Adaptive moving mesh algorithm based on local reaction rate.

An empirical mesh adaption algorithm is introduced for modeling one-dimensional reaction-diffusion systems with large moving gradients. Our new algorithm is based on the revelation, that in reaction-diffusion systems the high moving concentration gradients appear nearby to the region where the rate of reaction is maximal, thus the local reaction rate can be used to control the mesh adaption. We found, that the main advantage of such a method is its simplicity and easy implementation. As an example we study an acid-base diode, where large moving gradients appear. The mathematical model of the diode contains several parabolic PDEs, coupled with one elliptic PDE. An r-refinement technique is used and attached to the commercial finite element solver COMSOL. We investigated the time-dependent salt effects of the diode with our developed algorithm. Our mesh adaption method is advantageous for modeling of any reaction-diffusion systems with localized high concentration gradients.

Localization of Scattering Objects Using Neural Networks.

The localization of multiple scattering objects is performed while using scattered waves. An up-to-date approach: neural networks are used to estimate the corresponding locations. In the scattering phenomenon under investigation, we assume known incident plane waves, fully reflecting balls with known diameters and measurement data of the scattered wave on one fixed segment. The training data are constructed while using the simulation package µ-diff in Matlab. The structure of the neural networks, which are widely used for similar purposes, is further developed. A complex locally connected layer is the main compound of the proposed setup. With this and an appropriate preprocessing of the training data set, the number of parameters can be kept at a relatively low level. As a result, using a relatively large training data set, the unknown locations of the objects can be estimated effectively.

Design of equidistant and revert type precipitation patterns in reaction-diffusion systems.

In the past years considerable attention has been devoted to designing and controlling patterns at the microscale using bottom-up self-assembling techniques. The precipitation process proved itself to be a good candidate for building complex structures. Therefore, the techniques and ideas to control the precipitation processes in space and in time play an important role. We present here a simple and technologically applicable technique to produce arbitrarily shaped precipitation (Liesegang) patterns. The precipitation process is modelled using a sol coagulation model, in which the precipitation occurs if the concentration of the intermediate species (sol) produced from the initially separated reactants (inner and outer electrolytes) reaches the coagulation threshold. Spatial and/or temporal variation of this threshold can result in equidistant and revert (inverse) type patterns in contrast to regular precipitation patterns, where during the pattern formation a constant coagulation threshold is supposed and applied in the simulations. In real systems, this threshold value may be controlled by parameters which directly affect it (e.g. temperature, light intensity or ionic strength).

Pattern formation and self-organization in a simple precipitation system.

Various types of pattern formation and self-organization phenomena can be observed in biological, chemical, and geochemical systems due to the interaction of reaction with diffusion. The appearance of static precipitation patterns was reported first by Liesegang in 1896. Traveling waves and dynamically changing patterns can also exist in reaction-diffusion systems: the Belousov-Zhabotinsky reaction provides a classical example for these phenomena. Until now, no experimental evidence had been found for the presence of such dynamical patterns in precipitation systems. Pattern formation phenomena, as a result of precipitation front coupling with traveling waves, are investigated in a new simple reaction-diffusion system that is based on the precipitation and complex formation of aluminum hydroxide. A unique kind of self-organization, the spontaneous appearance of traveling waves, and spiral formation inside a precipitation front is reported. The newly designed system is a simple one (we need just two inorganic reactants, and the experimental setup is simple), in which dynamically changing pattern formation can be observed. This work could show a new perspective in precipitation pattern formation and geochemical self-organization.

Systematic front distortion and presence of consecutive fronts in a precipitation system.

A new simple reaction-diffusion system is presented focusing on pattern formation phenomena as consecutive precipitation fronts and distortion of the precipitation front. The chemical system investigated here is based on the amphoteric property of aluminum hydroxide and exhibits two unique phenomena. Both the existence of consecutive precipitation fronts and distortion are reported for the first time. The precipitation patterns could be controlled by the pH field, and the distortion of the precipitation front can be practical for microtechnological applications of reaction-diffusion systems.

Regular Liesegang patterns and precipitation waves in an open system.

We investigate the regular and moving Liesegang pattern formation phenomena in an open system. First, simulations have been performed at fixed coupling between the reactive medium and the reservoir, later this control parameter was varied during the simulations resulting in various phenomena. We predicted and monitored for the first time various--dynamically changing--precipitation structures and a spatial hysteresis phenomenon, which is beyond the scope of the Turing instability. The dynamics of the reaction is well detectable using specific quantities: the total amount of precipitate and its center of gravity.

A new universal law for the Liesegang pattern formation.

Classical regularities describing the Liesegang phenomenon have been observed and extensively studied in laboratory experiments for a long time. These have been verified in the last two decades, both theoretically and using simulations. However, they are only applicable if the observed system is driven by reaction and diffusion. We suggest here a new universal law, which is also valid in the case of various transport dynamics (purely diffusive, purely advective, and diffusion-advection cases). We state that p(tot) proportional X(c), where p(tot) yields the total amount of the precipitate and X(c) is the center of gravity. Besides the theoretical derivation experimental and numerical evidence for the universal law is provided. In contrast to the classical regularities, the introduced quantities are continuous functions of time.

Simulation of a crossover from the precipitation wave to moving Liesegang pattern formation.

Model simulations to investigate the precipitation wave phenomenon and a crossover from the precipitation wave to moving Liesegang patterns were performed. The chemical scheme contains four chemical species via the interaction of precipitation and redissolution (complex formation), in which the precipitation reaction term was based on Ostwald's supersaturation theory. In this article, for the first time, all the features and behaviors of the heterogeneous traveling waves are reproduced, which were observed experimentally in the work of Zrínyi et al. (Zrínyi, et al. J. Phys. Chem. 1991, 95, 1618.). The detailed investigation of the pattern formation showed three possible states of the system, which depend on the initial concentration of the inner and outer electrolytes, respectively. These are the precipitation wave (single moving precipitation band), the moving Liesegang pattern (moving precipitation bands), and the state where these two patterns coexist.

Precipitate pattern formation in fluctuating media.

Simulation of the Liesegang pattern formation in low concentration gradient is presented using concentration perturbation in a deterministic model. The precipitation process is based on ion-product supersaturation theory (Ostwald's model). In the classical experiments with high initial concentration gradients, appearance time and locations of the band formation are well reproducible. Decreasing initial concentration gradients results in a more stochastic pattern structure; this means that the reproducibility of the experiments becomes worse. The presented model and the results of the simulations exhibit the same trend, which were demonstrated and investigated experimentally by Kai et al.