Computational optimization of microfluidic devices via genetic algorithms

We discuss the development of new computational strategies based on evolutionary computing algorithms for the insilico design and optimization of MicroFluidic Devices (MFD) aimed at chemical synthesis. Evolutionary algorithms are stochastic global optimization methods inspired by the biological mechanisms of evolution and heredity. In particular we present an optmization procedure of a prototypical microfluidic device which is host to a generic second order chemical reaction A + B → C. This method is applied to the search of the best performing geometry, which assures a high yield in terms of product C at a fixed exit point of the microcircuit, given initial concentrations and flows of reactants A and B at two inlet points. The microfluidic circuit is represented by a binary code, as a succession of empty and filled elementary cells, and artificial evolution is employed to select the best performing specimen.