Felipe López-Isunza

5-Lump kinetic model for gas oil catalytic cracking

Abstract A new 5-lump kinetic model is proposed to describe the gas oil catalytic cracking (FCC) process. The model contains eight kinetic constants, including one for catalyst deactivation, taking into account LPG (combined C 3 –C 4 ), dry gas (C 2 and lighter) and coke yields separately from other lumps (unconverted gas oil and gasoline). Apparent activation energies were determined from experiments obtained in a microactivity reactor (MAT) at temperatures: 480°C, 500°C and 520°C; for a catalyst-to-oil ratio of 5 using vacuum gas oil and equilibrium catalyst, both recovered from an industrial FCC unit. Product yields predicted by this model show good agreement with experimental data.

Approach to the Analysis of the Dynamics of Industrial FCC Units

Abstract Although it is known that FCC units are operated at a controllable steady state, it is commonly assumed that this state is pseudo stable. An ignited steady state, theoretically stable, outside the operable domain has also been reported for the same set of operating parameters. The aim of this work is to introduce a theoretical method for the preliminary analysis of the controllability of an industrial FCC unit, at any set of conditions, using a nonlinear model directly. An approach for the analysis of the stability of the system zero dynamics is used to investigate the dynamic behaviour of two common operating policies when the unit is operating either in the standard or in the ignited steady state. It is used to predict some particular characteristics of the process, such as the presence of inverse response after the change in catalyst circulation rate, a common control action. The results derived from the analysis developed are discussed in terms of physical responses and verified by performing open loop dynamic simulations of an industrial FCC unit. The operation of the unit is simulated for both the standard and the ignited steady states.

Hydrodynamic Models for Packed Beds with Low Tube-to-Particle Diameter Ratio

In the last decade it has been a special interest to incorporate the hydrodynamics in packed bed reactor models. This seems to be important in the case of highly exothermic partial oxidation reactions normally performed in packed beds with low tube/particle diameter ratio (dt/dp< 5) because of the large void distributions in the radial and axial directions, which have a direct impact on the magnitude of radial, angular and axial profiles of the velocity field, and consequently on both, the temperature and concentration profiles in the catalytic reactor. A successful reactor model needs an adequate hydrodynamic description of the packed bed, and for this reason several models additionally incorporate empirical expressions to describe radial voidage profiles, and use viscous (Darcy) and inertial (Forchheimer) terms to account for gas-solid interactions, via Erguns pressure drop equation. In several cases an effective viscosity parameter has also been used with the Brinkmans viscous term. The use of these various approaches introduce some uncertainty in the predicted results, as to which extent the use of a particular radial voidage expression, or the use of an effective viscosity parameter, yield reliable predictions of measured velocity profiles.In this work the predictions of radial velocity profiles in a packed bed with low tube to particle diameter ratio from six hydrodynamic models, derived from a general one, are compared. The calculations show that the use of an effective viscosity parameter to predict experimental data can be avoided, if the magnitude of the two parameters in Erguns equation, related to viscous and inertial energy losses, are re-estimated from velocity measurements, for this particular packed bed. The predictions using both approaches adequately fit the experimental data, although the results are analyzed and discussed.

Comparison of two dynamic models for FCC units

Modelling of FCC units has been an interesting activity because of the complexity of the system and the economic incentives associated. Models have been classified depending on either the used kinetic scheme or the proposed reactor configuration. The main problems to assess are the elimination of coke when catalyst is regenerated and the delicate interaction between reactor and regenerator due to the global energy balance. In this work two models, each one using a different kinetic scheme, a different formulation for the catalyst activity and a different conception of the regenerator are compared when simulating the same FCC unit. Steady-state predictions and dynamic phenomena such as stability and multiplicity are simulated. In order to emphasise the ability for predicting the dynamic phenomena, the obtained results are compared in terms of accuracy and computing resolution time. It is pointed out that even though the complexity of a model could be a limiting factor for control purposes, in the case of dynamic studies this factor is not a constraint and it is possible to use more complex models. It is also emphasised that the most useful model will be the one which fulfils the requirements of the researcher, the situation to be modelled and the results that could be necessary.

Dynamic modelling of an industrial fluid catalytic cracking unit

Abstract A fluid catalytic cracking unit is described by a dynamic model which considers both reactor and regenerator behaviour and their interaction. A lumped kinetics were used for the cracking reactions and coke deposition which occur in the riser. The regenerator is described by a fluidized movingbed reactor with recycle, where coke is burned off with air. Simulation studies of the complex interactions between both reactors were performed for changes in the operating variables.

Modeling the Transient VOC (toluene) Oxidation in a Packed-Bed Catalytic Reactor

Abstract A model is developed to study the transient behavior of a non-isothermal, non-adiabatic packed-bed reactor during VOC (toluene) oxidation with air on a mixed-oxide catalyst via Mars-van Krevelen kinetic scheme. The aim is to find a safe reactor design and operating conditions for VOC elimination, which has been collected in a battery of adsorption units from dilute VOC streams. Once each adsorption column is saturated, a non-isothermal desorption takes place, and the gas stream exiting the sequence of VOC desorption columns feeds continuously the catalytic reactor for VOC elimination. The reactor model describes a 2D two-phase system interacting through the gas-solid interphase, including convection and axial and radial dispersions of mass and heat. The simulations show that the gas flow velocity, and reactor and particle diameters, are key parameters to achieve a safe design, and that traveling reaction fronts in the packed-bed exist when a series of reversible stepwise changes are performed in the concentration and temperature at the feed, as a result of the transient balance between heat generation and heat elimination along the packed-bed. When comparing the perturbation in VOC concentration at the feed versus those in temperature, a large parametric sensitivity is observed for the latter case without the presence of multiple steady states. Due to the uncertainty in the values of the effective heat transport parameters, transient responses of different magnitude are observed for the same operating conditions when using heat transport parameter of different magnitude.

Modelling of the Transient Behaviour of a Membrane-Attached Biofilm Reactor under Successive Pulses of a Synthetic Wastewater Substrate

This work deals with the theoretical and experimental study of the transient behaviour of a membrane-attached biofilm reactor (MARB) when it is exposed to a series of pulse injections of concentrated solutions of sodium acetate, used as a synthetic wastewater. The MARB is connected to a reservoir tank with full recirculation containing the synthetic wastewater, and oxygen permeates through the wall membrane to the biofilm attached to it. For the two experiments reported in this work air is also sparged into the residual water in the tank providing an extra source of oxygen that diffuses simultaneously from the membrane and from the liquid into the biofilm. A pseudo-heterogeneous model using Monod kinetics with dual substrate limitation was employed to predict the observed evolution of substrate and dissolved oxygen concentrations in the MABR. The model accounts for the counter-diffusion of substrate and oxygen as well as for the reaction within the biofilm. It also predicts biomass growth and the production of extra cellular polymers, which in turn causes the biofilm to grow. Transport and kinetic parameters previously estimated, are used in the model to predict the growth rates in the biofilm and allow the analysis of the relative contribution of the rates of mass transport by diffusion, convection and growth reaction.

An strategy to regulate temperature in FCC units with poor knowledge of chemical kinetics

In this work we study the temperature control in FCC units. Because FCC processes involve complex interactive dynamics which are difficult to operate and control as well as poorly known reaction kinetics, a model-based control controller which yields robust stabilization against chemical reactions uncertainties is provided. The control strategy makes use of estimates of uncertainties derived from calorimetric balances to give an input-output linearizing-like feedback.