Luciane da Silveira Ferreira

Modeling and Simulation of the Polymeric Nanocapsule Formation Process

Abstract Abstract In this work the modelling and simulation of nanoparticle formation according to the technique of nanoprecipitation was done. In this method, the particle is formed due to the further diffusion of solvent into the water, resulting in the aggregation of the associated polymer chains. In order to predict the characteristics of the nanoparticle and also to improve the process, it was developed a mathematical model that considers: (a) the type of polymer; (b) interaction between solvent and polymer; and, (c) dynamics of solvent diffusion. The diffusivity between polymer-solvent was modelled by means of the Vrentas & Duda Free Volume Theory, including the Sanchez-Lacombe equation-of-state. The model was written in terms of Partial Differential Equation, and solved with MAPLE for a given initial size distribution. Additionally, it is a moving boundary problem because the diffusion of the solvent out of the droplet leads to its size reduction. Based on a given initial droplet size distribution, the transient behaviour and the final droplet size distribution can be evaluated. The dynamic simulation shows both the evolution of the solvent inside the droplet and the variation of size in time. Additionally, the comparison between experimental and simulated results showed a very good agreement.

A growth kinetic model of Kluyveromyces marxianus cultures on cheese whey as substrate

This work presents a multi-route, non-structured kinetic model for determination of microbial growth and substrate consumption in an experimental batch bioreactor in which β-galactosidase is produced by Kluyveromyces marxianus growing on cheese whey. The main metabolic routes for lactose, and oxygen consumption, cell growth, and ethanol production are derived based on experimental data. When these individual rates are combined into a single growth rate, by rewriting the model equations, the model re-interpretation has a complexity similar to that of the usual variations of the Monod kinetic model, available in the literature. Furthermore, the proposed model is in good agreement with the experimental data for different growth temperatures, being acceptable for dynamic simulations, processes optimization, and implementations of model-based control technologies.

Tanneries: from waste to sustainability

The aim of this work was to look for an efficient and environmentally safe solution for the leather tanning industry through a problem of non-linear programming. The results were promising and showed a significant potential of wastewater reduction. Reuse of internal streams reduced around 70% wastewater. The solutions obtained could reduce significantly the amount of chemicals and consequently the amount of waste to be treated.

ARE MONOD MODELS ENOUGH FOR BIOREACTOR CONTROL? PART I – EXPERIMENTAL RESULTS

Abstract A model based substrate control system for yeast fed batch cultivations is presented. The control system consists of a FIA measurement system, which get a cell free sample stream from the culture broth due to a probing system with a time delay of 360 s. The glucose measurement values were utilized by a discrete-continuous extended Kalman filter, which applied the typical Monod model. The estimated values were applied to run a feedforward as well as feedback PI controller for set point control at 0.05 g/L glucose. The mean glucose measurement was 0.051 g/L with a standard deviation of 0.009 g/L. Therefore, the Monod model seems to be adequate for control, however the specific growth rate itself was not constant neither was the specific uptake rate of oxygen and the specific production rate of carbon dioxide constant, which indicate metabolic changes during the cultivation.

Modeling and Simulation of Nanoparticles Formation Process: A Diffusive Approach

Abstract In this work the modeling and simulation of nanoparticle formation according to the technique of nanoprecipitation was done. In this method, the particle is formed due to the further diffusion of solvent into the aqueous phase, resulting in the aggregation of the associated polymer chains. In order to predict the characteristics of the nanoparticle and also to improve the process, it was developed a mathematical model that considers: (a) the type of polymer; (b) interaction between solvent and polymer; and, (c) dynamics of solvent diffusion. The diffusivity between polymer-solvent was modeled according to the Vrentas & Duda Free Volume Theory, including the Sanchez-Lacombe equation-of-state. The model was written in terms of Partial Differential Equation, and solved with MAPLE® for a given initial size. Additionally, it is a moving boundary problem because the diffusion of the solvent out of the droplet leads to its size reduction. In this work only the diffusion behavior of one droplet was considered. The dynamic simulation shows both the evolution of the solvent inside the droplet and the variation of size in time.

ARE MONOD MODELS ENOUGH FOR BIOREACTOR CONTROL? PART II – SOME SIMULATION RESULTS

Abstract In this contribution a cybernetic model and a simple Monod based model were used to simulate the control of glucose and ethanol in Saccharomyces cerevisiae fed-batch cultivation. The main objectives were to investigate the type of modeling, the choice of the control variable (glucose or ethanol) and the kind of control action which is more suitable to control the studied system. It was found that the cybernetic model produces better qualitative and quantitative results and also that ethanol control shows a smooth behavior when compared to glucose control. According to the results obtained, an on-off control strategy based on ethanol measurements was suggested as an easier way to control the system without using effectively a controller.

Monitoring and Control Based on a FIA-Biosensor System with Automatic Correction

Abstract In this work, an enzymatic FIA-Biosensor system for lactose is used to on-line monitoring and control of a cultivation process to produce the enzyme lactase. Temperature effects on the output signal of the sensor are also studied. The whole system is modeled and, with this model, an automatic compensation is proposed to deal with temperature variations.