Deutsche Physikalische Gesellschaft e. V. (DPG)

WWW-Server für Frühjahrstagungen und E-Verhandlungen

Programm und Abstracts der Sitzung DY 46

Period orbit expansion for simple spin systems

•Sarah Hallerberg¹, Wolfram Just² und Günter Radons^1
1Institute of Physics, Chemnitz University of Technology, Chemnitz, Germany
²Department of Mathematics, Queen Mary/University of London, London, UK

We investigate the properties of zeta functions and the corresponding expansion in terms of periodic orbits for spin chain models. In particular we focus on mean field models which display phase transitions. The properties of the zeta function and the periodic orbit expansion in different phases and near the phase transition will be investigated by analytical methods.

Numerical Results for the Generalized Mittag-Leffler Function

•Hansjörg Seybold¹ und Rudolf Hilfer^1,2
1ICA-1 Universität Stuttgart, 70569 Stuttgart, Germany
²Institut für Physik, Universität Mainz, 55099 Mainz, Germany

Results of extensive calculations for the generalized Mittag-Leffler function E_a,b(z) are given in the complex plane: Discussion of the distribution of the zeros and the dependency of the parameters a > 0 and b Î R. This function is related to the eigenfunction of the fractional derivative operator of order a and type a+b(1-a) [1].

[1] R. Hilfer, Applications of fractional calculus in physics, World Scientific, Singapore (2000)

Characterization of Carbonates Using Local Porosity Theory

•Ali Moosavi¹ und Rudolf Hilfer^1,2
2Institut für Physik, Universität Mainz, 55099 Mainz, Germany
¹ICA-1, universität Stuttgart, 70569 Stuttgart, Germany

Local porosity theory is used to provide a quantitative description of two carbonate samples. The degree of heterogeneity, anisotropy and connectivity of the samples is discussed. Also, the effect of resolution on the results is studied. The results emphasize the necessity to take into accoun micro-nano scale pores when studying or modeling carbonates.

Universal energy distribution for interfaces in a random field environment

•Semjon Stepanow und Andrei A. Fedorenko
Universität Halle, Fachbereich Physik, 06099 Halle

We study the energy distribution function r(E) for interfaces in a random field environment at zero temperature by summing the leading terms in the perturbation expansion of r(E) in powers of the disorder strength, and by taking into account the non perturbational effects of the disorder using the functional renormalization group. We have found that the average and the variance of the energy for one-dimensional interface of length L behave as, < E > µ L ln L, DE µ L, while the distribution function of the energy tends for large L to the Gumbel distribution of the extreme value statistics.

Dynamics and shock formation in an exclusion process with creation and annihilation

•Róbert Juhász¹, Martin Evans² und Ludger Santen^1
1Theoretische Physik, Universität des Saarlandes, 66041 Saarbrücken, Germany
²School of Physics, University of Edinburgh, Mayfield Road, Edinburgh, EH9 3JZ, United Kingdom

We investigate an asymmetric exclusion process with creation and annihilation of particles in the bulk. We show how the continuum mean-field equations can be studied analytically and hence derive the phase diagrams of the model. Fluctuations of the shock position are also investigated by using a phenomenological random walk picture of the shock dynamics. The case where bulk creation- and annihilation rates vanish faster than the inverse of the system size N is also analyzed. We point out that shock localization is possible even for rates proportional to N^-a, 1 < a < 2. The maximum current phase is studied by means of scaling arguments and numerical simulation. Here, the length- and the time scale are found to be finite as opposed to the algebraically decaying correlations of the totally asymmetric simple exclusion process (TASEP). Critical exponents of the transition to the TASEP are determined.

Kritische Dynamik von Modell C

•Günter Moser¹ und Reinhard Folk^2
1Institut für Physik und Biophysik, Universität Salzburg, Österreich
²Institut für Theoretische Physik, Universität Linz, Österreich

Die kritische Dynamik eines relaxierenden n-komponentigen Ordnungparameters reller Relaxationskoeffizient G), der an eine erhaltene skalare Dichte (Onsagerkoeffizient l) koppelt wird durch das Modell C beschrieben. Es werden die feldtheoretischen Funktionen eines generalisierten Modell C (Modell C^*), mit komplexem Ordnungsparameter und mit komplexem Relaxationskoeffizienten G, in Zweischleifennäherung berechnet. Es stellt sich heraus, dass der Imaginärteil von Gamma nur den Fixpunktwert Null besitzt, und somit die Fixpunkte identisch zu jenen von Model C sind. Die Rechnungen korrigieren einerseits die bisherigen feldtheoretischen Ergebnisse [1] für Model C andererseits wird gezeigt, dass der anomale Bereich, in dem die Skaleneigenschaften unklar blieben, nicht existiert. Ferner wird der Fluss des dynamischen Zeitskalenverhältnisses w = G/l berechnet und gezeigt dass der asymptotische Bereich nur sehr langsam erreicht wird. Das bedeutet, dass auch für die Fälle n = 2,3, wo die erhaltene Dichte entkoppelt und damit die Asymptotik durch Model A bestimmt wird, Model C relevant ist. [1] E. Brezin und De Dominicis, Phys. Rev. B 12, 4954 (1975); V. Dohm, Phys. Rev. B 44, 2697 (1991) [2] B. I. Halperin, P. C. Hohenberg und S.-K. Ma, Phys. Rev. B 10, 139 (1974); 13, 4119 (1976)

Zusammenhang von a- und (langsamer) b-Relaxation in Glasbildnern

•Manfred Winterlich und Roland Böhmer
Experimentelle Physik III, Universität Dortmund, 44221 Dortmund

Bei einigen Glasbildnern existiert neben der a-Relaxation eine (langsame) b-Relaxation. Durch stimulierte Deuteronenechos kann die a-Relaxationszeit im Bereich von Millisekunden bis Sekunden gemessen werden. Hier wird gezeigt, dass bei manchen Glasbildnern (z. B. ortho-Terphenyl) eine Verringerung der Relaxationszeit des stimulierten Echos auftritt, falls vor dem Echoexperiment nicht der Kernspingleichgewichtszustand vorliegt. Die Relaxation in das Gleichgewicht des Spinsystems wird durch Prozesse auf der Mikrosekundenzeitskala (Larmorfrequenz) bewirkt, in der die (langsame) b-Relaxation liegt, wenn die a-Relaxation einige Millisekunden beträgt. Es wird gezeigt, dass die Beschleunigung der stimulierten Echorelaxation durch Annahme einer Relaxationszeitenverteilung qualitativ beschrieben werden kann, falls man eine Konnektivität zwischen a- und b-Relaxation unterstellt.

Einfluss der anharmonischen Wechselwirkung auf den Boson-Peak"

•Edith Maurer¹, Walter Schirmacher¹ und Markus Pöhlmann^1,2
1Phys.-Dept. E13, TU MMünchen, 85747 Garching
²Institut Laue-Langevin, Grenoble

Mit feldtheoretischen Methoden haben wir eine Mean-Field-Theorie für das Schwingungsspektrum ungeordneter und schwach anharmonischer Festkörper entwickelt. Im harmonischen Limes ergibt sich die niederfrequente Erhöhung der Schwingungszustandsdichte bezüglich des Debye-Gesetzes (Boson-Peak) als Folge der Unordnung in den elastischen Konstanten. Mit ansteigender Unordnung (Varianz der fluktuierenden elastischen Konstanten) verstärkt sich der Boson-Peak und verschiebt sich zu niedrigeren Frequenzen, wie es im Experiment beobachtet wird. Die Anwesenheit anharmonischer Wechselwirkung führt zu einem Anstieg des Spektrums mit der Temperatur unterhalb des Boson-Peaks, ebenfalls in Übereinstimmung mit den Experimenten. Die Theorie sagt einen Übergang zu einer nichtergodischen Phase voraus. Der Boson-Peak lässt sich als Precursor-Phänomen dieses Übergangs interpretieren. Die Bosonpeakeigenschaften der Zustandsdichte übertragen sich auf die Temperaturabhängigkeit der spezifischen Wärme. Unterhalb des Boson-Peaks sagt unsere Theorie eine anomale, von der Debye-Theorie abweichende Temperaturabhängigkeit voraus.

Identification of Ion Sites in Glasses from Equilibrium Molecular Dynamics Configurations

•Renata Brosi\'nska¹, Junko Habasaki² und Philipp Maass^1
2Tokyo Institute of Technology, Japan
¹Institut für Physik, Technische Universität Ilmenau, Germany

For an understanding of the complex ion dynamics in glasses it is important to identify and characterise the sites that are occupied by the mobile ions during their jump motion through the glassy network [1]. Recently this task was undertaken by means of molecular dynamics simulations [2,3]. In these simulations the number of ion sites and their occupation probabilities were determined by examining how many times a small region (cell) in the simulation box is occupied by a mobile ion. Here we investigate the question if an identification of the same sites is possible based on equilibrium configurations only. To this end we perform an analysis of the ionic potential energy surface with respect to the minima and their connectivity features.

[1] P. Maass, J. Non-Cryst. Solids 255, 35 (1999)
[2] H. Lammert, M. Kunow, and A. Heuer, Phys. Rev. Lett. 90, 215901 (2003)
[3] J. Habasaki and Y. Hiwatari, Phys. Rev. B, in print

Vacancy model for the mixed alkali effect

•Robby Peibst, Stephan Schott, Hartmut Grille und Philipp Maass
Institut für Physik, Technische Universität Ilmenau, Germany

In ionically conducting glasses, all properties associated with long-range ionic motion show large deviations from a simple additive behavior upon mixing of two different types of mobile ions. This phenomenon is called the mixed alkali effect and is of particular theoretical interest due to its crucial role for the general understanding of ion transport in glasses. Recent molecular dynamics simulations and further general considerations suggest that the number of vacant sites in the glassy network is rather small. Motivated by these findings and by taking into account the existence of different energies of the ion sites [3], we discuss possible szenarios to explain the mixed alkali effect by a vacancy mechanism.

[1] H. Lammert, M. Kunow, and A. Heuer, Phys. Rev. Lett. 90, 215901 (2003)
[2] J. Habasaki and Y. Hiwatari, Phys. Rev. B, in print
[3] Ph. Maass, J. Non-Cryst. Solids 255, 35 (1999)

Magnification control of concave/convex learning in self-organizing neural maps and neural gas

Thomas Villmann¹ und •Jens Christian Claussen^2
1Cl. for Psychotherapy, University Leipzig
²Theoretical Physics, University Kiel

The self-organizing neural maps (SOM) by Kohonen and the neural gas (NG) by Martinetz are paradigms of neural winner-take all feedforward computation and are widely applied for vector quantization tasks. A data representation maximizing mutual information corresponds to an magnification exponent of 1, which is not reached by SOM and (finite-dimensional) NG. Recently we presented magnification approaches for NG by an additive winner-relaxing term [1] (extending an approach for SOM [2,3,4]) as an alternative to localized learning [4].

In this work [5] an alternative learning rule, the concave/convex learning [6], is investigated with respect to its magnification behaviour. We present an analytical derivation of the magnification exponent for SOM (in 1D) and for NG in arbitrary dimensions from a continuum theory. The results are validated by numerical calculation of the entropy for a standardized test system. We observe a dependence on the learning exponent in agreement with the theoretical predictions.

[1] J.C.C. and T.V., Proc. ESANN 2003; J.C.C. and T.V. (subm.)

[2] J.C.C. cond-mat/0208414; [3] J.C.C. Complexity 8(4), 15 (2003).

[4] J.C.C.,

Math.Mod.Comp.Biol.Med. p.17, ed.V.Capasso, Bologna, 2003.

[5] M. Hermann, T.V., Proc. ICANN 1997.

[6] T.V. and J.C.C., Proc. WSOM 2003; T.V. and J.C.C. (subm.)

[7] Y.Zheng, J.F.Greenleaf, IEEE Trans. Neur. Netw. 7, 87 (1996).

The Dynamics of Learning Vector Quantization

•Michael Biehl^1,2, Ansgar Freking² und Georg Reents^2
1Institute for Mathematics and Computing Science, University of Groningen, P.O. Box 800, NL-9700 AV Groningen
²Institut fuer Theoretische Physik und Astrophysik, Universitaet Wuerzburg, Am Hubland, D-97074 Wuerzburg

Learning Vector Quantization (LVQ) is a widely used method for the implementation of classification schemes, in particular in the context of structured (clustered) data. Here, we analyse LVQ by means of the statistical physics theory of on-line learning in high-dimensional weight space.

We show how variations of LVQ can be interpreted as interpolating between unsupervised clustering (without a teacher) and the supervised learning of a rule. The focus is on the phenomen of symmetry breaking phenomena and the role of initial conditions for the learning process.

Parameters estimation of hidden diffusion processes

•Abdelhadi Benabdallah und Günter Radons
TU Chemnitz, Institut für Physik , D-09107 Chemnitz, Germany

Diffusion processes, continuous in time and space, often cannot be observed directly but only via functions reflecting the measurement process. Within this work we investigate the parameter estimation problem for hidden diffusion processes in a periodic potential. We propose a discrete model as an approximation of continuous partially observed diffusion Processes. Using the EM (Expectation-Maximisation) algorithm we estimate the drift and the diffusion parameters which characterise the diffusing state variable via an approximation by a discrete random walk.A comparison with the particle filter are made. Numerical results confirm our analytical estimation.

Duality Transformation for Quantum Ratchets

•Joël Peguiron^1,2 und Milena Grifoni^2
1Departement of Nanoscience, Delft University of Technology, 2628 CJ Delft, The Netherlands
²Institut für Theoretische Physik, Universität Regensburg, 93040 Regensburg

We investigate a duality transformation between weak-binding models and tight-binding models for quantum ratchet systems. Ratchets are periodic structures with broken spatial symmetry yielding the possibility of a directed current in presence of noise and unbiased external forces. The duality transformation would allow to relate results that we obtained on few-bands quantum ratchets in the incoherent tunneling regime [1] with existing results on weakly corrugated ratchets [2]. It would also make possible to extend the parameter range of applicability of these results and especially check their classical limit. We are also interested to see how the symmetry conditions under which the ratchet current vanishes in both models map onto each other.

[1] M. Grifoni et al., PRL 89, 146801 (2002)

[2] S. Scheidl and V. M. Vinokur, PRB 65, 195305 (2002)

Velocity fluctuations in ratchets

•Lukasz Machura^1,2 und Peter Hänggi^1
2Department of Theoretical Physics, Institute of Physics, University of Silesia, Poland
¹Theoretishe Physik I, Institut für Physik, Universität Augsburg, Germany

The quantity of foremost interest in the context of transport in periodic systems, like Brownian ratchets, is the mean velocity of the particle. However, the quality of the transport depends also on some other characteristics like effective diffusion or shape of the potential. In this work we use simulations to investigate the velocity correlation function in the inertia Brownian ratchet. The connection between its behavior and the optimal transport conditions is presented.

Dynamics and interaction of dissipative solitons

•Hendrik U. Bödeker¹, Andreas W. Liehr¹, Till D. Frank², Rudolf Friedrich², Ulf-Hendrik Hansen¹ und Hans-Georg Purwins^1
1Westfälische Wilhelms-Universität Münster, Institut für Angewandte Physik, Corrensstr. 2/4, 48149 Münster
²Westfälische Wilhelms-Universität Münster, Institut für Theoretische Physik, Wilhelm-Klemm-Str. 9, 48149 Münster

We present new results concerning the investigation of the dynamics and the interaction of dissipative solitons occurring as current filaments in a planar DC gas-discharge system with high-ohmic barrier. The dissipative solitons exhibit interesting particle-like properties like generation and annihilation, scattering, and formation of molecules. Processes in which the number of solitons stays preserved can be understood by analyzing the dynamics using means of stochastic data analysis. In this way, both intrinsically standing and propagating solitons can be found, corresponding to Brownian and active Brownian particles [1]. In addition, a distance-dependant interaction law between mutually interacting dissipative solitons can be determined, quantitatively explaining the experimentally observed interaction phenomena. Generation and annihilation processes taking place on a short time scale can be observed using fast cameras, revealing different basic scenarios.

[1] H. U. Bödeker, M. C. Röttger, A. W. Liehr, T. D. Frank, R. Friedrich and H.-G. Purwins, Noise-covered drift bifurcation of dissipative solitons in a planar gas-discharge system, PRE 67 (2003) 056220.

Characterisation of atrial fibrillation by analysis of ECG records

•Th. Hennig¹, A. Heinen², St. Heinrichs², D. Jäger³ und Ph. Maass^1
1Institut für Physik, Technische Universität Ilmenau, Germany
²Fachbereich Physik, Universität Konstanz, Germany
³Medizinische Klinik II - Kardiologie, Städtisches Krankenhaus Friedrichshafen, 88048 Friedrichshafen

We investigate the heartbeat of patients with atrial fibrillation (AF), which is a cardiac arrhythmia where the normal pacemaking by the sinus node is dysfunctional. The analysis of AF mechanisms and characteristics is an important task, since it allows one to decide for the right diagnosis and to improve possible therapies. Because standard ECG records taken in hospitals are influenced by many different processes, it is difficult to extract information from the ECG, which is particularly sensitive to AF. We present novel methods to tackle this problem by employing time series analysis and modeling, where AF is described by a stationary process of irregular electrical excitations and the ventricular excitations result from certain properties of the AV node. Our techniques allow us to identify physiological parameters that otherwise can only be determined invasively.

Stochastic resonance in colloidal systems

•Dusan Babic^1,2, Carmen Schmitt¹, Igor Poberaj² und Clemens Bechinger^1
12. Physikalisches Institut, Universität Stuttgart, Pfaffenwaldring 57, 70550 Stuttgart
²Faculty of mathematics and physics, University of Ljubljana, Jadranska 19, 1000 Ljubljana, Slovenia

The concept of stochastic resonance (SR) has been used to describe very different effects such as the periodic occurence of the ice ages, the feeding behavior of paddlefish or human balance control and visual perception. The essential feature of SR is that in nonlinear systems noise can improve the detection of weak periodic signals.

Although characteristic features of SR have been observed in several experiments, there are only few examples where the effect is studied on a microscopic level. Colloids - mesoscopic particles which are coupled to a thermal bath and therefore undergo Brownian motion - are a system with intrinsic noise where the time and length scales are conveniently accessible with experiments. As such they provide an excellent model system for fundamental experiments of SR. We investigate the dynamics of a colloidal particle in a double-well potential, the latter being created by optical tweezers. For asymmetric modulation of the potential (corresponding to a tilt) we observe the typical signatures of SR in the residence time distribution and in the power spectrum. In contrast, if we only modulate the height of the potential barrier we do not observe any synchronization between the modulation and the particle dynamics.

Exact Analytical Solutions and Unpredictability of the Lorenz System

•Oliver Strebel
Handjerystr. 31, 12159 Berlin

For the Lorenz equations [1] exact analytical solutions are given. Special care was taken to construct graphs of solution curves by evaluating the analytical formula up to the accuracy of the floating point arithmetic of numbers in the used computer system. In addition to an analytical check the solution and its time derivative are evaluated on a computer and the difference between the left and right hand side of the Lorenz equations is calculated. This results in a tiny difference of several hundreds times the floating point precision constant stemming from different rounding errors during the evaluations of the formulas for the solution and its time derivative.

For the chaotic case of the Lorenz system it is demonstrated that for a given accuracy of the initial conditions it becomes undetermined after a few time units whether the system will reside in the left or right half space having a x coordinate less or greater than zero. This indicates fast transition to unpredictability of the system. The dynamics has for large regions of initial conditions a close resemblence to a Smale horseshoe [2], where adjacent initial conditions generate completely different trajectories.

[1] C. Sparrow, The Lorenz Equations: Bifurcations, Chaos and Strange Attractors, Springer AMS 41, Berlin (1982).

[2] S. Wiggins, Global Bifurcations and Chaos, Springer AMS 73, Berlin (1988).

Online-Simulation von Fahrzeugen auf komplexen Straßennetzwerken mit einem Zellularautomatenmodell

•Florian Mazur, Roland Chrobok, Sigurdur F. Hafstein, Andreas Pottmeier und Michael Schreckenberg
Universität Duisburg-Essen, Physik von Transport und Verkehr, Lotharstr. 1, 47048 Duisburg

Das Konzept der Online-Simulation ist eine Abbildung des realen Verkehrs im Computer. Dabei wird jedes Fahrzeug mit Hilfe von gemessenen Verkehrsdaten virtuell nachgebildet. Sehr effektive Modelle mit verhältnismäßig wenig Rechenaufwand sind die so genannten Zellularautomaten-Modelle. Mit diesen Modellen ist es möglich, den Verkehr realistisch zu beschreiben und zu analysieren. Ein Ansatz, der auf dem von Nagel und Schreckenberg 1992 veröffentlichten Verkehrsmodell basiert, ist mittlerweile in der Lage, den kompletten Verkehr in einem komplexen Netzwerk realitätsgetreu zu simulieren. Um den Grad an Realität weiter zu erhöhen, wird der simulierte Verkehr mit Online-Daten abgeglichen, welche durch Induktionsschleifen jede Minute gemessen werden. Damit kann auch auf unvorhersehbare Ereignisse (z.B. Unfälle) reagiert werden.

Konzeption und Anwendung des Verkehrsinformationssystems OLSIM

•Florian Mazur, Roland Chrobok, Sigurdur F. Hafstein, Andreas Pottmeier und Michael Schreckenberg
Universität Duisburg-Essen, Physik von Transport und Verkehr, Lotharstr. 1, 47048 Duisburg

Seit Dezember 2003 kann man neben dem aktuellen Verkehrszustand und der 30-Minuten-Prognose im Internet unter www.autobahn.nrw.de eine 60-Minuten-Prognose des Autobahnverkehrs in Nordrhein-Westfalen abrufen. Hinter den Farben grün für freien Verkehr bis rot für Stau verbirgt sich eine komplizierte nichtlineare Dynamik. Da die Autobahnen in NRW nur lokal mit Zählschleifen ausgestattet sind, werden die Verkehrsstufen für das komplette Netzwerk in einer Simulation per Computer berechnet. Ebenso wird eine Verkehrsprognose aus einer in mehrfacher Echtzeit laufenden Simulation und Erfahrungswerten aggregiert. Unfälle, Baustellen und Spursperrungen beeinflussen diese Hochrechnungen, wie auch das Verhalten der mittlerweile über 300.000 Nutzer pro Tag, die ihre Reiseplanung (Routenwahl, Reisezeitpunkt) nach den veröffentlichten Prognosen erstellen. Inwieweit sich Staus in NRW zurückgebildet oder nur verlagert haben ist Gegenstand gegenwärtiger Forschungen.

Comparison of different coupling schemes for time-delayed feedback control

•Andreas Borck¹, Wolfram Just² und Günter Radons^1
1Institute of Physics, Chemnitz University of Technology, Chemnitz, Germany
²Department of Mathematics, Queen Mary/University of London, London, UK

For time-independent target states we investigate the control performance of time-delayed feedback control for different coupling schemes of the control force. Applying linear stability analysis we obtain, in dependence on the coupling scheme and the properties of the target state, the corresponding transcendental characteristic equation. Properties of the eigenvalue spectra are discussed by analytical methods. In this way an optimal control scheme can be designed.

Improvement of time-delayed feedback control by oscillating feedback

Andreas Fichtner¹, Wolfram Just², Günter Radons¹, •Andreas Fichtner¹ und Wolfram Just^2
1Institute of Physics, Chemnitz University of Technology, Chemnitz, Germany
²Department of Mathematics, Queen Mary/University of London, London, UK

By imposing time dependent modulation on the control amplitude the performance of time-delayed feedback control may be enhanced. The relative phase between the control loop and the target state plays the crucial role. We investigate these features by analytical methods in a simple time discrete model. The linear stability analysis clearly shows the dependence of the control properties on the relative phase of the periodic orbit.

Erregungswellen in quasi-zweidimensionalen, gekrümmten Reaktions-Diffusions-Systemen

•Niklas Manz^1,2 und Stefan C. Müller^2
1Institut für Experimentelle Physik, Otto-von-Guericke-Universität, D-39106 Magdeburg
²Department of Chemistry and Biochemistry, Florida State University, Tallahassee, FL 32306-4390, USA

Die chemische Belousov-Zhabotinsky-Reaktion stellt ein ideales System zur Untersuchung der Dynamik von Reaktions-Diffusions Mustern dar, die in den unterschiedlichsten biologischen, chemischen und physikalischen Systemen zu beobachten sind. Obwohl diese propagierenden Strukturen auch in gekrümmten Medien wie der Herzoberfläche oder der Retina vorkommen, befaßten sich die experimentellen Untersuchungen überwiegend mit planaren raumzeitlichen Strukturen. Wir stellen ein Verfahren zur Herstellung quasi-zweidimensionaler, nichtebener Medien vor [1]. Mit Hilfe dieser gleichmäßig bzw. nichtgleichmäßig gekrümmten Systemen werden propagierende Wellen und Spiralen in verschiedenen nichtebenen Systemen untersucht. Die experimentellen Ergebnisse werden im Rahmen der kinematischen Theorie der Ausbreitung von Reaktions-Diffusions-Wellen erörtert.

[1] N. Manz and S.C. Müller, Fabrication of quasi-two-dimensional, heterogeneously curved Belousov-Zhabotinsky systems, Rev. Sci. Instrum. 74(12) (2003), in print

Lyapunov Instabilities in Lennard-Jones Fluids

•Hongliu Yang und Günter Radons
Institut für Physik, Theoretische Physik I, Technische Universität Chemnitz, D-09107 Chemnitz

We present new results on Lyapunov spectra and Lyapunov vectors for Lennard-Jones fluids. The density and temperature dependence of these quantities is discussed in detail. The spatio-temporal behavior of the hydrodynamic Lyapunov modes is obtained by studying its dependence on the Lyapunov exponent, frequency and wave number. In contrast to this wave-like behavior, the Lyapunov vectors for large Lyapunov exponents show localized behavior in phase space, as is seen by evaluating the corresponding participation ratios.

Hydrodynamic Lyapunov Modes in Coupled Map Lattices

•Hongliu Yang und Günter Radons
Institut für Physik, Theoretische Physik I, Technische Universität Chemnitz, D-09107 Chemnitz

We present numerical results, which indicate that hydrodynamic Lyapunov modes also exist for Coupled Map Lattices. We discuss under which conditions these modes can be observed. The role of conserved quantities, dissipation and of boundary conditions is elaborated. The questions are investigated for standard maps and circle maps with different types of couplings. We find, for instance, that diffusive couplings do not support hydrodynamic Lyapunov modes. In contrast, for force-like interactions between maps such modes can be found.

The OFC earthquake model in d = 1 dimension

•Felix Wissel und Barbara Drossel
Institut für Festkörperphysik TU-Darmstadt Hochschulstrasse 6 64285 Darmstadt

We present analytical and numerical results for the one-dimensional version of the earthquake model by Olami, Feder and Christensen. In the analytical part, we prove that trajectories in state space approach each other when they have the same toppling sequence. This implies that the dynamics of the model cannot show chaos. In the numerical part, we study the transition time into the stationary state as function of the system size N and the coupling parameter a, and the statistical properties of the stationary state.

Magnetic Billiards as Hamiltonian Ratchets

•Manamohan Prusty und Holger Schanz
Max-Planck Institut für Strömungsforschung und Institut für Nichtlineare Dynamik der Universität Göttingen, Bunsenstr. 10, 37073 Göttingen

Hamiltonian ratchets are periodic systems which show directed and ballistic transport due to the presence of a mixed regular and chaotic phase space and a mechanism breaking time-reversal symmetry. We show that certain billiard chains can be a paradigm for this type of behaviour. A magnetic field perpendicular to the billiard plane separates in a suitable geometry regularly skipping trajectories from chaotic ones. Both sets of trajectories transport ballistically in opposite directions. We show how one can apply a classical sum rule for ratchet transport to predict the chaotic transport velocity analytically and confirm the result by numerical simulations. We study also the quantized versions of the billiard ratchets and discuss their spectral properties.

[1] H. Schanz et al., Phys. Rev. Lett. 87(01)070601.

Traveling waves in a reaction-diffusion system under periodic forcing

•E. P. Zemskov¹, K. Kassner¹ und S. C. Mueller^2
1Institut fuer Theoretische Physik, Otto-von-Guericke-Universitaet, Universitaetsplatz 2, 39106 Magdeburg
²Institut fuer Experimentelle Physik, Otto-von-Guericke-Universitaet, Universitaetsplatz 2, 39106 Magdeburg

A one-component reaction-diffusion system under external force is considered. The simplest case of a periodic forcing of cosine type is chosen. Exact analytical solutions for the two basic types of traveling waves, fronts and pulses, are obtained in the case of a piecewise linear approximation of the non-linear reaction term. Velocity equations are derived from the matching conditions. Restrictions that arise during the derivation of the pulse velocity equations are stated and their origin is explained. It is found that in the presence of nonconstant forcing there exists a set of wave solutions with different phases (matching point coordinates). The general characteristic feature is that nonmoving waves become movable under forcing. However, for specific choices of forcing parameters, the traveling waves are pinned (stopped). The pinning conditions are obtained and discussed. It is found that in the case of periodic forcing there are infinite sets of the pinning positions. The phase portraits of specific types of solutions are shown and briefly discussed.

References

E. P. Zemskov, K. Kassner, S. C. Mueller, Eur. Phys. J. B 34, 285 (2003).

Dynamics of colloidal particles in a toroidal optical trap

•Michael Reichert und Holger Stark
Universität Konstanz, Fachbereich Physik, D-78457 Konstanz, Germany

A theoretical study of the collective non-Brownian motions of colloidal particles circulating in a toroidal trap is presented. We consider equal-sized spheres in the regime of low Reynolds numbers whose interactions are solely of hydrodynamic origin. The concrete physical situation we have in mind are optical vortices, i.e., focussed helical light modes carrying an orbital angular momentum [1]. We mimick this in our simulations by applying a tangential driving force and a harmonic radial trap force on each particle.

The orbital velocities of the circulating particles depend on the particle number and the ring radius. We analyze them for different arrangements, e.g., regular symmetric or random configurations. It turns out that the random configurations lead effectively to a higher velocity compared to the regular arrangements due to ``drafting'' effects.

Furthermore, we study the time-dependent dynamics of the random configurations. The simulations show that, after an irregular transient regime, the system tends towards a periodic dynamic state of oscillating particle distances. During this synchronisation process, the viscous drag is lowered.

[1] J. E. Curtis, D. G. Grier, Phys. Rev. Lett. 90, 133901 (2003).

Dissipative Solitonen-Moleküle in Reaktions-Diffusions-Systemen

•A. W. Liehr, M. C. Röttger und H.-G. Purwins
Westfälische Wilhelms-Universität Münster, Institut für Angewandte Physik, Corrensstr. 2/4, 48149 Münster

Dissipative Solitonen sind großamplitudige lokalisierte Strukturen mit ausgeprägten Teilcheneigenschaften, die in diesem Beitrag anhand eines dreikomponentigen Reaktions-Diffusions-Systems untersucht werden. In der Nähe der Drift-Bifurkation kann die Dynamik und Wechselwirkung der dissipativen Solitonen mittels einen Teilchenansatzes durch Ordnungsparametergleichungen beschrieben werden, wobei als relevante Größen die Positionen und die Propagatormodenamplituden der dissipativen Solitonen erfasst werden [1]. Ausgehend von dieser reduzierten Dynamik diskutieren wir die Eigenschaften gebundener Zustände dissipativer Solitonen, sogenannte dissipative Solitonen-Moleküle, wobei sowohl dynamische Eigenschaften starrer Moleküle [2,3], als auch innere Freiheitsgrade größerer Moleküle behandelt werden.

[ 1] Bode, M. et al.: Interaction of dissipative solitons: particle-like behaviour of localized structures in a three-component reaction-diffusion system. Physica D 161, 2002, S. 45ff.

[ 2] Moskalenko, A. S. et al.: Rotational bifurcation of localized dissipative structures. Europhysics Letters 63, 2003, S. 361ff.

[ 3] Liehr, A. W. et al.: Transition from stationary to rotating bound states of dissipative solitons. In: E. Krause (Hrsg.), W. Jäger (Hrsg.), M. Resch (Hrsg.): High Performance Computing in Science and Engineering '03, Springer 2003, S. 225ff.

Universal behavior in complex front systems

•Andreas Amann, Andreas Wacker und Eckehard Schöll
Institut für Theoretische Physik, Technische Universität Berlin, Hardenbergstrasse 36, D-10623 Berlin

We show that the bifurcation scenario in a high-dimensional system with interacting fronts can be related to the universal U-sequence which is known from the symbolic analysis of iterated one-dimensional maps. This connection is corroborated for a model of a semiconductor superlattice [1], which describes the complex dynamics of electron accumulation and depletion fronts. By a suitable Poincaré section we reduce the dynamics to a low dimensional iterated map, for which in the most elementary case the bifurcation points can be determined analytically.

[1] A. Amann, K. Peters, U. Parlitz, A. Wacker, E. Schöll Phys. Rev. Lett. 91, 066601 (2003).

The effect of long-term correlations on clustering of extreme events

•Jan F. Eichner¹, Armin Bunde¹, Jan W. Kantelhardt¹ und Shlomo Havlin^2
1Institut für Theoretische Physik III, Justus-Liebig-Universität, Giessen, Germany
²Dept. of Physics and Minerva Center, Bar-Ilan University, Ramat-Gan, Israel

Long-term correlations, indicated by a power-law decay of the auto-correlation function C(s)  ~  s^-g, appear in many natural records (e.g. temperatures, river flows, and heartbeat intervals). In uncorrelated data the return intervals r between events above a certain threshold q are uncorrelated and follow the Poissonian statistics. We show that in the presence of long-term correlations, the return intervals are also correlated with the same correlation exponent g, and follow a stretched exponential distribution. As a consequence the extreme events show a tendency of clustering. We give many examples of observational and reconstructed data where this behavior is observed. The long-term correlations may present a natural mechanism for the observed clustering of hazardous floods between 1300 and 2000.

Transport von Sauerstoff in dünne Flüssigkeitsfilme

•Marcus J. B. Hauser und Stefan C. Müller
Otto-von-Guericke-Universität, Institut für Experimentelle Physik, Abteilung Biophysik, Universitätsplatz 2, 39106 Magdeburg

Der Transport von Sauerstoff aus der Gasphase in 1,5 mm dicken Filme aus Schwefelsäure bzw. Wasser wird untersucht. Dazu wird der zeitliche Verlauf der Sauerstoffkonzentration in den Flüssigkeitsfilmen höhenabhängig (im Abstand von 0,1 mm) gemessen. Daraus werden die Konzentrationsprofile berechnet. Während der Transport von Sauerstoff in einen Wasser-Film über Diffusion erfolgt, ist die Situation im Falle der Schwefelsäure-Filme komplexer. Der Sauerstofftransport in Schwefelsäure zeigt oberfl#achenspannungsgetriebene (Marangoni) Konvektion, die in den ersten 0,5 mm von der Grenzfläche (gasförmig / flüssig) auftritt. Der Transport in tieferen Schichten des Schwefelsäure-Films (0,5 mm leq h £ 1,5 mm) lässt sich gut durch Diffusion beschreiben. Über die Ursachen für dieses Verhalten wird berichtet.

Dynamics and Control of a 2-D Passive Dynamic Walker

•Joachim Haß¹, Norbert Mayer² und Michael Herrmann^1
2GMD - Japan Research Laboratory Collaboration Center, Kitakyushu, Japan
¹Max-Planck-Institut für Strömungsforschung, Göttingen, Germany

From the biomechanical point of view, bipedal locomotion can be considered as the dynamics of a set of rigid bodies, connected by joints, springs, and specifically placed actuators. A passive walker [1] is an even more simplified model which is capable of stable walking on a downhill slope with no actuation or control whatsoever.

Combining experimental, numerical and theoretical methods, stability and speed of a modified walker closer to human gait dynamics were examined. Both of them crucially depended on precise tuning of the walkers internal parameters, which were optimized using a genetic algorithm in the numerical simulations.

The basins of attraction in parameter space could be significantly broadened by incorporating actuators and nonlinear control algorithms into the model. Aside from increasing the walker's stability, this also enabled it to walk very efficiently on arbitrarily shallow slopes at a range of different speeds. Furthermore, we considered more biologically motivated forms of control, like synchronizing the body movements with a central pattern generator.

[1] McGeer T. (1993) Int. J. Robotics Res. 9:2, 62-82.

Calculation of periodic orbits and spectral correlations in pseudointegrable systems

•Jesper Mellenthin und Stefanie Russ
Institut für Theoretische Physik III, Universität Giessen, D-35392 Giessen, Germany

We investigate pseudointegrable quantum billiards of different genus numbers g. We calculate numerically the periodic orbits and compare them to the Fourier transform of the density of states. We find a very good agreement between both. From the periodic orbits we calculate the spectral correlation of the corresponding quantum system, i.e. the spectral rigidity D₃. We study how D₃ depends on the energy and investigate the difference between the diagonal approximation and the full semiclassical expression over diagonal and non-diagonal terms. We find that D₃ for pseudointegrable systems is intermediate between integrable and chaotic systems and that the contribution of very long orbits can be neglected.

Resonance-assisted tunneling in the kicked rotor

•Christopher Eltschka und Peter Schlagheck
Institut für theoretische Physik, Universität Regensburg

We investigate tunneling in the quantum kicked rotor, which manifests itself in small eigenphase splittings between symmetric and antisymmetric combinations of ``left-moving'' and ``right-moving'' quasi-modes. Our special interest lies in the role of nonlinear resonances, which appear as island chains in the classical phase space and which induce couplings that substantially enhance the tunneling process [1]. Semiclassical reproductions of the splittings, which are based on the resonance-assisted coupling mechanism, are compared with the exact quantum splittings, which are calculated by using multiple precision arithmetics.

[1] O. Brodier, P. Schlagheck, and D. Ullmo, Phys. Rev. Lett. 87, 064101 (2001), Ann. Phys. 300, 88 (2002).

Level dynamics in pseudointegrable billiards: an experimental study

•Yuriy Hlushchuk¹, Stefanie Russ¹, Ulrich Kuhl² und Hans-Jürgen Stöckmann^2
1Institut für Theoretische Physik III, Universität Giessen, D-35392 Giessen, Germany
²Fachbereich Physik, Philipps-Universität Marburg, Renthof 5, D-35032 Marburg, Germany

The level dynamics of pseudointegrable systems with different genus numbers g was studied experimentaly using microwave cavities. The eigenvalue velocity distribution is found to be Gaussian, like in GOE systems, and shows no dependances on the genus number g of the system. Although the distributions of large curvatures in our pseudointegrable systems correspond to the distribution in chaotic systems, the small curvatures show the intermediate behavior between chaotic and integrable systems. With growing genus number g the distribution of small curvatures goes from that one for integrable systems in direction of the distribution for pure GOE systems.
According to the conjecture of Yukawa the parametric motion of the eigenvalues in a system can be treated as a motion of particles in a 1D gas. The level dynamics of an integrable system corresponds to a 1D gas without interaction between particles, while a chaotic level dynamics corresponds to a 1D gas with a long-range interaction. The model of a short-rage interaction between eigenvalue particles for the description of the level dynamics in pseudointegrable systems is discussed.

Soft wall quantum dots modelled with microwave cavities

•Y.-H. Kim¹, U. Kuhl¹, H.-J. Stöckmann¹ und J. P. Bird^2
1Fachbereich Physik, Philipps-Universität Marburg, Renthof 5, D-35032 Marburg, Germany
²Department of Electrical Engineering and Center for Solid State Electronics Research, Arizona State University, Tempe, AZ 85287-5706, USA

By an adiabatic change of the height of flat microwave cavities quantum billiards with soft wall potentials can be modelled [1]. Eigenresonances and eigenfunctions are investigated for an harmonic oscillator like billiard, serving as a test example, and a billiard corresponding to a quantum dot with soft walls [2]. For the harmonic oscillator a good agreement between measurements and theory is found. In case of the quantum dot billiard the behaviour of the resonances corresponding to bouncing ball states is investigated in detail.

[1] H.M. Lauber, Ph.D. thesis, Universität Heidelberg, 1994.

[2] de Moura et al, Phys. Rev. Lett. 88, 236804 (2002).

Distribution of reflection coefficients for microwave cavities in dependence of coupling and absorption strength

•U. Kuhl¹, H.-J. Stöckmann¹, R. A. Méndez-Sánchez² und C. H. Lewenkopf^3
1Fachbereich Physik, Philipps-Universität Marburg, Renthof 5, D-35032 Marburg, Germany
²Centro de Ciencias Físicas UNAM, A.P 48-3, 62210, Cuernavaca, Morelos, México
³Instituto de Física, UERJ, R. São Francisco Xavier 524, 20550-900 Rio de Janeiro, Brazil

The reflection spectra of rectangular microwave cavities, with a half circular inset attached to one of the walls, have been measured in dependence of the position of the inset. To cover the range from weak up to strong absorption, for part of the measurements one wall was coated with a layer of absorbing material. The distribution of reflection coefficients was investigated in dependence of the antenna coupling strength and the absorption within the cavities. The results are in good agreement with numerical calculations, as well as with the theoretical predictions for the limiting case of strong absorption [1]. In spite of the fact that the system is time-reversal invariant, it is found that Gaussian unitary ensemble calculations can be used to describe the results, if only the coupling parameter is rescaled by a factor of two [2].

[1] E. Kogan, P.A. Mello, and H. Liqun, Phys. Rev. E 61, R17 (2000).

[2] D.V. Savin and H.-J. Sommers, Phys. Rev. E 68, 036211 (2003).

TRANSPORT AND DYNAMICAL PROPERTIES OF THE COSINE BILLIARD.THEORY V.S. EXPERIMENT.

•G.A. Luna-Acosta^1,2, U. Kuhl¹, J.A. Mendez-Bermudez³ und H.-J. Stöckmann^1
1Fachbereich Physik der Philipps Universität Marburg, Renthof 5,D-35032
²Instituto de Fisica, BUAP, Puebla, Mexico.
³Max Planck Institute für Strömungsforschung, Göttingen.

We investigate the properties of waves and particles reflected by and transmitted through 2D chaotic waveguides, using the cosine billiard as a prototype. We calculate the scattering probability matrix elements \mid S_n,m\mid², wave functions, and Husimi distributions for a set of geometrical parameters producing weak, mixed, and strong chaotic ray dynamics. Comparison of these quantities with their classical counterparts allows us, on the one hand, to understand important global properties of waveguides in terms of purely classical dynamics and, on the other, to unveil drastic differences between wave and particle behavior. Further, we compare our predictions with experimental results obtained in a microwave realization of the cosine billiard.

How optical microresonators radiate

•Martina Hentschel¹ und Henning Schomerus^2
1Department of Physics, Duke University, Durham, NC 27708-0305, U.S.A.
²MPI Physik komplexer Systeme, Noethnitzer Str. 38, D-01187 Dresden

Optical microresonators have broad application potential in optical communication devices, and are interesting model systems in the field of Wave Chaos. Most of our qualitative understanding of their radiation characteristics rests on the simple ray-optics picture, which however requires a wave length much smaller than the geometric features of the system. This is not realized in the most interesting applications. We highlight effects that lead to substantial deviations from the ray picture. Notably, at curved interfaces, the Goos-Haenchen effect causes deviations from both Snell's and Fresnel's laws. We analyse the resulting deviations in the radiation characteristics using four appropriately defined Husimi functions, corresponding to incident and emerging rays on either side of the interface. These four phase-space representations of the electromagnetic wave function naturally allow to read-off the radiation directions, and replace the single Husimi function used in hard-wall systems.

Stochastic Webs and Localization in a Quantum Mechanical Kick System

•Ulf Martin Engel und Peter Eckelt
Institut für Theoretische Physik, Universität Münster, Wilhelm-Klemm-Straße 9, D-48149 Münster

The classical kicked harmonic oscillator is characterized by a very special scenario of weak chaos: in the case of resonance between the eigenfrequency of the harmonic oscillator and the frequency of the periodic forcing, periodic stochastic webs in phase space are generated [1]. We investigate the quantum dynamics [2] of this model system and show that the resulting HUSIMI distributions in quantum phase space exhibit the same web-like structures as in the classical case. The quantum dynamics is characterized by diffusive energy growth - just as the classical dynamics in the channels of the webs. In the case of nonresonance, the classically diffusive dynamics is quantum mechanically suppressed. We explain this bounded energy growth (which corresponds to localization in quantum phase space) by mapping the system onto the ANDERSON model, in a way that is motivated by the standard treatment of the quantum kicked rotor [3].

[1] G. M. Zaslavsky et al., Weak Chaos and Quasi-Regular Patterns, Cambridge University Press, Cambridge 1991. [2] G. P. Berman, V. Y. Rubaev and G. M. Zaslavsky, Nonlinearity, 4 543 (1991). [3] S. Fishman, D. R. Grempel and R. E. Prange, Phys. Rev. Lett. 49, 509 (1982). [4] U. M. Engel, Dissertation Universität Münster (2003).

Path integral approach to the dissipative curve-crossing dynamics

•Alexey Novikov^1,2, Ulrich Kleinekathöfer^1,2 und Michael Schreiber^2
1International University Bremen, Bremen, Germany
²Institut für Physik, Technische Universität Chemnitz, Chemnitz, Germany

The dynamics of a curve-crossing system is considered using path integrals in a combined phase space and coherent state representation. The relevant system consists of two one-dimensional potential energy surfaces, the dissipation is described by its interaction with the bath of non-interacting harmonic oscillators. The Lagrangian function of the electronic subsystem is written in the coherent state representation using mapping approach while the vibrational degrees of freedom are treated by the Feynman-Vernon path integral. The non-Gaussian part of the path integral is calculated by introducing the effective bias and hopping constants for the electronic states. Finally, the dynamics of a Gaussian wave packet and the population of electronic levels is analyzed within the presented method.

Non-Markovian density matrix theories based on the decomposition of the spectral density

•U. Kleinekathöfer^1,2 und M. Schreiber^2
1International University Bremen, 28725 Bremen, Germany
²Institut für Physik, Technische Universität Chemnitz, 09107 Chemnitz

For the description of dynamical effects in quantum mechanical systems on ultra-short time scales memory effects play an important role. Here we propose two different theories [1] which are either based on the time-convolution Nakajima-Zwanzig projection operator formalism or the time-convolutionless Hashitsume approach. The proposed theories as well as the method by Meier and Tannor [2] are based on a numerical decomposition of the spectral density. For the example of the damped harmonic oscillator these non-Markovian theories are compared among each other, to Markovian approaches neglecting the memory effect and to path integral calculations. Some of the proposed non-Markovian theories treat explicitly time-dependent system Hamiltonians nonperturbatively. Therefore these method can be used for the description of experiments with arbitrary large laser fields.

[1

PHASE BEHAVIOUR AND ELASTICITY OF COLLOIDAL MODEL SUSPENSIONS

•Larysa Shapran, Patrik Wette, Hans Joachim Schöpe und Thomas Palberg
Johannes Gutenberg-University Mainz, Institute of Physics, Staudinger Weg 7, D-55099 Mainz

We have measured the phase behaviour and the elasticity of crystallizing systems of submicron polystyrene spheres in low salt aqueous suspension. Microscopy and static light scattering (SLS) yield the fluid-crystal transition lines including the coexistence region in the particle number density n salt concentration c plane. SLS shows further that the crystal phase is b.c.c. Torsional resonance spectroscopy [1] yields the static shear modulus G, which is observed to increase with n at constant c=0 and to decrease at n=const with increasing c. We compare our phase behaviour predictions from computer simulation [2,3] using the effective charge Z*(n,c) derived from the elasticity measurement. Use of the elasticity effective charge instead of previously often employed conductivity charge improves the consistency significantly.

[1] H.J.Schöpe, T.Palberg, J. Colloid Interface Sci. 234, 149-161 (2001)

[2] M.O.Robbins, K.Kremer and G.S.Grest, J. Chem. Phys. 88, 3286 (1988)

[3] E.J.Meijer and D.Frenkel, J. Chem. Phys. 94, 2269 (1991)

Solid friction at model metal interfaces studied via NEMD computer simulations

•Igor Stankovic¹, Martin Kröger ² und Siegfried Hess^1
1Institute of Theoretical Physics, PN 7-1, Technical University of Berlin, Hardenbergstr. 36, D - 10623 Berlin
²Polymer Physics, ETH Zurich, ML H18, Sonneggstr. 3, CH-8092 Zurich

The embedded atom method is adopted to study structural changes and mechanical alloying at model metal interfaces during dry solid friction [1,2,3]. The main constitutive properties of real metals, i.e., cohesive energy, elastic constants, and heat of solution, are reproduced by a set of a few basic model parameters. The model metal is subjected to shear deformation and strong flow via non-equilibrium molecular dynamics in order to discuss the origins of some qualitative properties observed more specific embedded atom potentials. The mechanisms leading to mechanical alloying at interface are observed and discussed [3]. Information about crystal structure inside NEMD configurations is obtained with common neighbor analysis, based on planar graphs [4].

1. M. S. Daw and M. I. Baskes, Phys. Rev. Lett. 50, 1285 (1983).
Phys. Rev. B 29, 6443 (1984).
2. I. Stankovic, M. Kröger, and S. Hess, Phys. Rev. E, accepted 2003.
3. M. Kröger, I. Stankovic, and S. Hess, Multiscale Model. Simul. 1, 25-39 (2003).
4. I. Stankovic, M. Kröger, and S. Hess, Comput. Phys. Commun. 145, 371 (2002).

Monte-Carlo Simulations of Structural and Elastic Properties in a two dimensional Model Colloidal Crystal

•Kerstin Franzrahe und Peter Nielaba
Lehrstuhl für Theoretische Physik, Fachbereich Physik, Universität Konstanz, D-78457 Konstanz

With the help of Monte-Carlo Simulations the elastic and structural properties of a two dimensional hard disk system are studied. Simulations in the NPT ensemble are done in order to analyse lattice formation in binary mixtures, while simulations in the NVT ensemble enable us to calculate the elastic constants of the system from microscopic strain fluctuations [1]. This method was used to study the influence of impurities on the elastic properties of a two dimensional hard disk system.

[1] S. Sengupta, P. Nielaba, M. Rao, K. Binder Phys. Rev. E 61,1072 (1999)

Computer simulation studies of structures and elastic properties of model colloids

•Peter Henseler und Peter Nielaba
Lehrstuhl für Theoretische Physik, Fachbereich Physik, Universität Konstanz, D-78457 Konstanz

We perform classical Monte-Carlo simulations to calculate isothermal elastic constants of crystal phases of mono-disperse hard-sphere colloids. The elastic constants are determined from microscopic fluctuations of the instantaneous local Lagrangian strain e_ij(r, t), measured with respect to a static reference lattice, and the use of a finite-size scaling theory. This approach is a generalization of the method of Sengupta et al. [1] to three dimensions.

We furthermore investigate structures of three dimensional colloidal mixtures with different composition and diameter ratios.

[1] S. Sengupta, P. Nielaba, M. Rao, K. Binder, Phys. Rev. E 61, 1072 (2000)

Laserinduziertes Gefrieren und Schmelzen in 3D

•Wolfram Quester, Wolfram Strepp und Peter Nielaba
Fachbereich Physik, Universität Konstanz

Seit den achtziger Jahren des vorigen Jahrhunderts werden die Phänomene des laserinduzierten Gefrierens und Schmelzens in 2D untersucht. Man versteht darunter zwei Phänomene, bei denen sich ein flüssiges System von Kolloiden in seiner kristallinen Phase anordnet, wenn es einem periodischen äußeren Potential ausgesetzt wird. Wird die Amplitude des äußeren Potentials vergrößert, so geht das System in eine modulierte Flüssigkeit über.

Es wurde das Auftreten dieser Phänomene in 3D für zwei Wechselwirkungspotentiale untersucht. Einmal handelt es sich um harte Kugeln, einmal um Lennard-Jones-Teilchen. Als Indikatoren für den Phasenübergang dienten hauptsächlich der Strukturfaktor S([k\vec]) und die Paar-Korrelationsfunktion. Im Lennard-Jones-Fall konnte sehr gut das Gefrieren beobachtet werden. Allerdings ergeben sich keine Anzeichen für ein Schmelzen des Systems bei Erhöhung des externen Potentials. Bei den harten Kugen verläuft das Gefrieren fast identisch wie im Lennard-Jones-System, allerdings erreicht der Ordnungsparameter ein Maximum, um dann mit zunehmendem Potential wieder leicht abzufallen. Diese Abnahme des Ordnungsparameters reicht allerdings nicht aus, um von einem Schmelzen zu sprechen.

Structure formation in ultrathin films of soft matter on chemically (in)homogeneous reactive and non-reactive substrates

•Uwe Thiele¹, Michael Bestehorn², Karin John¹ und Markus Bär^1
1Max-Planck-Institut für Physik komplexer Systeme, Nöthnitzer Str.38, 01187 Dresden
²Lehrstuhl für Statistische Physik und Nichtlineare Dynamik, BTU-Cottbus, Erich-Weinert-Strasse 1, 03046 Cottbus

A continuos model for an ultrathin partially wetting liquid film on a solid substrate is employed to study the structure formation of such a film on a heterogeneous substrate. We describe (1) film profiles on periodically patterned substrates, (2) sliding drops on substrates with an externally imposed wettability gradient and (3) moving droplets on a reactive substrate.

Construction of an optical line tweezer for colloidal particles

•Raquel Chulia Jordan, Hans Joachim Schöpe und Thomas Palberg
Johannes Gutenberg-Universität Mainz, Institut für Physik, Staudingerweg 7, D-55099 Mainz

Line tweezers are a valuable tool for investigations of pair interactions in colloidal suspensions. We report on the construction of such a device implementing a beam positioning by an acousto-optic modulator, a set of counter-propagating beams with l=532nm and an imaging optics with depolarized back scattered light. Construction principle, calibration and performance are discussed in detail.

Swelling Behaviour of Hydrogels

•Bernward A. Mann¹, Ralf Everaers², Christian Holm¹ und Kurt Kremer^1
2Max-Planck-Institut für Physik komplexer Systeme; Nöthnitzerstr. 38; 01187 Dresden; Germany
¹Max-Planck-Institute für Polymerforschung; Ackermannweg 10; 55128 Mainz; Germany

The equilibrium swelling behaviour of a charged polymer network and the dependence of its properties on the system parameters is studied in numerical experiments. The results of these computer simulations is related to current theoretical models and the scaling laws predict therein.

The influence of confining disordered materials on the equilibrium behavior of Stockmayer fluids

•Carsten Spöler und Sabine H.L. Klapp
Stranski-Laboratorium für Physikalische und Theoretische Chemie, Sekr. TC 7, Fakultät II, Technische Universität Berlin, Straße des 17. Juni 124, 10623 Berlin, Germany

The phase behavior of fluids is very sensitive to the nature of intermolecular interactions as well as to the presence of confining material on a nanometer scale. In the present work, which is based on integral equation theory (Reference Hypernetted Chain Approximation) in combination with the replica method we consider dipolar Stockmayer fluids exposed to disordered porous materials. The latter is modeled by a) a hard sphere matrix, b) a hard sphere matrix with additional Lennard-Jones fluid-matrix attraction c) a dipolar hard sphere matrix. As shown on this poster, we find for all systems a remarkable shift of the boundaries of the homogeneous, isotropic high temperature phase relatively to the bulk[1,2]. Our previous studies, however, require a cumbersome treatment of the orientational dependence of the dipolar potential. It is known for bulk Stockmayer fluids that angle-averaged potentials like the Keesom potential lead to remarkable well predictions of the condensation transition (as long as the dipole moment remains relatively small). Here we test the performance of the angle-averaged potential for the confined systems [e.g. model a)], considering particularly the behavior of the isothermal compressibility and adsorption isotherms.

[1] C. Spöler and S.H.L. Klapp, J. Chem. Phys. 118, 3628 (2003)
[2] C. Spöler and S.H.L. Klapp, submitted

Crystallization in one and two component charged colloidal systems

•Patrick Wette, Hans Joachim Schöpe und Thomas Palberg
Institut für Physik der Universität Mainz

We investigated the nucleation kinetics of aqueous suspensions of Polystyrene spheres in the deionized state using static light scattering and Avrami theory. Phase behaviour and growth kinetics were determined in parallel from microscopy. For a mixture of 68nm and 100nm sized particles the phase diagram shows a pronounced minimum at 20% 68nm particles which is reflected in both crystallization kinetics and elasticity.

Crystallization in one and two component charged colloidal systems

•Patrick Wette, Hans Joachim Schöpe und Thomas Palberg
Institut für Physik der Universität Mainz

We investigated the nucleation kinetics of aqueous suspensions of Polystyrene spheres in the deionized state using static light scattering and Avrami theory. Phase behaviour and growth kinetics were determined in parallel from microscopy. For a mixture of 68nm and 100nm sized particles the phase diagram shows a pronounced minimum at 20% 68nm particles which is reflected in both crystallization kinetics and elasticity.

Ruin & Recreate - A Method to Overcome Barriers in the Energy Landscape

•Johannes J. Schneider
Department of Physics, Johannes Gutenberg University, Staudinger Weg 7, D-55122 Mainz

When simulating or optimizing a system with Monte Carlo techniques, usually the so-called Local Search approach is used in which an initial configuration is iteratively altered by small changes. Therefore, the application of such a small move leads to a new configuration in the neighborhood of the current configuration, which is quite similar to it. We showed in [1] that the application of larger moves leads to better results in shorter times than the classic Local Search concept. However, these larger moves we used did not lead to thermal equilibrium, such that they could only be used for optimization purposes but not for simulating a system at a given temperature. Now we have altered this approach, such that the moves lead to a Boltzmann equilibrium at each temperature. Moreover, as these larger moves change the configuration to a much larger extent than the small moves, we are able to overcome barriers in the energy landscape more easily. We will present computational results both for the classic Local Search and the new Ruin & Recreate approach for a colloidal system.

[1] G. Schrimpf, J. Schneider, H. Stamm-Wilbrandt, and G. Dueck, J. Comp. Phys. 159, 139, 2000.

HP Proteins on Generalized Lattices

•Thomas Vogel, Michael Bachmann und Wolfhard Janke
Institut für Theoretische Physik, Universität Leipzig, Augustusplatz 10/11, 04109 Leipzig, Germany

We investigated HP proteins [1] in three dimensions on simple cubic lattices and on the fcc lattice. Employing the nPERM algorithm [2] to selected sequences, we searched for ground states on both lattices and compared the density of states and the heat capacity of sequences which are designed on the simple cubic lattice.

PERM is a chain growth algorithm which combines the Rosenbluth method with the ``Go with the Winners'' strategy. Population control is implemented by cloning and pruning growing conformations depending on variable thresholds which are computed adaptively during the simulation.

[1] K.F. Lau, K.A. Dill, Macromol. 22 (1989) 3986.

[2] H.-P. Hsu, V. Mehra, W. Nadler, P. Grassberger, J. Chem. Phys. 118 (2002) 444.

Stable multiarmed spiral structure and their interaction in excitable media

•Maria Inmaculada Rodriguez-Ponce und Franz Schwabl
Lehrstuhl 5 fuer theoretische Physik, Physik Department, Technische Universitaet Muenchen (Germany)

We study the dynamic properties and interaction of stable multiarmed spiral structures in excitable media. The model is based on the equations that describe the early stage of development in the self-organization of Dictyostelium discoideum cells. We discuss the stability of the different patterns, focusing on the interplay between the maximum number of arms in a multiarmed spiral and the presence of local hetereogeneities (defects) as well as between excitability properties. The interaction between structures is analyzed in terms of core drifts and frequency changes.

Poisoning of membrane patches

•Gerhard Schmid, Igor Goychuk und Peter Hänggi
Institut für Physik, Universität Augsburg

We consider the influence of intrinsic channel noise on the spontaneous spiking activity of poisoned excitable membrane patches by use of a stochastic generalization of the Hodgkin-Huxley model. Internal noise stemming from the stochastic dynamics of individual ion channels is known to affect the collective properties of the whole cluster [1]. In particular, there exists an optimal size of the membrane patch for which solely the internal noise causes a most regular spontaneous generation of action potentials. This phenomenon is termed intrinsic coherence resonance. In addition to the variation of the size of ion channel clusters, the living organisms may adopt the densities of ion channels in order to regulate the spontaneous spiking activity. In our model, we vary the densities resp. the number of the specific ion channels for a given patch size. Experimentally, this task could be achieved, e.g., by blocking the ion channels of a given sort with a specific channel poison. Interestingly, by poisoning the potassium, or the sodium ion channels in a certain proportion one can increase, or decrease the regularity of the spike train [2]. This work is supported by DFG (SFB 486).

[1] J.A. White, J.T. Rubinstein, and A.R. Kay, Trends Neurosci. 23, 131 (2000).

[2] G. Schmid, I. Goychuk, and P. Hänggi, Physica A.

Assortative random networks: Construction and properties

•Ramon Xulvi-Brunet und Igor M. Sokolov
Newtonstr 15, D-12489 Berlin

Since recently it got clear that many social networks exhibit assortative mixing [M. E. J. Newman, Phys. Rev. Lett. 89, 208701 (2002)] so that the predictions of uncorrelated models might be inadequate. To analyze the role of assortativity we introduce an algorithm which changes correlations in a network and produces assortative mixing to a desired degree. This degree is governed by one parameter p; changing this parameter one can construct networks ranging from fully random (p = 0) to totally assortative (p = 1). We apply the algorithm to the Barabási-Albert scale-free construction and show that the degree of assortativity is an important parameter governing geometrical and transport properties of networks. Thus the diameter of the network and the clustering coefficient increase remarkably with the degree of assortativity. Moreover, the concentration dependences of the size of giant cluster in the node percolation problem for uncorrelated and assortative networks are strongly different [R. Xulvi-Brunet, W. Pietsch, and I. M. Sokolov, Phys. Rev. E 68, 036119 (2003)].

Two phase flow in porous media

•Toby Joseph¹ und Rudolf Hilfer^1,2
2Institut für Physik, Universität Mainz, 55099 Mainz,Germany
¹ICA-1, Universität Stuttgart, 70569 Stuttgart, Germany

Our understanding of two phase flow in porous media is still very incomplete. Although Darcy's law has been modified to deal with two fluids flowing in a porous medium by introducing extra parameters like the relative permeabilities, its domain of validity is still a matter of scientific debate. Moreover, it is well known that the relative permeability curves are history dependent and even the residual saturations in an invasion process depends on the way in which the invasion is driven as is clear from the capillary saturation curves. We present here a micro-to-macro upscaling that takes into consideration some of these problems.

Macromolecule- Flow Interaction

•Achim Jung und Christian Wagner
Experimentalphysik, Universität des Saarlandes

We present measurements on the flow of diluted polymer solutions in basic flow geometries. Our goal is to connect the statistical conformation of the macromolecules with macroscopic sample parameters such as shear or elongational viscosity. An example of basic flow geometry is the extensional flow in a droplet detachment process where the sample is confined in the thread between droplet and nozzle. Existing theoretical models are not able to explain our findings on the elongational viscosity of diluted polyelectrolyte DNA-solutions in different salinities and thus with different polymer chain flexibilities. In order to gain more insight into the microscopic conformations of the macromolecules we are setting up optical tweezers. The focus of our microscopic studies will be the meaning of hydrodynamic interaction, an effect that has been neglected in most of the theoretical predictions of macroscopic material parameters.

Fluctuation Measurements of a Passive Scalar with Miniaturized Thermocouples

•Marco Munzel, J. Peinke und A. Kittel
Carl von Ossietzky University of Oldenburg, Faculty of Physics, 26111 Oldenburg, Germany

The measurement of thermal fluctuations is important for investigating transport features of a passive and active scalar in fluids. We present a thermal sensor based on a miniaturized thermocouple. Its coaxial setup results from the fabrication as a micropipette normally used in neurobiology. The glass micropipettes contain a core of platinum or constantan wire and are coated with gold. The achieved tip diameters are 1mm and less which enhance the spatial and temporal resolution significantly. Because of their chemically inert coating, these sensors are applicative for detecting temperature fluctuations in large variety of liquids and gases. The sensors were tested in turbulent free-jets of heated water and air. The presented results are compared to a commercially availiable cold wire setup.

Turbulente Geschwindigkeitsprofile über rauhen Ebenen

•Nikolaus von der Heydt und Irmgard von der Heydt
Institut für Umweltphysik Göttingen - Physik zum Leben - Landolfshausen

In atmosphärischen Luftströmungen wurden in Höhen bis zu 100 m erhebliche Abweichungen von der logarithmischen Höhenformel u(z) = \fracwkln(\fraczz₀) festgestellt: Die Anpassung der vermeintlich geländetypischen, sogenannten Rauhigkeitslänge z₀ ergab an ein und demselben Ort in Abhängigkeit von der Schubspannungsgeschwindigkeit w = Ö{\fracsr} (zwischen 0,2 und 1 m/s) Werte von 1 mm bis 1 m. Ausgehend von neueren Ergebnissen der Turbulenzforschung zeigen wir, dass Prandtl's Ansatz für die Wirbelzähigkeit n_Pr = kw z, der zur Log-Formel führt, verbessert werden kann, und leiten damit eine DGL für u(z) her, deren Lösung für z >> \fracnw (n kinemat. Viskosität) selbstähnlich ist und mit den Messdaten übereinstimmt.

Heterogeneous Nucleation of charged sphere suspensions on shear induced precursors

•Thomas Palberg, Andreas Stipp, Ralf Biehl, Hans Joachim Schöpe, Jianing Liu und Thorsten Preis
Institut f. Physik, Universität Mainz, Staudinger Weg 7, 55128 Mainz

Growth of oriented single crystals may be obtained from shear melts of colloidal particles after nucleation at the container walls. Using different microscopic and scattering techniques we here studied precursors of such nuclei induced by steady shear. In all cases we observed shear layer structures which were oriented with their densest plane parallel to the container wall and their densest packed direction parallel to the flow. A strong correlation between the modes of motion and the type of interlayer registering is observed. Depending on the shear geometry used, the equilibrium phase, the strength and the range of interaction such layer structures either disappear upon increase of the shear rate or grow to macroscopic objects. Stop of shear may lead to an interesting competition between the decay of these usually meta-stable structures in favour of the shear melt and their stabilisation through coverage by an epitaxially growing wall crystal. In the latter case the shear layer residues act as pre-structured substrate.

Lattice-Boltzmann scheme for the dendritic growth in presence of convection

•Dmitry Medvedev und Klaus Kassner
Institut fuer Theoretische Physik / Computerorientierte Theor. Physik, Otto-von-Guericke-Universitaet Magdeburg, Postfach 4120, D-39016 Magdeburg

Dmitry Medvedev, Klaus Kassner, Otto-von-Guericke Universität Magdeburg

A combined phase-field/lattice-Boltzmann scheme is proposed to simulate dendritic growth from a supercooled melt.

The phase change part of the problem is treated with the phase-field approach of Karma and Rappel, whereas the flow of the liquid is simulated by a standard lattice-Boltzmann-BGK (LBGK) method with interactions with solid and thermal convection incorporated. To simulate conductive and convective heat transfer we use the multicomponent LBE method. The step of flow simulation can be dropped out in the case of purely diffusional growth.

Test simulations without convection reproduce the dendrite tip velocity of the exact solution.

Depending on the level of anisotropy, dendrites or doublons were obtained in the simulations.

Dendritic growth in a shear flow was simulated for different flow velocities, as well as the growth with natural thermal convection with different orientations of the crystal in the gravitational field.

Kinetic roughening of laser deposited polymer films: Crossover from single particle to continuous growth

•Jörg Hachenberg¹, Christoph Streng¹, Erik Süske², Sebastian Vauth¹, Stefan G. Mayr¹, Hans-Ulrich Krebs² und Konrad Samwer^1
1I. Physikalisches Institut, Universität Göttingen
²Institut für Materialphysik, Universität Göttingen, Tammannstr. 1, 37077 Göttingen

Thin film samples are prepared by pulsed laser deposition (PLD) of Bisphenol-A (BPA) polycarbonate targets at room temperature and under UHV conditions. Atomic force microscopy (AFM) is used to investigate the development of the surface roughness as a function of the thickness and the deposition angle.

We find that the roughness increases proportional to the film thickness by the growth exponent of 1/4 between 20nm and more than 4 mm . This is interpreted within computer simulations and theoretical models. Similar results have been shown in amorphous metallic systems, but on different scales. For lower film thicknesses than 20nm we observed a crossover to even steeper slopes where the particle character dominates the roughening and a continuous description is no longer valid. Additional investigations under variation of the laser fluency have been performed.

This project is supported by the DFG, SFB 602, TP B3 and GRK 782.

Shape of mounds and scaling behaviour in unstable epitaxial crystal growth

•Markus Walther, Sebastian Weber und Michael Biehl
Institut für Theoretische Physik,Universität Würzburg, Am Hubland, 97074 Würzburg

We study the influence of an Ehrlich-Schwoebel like barrier at corners in the diffusion of adatoms along step edges. The presence of an infinite barrier determines the shape of mounds in unstable growth and the coarsening dynamics of the surface.

The investigation is realized by two different methods of Monte Carlo simulations leading to similar results: Kinetic Monte Carlo simulations and a single particle representation. We investigate the effect of a finite barrier and its role in a dynamical phase transition with respect to surface morphology and scaling behaviour.

Comparison of Three-dimensional Simulations of Dendrites with Experimentally Grown 3D Xenon Crystals

•Herman M. Singer und Jörg H. Bilgram
Laboratorium für Festkörperphysik, ETH Zürich, 8093 Zürich (Schweiz)

Dendritic growth from a pure substance into supercooled melt was simulated in two- and three-dimensional calculations of two different phase field models ([A.A. Wheeler, B.T. Murray and R.J. Schaefer, Physica D 66, 243 (1993)] and [A. Karma and W.-J. Rappel, Phys. Rev. E 57, 4323 (1998)]). By means of large scale parallel molecular dynamics simulations for a Lennard-Jones potential, estimates for the anisotropy of interfacial free energy for our experimental model substance were calculated. In our in situ experiments three dimensional xenon crystals during free growth can be observed. Apart from dendrites, doublon and seaweed morphologies as well as morphology transitions are observed. Those structures can be simulated by using special initial conditions. A morphology diagram for 2D and 3D simulations is presented and compared with the results of experiments and analytical studies.

Analysis of Solidification Dynamics by Wavelet Transformation Techniques

•Oliver Wittwer und Jörg H. Bilgram
Laboratorium für Festkörperphysik, ETH Zürich, 8093 Zürich (Schweiz)

Complex structures develop during solidification processes. Changes in growth morphology can be initiated in our experiments by changing the temperature distribution in the environment of a growing crystal [I. Stalder and J.H. Bilgram, Europhys, Lett. 56, 829 (2001)]. These transitions have gained interest since they have been discovered in metallic samples [K. Dragevski et al., Phys. Rev. Lett. 89, 215502 (2002)]. To study transitions in growth morphology and to monitor quasi stationary growth of dendrites we apply wavelet transformation techniques. We are able to detect changes in growth behaviour in a much earlier state as it is possible by determining growth parameters such as typical length scales or growth rates.

Temperaturabhängige Dynamik von Au-Clustern auf amorphen C-Substratfilmen

•Ralph Werner¹, Matthias Wanner², Dagmar Gerthsen² und Stefan-Sven Jester^3
1Institut für Theorie der Kondensierten Materie,Universität Karlsruhe (TH), 76128 Karlsruhe
²Laboratorium für Elektronenmikroskopie, Universität Karlsruhe (TH), 76128 Karlsruhe
³Institut für Physikalische Chemie, Universität Karlsruhe (TH), 76128 Karlsruhe

Kleine, deponierte Au-Cluster zeigen in verschiedener Hinsicht dynamisches Verhalten. Im einzelnen kann man (i) unterschiedliche kristallographischen Modifikationen (fcc-Kuboktaeder, Ikosaeder, Dekaeder), (ii) Translation und Rotation von Clustern in Abhängigkeit von Größe, Temperatur und Substratwechselwirkung, sowie (iii) Veränderung der Partikelgröße Infolge von Diffusionsprozessen beobachten. Es werden einige experimentelle Ergebnisse dargestellt sowie Monte-Carlo-Simulationen. Die Proben wurden durch Beschuss von mit amorpher C-Folie bespannten Cu-Netzen mit Au-Clustern präpariert, die in einer Laser-Ablationsapparatur erzeugt wurden.

Simulation of biaxial compression of two-dimensional granular materials

•Hans-Georg Matuttis und Shinichi Morita
The University of Electrocommunications, Department of Mechanical and Control Engineering, Chofu Chofugaoka 1-5-1,Tokyo 182-8585,Japan

For experimental triaxial compression of granular materials, the maximum of the yield curve corresponds to the maximum of the bulk density. After the failure sets in, the granular compressibility decreases, as does the bulk density.

In this work, we simulate a two-dimensional granular assembly of particles of various sizes and shapes (round, polygonal, elliptic, with various elongations) via molecular dynamics, hence the process corresponding to experimental tri-axial compression becomes bi-axial compression. For the given boundary conditions and size dispersions, only elongated particles with friction show the maximum of the yield curve, and a maximum of the bulk density. For round particles, we find neither a maximum in the yield curve nor in the bulk density.

Particle imaging velocimetry study of a horizontally shaken granular bed

•Olaf Brökmann und Achim Kittel
Energy and Semiconductor Research Group, Department of Physics, University of Oldenburg

The study of shaken granular matter has become very popular in recent years. Surprisingly most of these studies only investigate the behavior of vertical shaken particles. Thinking about transportation of grain e.g. horizontal shaking has a great practical importance and therefore has to be more investigated. Using particle image velocimetry (PIV) we studied the velocity fields of horizontal shaken granular matter. The experiments include a variation of the shaking frequency, shaking amplitude and the size of the investigated particles.

THE MAGNETIC SORET EFFECT

•Thomas Völker und Stefan Odenbach
ZARM, University of Bremen, Am Fallturm 28359 Bremen

Investigations were made to determine the Soret coefficient of magnetic particles in a ferrofluid under the influence of a magnetic field. This so called magnetic Soret effect was theoretically determined to be two to three orders of magnitude smaller than the conventional Soret effect. In contrast former experiments have shown that the magnetic Soret effect is much higher than the theoretical predictions. However in those experiments the influence of buoyancy and magnetic driven convection could not be excluded. So the question how large the magnetic Soret effect can be is still open. Thus an experimental setup was developed to minimize parasitic effects and thus to simplify the analysis of the experimental results. These results show that the magnetic Soret effect can even be higher than the conventional one and confirm therewith former experimental results.

Examination of the magnetisation dynamics of rotating ferrofluids

•Sonia May^1,2,3, Jan Embs^1,2, Christian Wagner¹, Klaus Knorr¹ und Manfred Lücke^2
1Technische Physik, Universität des Saarlandes
²Theoretische Physik, Universität des Saarlandes
³Department of Physics, University of Hull

Ferrofluids are colloidal suspensions of nanometre sized magnetic particles in a magnetically inert liquid which, being coated by polymeric surfactants do not agglomerate. External magnetic fields exert forces and torques on the particles that are transmitted via viscous friction on the ferrofluid as a whole. In equilibrium ferrofluids show (super) paramagnetic behaviour in a magnetic field H with magnetisation M and field H being parallel to each other.

Under non-equilibrium conditions where the ferrofluid rotates M and H are no longer parallel to each other. For such situations various theoretical models for the dynamic magnetisation equation have been proposed that are not yet properly tested; in particular in regimes where the (local) vorticity W is comparable to or larger than the Brownian rotational relaxation rate 1/t.

Experiments have been performed on ferrofluids rotating with constant frequency W. The magnetisation component normal to the applied field is measured using a Hall sensor as a function of applied fields of up to 60 kA/m and as a function of rotational frequencies of up to 2,500 rad/s.

Experimental results for the non-equilibrium situation are compared with theoretical models.

Bewegung von Domänenwänden

•Matthias Müller und Ingo Rehberg
Physikalisches Institut, Universität Bayreuth

Bei Elektrokonvektionsexperimenten in nematischen planar orientierten Flüssigkristallen treten unter anderem Domänen mit abnormalen Rollen auf. Diese lassen sich mittels Ellipsometrie sichtbar machen. Wir untersuchen die Bewegung der Domänenwände, die die zwei Sorten der abnormalen Rollen trennen.

Theoretische Berechnungen [1] sagen einen linearen Zusammenhang zwischen der Geschwindigkeit der Domänenwände und der treibenden Wechselspannung voraus, was für Bewegungsinstabilitäten ein ungewöhnliches Verhalten darstellt. Das Ziel unserer Arbeit ist es diesen Zusammenhang experimentell zu untersuchen.

[1] H. Zhao und L. Kramer, Phys. Rev. E 62, 5092-5100 (2000)

Growth rate of amplitudes of surface instabilities in ferrofluids

•Holger Knieling, Reinhard Richter und Ingo Rehberg
Universität Bayreuth, Lehrstuhl Experimentalphysik V, 95440 Bayreuth

The surface of magnetic fluids subjected to a normal magnetic field is becoming unstable when a certain threshold of the magnetic induction is surpassed and the initially flat surface exhibits a stationary array of peaks (Rosensweig or normal field instability). The more general case is a field, which is tilted towards the normal of the surface of the magnetic liquid. In that case you can observe at first liquid ridges instead of the peaks. Increasing further the magnetic induction a stretched hexagonal pattern of crests is emerging on top of the liquid ridges.

The amplitudes of all these instabilities and their relaxation can be measured with the help of a linear array of 32 Hall sensors which are placed directly under the ferrofluid. Sensors situated under a ridge (trough) detect higher (lower) values of the local magnetic induction. With a time resolution of the sensors of about a half millisecond the maximal growth rate of the modes of the Rosensweig instability can be measured and compared to a theoretical prediction [1].

[1] A. Lange, Europhys. Lett., 55 (3), 327-333 (2001)

Differences of the magnetic susceptibility in alternating and rotating fields

•Robert Krauß, Reinhard Richter und Ingo Rehberg
Experimentalphysik V, 95440 Bayreuth

Considering a layer of ferrofluid set in motion by a rotating magnetic field raises the question how the velocity depends on the frequency of the field. Investigations and calculations on this effect (referred to as magnetic pumping) yield that the velocity is a function of the real and imaginary parts of the complex magnetic susceptibility. In a first approach the susceptibility was measured in a toroidal channel with a linear alternating field. We found substantial differences between the measured velocity and the one calculated from the complex susceptibility. However taking into account the complex susceptibility measured in a rotating field, leads to a better understanding of the fluid motion. The ferrofluid applied shows a strong coupling parameter of l » 3. This fact indicates that aggregates in the fluid exist and are responsible for the non-linear behaviour obtained.

A phenomenological model for ferrofluid droplet dynamics in a shear flow

•Nana Sadowsky, Patrick Ilg und Siegfried Hess
Institut für Theoretische Physik, Technische Universität Berlin, Hardenbergstr. 36, D-10623 Berlin

Hydrostatics and hydrodynamics of ferrofluid droplets give valuable information on magnetic and dynamical properties of the magnetizable fluid [1].

While the equilibrium deformation of the ferrofluid droplet as a function of the applied magnetic field is known experimentally [1], the non-equilibrium dynamics remain rather unknown so far. Here, we propose a phenomenological model for the dynamics of ferrofluid droplets in the presence of shear flow. Generalizing the approach of [2], ferrofluid droplets are modeled as deformable and orientable ellipsoids of revolution.

The combined effects of a magnetic field and a shear flow on the orientation and deformation of the droplets are studied. It is found that the hydrodynamic drag leads to a flow alignment of the droplet orientation. Predictions for the alignment angle as a function of the applied field are presented. Some comparisons to experimental data [3] in the absence of shear are made.

[1] E. Blums, A. Cebers, M.M. Maiorov, Magnetic Fluids, deGruyter, Berlin 1997.

[2] P. Ilg and M. Kröger, Phys. Rev. E 66, 021501-1 (2002)

[3] A. Lebedev, A. Engel, K.I. Morozov and H. Bauke, New Journal of Physics 5, 57.1-57.20 (2003)

The corkscrew instability in a nematic liquid crystal

•Alberto de Lozar, Wolfgang Schoepf, Lorenz Kramer und Ingo Rehberg
Universitaet Bayreuth

A liquid crystal with slightly positive dielectric anisotropy is investigated in the planar configuration. This system allows for competition between electroconvection and the homogeneous Frèedericksz transition, leading to a rather complicated bifurcation scenario. We report measurements of a novel instability leading to the corkscrewpattern. This state is closely connected to the Frèedericksz state as it manifests itself as a regular modulation along a Frèedericksz domain wall, although its frequency dependence indicates that electroconvection must play a crucial role. It can be understood in terms of a pitchfork bifurcation from a straight domain wall. Quantitative characterization is performed in terms of amplitude, wavelenght and relaxation time. Its wavelenght is of the order of the probe thickness, while its ondulation amplitude is an order of magnitude smaller. The relaxation time is comparable to the one obtained for electroconvection.

Magneto-rheological study in inverse magnetic fluids using polystyrene particles

•Saldivar-Guerrero Ruben¹, Richte Reinhard¹, Rehberg Ingo¹ und Rodriguez-Fernandez Oliverio^2
1Universitaet Bayreuth, Experimentalphysik V
²Centro de Investigacion en Quimica Aplicada

Ferrofluids are colloidal suspensions of magnetic nanoparticles, which change their behaviour under control of an external magnetic field. But when for example, they are mixed with polystyrene particles, they present another interesting properties. The magneto-rheological fluid (MRF) obtained in this way, can show a solid-fluid transition by an application of a moderate magnetic field similar to electro-rehological fluids (ERF) under a electric field. It excels ERF's because only moderates currents instead of high voltages are necessary. We study the rheological influence of the diameter of the immersed polymer particle in MRF and the relation between the coupling parameter l and the yield point ty for different magnetic fields applied. The results show that the rheological properties are increased when the particle diameter increases in a constant magnetic field. Also we investigate structure formation in the fluid by video microscopy when a magnetic field is applied.

Smectic and crystalline order in colloidal monolayers on periodic one-dimensional substrate potentials

•Jörg Baumgartl¹, Matthias Brunner² und Clemens Bechinger^1
12. Physikalisches Institut, Universität Stuttgart, Pfaffenwaldring 57, 70550 Stuttgart
²Fachbereich Physik, Universität Konstanz, 78457 Konstanz

There is growing interest in the phase behavior of two-dimensional (2D) colloids in the presence of one-dimensional (1D) periodic substrate potentials. The phase behavior of such systems is determined by two factors: (i) the relative orientation between the 2D crystal and the periodic potential troughs which selects a set of Bragg planes running parallel to the troughs and (ii) the commensurability ratio p = a¢/d of the spacing a¢ between these Bragg planes to the period d of the periodic potential. So far, only the phase behavior for p = 1 was investigated and the following phases were observed for increasing (non zero) potential strength: modulated liquid and light-induced crystal. For other commensurability ratios additional, more complex phases are predicted.

We investigated experimentally the phase behavior for the case p = 2 with the 1D periodic potential being created by two interfering laser beams. We observed for the first time the theoretically predicted intermediate phase between modulated liquid and light-induced crystal which is termed locked smectic. By analyzing the structure factor and the pair correlation function, this phase is characterized by liquid-like behavior along the potential troughs. In contrast to the modulated liquid the ordering perpendicular to the troughs is similar to the crystalline state, with particles occupying only each second trough.

Commensurability effects of colloidal monolayers on hexagonally patterned substrates

•Stefan Bleil¹, Matthias Brunner² und Clemens Bechinger^1
2Fachbereich Physik, Universität Konstanz, 78457 Konstanz
¹2. Physikalisches Institut, Universität Stuttgart, Pfaffenwaldring 57, 70550 Stuttgart

Colloidal suspensions are excellent model systems for investigations of phase transitions in two-dimensional (2D) systems. While most of the previous work concentrated on homogeneous substrates, much less is known about the phase behavior on patterned substrates although the latter closely resembles the behavior of atomic adsorbates on crystalline surfaces. In our experiments we mimic 2D patterned substrates for colloidal particles by the interference pattern of several laser beams which lead to triangular substrate potentials. We present results where the number density of charge stabilized colloidal particles was systematically varied. This allowed us to study the effect of the filling fraction h (i.e. the ratio between the number of colloidal particles and the number of substrate wells) on the observed structure of the colloidal system. Depending on the filling fraction we observe commensurate and incommensurate structures. We also compare our results with recent numerical simulations.

Phasetransitions in Quantum fluids: PIMC - Studies

•Guido Günther und Peter Nielaba
Lehrstuhl für Theoretische Physik, Fachbereich Physik, Universität Konstanz, D-78457 Konstanz

Exploiting path integral Monte Carlo methods we studied the behavior of the fluid phase of quantum mechanical hard sphere- and disk- systems.

We focused on the superfluid phase transition of Bose systems analyzing, among other quantities, the superfluid fraction which in periodic boundary conditions is related to the winding number of the Trotter-chains [0].

[0] E.L. Pollock, D.M. Ceperley; Phys. Rev. B 36, 8343 (1987)

Microcanonical entropy of classical spin systems with a continuous symmetry

•Andreas Richter, Michel Pleimling und Alfred Hüller
Institut für Theoretische Physik I, Universität Erlangen-Nürnberg, D-91058 Erlangen

Critical phenomena of classical spin systems may be studied in many ways. In our approach we determine and analyse the microcanonical entropy S which is related to the density of states by S = k_B lnW, where k_B is the Boltzmann constant. Due to the huge amount of possible configurations it is in general impossible to calculate W exactly, even for systems of moderate size. The density of states can, however, be computed approximately from a numerical method using transition variables. Discrete classical systems, like the Ising or Potts model, were already studied with this method in the past [1,2]. We now extend this method to systems with continuous spins and apply it to the three-dimensional XY model. In addition, we discuss the determination of critical quantities directly from the numerically determined microcanonical entropy.

[1] A. Hüller, M. Pleimling, Int. J. Mod. Phys. C 13, 947 (2002)

[2] H. Behringer, J. Phys. A: Math. Gen 36, 8739 (2003)

A Topological Rule-of-Thumb for Percolation Thresholds

•Richard Neher und Herbert Wagner
Sektion Physik, Universität München

Randomly assembled structures exhibit a density driven percolation transition, beyond which an unbounded component is formed. Since properties of those structures change dramatically at the percolation threshold, its accurate knowlegde is important and various empirical formulas, predicting percolation thresholds, have been suggested. The mean Euler characteristic is a topological measure for such random sets and it has been suggested to use its zero crossing to estimate the threshold. In contrast to other formulas, this estimation does not rely on a fit to known thresholds.

We further investigated this matter and found, that for a great variety of two-dimensional lattices the site percolation threshold is bound from above and below by the zero and the inflection point of the Euler characteristic, respectively. In the case of site percolation on three-dimensional lattices, bond-site percolation and continuum percolation, the thresholds are bound from above by the zero of the Euler characteristic.

Phase transitions in periodic pinning arrays

•Konrad Mangold¹, Michael Köppl¹, Paul Leiderer¹ und Clemens Bechinger^2
1Universität Konstanz, Fachbereich Physik
²Universität Stuttgart, 2.Physikalisches Institut

We study by means of videomicroscopy the phase behavior of two-dimensional (2D) paramagnetic colloidal systems on periodic pinning arrays, the latter being created by a hollographic optical tweezer technique. When the particle interacion is decreased by changing an external magnetic field the colloidal monolayer shows a melting transition from an incommensurable solid to a liquid phase. An undisturbed 2D system exhibits a two-stage melting as described by the KTHNY theory. With an additional pinning array the 2D system shows a rich phase behavior. The details of the diffusion of interstitial particles through a matrix of pinned particles depend among other things on the geometry of the pinning lattice (square or hexagonal) and on density ratio of pinned to diffusing particles. We compare our results with numerical studies on vortex melting in periodic pinning arrays.

We also present first results where the particle transport through a channel under non-equilibrium conditions is studied. A topological channel is created by means of electron lithography and the motion of the particles are studied for different particle interactions and channel geometries.

Short-Time Dynamics

•Eric Lorenz und Wolfhard Janke
Institut für Theoretische Physik, Universität Leipzig, Augustusplatz 10/11, 04109 Leipzig, Germany

The 2d Ising model on a square lattice is quenched from a state at high temperature to the critical temperature and its dynamic relaxation is investigated with Monte Carlo methods. In the case of a small initial magnetization, a critical initial increase of the magnetization is observed during early time of the dynamic evolution. With power laws derived from the scaling relation being valid already in non-equilibrium, one can determine estimates for the critical exponents q, z, and b/n. This procedure of observing quantities in non-equilibrium prevents critical slowing down caused by the divergence of the spatial correlation length at the critical temperature in equilibrium. In this work special attention is paid to a comparison of algorithms. The dynamics produced by heat-bath, Metropolis and Glauber updates are found to be universal, despite surprising differences during the very early time evolution, which are investigated in some detail.

Damage Spreading in Ising Lattice

•Rodrigo Megaides und Wolfhard Janke
Institut für Theoretische Physik, Universität Leipzig, Augustusplatz 10/11, 04109 Leipzig, Germany

We study the ``damage spreading phenomenon'' in a two-dimensional Ising lattice of different sizes. The phenomenon consists in the propagation of a small modification, in the initial state of the lattice, to the whole system without changing the random noise. The method is first equilibrating the system and then performing a time average over final ``Hamming distances" (a measure of the number of spins in opposition in two similar systems) of initial pairs of lattices (the first one taken for such average and the second one being identical to the former except for the individual state of several spins, the so called ``damage"). In every individual measurement, we find one of these three types of ``final" Hamming distances (HD) : HD = 0 or healing (the damage disappears at the first sweeps), HD @ 0.5 or 0.5-spreading (the damage spreads until it reaches one half of the lattices, except for some fluctuations), HD = 1 or infinite spreading (the damage fills the lattice). The probabilities of such cases are obtained (for different temperatures and lattice sizes) from the time average, as well as a mean value of the Hamming distance which is easily calculated from the mentioned probabilities. Some scaling properties are proposed.

Desorption behaviour of binary quench-condensed noble gas mixtures

•M. Layer, A. Netsch, A. Fleischmann und S. Hunklinger
Kirchhoff-Institut für Physik, Universität Heidelberg, Im Neuenheimer Feld 227, D-69120 Heidelberg

Due to their simple van der Waals interaction noble gas crystals are commonly considered as model system for condensed matter. This holds in particular for binary mixtures. Freezing theories as well as molecular dynamical simulations predict the existence of structural ordered phases in good agreement with each other. We studied the desorption behaviour of quench-condensed noble gas films by means of high-frequency surface acoustic waves. In contrast to theoretical investigations quench-condensed films are in a state far away from thermal equilibrium. By annealing, these films are relaxing into thermodynamically more favorable phases. Over a wide range of preparation conditions (including substrate temperature, gas temperature, composition of the mixture) we found evidence for the formation of ordered film structures. Furthermore it turned out that in particular the ratio of diameters of the two components is relevant to the occurence of ordered film structures. Desorption processes of several condensed binary mixtures (Ne/Ar, Ne/Kr, Ar/Kr) are presented and the forming of possible van der Waals compounds is discussed.

Computer Simulations of the 3-Dimensional Abelian Higgs Model

•Sandro Wenzel, Elmar Bittner, Wolfhard Janke und Arwed Schiller
Institut für Theoretische Physik, Universität Leipzig, Augustusplatz 10/11, 04109 Leipzig, Germany

Using multicanonical and heat-bath Monte Carlo algorithms we investigated the 3-dimensional Abelian Higgs model [1] on the lattice with compact gauge fields and fluctuating Higgs amplitudes ( l ¹ ¥). Special consideration was given to the first-order regime of the confinement-deconfinement phase transition. The existence and location of a (tri-)critical point was explored using reweighting techniques.

[1] S.Mo, J.Hove, A. Sudbø, Phys. Rev. B 65 (2002) 104501.

3-Zustands Potts-Modell auf Voronoi-Delaunay Zufallsgraphen in 2 Dimensionen

•Goetz Kähler, Martin Weigel und Wolfhard Janke
Institut für Theoretische Physik, Universität Leipzig, Augustusplatz 10/11, 04109 Leipzig, Germany

Im Rahmen der Diplomarbeit erstellten wir Voronoi-Delaunay Zufallsgitter in verschiedenen 2D-Topologien (Torus, Kugeloberfläche). Auf diesen wurde dann ein modifiziertes, reichweitenabhängiges 3-Zustands Potts-Modell mit Hilfe des Wolff'schen Clusteralgorithmus auf vielen Realisierungen des Zufallsgitters simuliert. Ergebnisse dieser Arbeit, insbesondere in Bezug auf das kritische Verhalten von Magnetisierung und Suszeptibilität [1] werden hier vorgestellt.
[1] F.W.S. Lima, U.M.S. Costa, M.P. Almeida und J.S. Andrade Jr., Eur. Phys. J. B17, 111 (2000).

The Bragg glass - Vortex glass transition

•Roland Schorr, Ludger Santen und Heiko Rieger
Universität des Saarlandes, FR 7.1 Theoretische Physik, 66041 Saarbrücken

Many phase diagrams of high T_c superconductors obtained experimentally, e.g. in Bi₂Sr₂CaCu₂O_8+y (BSCCO) and YBa₂Cu₃O_7-d (YBCO), reveal a very complicated landscape of various phases. The most significant modifications applied to a naive idea of a type II superconductor, which discriminates only between Meissner (H < H_c1) and Abrikosov (H_c1 < H < H_c2) phase, have to take into account the impact of termal fluctuations and disorder. This leads to a water-like melting phenomenon in the regime of the Abrikosov phase with a melting line terminating at fields that are much lower than H_c2. The phase below the melting line is called Bragg glass whereas above the line one has to distinguish between Vortex glass and Vortex liquid. We focus on the phase transition between Bragg glass and Vortex liquid and Vortex glass, respectively, induced by the disorder at finite temperatures. As a starting point of our simulation we choose an elastic description of fluxlines fluctuating in a random potential. We perform a Monte Carlo Simulation applying cluster updates in order to sample the configuration space. Whereas local updating schemes seem to be problematic due to slowing down and a dynamical critical exponent z = 2, our cluster algorithm works well with z » 0. This enables us to realize the three phases for system sizes up to L = 32.

Monte Carlo Study of the Bond-Diluted 3D Ising Model

Pierre Emmanuel Berche¹, Christophe Chatelain², Bertrand Berche² und •Wolfhard Janke^3
1Groupe de Physique des Matériaux, Université de Rouen, 76821 Mont Saint-Aignan Cedex, France
²Laboratoire de Physique des Matériaux, Université Henri Poincaré, Nancy 1, BP 239, 54506 Vand uvre les Nancy Cedex, France
³Institut für Theoretische Physik, Universität Leipzig, Augustusplatz 10/11, 04109 Leipzig, Germany

We study by Monte Carlo simulations the influence of bond dilution on the three-dimensional Ising model. This paradigmatic model in its pure version displays a second-order phase transition with a positive specific heat critical exponent a. According to the Harris criterion the influence of disorder should hence lead to a new fixed point governed by new critical exponents. We have determined the phase diagram of the diluted model, starting from the pure model limit down to the neighbourhood of the percolation threshold. For the estimation of critical exponents, we have first performed a finite-size scaling study, where we concentrated on three different dilutions to check the stability of the disorder fixed point. Particular emphasis is placed on cross-over phenomena between the pure, disorder and percolation fixed points which lead to effective critical exponents dependent on the concentration. Furthermore, the temperature behaviour of physical quantities has been studied in order to characterize the disorder fixed point more accurately. The question of non-self-averaging at the disorder fixed point is also investigated and compared with recent results for the bond-diluted q = 4 Potts model.

Photoangeregte Nanopartikel im starken Nichtgleichgewicht

•Vassilios Kotaidis¹, Samuel Gresillion², Gero von Plessen³ und Anton Plech^1
1Fachbereich Physik der Universität Konstanz, Universitätsstr. 10, 78457 Konstanz
²ESPCI 10, rue Vauquelin, F-75005 Paris
³RWTH Aachen, I. Physikalisches Institut A, D-52074 Aachen

Metallische Nanopartikel stellen ideale Systeme dar, um elementare dynamische Prozesse im Nichtgleichgewicht zu untersuchen. Durch optische Anregung mit Femtosekundenlasern kann innerhalb einer Pikosekunde dem System geung Energie zugeführt werden, um Gittertemperaturen von einigen Tausend Kelvin zu erreichen. Mit der Kombination aus Laseranregung mit 150 fs Pulsen und Röntgenpulsabfrage werden irreversible Strukturänderungen in den metallischen Nanopartikeln aus Silber und Gold untersucht.

Schmelzprozesse bzw. Fragmentationsprozesse werden auf der 100 ps Zeitskala aufgezeichnet und in Beziehung zu der Ankopplung an das Medium,bzw. eine glatte Oberfläche gesetzt.

2-D Crystals on Curved Surfaces

•Peter Lipowsky, Hans-Georg von Ribbeck, Michael G. Nikolaides und Andreas R. Bausch
TU Muenchen, Physik-Department E22, James-Franck-Strasse, D-85747 Garching

Determining the minimum-energy configuration of repulsive particles on spherical surfaces is a rather difficult problem. Euler's theorem states that twelve pentagons are required to close a hexagonal network. Theory and computer simulation have shown that, with rising system size, the strain induced by single pentagonal disclinations will be lowered by the introduction of additional pairs of bound 5-7 defects. An experimental system to test the theoretical predictions has been developed. We use surface-modified microspheres self-assembled onto the surface of water droplets in an organic solvent. Crystalline structures are observed as soon as the area density of particles at the interface is high enough. As predicted by theory, we find that these crystals form distinctive high-angle grain boundaries. The number of excess defects in these scars grows linearly with the system size. Using digital video microscopy, particle tracking algorithms, and triangulation routines, the dynamical behaviour of the defects and the movements of single particles in the spherical lattice are explored. The elastic potential of the lattice, the Young modulus, and the diffusion constant are determined. First results demonstrating typical phase transitions between fluid and crystalline phases are presented. Binary mixtures formed by beads of two different diameters at the interface were investigated.

First-principle investigations of phonon spectra of crystalline ice

•Waheed Adeniyi Adeagbo¹, Peter Entel ¹ und Jürgen Hafner^2
2Material Physics Institute, University of Vienna, Austria
¹Institute of Physics, University of Duisburg-Essen, 47048 Duisburg, Germany.

Lattice dynamical calculations have been carried out for a tetrahedrally coordinated crystalline ice structure using Vienna Ab-initio Simulation Package (VASP). A one directional periodically replicated unit cell containing 24 molecules of water, whose constituents molecules are arranged according to Bernal-Fowler ice rules is employed in our phonon calculations. Dynamical properties were determined within the harmonic approximations by finite difference evaluation of the dynamical matrix from atomic forces. We investigated the effect of polarization arising from dipole-dipole interactions on our phonon spectra by carrying out two separate calculations with and without effect of induced polarization. The polariazation was taken into account by switching on the evaluation of the usual Berry phase expression for the electronic polarization of an insulating ground state system in the VASP code. This produces a significant effect in the splitting of longitudinal and transverse optic modes. Our results are discussed with respect to the neutron diffraction scattering data.

Akustisches Rauschen von Kondensatormikrofonen

•Volker Becker¹, Jürgen Breitlow² und Andreas Engel^1
1Carl von Ossietzky Universität Oldenburg
²Georg Neumann GmbH Berlin

Das akustische Gleichgewichtsrauschen von Kondensatormikrofonkapseln setzt eine untere Grenze für die erreichbare Güte von Studiomikrofonen, die nach grossen Fortschritten bei der Reduzierung des Rauschens in der äusseren elektrischen Beschaltung für technische Anwendungen relevant wird. Ein einfaches Modell zur theoretischen Beschreibung der Brownschen Bewegung der Mikrofonmembran liefert im mittleren Frequenzbereich zwischen 100Hz und 10kHz Ergebnisse für die Spektraldichte der Rauschenspannung, die sich in guter Übereinstimmung mit Messungen an Kapseln der Georg Neumann GmbH Berlin befinden. Von zentraler Bedeutung für die Modellierung ist die Beschreibung der speziellen Dämpfung der Membranschwingungen durch ein Luftpolster, die die verschieden Moden miteinander verkoppelt.

Orientational and positional phase transitions in geometric confined systems of alkanes - a numerical study

•Frank Oliver Pfeiffer und Heiko Rieger
Theoretische Physik, Universität des Saarlandes, 66041 Saarbrücken, Deutschland

A classical phenomenological atom-atom potential is used to study alkane chains in a single 2D hexagonal pore via the Monte Carlo method. We find different phases characterized by orientational and positional order and discuss the influence of (1) the surface structure of the pore wall and (2) the pore diameter on the correlation functions. Our calculations for the confinement are compared to the bulk scenario and to recent experiments on n-alkanes in nano-pores by Huber and Knorr.

[HOME]

[E-Verhandlungen 2004]

[Regensburg]

[Fachverband DY]

[Sitzung DY 46]

Zuletzt geändert am 27.02.2004

Fragen, Kommentare, Anregungen und Kritik