective spin models from a three band model for CuO-planes

The derivation of effective spin models describing the low energy magnetic properties of undoped CuO₂-planes is reinvestigated. Our study aims at a quantitative determination of the parameters of effective spin models from those of a multi-band model and is supposed to be relevant to the analysis of recent improved experimental data on the spin wave spectrum of La₂CuO₄. Starting from a conventional three-band model we determine the exchange couplings for the nearest and next-nearest neighbor Heisenberg exchange as well as for 4- and 6-spin exchange terms via a direct perturbation expansion up to 12th (14th for the 4-spin term) order with respect to the copper-oxygen hopping t_pd. Our results demonstrate that this perturbation expansion does not converge for hopping parameters of the relevant size. Well behaved extrapolations of the couplings are derived, however, in terms of Padé approximants. In order to check the significance of these results from the direct perturbation expansion we employ the Zhang-Rice reformulation of the three band model in terms of hybridizing oxygen Wannier orbitals centered at copper ion sites. In the Wannier notation the perturbation expansion is reorganized by an exact treatment of the strong site-diagonal hybridization. The perturbation expansion with respect to the weak intersite hybridizations is calculated up to 4th order for the Heisenberg coupling and up to 6th order for the 4-spin coupling. It shows excellent convergence and the results are in agreement with the Padé approximants of the direct expansion. The relevance of the 4-spin coupling as the leading correction to the nearest neighbor Heisenberg model is emphasized. Received 8 June 2001 / Received in final form 28 May 2002 Published online 19 July 2002     

Survival probability of a local excitation in a non-Markovian environment: Survival collapse, Zeno and anti-Zeno effects

The decay dynamics of a local excitation interacting with a non-Markovian environment, modeled by a semi-infinite tight-binding chain, is exactly evaluated. We identify distinctive regimes for the dynamics. Sequentially: (i) early quadratic decay of the initial-state survival probability, up to a spreading time t_S, (ii) exponential decay described by a self-consistent Fermi Golden Rule, and (iii) asymptotic behavior governed by quantum diffusion through the return processes, leading to an inverse power law decay. At this last cross-over time t_R a survival collapse becomes possible. This could reduce the survival probability by several orders of magnitude. The cross-over times t_S and t_R allow to assess the range of applicability of the Fermi Golden Rule and give the conditions for the observation of the Zeno and anti-Zeno effect.     

An examination of the shrinking-core model of sub-micron aluminum combustion

We revisit the shrinking-core model of sub-micron aluminum combustion with particular attention to the mass flux balance at the reaction front which necessarily leads to a displacement velocity of the alumina shell surrounding the liquid aluminum. For the planar problem this displacement simply leads to an equal displacement of the entire alumina layer, and therefore a straightforward mathematical framework can be constructed. In this way we are able to construct a single curve which defines the burn time for arbitrary values of the diffusion coefficient of O atoms, the reaction rate, the characteristic length of the combustion field, and the O atom mass concentration within the alumina provided that it is much smaller than the aluminum density. This demonstrates a transition between a ‘d  ²–t’ law for fast chemistry and a ‘d–t’ law for slow chemistry. For the spherical geometry, the one of physical interest, the outward displacement velocity creates not a simple displacement, but a stress field which, when examined within the framework of linear elasticity, strongly suggests the creation of internal cracking. We note that if the molten aluminum is pushed into these cracks by the high internal pressure characteristic of the stress field, its surface, where reaction occurs, could be fractal in nature and affect the fundamental nature of the burning law. Indeed, if this ingredient is added to the planar model, a single curve for the burn time can again be derived, and this describes a transition from a ‘d  ²–t’ law to a ‘d  ^ν–t’ law, where 0<ν<1.     

Transport coefficients of hard sphere fluids

New calculations have been made of the self-diffusion coefficient D, the shear viscosity η_s, the bulk viscosity η_b and thermal conductivity λ of the hard sphere fluid, using molecular dynamics (MD) computer simulation. A newly developed hard sphere MD scheme was used to model the hard sphere fluid over a wide range up to the glass transition (～0.57 packing fraction). System sizes of up to 32 000 hard spheres were considered. This set of transport coefficient data was combined with others taken from the literature to test a number of previously proposed analytical formulae for these quantities together with some new ones given here. Only the self-diffusion coefficient showed any substantial N dependence for N < 500 at equilibrium fluid densities (ε 0.494). D increased with N, especially at intermediate densities in the range ε ～ 0.3–0.35. The expression for the packing fraction dependence of D proposed by Speedy, R. J., 1987, Molec. Phys., 62, 509 was shown to fit these data well for N ～ 500 particle systems. We found that the packing fraction ε dependence of the two viscosities and thermal conductivity, generically denoted by X, were represented well by the simple formula X/X ₀ = 1/[1 ? (ε/ε₁)]^m within the equilibrium fluid range 0 > ε > 0.493. This formula has two disposable parameters, ε and m, and X ₀ is the value of the property X in the limit of zero density. This expression has the same form as the Krieger-Dougherty formula (Kreiger, I. M., 1972, Adv. Colloid. Interface Sci., 3, 111) which is used widely in the colloid literature to represent the packing fraction dependence of the Newtonian shear viscosity of monodisperse colloidal near-hard spheres. Of course, in the present case, X _o was the dilute gas transport coefficient of the pure liquid rather than the solvent viscosity. It was not possible to fit the transport coefficient normalized by their Enskog values with such a simple expression because these ratios are typically of order unity until quite high packing fractions and then diverge rapidly at higher values over a relatively narrow density range. At the maximum equilibrium fluid packing fraction ε = 0.494 for both the hard sphere fluid and the corresponding colloidal case a very similar value was found for η_s/η_o ?30–40, suggesting that the ‘crowding’ effects and their consequences for the dynamics in this region of the phase diagram in the two types of liquid have much in common. For the hard sphere by MD, D_o/D ～ 11 at the same packing fraction, possibly indicating the contribution from ‘hydrodynamic enhancement’ of this transport coefficient, which is largely absent for the shear viscosity. Interestingly the comparable ratio for hard sphere colloids is the same.     

Γ(Z → b ): A signature of hard mass terms for a heavy top

We calculate analytically the weak radiative corrections to the weak neutral current gauge boson-bottom fermion vertex, keeping the mass m_t of the internal fermion line for the relevant diagrams. We find, to order α, a hard mass-term dependence m_t²/M_W² of the amplitude, for large m_t values. Its origin comes from the unphysical charged Higgs coupling to fermions in the renormalizable gauge or, equivalently, from the longitudinal charged gauge boson couplings. The diagonal Z⁰ decay width to b-quarks decreases, due to these weak radiative corrections, by 0.6%–2.5% when the top mass m_t varies from 45 to 200 GeV.     

Theory of spin diffusion in liquid-phase polymer systems

A general theory of spin diffusion in condensed media is constructed by the method of Zwanzig-Mori projection operators using the superpositional approximation to decouple the many-particle correlation functions. The spin diffusion coefficient is expressed in the form D _sp=D _tr+D _f , where D _tr is the contribution associated with translational displacements of the molecules and D _f is the contribution caused by intermolecular flip-flop processes. The expression for D _tr differs from the well-known Kubo-Green formula for the self-diffusion coefficient D _sd in that the integrand contains an additional factor P _f (t), which is the probability of the molecular spins not participating in intermolecular flip-flop transitions over the time t. A microscopic expression is obtained for D _f in the form of a time integral of the intermolecular dipole-dipole dynamic correlation functions. For liquid-phase polymer system with fairly high molecular mass the condition D _sp≫D _sd is satisfied. Zh. éksp. Teor. Fiz. 114, 538–554 (August 1998)     

A diffusion model of metal surface modification during plasma nitriding

A diffusion model of metal surface modification by plasma nitriding has been developed. This model takes into account the erosion effects at the plasma/solid interface occurring due to the ion bombardment of the surface. For constant sputtering rate, which is the usual situation during plasma nitriding, the growth of the sub-layers is well described by the analytical expressiong(t) =g ₀,f ^–1 (t/t ₀), whereg(t) is the sub-layer thickness at timet,g ₀ andt ₀ are parameters which depend on the treated material and plasma characteristics,f ^–1 is the inverse of the function — In(1 - x) + x), 0 x 1. Under negligible erosion effects, the expression forg(t) reduces to the parabolic law. The diffusion zone (substratum) growth does not follow the parabolic law as well. However, the deviation occurs after long plasma nitriding time. The model can be used for experimentally determining the effective diffusion coefficients and the erosion rate during plasma nitriding of metal surfaces.