A kinetic model for hydroisomerization of long chain paraffins has been developed by extending the single events modeling methodology to very large reaction networks. The lumped coefficients of the reaction network can now be directly calculated based on a decomposition strategy of the various intervening species into lateral chains and active centers. This approach advantageously avoids the time-consuming generation of the complete exhaustive list of molecules and single events reactions. The properties of lateral chains are computed through recursive formulas. The identified model is confronted to experimental data of squalane hydrocracking.

No abstract

An integrated experimental and modeling approach for the design of high productivity protein A chromatography is presented to maximize productivity in bioproduct manufacture. The approach consists of four steps: (1) small-scale experimentation, (2) model parameter estimation, (3) productivity optimization and (4) model validation with process verification. The integrated use of process experimentation and modeling enables fewer experiments to be performed, and thus minimizes the time and materials required in order to gain process understanding, which is of key importance during process development. The application of the approach is demonstrated for the capture of antibody by a novel silica-based high performance protein A adsorbent named AbSolute. In the example, a series of pulse injections and breakthrough experiments were performed to develop a lumped parameter model, which was then used to find the best design that optimizes the productivity of a batch protein A chromatographic process for human IgG capture. An optimum productivity of 2.9 kg L⁻¹ day⁻¹ for a column of 5mm diameter and 8.5 cm length was predicted, and subsequently verified experimentally, completing the whole process design approach in only 75 person-hours (or approximately 2 weeks).