Three-Dimensional Simulation on Liquid Flow in a Rotating Packed

Subscriber access provided by UNIV OF NEW ENGLAND ARMIDALE

Article

3D Simulation on Liquid Flow in a Rotating Packed Bed Reactor Tian-Yu Guo, Kun-Peng Cheng, Li-Xiong Wen, Ronnie Andersson, and Jianfeng Chen Ind. Eng. Chem. Res., Just Accepted Manuscript • DOI: 10.1021/acs.iecr.7b01759 • Publication Date (Web): 29 Jun 2017 Downloaded from http://pubs.acs.org on July 1, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Industrial & Engineering Chemistry Research is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Industrial & Engineering Chemistry Research

3D Simulation on Liquid Flow in a Rotating Packed Bed Reactor

Tian-Yu Guoa, Kun-Peng Chenga, Li-Xiong Wena,b,*, Ronnie Anderssonc,*, Jian-Feng Chena,b a

State Key Laboratory of Organic-Inorganic Composites; Beijing University of Chemical Technology, Beijing 100029, China

b

Research Center of the Ministry of Education for High Gravity Engineering and

Technology, Beijing University of Chemical Technology, Beijing 100029, China c

Department of Chemistry and Chemical Engineering, Chalmers University of Technology, Gothenburg, SE-41296, Sweden

*

Corresponding author.

E-mail: [email protected] (L.X. Wen) [email protected] (R. Andersson)

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Abstract Rotating packed bed (RPB) reactors, with strong centrifugal acceleration up to several hundred times gravitational acceleration, can greatly intensify the gas-liquid mass transfer efficiency. With the improvement of high performance computer clusters and simulation models, a more precise and comprehensive simulation on the gas-liquid flows in RPB has been achieved by a three-dimensional model. The volume of fluid (VOF) multiphase model, sliding model (SM), and different turbulence models were used to compute the velocity fields and capture the evolution of the gas-liquid interface in the RPB reactor. The liquid flow behavior, droplet size, liquid phase distribution, specific surface area and mean residence time (MRT) of the liquid phase within RPB were studied. Compared with 2D simulations, the 3D simulation model can not only describe the liquid breakage and coalescence processes within RPB more clearly, but also obtain results in more satisfactory agreement with experimental data. Keywords: Rotating packed bed (RPB); CFD; Liquid flow; Three-dimensional simulation; VOF

1. Introduction Rotating packed bed (RPB) reactors, one of the most efficient process intensification apparatuses, can greatly enhance the gas-liquid mass transfer efficiency by strong centrifugal acceleration.1 It has been used in many industrial processes, such as absorption,2 distillation,3 ozone oxidation4 and synthesis of nanoparticles,5 due to


Page 2 of 41

Page 3 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


the advantages of less capital cost, smaller device size and environment-friendly.6 Within the RPB reactor, as liquid streams being sprayed into the RPB packing through distributors, they are split continuously by the high speed rotating packing and begin to break up into thin film and small droplets along the path direction. Consequently, the surface area of the liquid phase will be increased dramatically and, furthermore, the liquid surface will be updated constantly, which could significantly enhance the gas-liquid mass transfer. The effective interfacial area for mass transfer is provided by the updated liquid film and droplets between and over the packing material. It was reported that the effective interfacial area of gas-liquid mass transfer depends on the contact device, operational conditions and the properties of the gas-liquid system.7 Therefore, deep understanding of the flow behaviors and characteristics of the liquid phase within RPB is essential for its design, optimization and scaling up. In recent years, many experimental investigations have been performed to study the characteristics of fluid flow within RPB reactor. The liquid flow has been assumed to be laminar film flow on the packing surface.8,9 Three kinds of flow patterns (film flow, droplet flow and pore flow) and the maldistribution of flow within the RPB were reported by Burns and Ramshaw.10 By mounting a camera on the rotor, the end effect region was confirmed,11 and it turned to be the most intensive part of RPB in processes of liquid impinging, deforming and mixing. With the help of an image analyzer and two electro-conductivity sensors, the thickness of liquid film and the mean residence time (MRT) of liquid within a RPB were obtained, respectively.12 The



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

liquid diameter was measured in the range of 0.15~0.9 mm and droplet velocities in the cavity regions (outer periphery of the packing) were obtained by a high-speed camera.13 By using a chemical absorption process of CO2 in NaOH solution, the effects of radial packing thickness, rotating speed, and the liquid and gas volumetric flow rates on the effective interfacial area within a multi-stage centrifugal RPB were studied.14 However, it is difficult to acquire the flow characteristics within RPB reactors by experiments due to their complicated structure and the limitation of measuring instruments. Therefore, a high resolving computational fluid dynamics (CFD) model is developed in this work that allows studies on how the fluid phases and the packing material inside RPB interact. It is a powerful tool for analyzing the flow patterns, phase distribution and mass transfer processes in numerous reactors.15-18 As yet, only a few numerical models, by solving Navier-Stokes equations, have been published for simulating the fluid behavior in RPB. A 3D non-steady state turbulent rotating single-phase flow was simulated with CFD and the overall dry pressure drop of RPB was calculated.19 Simple 2D and 3D models were developed by Yang et al. to predict the single liquid phase flow velocity at different rotating speeds.20 For gas-liquid two phase flow, a 2D VOF model was developed by Shi.21 The multiphase flow in RPB reactors was limited to 2D simulations in the previous CFD modeling. In 2D RPB model, the actual interfacial tension and formation of different kinds of flows could not be described precisely since the effect of the third dimension on velocity, fluid stress and phase fields was neglected. Therefore, the understanding of liquid flow


Page 4 of 41

Page 5 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


behaviors in RPB in terms of three-dimension is absolutely necessary. The volume of fluid (VOF) model, along with a continuum interfacial tension force model, was used in this work to capture the interface of immiscible fluids by solving a single set of momentum equations. Numerical studies confirmed that the volume of fluid (VOF) approach is well suitable for tracking the volume fraction of each phase and can yield good predictions,22,23 including interface dynamics and evolution of droplets in the micro- to millimeter range.24 A fully automated process, dynamic gradient adaption, was used in this study. The dynamic gradient adaption executes the gradient adaption automatically, allowing simulation of flows where the shape of the domain is changing in time. By using the dynamic adaption function, the number of mesh points is reduced, providing high resolution in regions of the evolving interface, thereby making the speed of transient 3D multiphase simulation acceptable. In this study, the VOF method combined with sliding model (SM) and the realizable k-ε model was used to simulate the velocity field and liquid volume fraction distributions in the rotating packing area. In addition, a comparative study using different turbulence model, k-ω, was done. The three-dimensional computational framework of RPB reactor was developed, meshed and run using ANSYS workbench 16.0 in order to obtain better understanding of the behavior of multiphase flows within RPB reactors. The characteristics of the RPB reactor was evaluated with respect to the liquid flow behavior, the mean diameter of liquid, the size distribution of liquid droplets, the specific area and residence time distribution (RTD) of liquid in the packing area. Finally, the numerical results were validated with experimental data



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

and compared with 2D simulation results.

2. CFD simulations 2.1 Geometrical model and mesh generation Some necessary simplifications were applied in this 3D simulation study due to the complexity of packing structure within RPB reactor. The multi-layer wire mesh packing was assumed to be composed of numerous straight wires parallel to the rotor axis. Due to the limit of the computer capacity and the fact that the effects of the parallel round wires along the axial height direction on a specific liquid flow element were less significant compared to those caused by the round wires in the coaxial direction (wires within the same cross-section perpendicular to the rotor axis),21 the axial height of packing was chosen as 5 mm. Detailed dimensions of the RPB model are presented in Figure 1. The center distance between two adjacent layers of packing in radial direction, and the center distance between two adjacent packing wires (squares) in the circumferential direction were both 1.5 mm. The cross-sectional area of a single wire was 0.25 mm2. In this work, the geometry of the computational domain was generated in Design Modeler, and meshed in Meshing-ICEM. The cells were a mixture of hexahedron (~99%) and tetrahedron (~1%) with a size between 0.1~0.25 mm (Figure 2). The orthogonality of the cells was high for all cells and the skewness of the tetrahedral cells was low, confirming high quality of the mesh. Grid independence was explored with different cells in each case. Results of 54 mm thickness packing in RPB (Figure 3) using 640,000, 860,000, 1,150,000 cells and up


Page 6 of 41

Page 7 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


to 3 levels dynamic mesh adaption were analyzed and it showed that, a mesh containing at least 860,000 cells with 2-levels of dynamic mesh adaption, corresponding to 1.6 million cells in total, was required to resolve the flow field sufficiently. The simulations were run in parallel on a computer cluster to achieve simulation time below 60 hours in each analysis. Grid independent results of RPB with other thickness packing were also investigated, and a 2-levels of dynamic mesh adaption turned to be required for all cases.

Liquid inlet

Φ0 5 mm

Φ

(a)

1 3

2

4

(b)



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 1. (a) 3D physical model diagram of RPB reactor; (b) Top view of the physical model of a RPB and partial schematic illustration of the packing arrangement (1, packing area; 2, cavity area; 3, center distance in the circumferential direction; 4, center distance in the radial direction).

Figure 2. Three dimensional meshed geometry of RPB reactor.

2.2 Mathematical model The flow in RPB reactor can be treated as a transient, incompressible gas-liquid phase flow according to the dynamic behaviors observed from the experiments. Since the drop-wall interaction controls the interfacial area at low turbulent stresses, there is no need to resolve the individual turbulence structures. Therefore, a RANS-VOF, instead of a LES-VOF model, was chosen to investigate the liquid-gas flow in RPB reactor in previous study. The VOF model has also been confirmed suitable for the prediction of jet breakup, the motion of liquid after a dam break, and the steady or transient tracking of any gas-liquid interface.25 The mass conservation and momentum equations for the volume fraction function in the computational domain can be


Page 8 of 41

Page 9 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


expressed as follows, respectively:26

∂ (α q ρ q ) ∂t

( )

+ ∇ ⋅ α q ρ q u q  = 0   with q = g or l

(1)

( )

(2)

where ρ is the fluid density,

u is the velocity vector, µ is the dynamic viscosity, p is

T ∂ ρu + ∇ ⋅ ρ u u = −∇ P +  µ  ∇u + ∇u  + FV  ∂t  

pressure, subscripts l and g represent the liquid and gas phases, respectively, t is the time, and αq is the volume fraction for phase q. The volume-fraction-averaged material properties take the following form:

ρ =

∑α ρ

q

(3)

µ =

∑α µ

q

(4)

q

q

In addition, the volume fraction equation will be solved in each computational cell throughout the domain only for the second phase in this simulation, and the first phase will be computed based on the following constraint: n

∑α q =1

q

= 1 (5)

where αq as 1 represents the computational cell full of the qth fluid. The interface between the qth fluid and one or more other fluids can be found in the cells with 0

Three-Dimensional Simulation on Liquid Flow in a Rotating Packed

Recommend Documents