R. Schneider

Reactive absorption : Optimal process design via optimal modelling

Optimal design of complex reactive separations is inconceivable unless reliable process models are available. Such models have to be both rigorous enough in order to reflect the process complexity and simple enough in order to ensure feasibility of process simulations. In this respect, an optimal model should represent a kind of a consensus between the rigour and simplicity, and its development requires a comparison of both detailed and simplified process models with experimental data. From this viewpoint, reactive absorption operations with their pronounced kinetic character and complex non-ideal behaviour provide one of the best objects for studying. This work gives a detailed analysis of the problem supplemented by the examples of industrial importance.

Dynamic Modelling of Reactive Absorption with the Maxwell-Stefan Approach

Modelling and design of reactive absorption are based on the theoretical description of the reaction and mass transport in multicomponent systems. The multicomponent nature of these phenomena leads to complex process behaviour due to the superposition of many driving forces multicomponent diffusion, chemical interactions, convective flows, multicomponent thermodynamic interplay, etc. For this reason, adequate theoretical description of multicomponent reactive systems calls for the application of the Maxwell-Stefan equations and, further, for the use of coupled mass transfer equations together with the relevant reaction kinetics. On this basis, a two-phase, gas-liquid reactive system is considered and a general dynamic model is developed for its design. Both the film and bulk reaction mechanisms are allowed for. This dynamic rate-based approach leads to a system of partial differential equations, which have to be discretized in the axial direction. The resulting DAE system is solved numerically. As an application example, the reactive absorption of sour gases in an air purification process with packed columns is simulated. For this case, an additional account of the electrical potential gradient is involved because of the presence of electrolytes. Simulation results are presented for the H 2 S scrubber with three liquid distributors and a structured packing section. For the validation of the model, pilot plant steady state experiments were carried out at Thyssen Still Otto in Duisburg, Germany. The simulation results are in good agreement with the experimental data.

Rigorous dynamic modelling of complex reactive absorption processes

Reactive absorption processes usually represent kinetically controlled operations in which complex reactions are combined with multicomponent mass transfer. For this reason, an adequate theoretical description of reactive absorption should be based on the rate-based approach, with the direct consideration of the process rates. Along these lines, a rigorous dynamic model allowing for multicomponent two-phase mass transfer, film and bulk reaction mechanisms, thermodynamic nonidealities, fluid dynamics and heat effects is developed. For systems including electrolytes, an additional account of the electrical potential gradient is involved. The whole dynamic rate-based model represents a system of partial differential equations which has to be solved numerically. To validate the model, the reactive absorption of sour gases in an air purification process with packed columns is simulated. Several simulation results including different sensitivity studies are presented. The model is also tested against pilot-plant steady-state experiments. The simulation results are in good agreement with the experimental data.

Dynamic modelling and simulation of reactive batch distillation

Abstract Reactive batch distillation offers an innovative possibility to combine the benefits of reactive distillation as an integrated process with a significant reduction of capital investment and batch processes, providing additional flexibility. The predictive process design requires a detailed model considering both, the dynamic character of batch processes and the influence of chemical reactions on the mass transfer. Therefore, a rigorous dynamic rate-based approach has been developed including the explicit calculation of heat and mass transfer rates. Diffusion phenomena in multicomponent mixtures resulting from molecular interactions are taken into account via the Maxwell-Stefan equations. Thermodynamic non-idealities have been considered in the description of the phase equilibrium and the transport equations. The chemical reaction kinetics have been implemented using a quasi-homogeneous approach which is also suitable for the heterogeneous catalysis. The resulting highly nonlinear system of equations had to be discretized and implemented into a numerical solver. The developed model has been applied to a semi-batch process for the production of methyl acetate, performed in a packed column with a new catalytic packing. For the model validation, several experiments have been carried out in a pilot plant column. A comparison of the simulation results and the experimental data and numerical investigations for the determination of the most sensitive parameters are presented.

Dynamic simulation of industrial reactive absorption processes

Accurate dynamic models for industrial reactive absorption processes have to be both rigorous enough in order to reflect the process complexity and simple enough in order to ensure feasibility of process simulations. A comparison of different model approaches revealed that traditional equilibrium stage models and efficiency approaches are inadequate in this context. Therefore, we have developed a new rigorous dynamic two-phase model based on the two-film theory, which serves as a reference description and takes into account the influence of chemical reactions and additional driving forces in electrolyte systems on mass transfer, thermodynamic non-idealities and the impact of structured packings and liquid distributors on the process hydrodynamics. For the model-based control and dynamic on-line simulation some model reductions are required to limit the calculation time for the numerical solution. According to sensitivity study results, simplifications for some physical parameters and a linearisation of the film concentration profiles have been accomplished leading to a reduction of the total number of equations by half. The model has been applied to an industrial sour gas purification process with many parallel and consecutive reversible chemical reactions within a multicomponent system of aqueous electrolytes and validated by pilot plant experiments.

Multicomponent unsteady-state film model : a general analytical solution to the linearized diffusion-reaction problem

In this paper, a classical film model is extended to account for unsteady-state multicomponent diffusion and reaction coupling. The governing matrix-form partial differential equation is complemented by the initial and boundary conditions formulated as general vector functions. An additional linearization of the reaction source term results in the formulation which allows an analytical handling of the problem. The latter is found to be successful even with the generally set initial and boundary conditions. The solution approach combines the superposition principle and the method of the separation of variables extended for the matrix operations. The exact analytical matrix solution obtained is a generalization of many simpler problems and either can be employed by itself or provides suitable preset values for relevant numerical simulations of industrial-scale reactive separation operations.

Dynamic and steady state simulation of coke oven gas purification

Abstract The purification of coke oven gas represents a complex multicomponent separation process combined with many parallel reversible chemical reactions. For the theoretical description of this process, traditional equilibrium stage models and efficiency approaches are usually inadequate. Therefore, a rigorous dynamic two-phase model based on the two-film theory has been developed, which takes into account diffusional interactions, the influence of chemical reactions on mass transfer, thermodynamic non-idealities and the impact of structured packings and liquid distributors on the process hydrodynamics. In addition, the influence of driving forces in electrolyte systems like electrical potential gradients are considered. This rigorous dynamic rate-based approach leads to a system of algebraic and differential equations, which has to be solved numerically. For the H 2 S scrubber, steady-state and dynamic simulation results are presented. The model validation and the determination of parameters require steady-state and dynamic experiments, which have been carried out in a pilot-scale gas scrubber, equipped with a process control system. The simulation results are in good agreement with the experimental data and hence, the developed dynamic model serves as a basis for the model-based control.

A cascade screening approach for the identification of Bcr-Abl myristate pocket binders active against wild type and T315I mutant

The major clinical challenge in drug-resistant chronic myelogenous leukemia (CML) is currently represented by the Bcr-Abl T315I mutant, which is unresponsive to treatment with common first and second generation ATP-competitive tyrosine kinase inhibitors (TKIs). Allosteric inhibition of Bcr-Abl represent a new frontier in the fight against resistant leukemia and few candidates have been identified in the last few years. Among these, myristate pocket (MP) binders discovered by Novartis (e.g. GNF2/5) showed promising results, although they proved to be active against the T315I mutant only in combination with first and second generation ATP-competitive inhibitors. Here we used a cascade screening approach based on sequential fluorescence polarization (FP) screening, in silico docking/dynamics studies and kinetic-enzymatic studies to identify novel MP binders. A pyrazolo[3,4-d]pyrimidine derivative (6) has been identified as a promising allosteric inhibitor active on 32D leukemia cell lines (expressing Bcr-Abl WT and T315I) with no need of combination with any ATP-competitive inhibitor.

Dynamic Simulation of Reactive Absorption Processes for the Purification of Coke Oven Gases

Reactive absorption processes represent complex mass transfer of multicomponent systems combined with chemical reactions. Therefore, traditional equilibrium stage models and efficiency approaches are usually inadequate and a rigorous two-phase model, based on the two film theory had to be developed. This dynamic model takes into account diffusional interactions and the influence of chemical reactions on mass transfer as well as thermodynamic non-idealities and considers the impact of special column internals like structured packings on hydrodynamics. By making use of the Nernst-Planck equation for the description of mass transfer, the influence of additional driving forces in electrolyte systems like electrical potential gradients are expressed. This rigorous dynamic rate-based approach leads to a system of partial differential equations, which had to be discretized in axial direction. The resulting DAE system has been implemented into SpeedUp® and solved numerically. For the H 2 S scrubber some simulation results are presented with the consideration of three liquid distributors and a structured packing section. For the validation of the model, pilot plant steady state experiments have been carried out at Thyssen Still Otto in Duisburg. The simulation results are in good agreement with the experimental data.