ii
SIMULATION OF PALM OIL AND METHANOL MIXING IN
STIRRED TANK BY IMPLEMENT A NORMAL AND FRACTAL
BAFFLES
ALI.H.ASMAYOU
This project report presented in partial
fulfillment of the requirements for the award of
the Degree of Master of Mechanical and Manufacturing Engineering
Faculty of Mechanical and Manufacturing Engineering
Universiti Tun Hussein Onn Malaysia
JUNE, 2014
vi
ABSTRACT
Numerical analysis of mixing and dissolution processes is becoming great significant
for achieving a process of consideration and optimizing the production. Owing to the
rising costs and deficiency of raw materials, “palm oil” processes are presently
employed together with standard experimental analysis. Numerical simulations have
confirmed to be a valuable instrument for understanding and enhancing industrial
mixing problems. However, such simulations are still in the research phase. Even
though Computational Fluid Dynamics (CFD) is a powerful and validated method,
however, for more complicated applications (e.g., industrial mixing), more work is
still required to gain reliable results quickly enough. In this project concentrated on
the simulation of two type of tank to compare between normal baffle and fractal
baffle in mixing tank to get which one is the best of homogenizing. Approximations
were made with respect to improving batch sizes, tank geometry, impeller type,
baffle type, and placement and process variables, such as the impeller agitation
speed. In addition, the feeding position of the methanol. Finally, a quantitative
comparison of different stirring systems and scale-up studies was set.
vii
CONTENTS
TITLE ii
DECLARATION iii
DEDICATION iv
ACKNOWLEDGEMENT v
ABSTRACT vi
CONTENTS vii
LIST OF TABLES xi
LIST OF FIGURES xiii
LIST OF SYMBOLS xvi
LIST OF APPENDICES xvii
CHAPTER 1 INTRODUCTION 1
1.1 Definition 1
1.2 Biodiesel 1
1.3 Problem statement 2
1.4
1.5
Objectives
Scope of research
3
3
CHAPTER 2 LITERATURE REVIEWS 5
2.1
2.2
Biodiesel
Mixing process in tank
5
6
2.3 Multiple phase system of the mixing 7
2.3.1 Gas-liquid dispersion 7
2.3.2 Solid-liquid suspension 8
2.3.3 Liquid-liquid emulsions 9
2.4 Flows in the tank 10
2.4.1 Radial Flow 11
2.4.2 Axial flow 11
viii
2.5 Impeller clearance 13
2.6 mixing time 13
2.7 Reynolds number 14
2.7.1 Laminar region 14
2.7.2 Transition regime 15
2.7.3 Turbulent regime 15
2.8 Baffle 15
2.8.1 Effects on baffling 16
2.9 Fractal square grid 17
2.10 Coefficient of variance (COV) 18
2.11 Palm kernel oil 20
2.12 Methanol and ethanol 21
2.13 Computational Fluid Dynamics (CFD) 22
CHAPTER 3 METHODOLOGY 23
3.1
3.2
Introduction
Methodology flow chart
23
23
3.3 Pre processor 26
3.3.1 Model geometry 27
3.3.2 Models 3D of normal impeller 27
3.3.3
3.3.4
3.3.5
Models 3D of normal and fractal baffle
Configuration of normal baffle tank
Configuration of fractal baffle tank
29
31
32
3.4
3.5
3.6
3.7
3.8
3.9
ANSYS Fluent
Grid of normal and fractal baffle tank
Domain and boundary condition
Solver setup
Post processing
Coefficient of variation (COV)
33
33
36
36
37
37
CHAPTER 4 RSULTS AND DISCUSSION 37
4.1 Introduction 37
4.2 Post processing (Simulation Result) 38
4.3
Turbulent Flow Simulations of normal baffles in mixing
tank, (Time =1000 s) by ANSYS fluent Software.
39
ix
4.4
4.5
4.6
4.7
4.8
4.9
4.10
4.11
4.12
4.13
4.14
Turbulent Flow Simulations of normal baffles in mixing
tank, (Time =1200 s) by ANSYS fluent Software for.
Re = 4965.
Turbulent Flow Simulations of normal baffles in mixing
tank, (Time =1400 s) by ANSYS fluent Software for
Re = 4965.
Turbulent Flow Simulations of normal baffles in mixing
tank, (Time =3600 s) by ANSYS fluent Software for
Re = 4965.
The COV with homogeneous for normal baffle tank at
time (1000, 1200, 1400 and 3600) s and Re of 4965.
Turbulent Flow Simulations of fractal baffles in mixing
tank, (Time =1000 s) by ANSYS fluent Software for
Re = 4965.
Turbulent Flow Simulations of fractal baffles in mixing
tank, (Time =1200 s) by ANSYS fluent Software for
Re = 4965.
Turbulent Flow Simulations of fractal baffles in mixing
tank, (Time =1400 s) by ANSYS fluent Software for
Re = 4965.
Turbulent Flow Simulations of fractal baffles in mixing
tank, (Time =3600 s) by ANSYS fluent Software for
Re = 4965.
The COV with homogeneous of fractal baffled tank at
(1000, 1200, 1400 and 3600) s and Re of 4965.
Comparison between the COV with homogeneous of
normal and fractal baffle at only plan1 ZX= [0.01] m,
(Time =1000, 1200, 1400 and 3600 s) , 20 iteration,
Re = 4965.
Turbulent flow confirmation
41
43
45
47
48
50
52
54
56
57
58
x
CHAPTER 5 CONCLUSIONS AND RECOMMENDATIONS 60
5.1 Conclusion 60
5.2 Recommendation 62
REFERENCES 63
Appendices A
Appendices B
66
69
xi
LIST OF TABLES
3.1 Specifications dimensions tank 27
3.2 Specifications dimensions of impeller 28
3.3 Specifications dimensions of normal and fractal baffle 30
3.4
3.5
4.1
4.2
4.3
4.4
4.5
4.6
4.7
Element and nodes in normal baffle tank
Element and nodes in fractal baffle tank
Total nodes and element for grid independent test.
Homogeneity level of the local volume fraction for the
dispersed phase, methanol in palm oil, the normal baffles
in mixing tank, Re = 4965.
Homogeneity level of the local volume fraction for the
dispersed phase, methanol in palm oil, the normal baffles
in mixing tank, at ,(Time =1200 s).
Homogeneity level of the local volume fraction for the
dispersed phase, methanol in palm oil, the normal baffles
in mixing tank, at ,(Time =1400 s), Re =4965
Homogeneity level of the local volume fraction for the
dispersed phase, methanol in palm oil, the normal baffles
in mixing tank, at ,(Time =3600 s), Re =4965
Homogeneity level of the local volume fraction for the
dispersed phase, methanol in palm oil, the fractal baffles
in mixing tank, at (Time =1000 s) Re =4965
Homogeneity level of the local volume fraction for the
dispersed phase, methanol in palm oil, the fractal baffles
in mixing tank, at (Time =1200 s)Re =4965.
34
35
37
40
42
44
46
49
51
xii
4.8
4.9
4.10
Homogeneity level of the local volume fraction for the
dispersed phase, methanol in palm oil, the fractal baffles
in mixing tank, at (Time =1400 s) Re =4965
Homogeneity level of the local volume fraction for the
dispersed phase, methanol in palm oil, the normal baffles
in mixing tank, at ,(Time =3600 s), Re =4965
Parameter values of Reynolds Numbers
53
55
59
xiii
LIST OF FIGURES
2.1
2.2
2.3
2.4
2.5
2.6
2.7
2.8
2.9
2.10
2.11
3.1
3.2
3.3
3.4
3.5
3.6
3.7
3.8
3.9
3.10
4.1
process mixing tank
Typical arrangement of Rushton radial-flow R100 flat-
blade turbine for gas-liquid mass transfer
Solid-liquid reaction.
Method to obtained light phase and dense phase Radial
Redial Flat-blade turbine.
Marine-type mixing impeller and Pitched-blade turbine
Turbulent flow pattern for an axial impeller type
Increased baffling increases the power draw of an
agitator
Baffling also reduces blend time
Construction of a plane fractal square grid based on three
fractal iterations.
Global consumption and major users of oil palm 1995-
2010.
Methodology flowchart (1)
Methodology flowchart (2)
Dimensional of tank (mm)
Dimensional of impeller (mm) (mm)
Dimensional normal baffle (mm)
Dimensional fractal baffle (mm)
Geometry of normal baffles tank (mm)
Geometry of fractal baffles tank (mm)
Mesh for normal baffle tank
Mesh for fractal baffle tank
Graph of velocity versus distance for three type of mesh
6
8
9
10
11
12
12
16
17
18
20
24
25
27
28
29
30
31
32
34
35
38
xiv
4.2
4.3
4.4
4.5
4.6
4.7
4.8
4.9
4.10
4.11
Distribution of the local volume fraction for the
dispersed phase, methanol in palm oil, of normal baffles
in mixing tank, plan 1 at XY= [0.0] m.
Distribution of the local volume fraction for the
dispersed phase, methanol in palm oil, of normal baffles
in mixing tank, at ZX= [0.01,0.1,0.2 and 0.3] m .
Distribution of the local volume fraction for the
dispersed phase, methanol in palm oil, of normal baffles
in mixing tank at (Time =1200 s)
Distribution of the local volume fraction for the
dispersed phase, methanol in palm oil, of normal baffles
in mixing tank at (Time =1200 s)
Distribution of the local volume fraction for the
dispersed phase, methanol in palm oil, of normal baffles
in mixing tank at (Time =1400 s) and20 iteration, Re =
4965
Distribution of the local volume fraction for the
dispersed phase, methanol in palm oil, of normal baffles
in mixing tank at (Time =1400 s)
Distribution of the local volume fraction for the
dispersed phase, methanol in palm oil, of normal baffles
in mixing tank at (Time =3600 s) and20 iteration, Re =
4965
Distribution of the local volume fraction for the
dispersed phase, methanol in palm oil, of normal baffles
in mixing tank at (Time =3600 s)
Graph of COV with four types of distance plans and
three deference times for normal baffle tank
Distribution of the local volume fraction for the
dispersed phase, methanol in palm oil, of fractal baffles
in mixing tank, plan 1 at XY= [0.0] m .
39
40
41
42
43
44
45
46
48
48
xv
4.12
4.13
4.14
4.15
4.16
4.17
4.18
4.19
4.20
Distribution of the local volume fraction for the
dispersed phase, methanol in Mixing fractal baffle tank,
(Time =1.0000e+03).
Distribution of the local volume fraction for the
dispersed phase, methanol in palm oil, of fractal baffles
in mixing tank at (Time =1200 s) and20 iteration, Re =
4965.
Distribution of the local volume fraction for the
dispersed phase, methanol in Mixing fractal baffle tank,
(Time =1200 s).
Distribution of the local volume fraction for the
dispersed phase, methanol in palm oil, of normal baffles
in mixing tank at (Time =1400 s) and20 iteration, Re =
4965
Distribution of the local volume fraction for the
dispersed phase, methanol in Mixing fractal baffle tank,
(Time =1400 s).
Distribution of the local volume fraction for the
dispersed phase, methanol in palm oil, of fractal baffles
in mixing tank at (Time =3600 s) and20 iteration, Re =
4965
Distribution of the local volume fraction for the
dispersed phase, methanol in palm oil, of fractal baffles
in mixing tank at (Time =3600 s)
Graph of COV with four types of distance plans and
three deference times for fractal baffle tank.
Graph the result compare between normal and fractal
baffle of COV
49
50
51
52
53
54
55
57
59
xvi
LIST OF SYMBOLS
Mixing time
cycling time
, dimensionless blending time
N
D
N
X
COV
Reynolds number
Fluid viscosity
Fluid density
Rotational speed
Impeller diameter
Standard deviation
The number of data points
The mean of the xi
Coefficient of variation
Mean concentration
xvii
LIST OF APPENDICES
APPENDIX TITLE PAGE
A Overall result for volume fraction of Methanol in
normal baffle mixing tank
66
B Overall result for volume fraction of Methanol in
fractal baffle mixing tank
69
1
CHAPTER 1
INTRODUCTION
1.1 Definition
Stirring, mixing, blending, homogenization, and emulsification are necessary unit
operations in the manufacture of many pharmaceutical, fuels products and are in most
cases carried out in agitated tanks made of steel and sometimes equipped with a glass
liner. Another significant operation is the dissolution of solids in a liquid phase.1 A large
number of studies have therefore focused on analyzing the (multiphase) flow in agitated
vessels, as well as numerical and experimental studies. Different flow regimes have been
recognized, depending on the range of the Reynolds (Re) number, and also different
systems including various impeller types and fractal baffle type system and with normal
baffle type system at all. According to, (Thomas Heormann, et al., 2011)
1.2 Biodiesel
Considered Biodiesel an alternative fuel and become more attractive as of late in view of
it is ecological profits fuel which is made from renewable biological sources such as
2
vegetable oils and animal fats. The cost of biodiesel, is the primary obstacle to
commercialization of the item .The utilized cooking oils are utilized as crude material,
adaption of uninterrupted transesterification procedure and recovery of high quality
glycerol from biodiesel by-item (glycerol) are primary cucumber to be considered to
minimize the expense of biodiesel. (Fangrui Ma et. al, 1999). Malaysia has left on an
exhaustive palm bio fuel customized since 1982 and has effectively settled the utilization of
palm methyl esters and the mix of transformed palm oil (5%) with petroleum diesel (95%)
as a suitable fuel for the transport and industrial parts. At present, the real sympathy toward
biodiesel handling is financial achievability. Biodiesel creation will not be supported
without tax exemption and subsidy from government; as the generation expense is
higher than fossil inferred diesel (Demirbas & Balat, 2006).
1.3 Problem Statement
A mixing tank is one of gadgets in designing commercial ventures that is utilized for the
persistent blending of liquid materials. By and large, mixing tank is utilized to blend
fluid; however it can additionally be utilized to blend gas streams, scatter gas into fluid
or mix immiscible fluids.
In industry, there are numerous sorts of mixing tank have been composed and it
is utilized broadly within industry. Notwithstanding, there are numerous mixing tank
that have been proposed in industry having unpredictable and muddled in configuration.
The sort of mixing tank that ordinarily used as a piece of industry is CSTR, PFR and BR
tank. Each of mixing tank has their approach and state of blender that is formed in order
to enhance benefit to mix the fluid homogeneously. Meanwhile, each of the
arrangements obliges high cost of amassing and need to put a huge amount of time in
gathering and station. This exploration study will turn out with straightforward
configuration of blending tank and in the meantime having standard effectiveness of
blender keeping in mind the end goal to decrease current expense of assembling yet
processing same consequences of blending liquid as other blending tank. So as to outline
3
ideal blender geometries, proper devices and techniques are required to describe the
stream conditions and their impact on the blending procedure.
In this study, COV will be connected keeping in mind the end goal to measure
for showing the consistency of fixation at a cross area of blending tank. The recreation
of blending liquid could be reproduced by utilizing computational liquid element (CFD)
programming. The reenactment will anticipate the conduct of liquid course and blending
inside the tank. This study will concentrate on recreation of liquid flow and blending
inside the Tank at particular separation of investment.
1.4 Objective
This research study embarks on the following objectives:
i. To propose a new approach of fractal concept (square grid fractal) for baffle in
mixing tank.
ii. To assess a capability of fractal pattern in mixing process by determining the
coefficient of variation (COV).
iii. To make a recommendation for new concept of mixing in tank by using a fractal
concept based on square grids fractal.
1.5 Scopes of Study
To conduct this research study, several scopes have been outlined:
i. The simulation is done primarily in mixing tank with two kind of baffle,
normal and fractal to get the best mixing by using fluent 14.5 ANSYS.
4
ii. This study will be implemented by fully numerical simulations.
iii. Three different time steps of 1000, 1200, 1400 and 3600 second will be use in
this simulation.
iv. Use the 3 dimension model.
v. The Methanol (C𝐻!O) will be used in the inlet feed.
vi. The propeller rotation speed is 150 RPM.
vii. The flow will be turbulent with a Reynolds numbers, Re of 4965.
viii. In order to test the quality of the new modeling approach, the numerical
simulations will be done by comparing their results of volume fraction with
normal baffles simulation to evaluate local values of mixing tank.
5
CHAPTER 2
LITERATURE REVIEW
2.1 Biodiesel
Bio diesel is a renewable asset serving to decrease the reliance of the economy on
constrained assets and imports, make a business sector for ranchers and lessen the
measure of waste oil, fat and oil being dumped into landfills and sewers. Moreover,
Biodiesel is a light to dull yellow fluid immiscible with water, with high ebullition point
and low vapor pressure. It additionally alludes to a diesel – proportionate handled fuel
determined from biodiesel sources, (for example, vegetable oils), which could be
utilized as a part of unmodified diesel – motors vehicles. It is additionally biodegradable,
non-dangerous and regularly transforms something like 60% less net carbon dioxide
(Co2) emanations than petroleum – based diesel. Ecological Researchers have reported
that global warming change by humanly induced. Its head reason incorporates blazing of
fossil powers, for example, coal, oil and characteristic gas via cars which ceaselessly
discharges carbon dioxide into the atmosphere (Idusuyi, N. et al 2012).
6
2.2 Mixing process in tank
Liquid blenders cut crosswise over very nearly every preparing industry including the
concoction process industry; minerals, mash, and paper; waste and water treating and
very nearly every individual methodology area. The specialist working with the
requisition and configuration of blenders for a given procedure has three essential
hotspots for data. One is distributed writing, comprising of a few thousand distributed
articles and a few as of now accessible books, and pamphlets from supplies sellers.
Figure :(2:1) shows process mixing tank ( James Y. Oldshue, 1983). They have a
considerable measure assortment of employments because of the operation adaptability.
It could be worked in laminar and turbulent blending administrations. Stirred tank might
be utilized for a fluid and gas, a fluid stage mixture, a three stage mixture or fluid with
suspended solids mixture.
Figure (2.1): process mixing tank. (James Y. Oldshue, 1983).
7
It likewise accompanies part of size and utilized focused around the creation and
interest. In biodiesel production mixing process is an essential process considered in
biodiesel generation. This production is exceptionally intricate and it generates from
convection and turbulent trades in a mixing tank (G.r. Kasat et. al, 2008).
For blending process, a couple of challengers must to think about on the grounds
that this methodology obliged long living arrangement time, high working expense, high
vitality utilization and low of preparation proficiency. To comprehend all the
challengers, thinks about on biodiesel blend are creating focused around the
strengthening innovations. From the research, the analyst found that to attained the best
reactor plans, there have a couple of imperative components must to think about
including the measure of reactor, tumult framework, hydrodynamic, physical properties
of reactants, development material, sharpened steel model, impeller sort, size and speed
and confound plan and the systems for hotness expelling and supplying from reactor (M.
Hosseini, et al, 2012).
2.3 Multiple phase system of the mixing
Multiphase systems are habitually encountered in an assortment of modern methods
including a.o. covering, granulation, drying and blend of powers (Fischer Tropsch) and
base chemicals. The hydrodynamics of multiphase systems are managed by the
movement towards the individual stages and the complex shared collaborations and as a
direct an immediate thereof CFD-based modeling of these frameworks has demonstrated
so difficult (Niels G. DEEN ,et al 2006).
2.3.1 Gas-liquid dispersion
More often than not, to disband the gas in the fluid is the primary motivation behind gas
scattering. The gas generally is injected into the tank from the down of reactors, or
8
someplace else nears the impeller to give the diffusion of gas. Figure (2.2) shows how
gas dispersed in the liquid state. (James Y. Oldshue, 1983).
Figure (2.2): Typical arrangement of Rushton radial-flow R100 flat-blade turbine for
gas-liquid mass transfer. (James Y. Oldshue. 1983)
2.3.2 Solid-liquid suspension
Generally speaking in a powerful liquid reaction, the reaction was slower than the
ordinary and it is happened at the bottom outlet of the tank. Normally, most of the
requisition in blending engineering is identified with the suspension of the solids-fluid
stage. The solids particles in the tank all around the mixing methodology are denser than
the pass on fluid achieving tenacious settling of the particles towards the base of the
tank. Along these lines, to stay far from the interminable settling of solids and to
procure an adequate mass trade flux throng the powerful surface all around mixing
process, the technique is given to keep the solids is suspension (Hinze and J.o, 1975).
Solid dissolving issues and insufficient blending for the off-bottom suspension were
9
comprehended by utilizing a pitched cutting edge turbine impeller and full baffled. The
affect ability of response rate to both molecule size and impeller velocity was promptly
settled. Figure (2.3) shows the example of a solid- liquid reaction inside an instigator in
tank. (J. Derksen, 2010)
Figure (2.3): Solid-liquid reaction,(J. Derksen, 2010)
2.3.3 Liquid-liquid emulsions
The prediction of drop sizes in liquid-liquid systems is difficult. Most of the studies have
used very pure fluids as two of the immiscible liquids, and in industrial practice there
almost always are other chemicals that are surface-active to some degree and make the
prediction of absolute drop sizes very difficult. In addition, techniques to measure drop
sizes in experimental studies have all types of experimental and interpretation variations
and difficulties so that many of the equations and correlations in the literature give
contradictory results under similar conditions.
Experimental difficulties include dispersion and coalescence effects, difficulty
of measuring actual drop size, the effect of visual or photographic studies on where in
10
the tank you can make these observations, and the difficulty of using probes that
measure bubble size or bubble area by light or other sample transmission techniques
which are very sensitive to the concentration of the dispersed phase and often are used in
very dilute solutions. (James Y. Oldshue, 1983). The impeller should be placed in the
phase which is going to be the continuous phase to create a table emulsion between a
light and also a dense liquid phase such as Figure (2.4).
Figure (2.4): Method to obtained light phase and dense phase dispersed
(James Y. Oldshue, 1983)
2.4 Flows in the tank
Flow is described as one or two components resulting from the of the mixer impellent in
the tank . Often the stream pattern formed by an instigator are become into the first sign
of its suitability for a specific process. The allocation of the scattering of gas and solid
molecule in a fluid usually relies on the kind of the stream pattern that prepared by the
specific instigator in a given tank (R.Zadghaffari et al., 2008). Plus that, the suitability of
a specific instigator for a given mixing procedure is relies on upon its capacity which is
to affect the liquid substance of the tank into the sturdy rotational, keeping away from
the circumstances where areas of the liquid are inadequately blended or not blended
11
whatsoever, while in the meantime abstaining from harming the different species
scattered in it. (Hinze and J.o, 1975).
2.4.1 Radial Flow
Radial-flow impellers have sharpened pieces of steels which are parallel to the pivot of
the drive shaft. The littler multi bladed ones are known as turbines; bigger, slower-speed
impellers, with two or four edges, are frequently called oars. The width of a turbine is
ordinarily between 0.3 and 0.6 of the tank breadth. Turbine impellers arrive in a mixture
of sorts, for example, bended cutting edge and even sharpened steel, as delineated in
Figure (2.5). (James Y. Oldshue, 1983).
Figure (2.5): Radial Flat-blade turbine. (James Y. Oldshue, 1983)
2.4.2 Axial flow
Impellers incorporate all impellers in which the razor sharp edge makes a plot of short of
what 90° with the plane of pivot. Propellers and pitched-razor sharp edge turbines, as
represented in Fig (2.6), are illustrative pivotal stream impellers.
12
Figure (2.6): Marine-type mixing impeller and Pitched-blade turbine
(James Y. Oldshue, 1983)
Versatile blenders may be braced as an afterthought of an open vessel in the precise, off-
center position or darted to a rib or plate on the highest point of a shut vessel with the
pole in the same rakish, unbalanced position. This mounting brings about a solid start to
finish dissemination. (James Y. Oldshue, 1983) .The fan create parallels stream to the
impellers. In addition that, if impeller set close to the lowest part of the vessel, radial
drainage may happen. when utilized the propeller of axially kind , liquids stream moves
upward and downward in the tank .axial stream propeller distribute material chiefly
axially .In standard case of reactor without baffles, the rotational time to be lower than in
a baffles time. (Tatterson and J.b, 1991). Figure (2.7) show the diverse of the stream
design for axial impeller.
Figure (2.7): Turbulent flow pattern for an axial impeller type, (Tatterson and J.B, 1991)
13
2.5 Impeller clearance
Most studies have been completed at the standard impeller leeway (one-third of the tank
diameter). Furthermore, it has been demonstrated that the impeller clearance does
influence the liquid stream modality. In any case, it has been accounted for that the
stream example created by the Ruston turbine transformed from the average two circles
at a standard clearance to a solitary circle design at a low clearance. For a multiple
impeller system, the most favorable clearance at which there is minimal intrusion among
the flow generated by the top and lower impellers is the same to the tank diameter. Well-
spaced impellers generate smooth and high fluid flows, which are characterize by mean
velocity of the fluid and turbulence concentration. (A. Ochieng ,et al ,.2009)
2.6 Mixing Time
Mixing time, 𝑡! , is the time needed to accomplish a given deviation from the
completely blended state from the occasion of a tracer data. Mixing time is a global
indicator of blending and it is influenced by pivotal and outspread blending and the
impacts of bulk stream. Mixing in airlift reactors is in some cases depict regarding a
dimensionless blending time 𝜃!, describe as follows:
𝜃! = !!!!
2.1
Where 𝑡! is the cycling time, or the time needed for one section through the
course circle.The dimensionless blending time is by and large seen to depend just on the
reactor geometry and not on gas speed where tc is the cycling time, or the time needed
for one section through the course circle (A.sa nchez miro N, et al 2004).
14
2.7 Reynolds number
The definition of Reynolds number for a mixing impeller differs from a pipe or particle
Reynolds number in that the product of N D is used to represent velocity. Reynolds is:
N!" = !²!!!
2.2
Where is the fluid viscosity, ρ is the fluid density, N is the rotational speed and D is
the impeller diameter .Again, coherent units or appropriate conversion factors will make
the Reynolds number dimensionless .Value for the impeller Reynolds number are also
different from other forms of Re, in that turbulent conditions usually exist for Re >
20,000 and laminar conditions occur for Re < 10. The large transition range between
turbulent and laminar condition represents a gradual transition from turbulent conditions
near the impeller to laminar conditions near the tank wall. In the transition range, the
turbulent regions, especially in the impeller discharge, gradually diminish in size as the
Reynolds number becomes smaller (D. S. Dickey et al., 2004).
2.7.1 Laminar region
The laminar regime corresponds to (Re) i < 10 for many impellers; for stirrers with very
small wall-clearance such as the anchor and helical-ribbon mixer, laminar flow persists
until (Re) i = 100 or greater. In the laminar regime:
𝑁! ∝ 1/(𝑅𝑒)! 2.3
15
2.7.2 Transition regime
Between laminar and turbulent stream lies the move administration. There is normally a
progressive move from laminar to completely created turbulent stream in mixed tanks;
the stream example and Reynolds-number range for move rely on upon framework
geometry (Mohan, p et al., 1992).
2.7.3 Turbulent regime
The marvel of turbulent mixing is an immediate aftereffect of turbulent liquid stream,
which is portrayed by an arbitrary variance of the liquid speed at any given point within
the framework. The liquid speed at a given moment may be communicated as the vector
aggregate of its segments in the x, y, and z directions. With turbulence, these directional
segments vary arbitrarily about their individual mean values, as does the speed itself. As
a rule, with turbulence, the liquid has distinctive immediate speeds at diverse areas in the
meantime (Dr. Bhawna Bhatt, Prof. S.S. Agrawal 2007).
2.8 Baffle
During agitation of a low-viscosity liquid, the rotating impeller imparts tangential
motion to the liquid. Without baffling, this swirling motion approximates solid-body
rotation in which little mixing actually occurs. Think about stirring a cup of coffee or a
bowl of soup: The majority of the mixing occurs when the spoon is stopped or the
direction of stirring is reversed. The primary purpose of baffling is to convert swirling
motion into a preferred flow pattern to accomplish process objectives. The most
common flow patterns are axial flow, typically used for blending and solids suspension,
and radial flow, used for dispersion. However, baffling also has some other effects, such
16
as suppressing vortex formation, increasing the power input and improving mechanical
stability (Kevin J. Myers,et al., 2002).
2.8.1 Effects on baffling
During agitation of a low-viscosity liquid, the rotating impeller imparts tangential
motion to the liquid. Without baffling, this swirling motion approximates solid-body
rotation in which little mixing actually occurs. The primary purpose of baffling is to
convert swirling motion into a preferred flow pattern to accomplish process objectives.
The most common flow patterns are axial flow, typically used for blending and solids
suspension, and radial flow, used for dispersion. However, baffling also has some other
effects, such as suppressing vortex formation, increasing the power input and improving
mechanical stability, Figure (2.8) shows that the impeller power number increases as the
number of standard width baffles (T/12) is increased. (Kevin J. Myers,et al., 2002).
Fig (2.8): Increased baffling increases the power draw of an agitator
(Kevin J. Myers,et al., 2002).
Data are presented for three impeller styles: radial- flow impellers, such as straight-blade
and Rushton turbines, mixed-flow impellers, such as pitched-blade turbines, and axial-
flow impellers, such as high-efficiency impellers. All data in figure (2.9) are normalized
17
with respect to the unbaffled condition, with each impeller style being normalized
individually, rather than with respect to a common reference (Kevin J. Myers,et al.,
2002).
Fig (2.9): Baffling also reduces blend time (Kevin J. Myers, et al., 2002).
“Normalized” means that, if the unbaffled radial-flow power number is 2.5 and the
unbaffled high-efficiency impeller number is 0.2, then all of the radial-flow impeller
data are divided by (normalized) 2.5 and all of the high-efficiency impeller data are
divided by 0.2 (Kevin J. Myers ,et al., 2002).
2.9 Fractal square grid
Homogeneous isotropic turbulence has been widely studied both experimentally in wind
or water tunnels and numerically by Direct Numerical Simulations (DNS).Recently,
used various multi scale (fractal) grids to generate turbulence in wind tunnels and found
that complex multi scale boundary/initial conditions can drastically influence the
behavior of a turbulent flow, especially when a fractal square grid (see Figs.2.10) is
placed at the entry of a wind tunnel test section. According to, (Sylvain Laizet • John
Christos, 2011).
Very recently, DNS of full three dimensional turbulent flows generated by three
different fractal grids but at relatively small Reynolds numbers they have been able to
18
recover some of the experimental results of, such as the higher turbulent intensities that
fractal square grids generate by comparison to regular grids of same or even larger
blockage ratio.
Fig (2.10): Construction of a panel fractal square grid based on three fractal iterations.
According to, (Sylvain Laizet · John Christos, 2011)
In these simulations the fractal stirrers were mainly based on a square pattern with three
multi scale iterations and with a relatively small computational domain in the stream
wise direction. There were also successfully performed DNS of turbulence generated by
strategy. According to, (Sylvain Laizet • John Christos, 2011).
2.10 Coefficient of variance (COV)
According to (Fan 1990), the homogeneous mixture is a mixture that having mixing
element uniformly. There are several factors that influence the homogeneous mixture
which are mechanisms of mixing, characteristics of materials to be mixed and
characteristics of the mixer. To analyze a mixture, samples are taken from the mixture at
random which represent of the state of the mixture. The level of acceptable homogeneity
of the mixture depends on each application. It can be specified in terms of variation
19
coefficient, COV. Usually a COV value between 0.01 and 0.05 is a reasonable target of
completely homogeneous mixing fluid for most applications. The lower value of COV
represents better quality of the homogenous mixture. The standard COV value equal to 0
represents a complete distributive mixing, while COV value more than or equal to 1
represents total segregation. According to (R.V., 2004). The COV is defined as the
standard deviation of concentration, σ over the mean concentration, 𝑥 for a given set of
data points.
The standard deviation of concentration is:
σ =!!!! !!
!!!!!!
(2.4)
Where;
σ = Standard deviation
n = the number of data points
𝑥 = the mean of the 𝑥!
𝑥!= Each of the values of the data
An alternative measure of homogeneity is provided by the coefficient of variation:
COV = !
! (2.5)
Where:-
COV = Coefficient of variation
σ = Standard deviation
𝑥 =Mean concentration
dd t
20
2.11 Palm kernel oil
Palm oil is the world’s cheapest edible oil, and increasingly one of the most popular. As
global demand continues to grow so has the vigorous search for land for new plantations
by investors and industry. When it is done well and is properly managed, palm oil
production can be of potential benefit to the populations of developing countries by
providing sustainable livelihoods. Oil palm cultivation also has a greater oil yield per
hectare than any other oil crop, which in theory means it should require less land.
Fig (2.11): Global consumption and major users of oil palm 1995-2010 (Basiron, Y.,
2010)
On the other hand, unchecked large-scale expansion of the industry could lead to
environmental devastation, and precipitate social and economic havoc for African
people. Some acquisitions put forests, ecosystems and the climate at risk, and threaten
21
the livelihood of the people depending on the land, global consumption and major users
of oil palm 1995-2010 as in fig 2.11 (Basiron, Y., 2010).
Palm oil, produced within well-managed and diverse agro forestry systems,
would not only help ensure food security for millions of Africans, provide people with a
living grow local economies; it can also help protect the region’s last remaining
rainforests. In turn, food production will become more diverse and more resilient,
helping to offset the impact of climate change, (Basiron, Y., 2010).
2.12 Methanol and ethanol
Actually they have deferent between , methanol and ethanol, the ethanol is reacted with
the triglycerides molecules in the oil, producing glycerin and alkyl esters of fatty acids
(biodiesel) Worldwide, methanol is the most commonly used alcohol for biodiesel
production, but the large-scale production of sugarcane in Brazil makes ethanol cheaper
and easier to use for biodiesel production. Since ethanol is obtained from plant sugars
and methanol is commonly produced from natural gas (or petroleum), the production of
biodiesel by mixing of oils with ethanol (ethanolysis) represents a more sustainable
pathway for this befoul production than that employing methanol (methanolysis).
Methanolysis is a multiphase reaction, which occurs only at the boundaries of the
methanol droplets dispersed in the vegetable oil phase. Ethanolysis, on the other hand,
can be approximated by a single-phase reaction while in stirred tank reactors. Modeling
and simulation of multiphase reactions like methanolysis usually requires prohibitive
computational resources, while ethanolysis requires only a commercial CFD code in
which equations for the reaction kinetics can be introduced as source terms in the mass
conservation equations implemented in the software ( Demirbas, A. ,2005).
22
2.13 COMPUTATIONAL FLUID DYNAMICS (CFD)
The CFD simulation procedure embodies mesh creating, comparison solving and result
handling modules. A lot of exertion has been committed to the simulation and modeling
of an impeller. The absolute most normal methodologies to modeling an impeller will be
impeller limit condition, preview, sliding network, various casings of reference and
inner–outer. In an extraordinary number of these studies, impellers with basic
geometries, for example, the Rushton turbine, have been modeled. Modeling of curved
blade impellers poses a great challenge with regards to grid generation (A. Ochienga et
al., 2009).
In the Reynolds-normal Navier-Stokes approach, the mathematical statements
are solved to foresee mean stream properties. Actually, with respect to any numerical
result of incomplete differential mathematical statement, the area is speak to by the
discrete and its limits assumes a huge part in characterizing the right stream features or
examples around the impeller blade. (A.Egedy et al., 2011).
23
CHAPTER 3
METHODOLOGY
3.1 Introduction
Methodology is a method or process, or facts that involve an array of measures of work
that should be in a scientific study. For the method the main essential aspect that should
have to concern is on the modeling of the baffled stirred reactor and kinetics model for
biodiesel reactor which will be a way to reach our goals and the results accepted by
using commercial CFD code ANSYS Workbench 14.5 is employed as it is the most
effective way to determine such geometrical parameters.
3.2 Methodology flow chart
Research method or methodology is the process shown in Figure 3.1 and the procedure
and steps of the simulation process is shown in form of flowchart in Figure 3.2
24
Determine title
Objective/scope
Literature review
Discuss the literature review content
Collecting data
Simulation using ANSYS Fluent
Data analysis
No
Yes
No
Results
Validation and discussion
Conclusion
End
Yes
Figure 3.1: Methodology flowchart
Start
63
REFERENCES
A. Egedy, T. Varga, T. Chovan (2011). Application of Models with the different
complexity for a stirred tank Reactor.
A. Ochienga, M. Onyangob and K. Kiriamitic, Experimental measurement and
computational fluid dynamics simulation of mixing in a stirred tank , South
African Journal of Science 105, November/December 2009
A.Sa Nchez Miro N, M.-C. Cero N Garcia, F. Garcia Camacho, E. Molina Grima
And Y. Chisti.( 2004) mixing in bubble column and airlift reactors, Trans
IChemE, Part A, October 2004 , Chemical Engineering Research and Design, 82
Basiron, Y. (2010). Palm oil. Bailey's Industrial Oil and Fat Products.
D. S. Dickey, K. J. Bittorf, C. J. Ramsey and K. E. Johnson(2004). Understand Flow
Patterns in Glass-Lined Reactor. www.cepmagazine.org.
Demirbas, A. (2005). Biodiesel production from vegetable oils via catalytic and non-
catalytic supercritical methanol transesterification methods. Progress in energy
and combustion scienc.
Demirbas, M.F. & Balat, M. (2006). Recent advances on the production and utilization
trends of biofuels: a global perspective. Energy Conversion and Management,
47, pp. 2371–2381.
64
Dr. Bhawna Bhatt, Prof. S.S. Agrawal (2007), Mixing, Pharmaceutical engineering,
www.spxprocessequipment.com
Fan, C. a. (1990). Recent Developments in Solids Mixing, Powder Technology. 255–
287.
Fangrui Ma, Milford A. Hanna., Biodiesel production: a review1a Department of
Food Science and Technology, University of Nebraska, Lincoln,NE,USA,
Bioresource Technology 70 (1999) 1-15
G.R.Kasat, A.R.khopar, V.V.Ranade, A.B.Pandit., CFD Simulation of Liquid, Phase
mixing and Liquid stirred reactor. Mumbai : Elsevier Ltd, 2008, Vol. 63. 3877 --
3885.
Hinze and J.O, Turbulent, Second Edition, McGraw-Hill Book Company, 1975.
Idusuyi, N, Ajide. O.O., Abu, R.(2012) .Biodiesel as an Alternative Energy Resource
in Southwest Nigeria Mechanical Engineering Department, University of
Ibadan, Nigeria. ISSN 2224-3577,http://www.ejournalofsciences.org 323
J. Derksen, Solid-liquid suspension in a stirred tank retrieved April 22, 2010 from the
University of Alberta Chemical Engineering instructor website,
http://www.ualberta.ca/~jos/mv.html
James Y. Oldshue, Ph.D., McGraw-Hill, New York, 1983. Liquid-Solid Operations and
Equipment, ria.servinit.cl/grabiel/libros/Perry/CHAP18.pdf
Kevin J , Myers,., Mark F. Reeder, and Julian B. Fasano. "Optimize mixing by using
the proper baffles." Chem. Eng.FEB 2002
M.Hosseini, A.M.Nikbakht, M.Tabatabaei., Biodiesel Production in Batch Tank
Reactor Equipped to Helical. s.l. : Modern Applied Science, 2012, Vol. 6. ISSN
1913-1844.
65
Mohan, P., Nicholas Emery, A., & Al-Hassan, T. (1992). Review heat transfer to
Newtonian fluids in mechanically agitated vessels. Experimental thermal and
fluid science.
Niels G. DEEN, Martin VAN SINT ANNALAND And J.A.M. KUIPERS. Direct
numerical simulation of particle mixing in dispersed gas-liquid-solid flows using
a combined volume of fluid and discrete particle approach, University of
Twente, Fac. of Science and Technology, PO Box 217, 7500 AE. 13-15
December 2006
R.V., L. D. (2004). Immiscible liquid-liquid systems. In V. A.-O. E.L. Paul, Handbook
of Industrial Mixing (pp. 689-753). Hoboken, New Jersey: John Wiley & Sons
Inc.
R.Zadghaffari, J.S. Moghaddas and J.Revstedt. A Study of Liquid-liquid Mixing in a
Stirred Tank with 6- Blade Rushton Turbine. 2008.
Sylvain Laizet • John Christos Vassilicos,(2011) , DNS of Fractal-Generated
Turbulence. Flow Turbulence Combust (2011) 87:673–705
Tatterson, G.B, Fluid Mixing and Gas Dispersion in Agitated Tanks, McGraw-Hill,
1991
Thomas Heormann, Daniele Suzzi, and Johannes G. Khinast,(2011), Mixing and
Dissolution Processes of Pharmaceutical Bulk Materials in Stirred
Tanks,Research Center Pharmaceutical Engineering GmbH, Inffeldg. 21A,
Graz, Austria