Modelagem estocástica e simulação da polimerização radicalar livre de estireno em microcanais usando um algoritmo híbrido de Gillespie.

Abstract

Most recently, advanced studies have been carried out on the production of polystyrene by Free Radical Polymerization (F RP ) via microchannels. This has been a subject of core interest primarily due to the efficiency of a microreactor in terms of process intensification. In addition, especially in pilot experimentations, a micro or milli-reactor has been known widely to be efficient in monitoring the microstructural end-use features or properties of the polystyrene polymer as the chain grows and ultimately terminates. However, a critical problem that occurs in milli-and micro-reactors is the clogging of the microchannels.In this work, the synthesis of polystyrene via F RP through microchannels is simulated using a robust and time-efficient Hybrid Stochastic Simulation Algorithm based on the Gillespie Algorithm. This algorithm not only simulates the chain growth polymerization but also allows a simultaneous parallel deterministic computation of the same chain growth reaction system. The produced deterministic profiles at various conditions were compared to the respective stochastic trajectories.To validate the model, the obtained results of the end-use properties of polystyrene such as Monomer Conversion (X), Polydispersity Index (P DI), Number-Average Molar Mass (M n) and Weight-Average Molar Mass (M w) were compared to experimental data. Also, the residence times deployed for the simulation was from 5 to 80 minutes and as well as varying operating conditions of initial Monomer (M ), Solvent (S) and Initiator (I) which includes temperatures ranging from 100 and 140 degrees Celsius.The Average Percentage Error (AP E) obtained from the deterministic simulation of the hybrid stochastic simulation algorithm (HSSA) was close to the results found in literature, thus validating the efficiency of the algorithm to simulate the polymerization via F RP of the styrene monomer. Thus, applying different input conditions, the algorithm was used to simultaneously predict deterministic profiles and stochastic trajectories. The stochastic model allows us to understand the physical phenomena that occur inside a microreactor.