Short-Cut Models for Precipitation in Flowsheet Simulation: Experimental and Numerical Investigation of a Two-Zone Model for Stirred Tank Reactors

Short-Cut Models for Precipitation in Flowsheet Simulation: Experimental and Numerical Investigation of a Two-Zone Model for Stirred Tank Reactors

H. Rehage¹, M. Kind¹

¹Karlsruhe Institute of Technology, Department of Thermal Process Engineering, Karlsruhe, Germany

Keywords: flow sheet simulation, precipitation, numerical simulation, stirred tank reactor

Introduction

Precipitation is an important operation in solid process engineering, used in the production of many products related to the chemical or pharmaceutical industry. Typical size ranges for precipitated particles are nm to µm. Characteristic for many precipitation systems is the mixing limitation, due to the the fast solid formation step. Therefore process conditions (e.g. flow rate or mixing geometry) can crucially influence particle properties like size distribution or morphology.

The fast solids formation processes (~ms) in contrast to slowly changing process conditions (min – h) pose a serious challenge for flowsheet simulation. The complexity of precipitation modelling depends very much on the chosen mixing geometry. For Confined Imping Jet Mixers (CIJM), which allow a defined residence time distribution, several research groups [1–3] achieved simulation results which agreed well with simulation results in the past. However, the stirred tank reactor, operated in semi-batch, is the standard apparatus for industrial precipitation. The complexity of the flow and the residence time distribution make CFD simulations of the whole reactor possible only for short process time scales. Several zone models for simplification of the complex current in the feed region of stirred tanks have been developed in the past [4–9]. Most of these models are based on the determination of the turbulent dissipation rate ε in the mixing region coupled with a micro mixing model (e.g. Engulfment Model). However, these types of models only hold under fully turbulent conditions and a negligible influence of the feed pipe momentum. Since in many processes these two conditions are not fulfilled, our aim is therefore the development of a new zone-model method for the semi-batch precipitation in stirred tank reactors.

In our new short-cut model approach, the semi-batch stirred tank reactor is divided into two zones, a mixing zone and an ideally mixed holdup. The small mixing zone includes the feed pipe and its immediate environment and accounts for mixing, nucleation and particle growth. In contrast to common models based on dissipation of turbulent energy, in our model the mixing zone is not assumed to be ideally mixed. Instead it is parameterized as a Jet in Cross Flow (JICF) or Coaxial Flow (CF) reactor, dependent on the main flow direction relative to the feed pipe.

This mixing zone is connected with the bulk volume which is not close to the feed pipe. The fluid in the bulk volume is assumed to not take part in the precipitation process. The ratio of bulk volume mixed with the feed volume depends on the stirrer type, the size of the stirrer and its speed. For this reason, the specific flow number of the stirrer is one input parameter in our model. A central advantage of this zone model is the possibility of experimental validation. This validation is done through comparison between standard semi-batch stirred tank precipitation and an experimental replication of the two-zone-system with JICF or CF reactor.

Materials and Methods

For the experimental investigation of our zone-model approach, an experimental set-up is chosen, which allows for the comparison between particle attributes of a standard semi-batch stirred tank precipitation and the results of semi-batch reactor connected to a separate mixing compartment. This feed region is designed as Jet in Cross Flow/Coaxial Flow reactor, dependent on the chosen stirrer in the bulk reactor. For validation of the two-zone model, experiments with different stirrer speeds, stirrer types and feed-inlet velocities were carried out with barium sulfate as precipitation system at an initial supersaturation of 1000 (calculated. with Pitzer model [10]). Simulative methods using OpenFoam 4.1 are used to investigate mixing and supersaturation build-up in the Jet in Crossflow arrangement. These numerical simulations are used to ‘measure’ supersaturation as driving force of precipitation [11,12]. The results of this “measurement” are transferred to a 1-D-Population balance solver for the calculation of the particle size distribution. Different stirrer speeds, feed pipe diameters, free lattice ion ratio and supersaturations are investigated.

Results:

The experimental results show, that with our zone model the particle size can be approximated in the mixing limited region. Since experimental validation of zone models for precipitation is a new research field to the author’s knowledge, we tested different pumping setups and found out that in experimental validation the chosen pump is a key influence factor for a successful validation. Therefore, different experimental setups are introduced and compared. Furthermore it could be shown, that pulsation of the pump is of crucial importance. This effect is further investigated using a specific pulsation pump.

With the developed Open FOAM-PBE coupled method it was possible to show the same trends for the jet velocity influence on the precipitation compared to the semi-batch reactor experiment. This allows a promising outlook on the future work to find a short-cut model for precipitation in stirred tanks for the flowsheet simulation framework DYSSOL [13].

Acknowledgments: Financial support for this work by the DFG (German Research Foundation, SPP 1679) is gratefully acknowledged.

References

[1] L. Metzger, M. Kind, Influence of mixing on particle formation of fast precipitation reactions—A new coarse graining method using CFD calculations as a “measuring” instrument, The 15th European Conference on Mixing 108 (2016) 176–185.

[2] D.L. Marchisio, L. Rivautella, A.A. Barresi, Design and scale-up of chemical reactors for nanoparticle precipitation, AIChE J. 52 (5) (2006) 1877–1887.

[3] J. Gradl, H.-C. Schwarzer, F. Schwertfirm, M. Manhart, W. Peukert, Precipitation of nanoparticles in a T-mixer: Coupling the particle population dynamics with hydrodynamics through direct numerical simulation, Particulate Processes A Special Issue of Chemical Engineering and Processing 45 (10) (2006) 908–916.

[4] L. Metzger, M. Kind, Compartment Method for Dynamic Multi-Scale Simulation of Precipitation Reactors, 2014.

[5] V. Alopaeus, J. Koskinen, K.I. Keskinen, Simulation of the population balances for liquid–liquid systems in a nonideal stirred tank. Part 1 Description and qualitative validation of the model, Chemical Engineering Science 54 (24) (1999) 5887–5899.

[6] J. Baldyga, W. Podgórska, R. Pohorecki, Mixing-precipitation model with application to double feed semibatch precipitation, Chemical Engineering Science 50 (8) (1995) 1281–1300.

[7] J.R. Bourne, S. Yu, Investigation of micromixing in stirred tank reactors using parallel reactions, Ind. Eng. Chem. Res. 33 (1) (1994) 41–55.

[8] A. Jones, S. Rigopoulos, R. Zauner, Crystallization and precipitation engineering, in: A. Barbosa-Póvoa and H. Matos (Ed.), Computer Aided Chemical Engineering European Symposium on Computer-Aided Process Engineering-1437th European Symposium of the Working Party on Computer-Aided Process Engineering, Elsevier, 2004, pp. 75–86.

[9] R. Zauner, A.G. Jones, On the influence of mixing on crystal precipitation processes—application of the segregated feed model, Chemical Engineering Science 57 (5) (2002) 821–831.

[10] K.S. Pitzer, Thermodynamics of electrolytes. I. Theoretical basis and general equations, J. Phys. Chem. 77 (2) (1973) 268–277.

[11] L. Metzger, M. Kind, On the mixing in confined impinging jet mixers – Time scale analysis and scale-up using CFD coarse-graining methods, Chemical Engineering Research and Design 109 (2016) 464–476.

[12] L. Metzger, M. Kind, The influence of mixing on fast precipitation processes – A coupled 3D CFD-PBE approach using the direct quadrature method of moments (DQMOM), Chemical Engineering Science.

[13] V. Skorych, M. Dosta, E.-U. Hartge, S. Heinrich, Dyssol – Modeling Framework for the Dynamic Flowsheet Simulation of Solids Processes, Chemie Ingenieur Technik 88 (9) (2016) 1357.