CORDIS - EU research results

Exploiting Chemical Self-Organisation in Materials Science

Final Report Summary - ECSOMS (Exploiting Chemical Self-Organisation in Materials Science)

In this project we set out to develop new methods for the generation of chemical patterns. The research was inspired by nature, in particular the work of Turing on the chemical basis of morphogenesis in which he envisaged that the growth of form in biological systems might arise from a chemical pre-pattern. The pre-pattern arose from a combination of chemical feedback, or autocatalysis, and differential transport of chemical species between cells. Our goal was to develop ways by which these pattern forming mechanisms might be exploited in materials science.
In order to achieve this goal, we addressed several issues. Firstly, to date most of the chemical feedback that has been characterised involves harsh inorganic redox processes with species such as bromate. These species are unsuitable for coupling with soft-matter systems as they are too reactive. New sources of feedback were required that could be exploited for use with pH-sensitive polymers. There are many groups working with pH sensitive far-from-equilibrium processes that would be interested in the development of the first pH oscillator in a closed system, particularly if the system is benign and biocompatible. Therefore we investigated the use of enzyme catalysed and acid/base catalysed ester/amine hydrolysis as a source of feedback for pH patterns. Secondly, new scenarios for the formation of chemical patterns far-from-equilibrium had to be developed as the current methods involved a continuous flow unstirred reactor, (CFUR) – essentially a flat gel disk or membrane coupled to continuous stirred tank reactor – were somewhat restrictive. With the use of new flow reactors and catalyst loaded microparticles, we are now able to obtain far-from-equilibrium responses driven by pH feedback.
The methodology for enzyme catalysed feedback involves exploiting the bell-shaped rate-pH curve of enzyme reactions coupled with the production of acid/base, this has been implemented experimentally in only a few systems. We examined the urea-urease reaction as the hydrolysis of urea in acid (pH 4) produces ammonia and the pH of the reaction increases which in turn increase the enzymatic activity, thus the rate accelerates as the reaction proceeds. We were able to obtain propagating fronts in a thin layer of the reaction medium. The relationships between front speed and initial concentrations were reproduced qualitatively in simulations. These were the first acid to base fronts observed in an aqueous phase enzyme catalysed reaction. A 2-variable model was developed to investigate the possibility of oscillations in the urea–urease reaction. An open system was considered in which acid and urea were transported to a cell containing the enzyme. Using linear stability analysis we determined the range of transport coefficients limit cycles may exist for and showed that differential transport is required for oscillations in a class of compartmentalised enzyme processes similar to the urea–urease system. The open system that we considered is representative of a flow reactor in which the reaction is maintained far-from-equilibrium by exchange (kex) of species with the surroundings, however our results suggested that oscillations could not be obtained in a typical flow reactor as the inflow rates of species are the same. However, the equations are appropriate for both an open reactor or a uniform enzyme loaded particle in a bath of reactants, using the pool chemical approximation. More-over, this latter scenario was is in the spirit of the original Turing paper in which stationary patterns are observed in a ring of cells undergoing autocatalysis and exchange of chemicals with their neighbours. We therefore turned our attention to the urea-urease reaction in polymer particles. In the model, we considered a polymer particle loaded with enzyme and placed in a reservoir containing substrate and acid. As the diffusion of acid is greater than that of substrate, using reasonable experimental parameter values we were able to obtain oscillations and bistability in the particles and complex behaviours in groups of particles. We also have preliminary results suggesting the possibility of stationary pH patterns.
In order to obtain these results in experiments we used urease-loaded alginate particles. The particles were prepared by syringing a solution of sodium alginate containing urease and a pH indicator, cresol red, into a solution of calcium chloride, then coating with charged polyelectrolyes such as PSS and PAH using a layer by layer technique (LBL) in order to reduce diffusional loss of the enzyme. The urease-alginate particles were placed in a solution of urea with the initial pH adjusted by sulphuric acid to 4 and monitored using optical microscopy to observe the switch to high pH in the particles. We obtained evidence of particle size-dependent activity, propagating pH patterns, stationary patterns and group activity including propagating pH waves. Enzyme microparticles have been designed for a number of biotechnological applications, such as sensors, drug delivery devices and bio-reactors, however the wealth of behaviour associated with autocatalytic reactions has yet to be exploited. Enzymes offer specificity, efficiency and biocompatibility for these applications; feedback gives potential advantages such as fast robust response over a wide range of conditions and in the presence of noise; periodic release of a chemical or synchronised activity overcoming diversity in a group of particles and self-motion of a particle and we are now able to explore these possibilities.
A new flow reactor was designed in order to obtain the chemical patterns on cylinders that we predicted in simulations. Numerical simulations were produced using the 2-variable CDIMA reaction in three spatial dimensions in thin cylindrical layers. We found that the resultant pattern could be effectively controlled by the ratio of the cylinder circumference to the intrinsic wavelength of the pattern. Stripes were observed on narrow cylinders, irrespective of the instrisic pattern (spot or stripe) and spots only emerged on sufficiently large domains. As the ratio of the circumference to the wavelength was increased, complex patterns appeared including intertwined twisted waves. In order to obtain pH patterns in the new flow reactor that might eventually be coupled with polymerisation processes, we also examined a number of realistic networks based on enzyme catalysed hydrolysis or dehydration processes. We focussed on hydrolysis of esters to produce carboxylic acids (acid switch) and amides to produce amines (base switch, in cases where the net pH increases i.e. urea derivatives). In numerical simulations, the conditions for a pH switch from high to low pH and bistability between high and low pH values in a flow reactor were obtained using a general model of ester hydrolysis with acid catalysis and with enzyme catalysis.
During the investigation of feedback in acid catalysed ester hydrolysis it was noted that crystals of aspirin (o-acetylsalicylic acid) underwent autonomous motion and interactions at the air-water interface. We decided to investigate this phenomenon as much attention has been paid to the possibility of combining autonomous motion and self-assembly of nano- to millimeter sized particles with potential applications in materials science of drug delivery. The crystals motion was influenced by the addition of other ions that alter the surface tension; translational, rotational and intermittent motion was observed and motion ceased outside a bounded range of surface tensions. The complex surface flows generated by the organic crystals led to the formation of clusters of two or more crystals in specific orientations that also underwent motion. The behavior of groups of crystals can be explained by a combination of capillary interactions and Marangoni effects, although electrostatic interactions during collisions cannot be ruled out. Anisotropic particles have capillary profiles that can lead to alignment in specific orientations. The capillary profile of the dense aspirin particle shows dips and rises that may result in attractive or repulsive interactions depending on the orientation of the crystals. Some aspirin crystals are capable of turning and re-aligning in a series of bouncy collisions as the complex surface deformations in the vicinity of each crystal result in sharp changes between like and unlike interactions. Like interactions lead to the formation of dimers, however the attractive forces are often overcome by surface flows as result of the Marangoni effect in the vicinity the crystals.