Jet loop reactors are used for performing fast gas-liquid reactions such as hydrogenations, chlorinations, phosgenations and so on. Most often, in these reactions, gas phase contains only reacting gas and reactors are operated in a dead end manner (gas is fed to the reactor as per the consumption via pressure regulator). The performance of such reactors is controlled by gas-liquid mass transfer. It is therefore essential to improve gas-liquid mass transfer for enhancing performance of these reactors.

In jet loop reactors, external pump is used to circulate liquid (along with the catalyst and often some gas) through ejector type nozzle. The high velocity flow through throat of the nozzle generates lower pressure. The gas then gets induced through this low pressure region and very fine gas-liquid dispersion in generated in the nozzle. This fine dispersion provides excellent gas-liquid mass transfer. Despite significant potential for commercial applications, there are relatively few studies available in the open literature on hydrodynamics, gas induction and mass transfer in such jet loop reactors.

Gas inducing nozzle (ejector) is the critical design component of the jet loop reactor. Bhutada and Pangarkar^[1] and Brahim et al.^[2] have discussed influence of design parameters like liquid nozzle diameter, the throat diameter of the ejector, length of mixing tube etc. on gas induction and mass transfer. It was observed that gas induction rate and effective mass transfer rate realized in the reactor has a strong correlation. Despite the available studies, general understanding on relationship between nozzle configuration and gas induction performance is still not adequately established. Several new ideas such as swirling devices, introducing additional open area in ejector nozzles ^[3] have been proposed. The present work is focused on developing a CFD based methodology for estimating gas induction in nozzles used in jet loop reactors. The objective is to develop an approach which can then be used to evaluate different ideas of nozzle configurations and to help identification of most promising configurations for further investigations. Specific objectives of this work are:

1. Develop a methodology & model for estimating gas induction rates as a function of liquid flow rate based on nozzle geometry

2. Use the models to quantify influence of key geometric parameters on gas induction

3. Develop and identify promising ideas for enhancing gas induction

Methodology and preliminary results are discussed in the following.

Methodology

 Recently Kandakure et al^.[4] have developed CFD model for estimating gas induction. They have used two phase mixture model for this purpose. Notwithstanding some of the uncertainties associated with the mixture model, they have reported very good success in estimating gas induction. In this work, we further simplify the approach and have use a strategy to use single phase flow simulations to estimate gas induction rates. The motivation behind this is that single phase simulations require order of magnitude lower computing resources and will allow quick evaluation of large number of nozzle configurations. The methodology is developed to estimate gas induction using the pressure field simulated using the single phase CFD models. This is verified and validated by comparing the estimated (with one fitted parameter) gas induction rates with the published experimental data. After verifying the methodology, the model was used to screen several nozzle configurations for identifying most promising configurations.

 Results and Discussion

 Single phase simulations of flow in nozzles considered by Bhutada and Pangarkar^[1] were carried out. The simulated pressure fied was used to estimate gas induction rates. The approach and model were then extended to evaluate different nozzle configurations. The simulations were used to identify promising configurations of nozzles. These designs were then tested by carrying out experiments. An experimental set-up was established for this purpose. Gas induction rates were measured for identified nozzle designs. The newly proposed designs were compared with the reference configuration (of Bhutada and Pangarkar^[1]).

The details of computational model, experimental set-up and measurement procedures, new ideas for nozzle configuration and identified most promising nozzle configuration giving higher gas induction rates are discussed in the full manuscript. The approach and results will be useful to develop improved nozzle designs for jet loop reactors.

 References

[1]       S. Bhutada, V. Pangarkar, Chem. Eng. Comm. 61 (1987) 293-258

[2]       A. Ben Brahim, M. Prevost, R. Bugrael, Int. J. Multiphase Flow 10 (1984) 79-94

[3]       Y. Suryavanshi and V.V. Ranade (2005), NCL Internal report

[4]       Kandakure, M.T., V.G. Gaikar and A.W. Patwardhan, Chem. Eng. Sci., 60 (2005) 6391