CFD modeling of turbulent reacting flow in a semi-batch stirred-tank reactor

Xiaoxia Duan, Xin Feng, Chao Yang, Zaisha Mao

PII: S1004-9541(17)30310-5
DOI: doi:10.1016/j.cjche.2017.05.014
Reference: CJCHE 841

To appear in:

Received date: 15 March 2017

Revised date: 15 May 2017
Accepted date: 18 May 2017

Please cite this article as: Xiaoxia Duan, Xin Feng, Chao Yang, Zaisha Mao, CFD
modeling of turbulent reacting flow in a semi-batch stirred-tank reactor, (2017),
doi:10.1016/j.cjche.2017.05.014

This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.
 ACCEPTED MANUSCRIPT

Fluid Dynamics and Transport Phenomena

T
R IP
CFD modeling of turbulent reacting flow in a semi-batch

SC
☆
stirred-tank reactor

NU
Xiaoxia Duan1,2, Xin Feng1,2*, ChaoYang1,2*, Zaisha Mao1
MA
1. Key Laboratory of Green Process and Engineering, Institute of Process Engineering, Chinese
Academy of Sciences, Beijing 100190, China
2. University of Chinese Academy of Sciences, Beijing 100049, China
D
TE

Abstract: For the mixing-sensitive reactions, both chemical kinetics and mixing
P

conditions of the reactants determine the distributions of products. The direct

CE

quadrature method of moments combining with the interaction by exchange with the
AC

mean micro-mixing model (DQMOM-IEM) has been validated for the chemical

reacting flows in micro-reactors. Quite encouraging simulation results offer great

promise, but the applicability of this method is needed to be explored furthermore,

such as in stirred reactors. In this work, the two-environment DQMOM-IEM model

☆Supported by the National Basic Research Program of China (2012CB224806), the

National Natural Science Foundation of China (21306197, 91434126), the Major
National Scientific Instrument Development Project (21427814) and Jiangsu National
Synergetic Innovation Center for Advanced Materials.
*
Corresponding authors. E-mail addresses: xfeng@ipe.ac.cn (X. Feng), chaoyang@ipe.ac.cn (C.
Yang) .
1
 ACCEPTED MANUSCRIPT

was created with C language and used to customize Fluent through the user-defined

functions. The mixing effects on the course of parallel competing chemical reactions

carried out in a semi-batch single-phase stirred reactor were predicted. The simulation

T
IP
results show that the rising feed velocity enlarges the volume of reaction zone and

R
maximize the yield of the by-product, which also indicates that the feed stream is

SC
more difficultly dispersed into the main stream and the zone surrounding feedpipe exit

with high turbulent kinetic dissipation rate cannot be efficiently used.

NU
MA
Keywords: CFD; Mixing; Segregation; DQMOM-IEM; Mixture fraction; Stirred

tanks
D
P TE

1. Introduction
CE

Semi-batch stirred reactors are often encountered in chemical and pharmaceutical

industries, the particular reagent is added into the reactor with a feed-pipe. The scale
AC

of segregation of the feeding stream is reduced under the bulk inertial motion,

followed with the stretch, deformation and engulfment of feed elements [1]. At last,

molecular diffusion reduces the intensity of segregation down to zero when the

reagents achieve the molecular-scale homogeneity [2,3]. For the mixing-sensitive

multiple chemical reactions, if the characteristic reaction time is shorter than the time

required for bringing homogeneity at molecular scale, micro-mixing can significantly

influence the product quality and selectivity [4]. Mixing effects get more critical in

2
 ACCEPTED MANUSCRIPT

reactor scale-up because the chemical reactions are scale-independent, while the local

mixing time are sensitive to scale and position when the power per unit volume

remains constant [2]. Therefore, the studies of flow, mixing and chemical reaction that

T
IP
simultaneously occur in the stirred tank are significant to the design, optimization and

R
scale-up of reactors.

SC
In the CFD modeling of turbulent chemical reacting flow, the key problem is to

select an appropriate micro-mixing model that can describe the local concentration

NU
information to close the chemical source term. Different mixing models have been
MA
developed and the most commonly used models include the engulfment model [5,6]

and the probability density function methods, e.g. finite-node PDF [7,8],
D

multiple-time-scale turbulent mixer model [9-11], DQMOM-IEM model [12]. The

TE

DQMOM-IEM micro-mixing model is first presented by Fox [12] and has been
P

successfully implemented to model the precipitation process [13-16] and parallel

CE

reactions scheme [17-19], combustion [20,21], etc. This method, compared with the
AC

transport-PDF method, is easily incorporated into existing CFD codes and does not

introduce statistical errors [22].

Those successful applications validate that the DQMOM-IEM model can be used

to simulate the reacting flow in the confined impinging jet reactor (CIJR)

[14,15,17-19], plug-flow reactor [13], tubular reactor [23,24], etc. However, this

model has not been used to model turbulent reacting flow in the stirred-tank reactors.

The simulations of turbulent reacting flow are of great differences between semi-batch

stirred tank and CIJR both in geometry and operating conditions, which are given

3
 ACCEPTED MANUSCRIPT

below: (1) Size of geometries: the diameters of stirred tank are often hundreds of

times larger than the inlet diameters of CIJR that are a few millimeters [14,15,17-19].

Furthermore, the interactions between the rotating impeller and stationary baffles

T
IP
produce complicated 3D-flow field, so the principal length scales and time scales of

R
turbulent mixing process cover a wide range; (2) Ratio between different feed streams:

SC
CIJR is carried out in a continuous operation, the flow rates of two feed streams are

often equal [17-19], but for the semi-batch stirred-tank reactors, the feed volumes is

NU
only 2% of the bulk flow which originally exists in the tank [5,10]; Besides, the
MA
characteristics of feed ratio influence the initial concentrations if the quantities of all

reagents are chemically equal [5,10]; (3) Feed time: the residence time of CIJR is far
D

less than the feed times tf of semi-batch stirred tank. tf is often on the scale of several
TE

minutes which also makes the unsteady simulation time very long. All of these
P

differences increase the difficulty of simulation turbulent reacting flow in stirred

CE

tanks.
AC

The aim of this work is to employ the DQMOM-IEM model to predict the mixing

effects on the course of parallel competing chemical reactions carried out in a

semi-batch single-phase stirred-tank reactor. The simulation results are compared with

the experimental data reported by Baldyga and Makowski [10]. Some numeircal

strategies are adopted to solve the problems present in simulation of stirred tanks with

complex geometry and different operations. In addition, the mixture fraction variance

can be used to describe the degree of the local micro-mixing [25], so the state variable

is used to mark the characteristics of reaction region in the semi-batch stirred-tank

4
 ACCEPTED MANUSCRIPT

reactors.

2. Details of DQMOM-IEM micro-mixing model

T
IP
DQMOM-IEM is a presumed probability density function model. This

approach assumes that the joint composition PDF f  ; x, t  can be approximately

R
SC
expressed as the sum of a finite number of multi-dimensional delta functions [26]:

Ne Ns
f  ; x, t   f  1 , , k ; x, t    pn  x, t     k   n   x, t  

NU
(1)
n 1  1

where N e is the total number of environments, pn (x, t ) is the probability or volume

MA
fraction of environment n,  n  x, t  is value of scalar α in environment n, N s is the

number of scalars, and  is the composition vector.

D
TE

For the presumed PDF method, the non-reacting mixture fraction scalar and

dimensionless reaction progress variable are useful to describe the mixing process
P
CE

between multiple non-premixed feed streams.

Considering the following second-order parallel competitive reactions system

AC

used in this work:

A  B 
k1
R (2)

A  C 
k2
S (3)

One stream contains reactant A with the initial concentration cA0 , the other

stream carries the pre-mixture B ( cB0 ) and C ( cC0 ) in the tank.

The mixture fraction is related to the local concentrations of reactants A, B and C ,

(i.e., cA , cB and cC ) defined by [10]

5
 ACCEPTED MANUSCRIPT

cA  cB  cC  cB0  cC0
 (4)
cA0  cB0  cC0

The mixture fraction ranges from 0 to 1.   1 is for the pure stream with

T
concentration cA0 .   0 is for the pre-mixture inside the reactor.

IP
The stoichiometric mixture fractions  s1 and  s 2 are defined respectively [12]

R
cB0 cC0
 s1  , s 2 

SC
(5)
cA0  cB0 cA0  cC0

For the reaction system, the chemical species concentrations have to be expressed

NU
as the mixture fraction coupled with the reaction progress variable Y. The
MA
dimensionless reaction progress variables Y1 and Y2 for reactions (2) and (3) are

defined
D

cB cC
Y1  , Y2  (6)
TE

cA 0 s 1 cA0 s 2

According to the definitions of Eq.(6), the reaction progress variables are

P
CE

normalized by the stoichiometric concentrations cA0 s1 and cA0 s 2 . Furthermore, Y1

and Y2 are equal to 0 in the inlet and initial conditions.

AC

Thus, chemical species concentrations can be expressed in terms of one mixture

fraction variable  and two reaction progress variables ( Y1 and Y2 ).

cA c
    s1Y1   s 2Y2 ， B  1     1   s1  Y1 ,
cA0 cB0

cC c c
 1     1   s 2  Y2 , R  1   s1  Y1 , S  1   s 2  Y2 (7)
cC0 cB0 cC0

As the micro-mixing test reactions, the first reaction is instantaneous and the

second reaction (side reaction) is a finite-rate reaction. When the feed stream of

limited reactant A meets the other stream containing the premixed B and C, reactant A
6
 ACCEPTED MANUSCRIPT

and B can not coexist at the same spatial location. Thus, the first reaction progress

variable Y1 can be evaluated by the limiting value Y1 as Eq. (8).

 1    s 2 
Y1  min  ,  Y2 

T
(8)
1  s1 s1 s1 

IP
where 0  Y2    s 2 because the reaction progress variable is non-negative.

R
If the two environments DQMOM-IEM is employed to this reaction system, five

SC
transportation equations need to be solved.

NU
1) The transport equation for the probability of environment 1 ( p1 )：

p1 p1   p 
MA
 uj   Γ T 1   0 (9)
t x j x j  x j 

where the turbulent diffusivity Γ T is defined as Eq. (10):

D

C k 2
ΓT 
TE

(10)
ScT 

where C  0.09 , turbulent Schmidt number ScT 0.7 , k is the turbulent kinetic
P
CE

energy and  is the rate of energy dissipation.

Furthermore, the sum of probabilities is unity. Thus, the probability of the second
AC

environment can be calculated by p2  1  p1 .

2) The transport equations for the weighted mixture fraction in environments 1

( p11 ) and 2 ( p22 ):

p11 p11   p11 

 uj   ΓT 
t x j x j  x j 
(11)
Γ T  1 1   
  p1 p2 2  1    p1  p2 2 2 
1  2  x j x j x j x j 

7
 ACCEPTED MANUSCRIPT

p22 p22   p22 

 uj   Γ T 
t x j x j  x j 
(12)
Γ T  1 1   
  p1 p2 1  2    p1  p2 2 2 
2  1  x j x j x j x j 

T
IP
where the first and second terms on the right hand of the transport equations are the

R
micro-mixing terms and correction terms, respectively. The micro-mixing rate is

SC
defined as:

C 

NU
 (13)
2 k

where the micro-mixing rate constant C is a function of the local Reynolds number,
MA
the detailed formulation can be referred to Liu and Fox [17].
D

3) The transport equations for the weighted reaction progress variable of the side
TE

chemical reaction in environments 1 ( p1Y21 ) and 2 ( p2Y22 ):

p1Y21 p1Y21   p1Y21 

P

 uj   Γ T    p1 p2 Y22  Y21 
t x j x j  x j 
CE

(14)
Γ T  Y21 Y21 Y Y 
  p1  p2 22 22   p1S2  1 , Y21 
Y21  Y22  x j x j x j x j 
AC

p2Y22 p2Y22   p2Y22 

 uj   ΓT    p1 p2 Y21  Y22 
t x j x j  x j 
(15)
Γ T  Y21 Y21 Y Y 
  p1  p2 22 22   p2 S2  2 , Y22 
Y22  Y21  x j x j x j x j 

where the chemical source terms S2 1,Y21  and S2 2 ,Y22  are :

k2     1  1 
S2  1 , Y21   cA1cC1  k2 2  1  s1 Y11  Y21    Y21 
cA0 s 2  s 2 s 2   1  s 2  (16)

k2     1  2 
S2  2 , Y22   cA2cC2  k2 2  2  s1 Y12  Y22    Y22 
cA0 s 2  s2 s2   1  s2  (17)

8
 ACCEPTED MANUSCRIPT

with

cA0cC0
2  (18)
cA0  cC0

T
and the limiting value of the first chemical reaction progress variable in environments

IP
1 ( Y11 ) and 2 ( Y12 ) :

R
 1  1 1  s 2 
Y11  min  , 

SC
Y21  (19)
1   s1  s1  s1 

 1  2 2  s 2 

NU
Y12  min  ,  Y22  (20)
1   s1  s1  s1 

The reactant concentrations ( cAn , cBn , cCn ) and product concentrations ( cRn ,
MA
cSn ) in the nth environments can be predicted by solving the transport equations of
D

mixture fraction and reaction progress variables. Afterwards, the mean concentrations
TE

can be calculated using Eq. (21):

cA p1cA1 p2cA2 ， cB p1cB1 p2cB2 ， cC p1cC1 p2cC2

P
CE

cR p1cR1 p2cR2 ， cS p1cS1 p2cS2 (21)

AC

3. Numerical simulation
3.1 Stirred tank and reaction system

As shown in Fig. 1, the stirred tank used for simulation in this work is the same

as that used by Baldyga and Makowski [10]. This vessel is equipped with a standard

six-bladed Rushton turbine.

9
 ACCEPTED MANUSCRIPT

T
R IP
SC
NU
Figure 1. Details of the stirred tank
MA
The reaction system for characterizing mixing efficiency is hydrolysis of ethyl
D

chloroacetate in competition with neutralization of sodium hydroxide. The reaction

TE

rate constant k1=1.3×108 m3·mol-1·s-1 and k2=0.023 m3·mol-1·s-1 at 20 ℃.

NaOH  HCl 
k1
 NaCl  H 2O (22)
P

(A) (B) (R) (P)

CE

NaOH  CH 2ClCOOC2 H 5 
k2
 CH 2ClCOONa  C2 H 5OH (23)
(A) (C) (Q) (S)

The NaOH solution with a concentration of 2000 mol·m-3 is slowly fed into the
AC

stirred tank reactor which initially contains the premixture of HCl and

CH2ClCOOC2H5, both with a concentration of 40 mol·m-3. The injection volume VA0

is equal to 4.16×10-4 m3, and the initial volume of stirred tank VBC,0 is 0.02079 m3.

Four different feed times are 8, 10, 15 and 20 min, respectively. The inner diameter of

feedpipe is d=1 mm. The height of the feed point, located at the middle plane between

two adjacent baffles, is equal to the impeller clearance off the tank bottom and the

radial position r=0.09 m. The segregation index XS is used to quantify the product

10
 ACCEPTED MANUSCRIPT

distribution and calculated according to the volume-average concentration cC after

the experiment [10].


X S  cC0  cC  cA0 (24)

T
IP
where cC0 and cA0 are equal and defined as:

cA0  cC0  cA0VA0 VA0  VBC,0 

R
(25)

SC
3.2 Simulation strategy

NU
Numerical simulation is performed with the CFD commercial software ANSYS
MA
Fluent. The origin of three-dimensional coordinate s set at the center of the stirred

tank bottom. The feed-pipe is created and the inlet is set as the velocity inlet boundary
D

condition. The semi-batch stirred tank is operated at a long feed time, so the velocity
TE

outlet is set at the bottom of the stirred tank to maintain the liquid level through
P

draining out the equivalent amount working material. The computational domain
CE

which contains the whole tank geometry is divided into two sub-domains. The inner
AC

zone encompasses the rotating Rushton impeller and the multiple reference frames

(MRF) approach is applied to model the motion of impeller. The rest part of the tank

is solved with the stationary reference frame equations. The boundaries between the

rotating domain and the stationary domain are located at r=0.075 m and 0.06 m≤z≤

0.14 m. The whole domain is discretized with tetrahedral and hexahedral grids using

the GAMBIT mesh generation tool. The equisize skew is smaller than 0.75 to ensure

good grid quality. The grids are refined around the feed-pipe exit and the impeller

where the variables have large gradient.

11
 ACCEPTED MANUSCRIPT

First, the flow field is solved with the Reynolds stress model (RSM) at

steady-state. The RSM, which builds the model equations for the unclosed Reynolds

stress and accounts for the anisotropy of turbulence, also belongs to RANS method.

T
IP
The previous simulation results [5-6] indicate that the RSM is relatively superior to

R
the standard k-ε model on the prediction of turbulent flow field. Although LES

SC
approach better predicts the turbulent kinetic energy around the impeller than the

RSM [27], the RSM method makes the computational cost more acceptable. For the

NU
sake of compromise between simulation accuracy and total computing time, the RSM
MA
approach is adopted to simulate the turbulent flow field in this work. The SIMPLEC

algorithm is used to couple pressure and velocity. The second-order upwind scheme is
D

adopted for the spatial discretization of transport equations. Then, the two
TE

environments DQMOM-IEM micro-mixing model is loaded into the ANSYS Fluent

P

through the user-defined functions (UDFs). Based on the Fluent platform, the
CE

program written in the C programming language realizes the customization. The

AC

transport equations for five additional user-defined scalars used in the DQMOM-IEM

model are:

X1 p1 , X 2  p11 , X 3  p22 , X 4  p1Y21 , X 5  p2Y22 (26)

Those scalars are solved time-dependently with the second-order implicit time

difference format.

The initial boundary conditions at the velocity inlet where reactant A is fed are:

X1 X2 1.0 , X 3 X4 X5 0.0 (27)

and those for the computational domain inside the tank are:

12
 ACCEPTED MANUSCRIPT

X1 X2 X3 X4 X5 0.0 (28)

Due to the long feed time, variable timestep approach is adopted to save the

unsteady simulation time. The time step is set to 10-4 s at the beginning of the

T
IP
simulation to overcome the numerical difficulties.

R
4. Results and Discussion

SC
4.1 Verification of DQMOM-IEM procedure

NU
Although the DQMOM-IEM model has been used successfully for the simulation

of turbulent reacting flow in the CIJR etc, the procedure created in this work has to be
MA
verified before exploring the applicability of DQMOM-IEM model in stirred tank

reactor. The effect of mixing on the selectivity of parallel competitive reactions

D

carried out in the CIJR is simulated with two environments DQMOM-IEM model.
TE

The detailed descriptions about geometry and operating parameters can be referred to
P

Liu and Fox [17]. The chemical reaction sources added in the transport equations of
CE

weighted reaction progress variable and the computational boundaries are set

according to the reaction system employed in CIJR [17]. As shown in Fig. 2 and Fig.
AC

3, the predicted results about the mixture fraction variance  '2 and the segregation

index X are in agreement with the experiment data [17] and the simulation results got

with the same model by Liu and Fox [17], which confirm the validity of

DQMOM-IEM procedure.

13
 ACCEPTED MANUSCRIPT

T
R IP
SC
Figure 2 Profiles of  '2 NU
along the feedpipe axis predicted with different methods
MA
（Rej=400, dj=0.5 mm）
D
P TE
CE
AC

Figure 3 Effect of Reynolds number Rej on the segregation index X

(dj=0.5mm, tr=4.8 ms)

4.2 Grid independence test

Except the verification of DQMOM-IEM procedure, it is also essential to analyze

the sensitivity of the simulation results of turbulent flow field on the grid number. The

14
 ACCEPTED MANUSCRIPT

turbulent flow field simulations are carried out with five different grids to achieve

grid independence results. Grid 1 to grid 5 consists of 366381, 601492, 808938,

1072953 and 1439368 cells, respectively. Fig. 4 shows the axial profiles of

T
IP
dimensionless velocities components and turbulent kinetic energy at r=0.063 m. The

R
results indicate that the turbulent property is more difficult to achieve the grid

SC
independent in comparison with the velocities. According to the numerical predictions,

grid 4 is finally used for the simulations in this work.

NU
MA
D
P TE
CE
AC

(a) Dimensionless axial velocity

15
 ACCEPTED MANUSCRIPT

(b) Dimensionless radial velocity

T
R IP
SC
NU
MA
(c) Dimensionless tangential velocity
D
P TE
CE
AC

(d) Dimensionless turbulent kinetic energy

Figure 4 Comparison of simulation results for five different grids

(N=214 rpm, r=0.063 m)

5. Results and Discussion

5.1 Contour of volume fraction for environment

The ratio between reactor volume and feed volume is large for the semi-batch
16
 ACCEPTED MANUSCRIPT

stirred-tank reactor. The feed stream consisting of reactant A slowly enters into the

reactor from the feed inlet. Thus, the geometry and feeding characteristics of stirred

tank determine the distribution of volume fraction of environment.

T
IP
In Fig. 5, contour plot of the volume fraction of environment 1 ( p1 ) for the case

R
N=214 rpm, tf=8 min is presented. At the beginning of the time-dependent simulation,

SC
the reactor is occupied by the environment 2 except the velocity inlet. During the

feeding period, the environment 1 gradually enters into the reactor. The large gradient

NU
of p1 decreases rapidly for a few seconds after the beginning of feeding and the
MA
system reaches the quasi-steady state due to the long feed time. As seen in Fig. 5, the

environment 1 concentrates in the zone surrounding the feed-pipe exit and its volume
D

fraction remains relatively small in the other region.

P TE
CE
AC

Figure 5. Contour plot of volume fraction of environment 1 (N=214 rpm, tf=8 min )

5.2 Distributions of mixture fraction variance

17
 ACCEPTED MANUSCRIPT

By solving the transport equations of the probability of different environment

( p n ) and its weighted mixture fraction ( pnn ), we can compute the mean mixture

fraction  and its variance  '2 according to Eqs. (29) and (30) [17]:

T
  p11  p22

IP
(29)

R
 '2  p112  p22 2  
2
(30)

SC
The mean mixture fraction  is a non-reacting scalar which indicates the level

of the macro-mixing and the mixture fraction variance  '2 can be used as a

NU
measure for the small-scale segregation of the fluid [11]. For the perfect mixing, the
MA
variance is zero. If the segregation is complete, it is equal to one. Fig. 6 shows the

segregation zone marked with different values of the variance ratio  '2  '2 max
D

between  '2 and its maximum  '2 in the whole tank. With the increase of
TE

max

 '2  '2 max

, the highlighted zone extremely shrinks and concentrates on the
P

region close to the exit of feed-pipe. This result agrees with the conclusions by Vicum
CE

et al. [11] and Duan et.al [28].

AC

(a) variance ratio ≥ 0.001

18
 ACCEPTED MANUSCRIPT

T
R IP
SC
(b) variance ratio ≥ 0.01

NU
MA
D
P TE
CE

(c) variance ratio ≥ 0.1

AC

Figure 6. Contours plot of mixture fraction variance (N=106 rpm, tf=20 min)

5.3 Distributions of reactant concentrations

Initially, environment 1 contains pure reactant A. It is injected into the

stirred-tank reactor from the beginning of time-dependent micro-mixing simulation.

Environment 2 is originally filled with the mixture of reactants B and C. The

interaction between environment 1 and environment 2 causes the compositional

change, so the chemical reactions occur in both environments. Fig. 7 shows the

concentration distributions of reactants B and C in the environments 1 and 2 (at

19
 ACCEPTED MANUSCRIPT

N=214 rpm，tf=8 min), respectively. The micro-mixing test reactions are fast and the

feed position is in the discharge region of the Rushton impeller. Thus, reactants B and

C are consumed almost in the vicinity of the feed-pipe exit. As shown in Fig. 7, the

T
IP
consumption zone of reactant C is more concentrated than that of reactant B on the

R
region where the value of the mixture fraction variance is high. This also indicates

SC
that the side reaction mainly occurs in the large segregation zone [28]. Furthermore,

the reaction zone extends to the tank wall due to the strong radial discharge flow. The

NU
characteristics of concentration distributions reflect the flow pattern generated by the
MA
standard Rushton turbine.
D
P TE
CE
AC

(a) concentration of reactant B in Environment 1

20
 ACCEPTED MANUSCRIPT

T
R IP
SC
NU
(b) concentration of reactant C in Environment 1
MA
D
P TE
CE
AC

(c) concentration of reactant B in Environment 2

(d) concentration of reactant C in Environment 2

21
 ACCEPTED MANUSCRIPT

Figure 7. Contour plots of reactant concentrations in different environments (N=214

rpm, tf=8 min, in the y=0 plane)

T
5.4 Effect of agitation speed and feed time on segregation index

IP
The effect of agitation speed on the reaction process for two cases (feed time

R
SC
tf=20 min and tf=15 min, respectively) are reported in Figs. 8 and 9 in term of the

segregation index XS which represents the selectivity of by-product. For the parallel

NU
competitive reactions given by Eqs. (22) and (23), the acid-base neutralization
MA
reaction occurs only and the yield of the undesired product tends to zero in the well

mixed condition, because there is a great disparity between the reaction rate constants
D

k1 and k2. However, in the complete segregation, the initial reactants concentrations
TE

determine the distribution of products. According to experimental parameters listed in

P

the Sect. 3.1, the segregation index is 0.5 that is also the upper limit.
CE

The segregation indexes predicted by the DQMOM-IEM micro-mixing model are

AC

compared with the experimental data and the numerical simulation results in literature

[10]. Model I is the multiple-time-scale turbulent mixing model and Model III

neglects the concentrations fluctuations. As shown in Figs. 8 and 9, the predicted

segregation index decreases with the increase of agitation speed because the

micro-mixing performance is intensified by the enhanced turbulence level.

There is apparent difference between the predicted results with Model III and

micro-mixing model (either DQMOM-IEM model or multiple-time-scale turbulent

mixing model). This indicates that the effect of micro-mixing on the course of parallel

22
 ACCEPTED MANUSCRIPT

competing chemical reactions is significant. The suitable closure models have to be

used to calculate the non-linear chemical source term and describe the sub-grid

mixing.

T
IP
In addition, the predicted results with Model I are really in good agreement with

R
the experiment data as well. It is a presumed PDF method and the beta PDF is

SC
employed to represent the mixture fraction PDF. For the non-premixed turbulent

flows with multiple feed streams, the mixture fraction PDF is poorly approximated by

NU
a beta PDF [12]. In this work, the semi-batch stirred tank is predicted and there is only
MA
one feed stream. For this binary mixing, the mixture fraction PDF is well

approximated by a beta PDF [12]. In order to select the most appropriate model to
D

simulate different cases, the comprehensive comparison about the predicted ability
TE

between the DQMOM-IEM model and Multi-time-scale turbulent mixing model

P

should be conducted in more CFD studies.

CE
AC

Figure 8. Effect of agitation speed on XS at tf=20 min

23
 ACCEPTED MANUSCRIPT

T
R IP
SC
NU
Figure 9. Effect of agitation speed on XS at tf=15 min
MA
Altering the feed time at N=214 rpm, the predicted results are plotted in Fig. 10

and have the same trend with experiment data. For the fixed volume of reactant A, the
D

feed velocity is inversely proportional to the feed time. At the relatively low feed
TE

velocity, the feed stream is more easily dispersed into the mainstream which is benefit
P

to maximize the yield of the desired product.

CE
AC

Figure 10. Effect of feed time on XS at N=214 rpm

The visualization of reaction plume at the end of the feeding is also presented

24
 ACCEPTED MANUSCRIPT

with the mean concentration of reactant A in Fig. 11. As it can be seen, the volume of

zone marked with cA >0.0001 mol/m3 at the feed time tf=20 min is also smaller

than it at tf=8 min and the reaction zone is more close to the feed point with the

T
IP
increase of feed time. These visualized results also prove that the injected reactant A

R
can be more easily dispersed and consumed by the reactants B and C at low feed

SC
velocity.

NU
MA
D
P TE
CE

(a) N=214 rpm, tf=8 min

AC

(b) N=214 rpm, tf=20 min

25
 ACCEPTED MANUSCRIPT

Figure 11. Visualization of reaction plume with cA >0.0001 mol·m-3 at the end of

the feeding

T
5. Conclusions

IP
In this work, the two-environment DQMOM-IEM model coupled with CFD is

R
implemented to predict and quantify the effects of mixing on the distributions of

SC
products in a semi-batch stirred-tank reactor. The segregation index of parallel
competitive chemical reactions decreases with the increase of feed time. The higher

NU
the agitation speed is, the lower the yield of by-product. In the semi-batch stirred tank,
the reaction zone mainly concentrates on the region close to the feed-pipe exit where
MA
the mixture fraction variance is large. The concentration distributions of reagents and
the visualization of reaction plume can be presented with the two-environment
D

DQMOM-IEM micro-mixing model, which can not be achieved by the empirical and
TE

theoretical models solved in the Lagrangian framework. Although the simulation

results agree with the experimental data, it is essential to investigate the predictive
P

behavior of DQMOM-IEM micro-mixing model used for stirred tanks in other

CE

geometry parameters and operating conditions in the further works.

AC

Nomenclature

ci concentration of specie i, mol·m-3

cin concentration of specie i in environment n, mol·m-3

<ci> mean concentration of specie i, mol·m-3

C impeller off-bottom clearance, m

dj diameter of the impinging jets, m

D impeller diameter, m

H height of stirred tank, m

26
 ACCEPTED MANUSCRIPT

k turbulent kinetic energy, m2·s-2

N agitation speed, r·min-1

p probability

T
r radial distance from Z-axis, m

IP
Rej Reynolds number in the CIJR

R
T stirred tank diameter, m

SC
t time, s

tf feed time, s

tr
NU
characteristic reaction time, ms
MA
X segregation index

XS segregation index
D

x spatial position vector

TE

Y reaction progress

z axial distance from tank bottom, m

P

 micro-mixing rate, s-1

CE

ε rate of energy dissipation, m2·s-3

AC

 kinematic viscosity, m2·s-1

 mixture fraction

1 mixture fraction in environment 1

2 mixture fraction in environment 2

Subscripts

A, B, C, R, S chemical species

j direction

0 initial value
27
 ACCEPTED MANUSCRIPT

References

[1] J. Baldyga, J.R. Bourne, A Fluid Mechanical Approach to Turbulent Mixing and

T
IP
Chemical Reaction Part II Micromixing in the Light of Turbulence Theory,

R
Chem. Eng. Commun. 28(4-6) (1984) 243-258.

SC
[2] E.L. Paul, V.A. Atiemo-Obeng, S.M. Kresta, Handbook of Industrial Mixing:

Science and Practice, John Wiley & Sons, USA, 2004.

NU
[3] P.V. Danckwerts, The effect of incomplete mixing on homogeneous reactions,

Chem. Eng. Sci. 8(1-2) (1958) 93-102.

MA
[4] J.C. Cheng, X. Feng, D. Cheng, C. Yang, Retrospect and perspective of

micro-mixing studies in stirred tanks, Chin. J. Chem. Eng. 20(1)(2012)

D

178-190.
TE

[5] O. Akiti, P.M. Armenante, Experimentally-validated micromixing-based CFD

P

model for fed-batch stirred-tank reactors, AIChE J. 50 (3) (2004) 566-577.

CE

[6] Y. Han, J.J. Wang, X.P. Gu, L.F. Feng, Numerical simulation on micromixing of

viscous fluids in a stirred-tank reactor, Chem. Eng. Sci., 74 (0) (2012) 9-17
AC

[7] D.L. Marchisio, A.A. Barresi, CFD Simulation of mixing and reaction: the

relevance of the micro-mixing model, Chem. Eng. Sci. 58 (16) (2003) 3579-3587.

[8] Z. Wang, Q.H. Zhang, C. Yang, Z.S. Mao, X.Q. Shen, Simulation of barium

sulfate precipitation using CFD and FM-PDF modeling in a continuous stirred

tank, Chem. Eng. Technol. 30 (12) (2007) 1642-1649.

[9] J. Baldyga, M. Henczka, L. Makowski, Effect of Mixing on Parallel Chemical

Reactions in a continuous-flow stirred-tank Reactor, Chem. Eng. Res. Des. 79 (8)

(2001) 895-900.

[10] J. Baldyga, L. Makowski, CFD modeling of mixing effects on the course of

28
 ACCEPTED MANUSCRIPT

parallel chemical reactions carried out in a stirred tank, Chem. Eng. Technol. 27

(3) (2004) 225-231.

[11] L. Vicum, S. Ottiger, M. Mazzotti, L. Makowski, J. Baldyga, Multi-scale

T
modeling of a reactive mixing process in a semibatch stirred tank, Chem. Eng. Sci.

IP
59 (8-9) (2004) 1767-1781.

R
[12] R.O. Fox, Computational models for turbulent reacting flows, Cambridge

SC
University Press, Cambridge, 2003

[13] L. Wang, R.O. Fox, Comparison of micromixing models for CFD simulation of

NU
nanoparticle formation, AIChE J. 50 (9) (2004) 2217-2232.
MA
[14] E. Gavi, L. Rivautella, D.L. Marchisio, M. Vanni, A.A. Barresi, G. Baldi, CFD

Modelling of Nano-Particle Precipitation in Confined Impinging Jet Reactors,

D

Chem. Eng. Res. Des. 85 (5) (2007) 735-744.

TE

[15] E. Gavi, D.L. Marchisio, A.A. Barresi, On the Importance of Mixing for the

Production of Nanoparticles, J. Dispersion Sci. Technol., 29 (4) (2008) 548-554.

P

[16] J. Sierra-Pallares, D.L. Marchisio, M.T. Parra-Santos, J. García-Serna, F. Castro,

CE

M.J. Cocero, A computational fluid dynamics study of supercritical antisolvent

AC

precipitation: mixing effects on particle size, AIChE J. 58 (2) (2012) 385-398.

[17] Y. Liu, R.O. Fox, CFD predictions for chemical processing in a confined

impinging-jets reactor, AIChE J. 52 (2) (2006) 731-744.

[18] E. Gavi, D.L. Marchisio, A.A. Barresi, CFD modelling and scale-up of confined

impinging jet reactors, Chem. Eng. Sci. 62 (8) (2007) 2228-2241.

[19] D.L. Marchisio, Large Eddy Simulation of mixing and reaction in a confined

impinging jets reactor, Comput. Chem. Eng. 33 (2) (2009) 408-420.

[20] V. Raman, H. Pitsch, R.O. Fox, Eulerian transported probability density function

sub-filter model for large-eddy simulations of turbulent combustion, Combust.

29
 ACCEPTED MANUSCRIPT

Theor. Model. 10 (3) (2006) 439-548.

[21] R. Yadav, A. Kushari, V. Eswaran, A.K. Verma, A numerical investigation of the

Eulerian PDF transport approach for modeling of turbulent non-premixed pilot

T
stabilized flames, Combust. Flame., 160 (3) (2013) 618-634.

IP
[22] J. Lee, Y. Kim, Transported PDF approach and direct-quadrature method of

R
moment for modeling turbulent piloted jet flames, J. Mech. Sci. Technol. 25 (1)

SC
(2011) 3259-3265.

[23] J. Akroyd, A.J. Smith, L.R. McGlashan, M. Kraft, Comparison of the Stochastic

NU
Fields Method and DQMOM-IEM as Turbulent Reaction Closures, Chem. Eng.
MA
Sci. 65 (20) (2010) 5429-5441.

[24] J. Akroyd, A.J. Smith, L.R. McGlashan, M. Kraft, Numerical Investigation of

D

DQMOM-IEM as a Turbulent Reaction Closure, Chem. Eng. Sci. 65 (6) (2010),

TE

1915-1924.

[25]R.O. Fox, CFD models for analysis and design of chemical reactors, Adv. Chem.
P

Eng. 31 (2006) 231-305.

CE

[26] R.O. Fox, On the relationship between Lagrangian micro-mixing models and
AC

computational fluid dynamics, Chem. Eng. Process. 37 (6) (1998) 521-535.

[27] B.N. Murthy, J.B. Joshi, Assessment of standard k–ε, RSM and LES turbulence

models in a baffled stirred vessel agitated by various impeller designs, Chem.

Chem. Eng. Sci. 63(22) (2008) 5468-5495.

[28]X.X. Duan, X. Feng, C. Yang, Z.S. Mao, Numerical simulation of micro-mixing

in stirred reactors using the engulfment model coupled with CFD, Chem. Eng. Sci.

140 (2) (2016) 179-188.

30
 ACCEPTED MANUSCRIPT

Type of Article: Research articles

Title: CFD modeling of turbulent reacting flow in a semi-batch stirred-tank reactor

T
Graphical Abstract:

IP
The direct quadrature method of moments combining with the interaction by
exchange with the mean model (DQMOM-IEM) created with C language customizes

R
Fluent software and it is first used to predict the micro-mixing effects on the course of

SC
parallel competing chemical reactions carried out in a semi-batch stirred tank. The
predicted segregation indexes are in good agreement with the experimental data.

NU
MA
D
P TE
CE
AC

31