Search

The Computational Fluid Dynamics (CFD) technique was used to study the effect of the geometrical parameters on the intensification of hydrodynamic parameters and Carbon dioxide (C O 2) absorption efficiency by Mono-Ethanol-Amine (MEA) solution in a novel Arc-blade Rotating Packed Bed reactor (Arc-RPB). The VOF multiphase approach and RNG k - ε turbulent model were used to study the multiphase and turbulent behavior of the flow field in the computational domain, respectively. The effects of liquid flow rate, the bed wire mesh diameter, and number of the mesh layers were studied on the liquid holdup (α L) , gas-liquid interfacial area (A int) and liquid dispersion index (I d). Considering the Arc-RPB bed wire mesh diameter from 0.2 mm to 1.6 mm showed that α L , A int and I d decrease 20%, 29% and 11%, respectively. By examining the effect of the bed layers enhancement, the liquid holdup α L decreased by about 36% but, A int and I d reached a maximum value at 16 bed layers. Finally, with study the effect of the bed geometrical parameters on C O 2 absorption efficiency, results showed the efficiency reduces about 7.5% with increase in the bed mesh diameter and shows a maximum at 16 bed layers. [Display omitted] • Study of geometrical parameters effects on the hydrodynamic characteristics of a novel Arc-RPB by CFD simulation. • High effect of bed wire mesh and bed layer number in intensification of absorption efficiency of C O 2 by MEA solution. • Dependence of absorption efficiency to the liquid dispersion index. [ABSTRACT FROM AUTHOR]

In the present study, the Eulerian-Eulerian multiphase approach was implemented to simulate the hydrotreating processes (hydrodesulfurization (HDS) and hydrodearomatization (HDA)) in the trickle bed reactor (TBR) by means of computational fluid dynamics (CFD) technique. A new gas-liquid interphase heat transfer coefficient was used in the CFD model to predict the reactor performance at non-isothermal conditions. The effects of feed inlet temperature, gas and liquid velocities, operational pressure and hydrogen sulfide concentration of the gas phase were investigated to calculate the reactions conversions and the bed temperature distribution. Also, the influence of bed porosity and adiabatic operational conditions on the reactor temperature and HDS reaction conversion was discussed. The results showed at adiabatic reactor, HDS reaction conversion increased about 9% compared to constant wall temperature condition. Furthermore, it was found that neglecting the gas-liquid interphase heat transfer effect in the CFD model increases the average relative error between numerical and the experimental data at prediction of HDS conversion about 5% and also affects the bed temperature pattern. [ABSTRACT FROM AUTHOR]

In the present work, two different models (micro-scale and meso-scale) were developed to investigate the heat transfer between the gas and liquid phases in the trickle flow regime of a packed bed reactor. In the micro-scale model, a simplified description of bed geometry, known as the double-slit model, was implemented to study the effects of different operating parameters, in terms of gas and liquid Reynolds, Prandtl and Eötvös numbers, on the interfacial Nusselt number. In the meso-scale model, the Volume-of-Fluid (VOF) approach was used to simulate trilobe, cylindrical and spherical catalyst shapes and accurately predict the effects of interface morphology and bed geometry on interfacial heat transfer. To validate the implemented methods, a simple packed bed reactor with spherical catalysts was developed to experimentally investigate the interfacial heat transfer of co-current, downward gas–liquid film flows. The results obtained from CFD simulations and experimental data were in agreement and accurately predicted bed reactor temperature profiles with a mean relative error of 2.15%. In both the micro- and meso-scale models, an increase in the Reynolds and Prandtl numbers increased the interfacial Nusselt number; whereas, an increase in dimensionless groups of the liquid phase or the Eötvös number caused the opposite effect. Finally, a new correlation was proposed that evaluated the gas–liquid Nusselt number in the trickle flow regime with a standard deviation of 7.19% compared to results acquired using the micro-scale model. [ABSTRACT FROM AUTHOR]