Journal of Complex Flow, Vol. 1 No.2 (2019) p.

10-13

Journal of Complex Flow

Journal homepage : www.fazpublishing.com/jcf

e-ISSN : 2672-7374

Implementation of Normal and Fractal Baffles for Palm Oil

and Methanol Mixing in Stirred Tank
Asmayou, H. Ali.1,*, Azwan Sapit2, Amir Khalid3, Thariq Wijanarko4
1
School of Mechanical Engineering, Faculty of Engineering
Universiti Teknologi Malaysia, MALAYSIA
2
Flow Analyses, Simulation, and Turbulence Research Group (FASTREG),
Faculty of Mechanical and Manufacturing Engineering, Universiti Tun Hussein Onn Malaysia, MALAYSIA
3
Advanced Technology Centre (ATC), Faculty of Engineering Technology
Universiti Tun Hussein Onn Malaysia, MALAYSIA
4
Departments of Mechanical Engineering, Faculty of Engineering,
Universitas Muhammadiyah Surakarta, INDONESIA

Received 4 July 2019;

Accepted 30 August 2019; Abstract: Mixing is a vital process in a biodiesel reactor. It is known that turbulent flow is
Available online 30 September recommended for mixing process since it could mix the mixture thoroughly in an effective
2019 manner. Previous studies have proven that flow pattern, Reynolds’s number are influenced by the
impeller design especially blade number, blade type, and blade size. This study of comparison and
stimulates the flow characteristics between normal baffled and fractal baffled stirred in mixing
biodiesel reactor by using ANSYS Fluent software. The geometry of the model was modelled in
Solidworks then imported to ANSYS Fluent software. The CFD simulation then was run with
medium mesh and the k-epsilon turbulence model. The simulation run by Rushton impeller with a
rotational speed of 150 rpm and number time steps are 1000, 1200, 1400 and 3600 seconds with
20 iterations. The satisfactory result of comparisons shows that different baffled to influence and
strongly affects the flow characteristics at a different type of baffle. Based on the simulation
result, the fractal baffles give better performance in mixing process compare to the normal baffle
for this application.

Keywords: Bio-diesel, CFD, fluid mixing, mixing tank, Rushton blade

1. Introduction preparation proficiency. To comprehend all the challengers

Considered Biodiesel an alternative fuel and become thinks about on biodiesel blend are creating focused around
more attractive as of late in view of it is ecological profits fuel the strengthening innovations [4]. Flow is described as one or
which is made from renewable biological sources such as two components resulting from the mixer impellent in the
vegetable oils and animal fats. The cost of biodiesel is the tank. Often the stream pattern formed by an instigator
primary obstacle to commercialization of the item [1]. becomes into the first sign of its suitability for a specific
Keywords: Keyword 1, keyword process.
2, number Theof allocation
keywords isof usually 3-7, but of
the scattering more
gasisand
allowed
solid if
Stirring, mixing, blending, deemed
homogenization,
necessary and
emulsification are necessary unit operations in the molecule in a fluid usually relies on the kind of the stream
manufacture of many pharmaceuticals, fuels products and are pattern that prepared by the specific instigator in a given tank
in most cases carried out in agitated tanks made of steel and [6].
sometimes equipped with a glass liner [2]. Most studies have been completed at the standard
Mixing flow cut crosswise over very nearly every impeller leeway (one-third of the tank diameter). Furthermore,
preparing industry including the concoction process industry; it has been demonstrated that the impeller clearance does
minerals, mash, and paper; waste and water treating and very Influence the liquid stream modality. In any case, it has been
nearly every individual methodology area. The specialist accounted for that the stream example created by the Ruston
working with the requisition and configuration of blenders for turbine transformed from the average two circles at a standard
a given procedure has three essential hotspots for data. One is clearance to a solitary circle design at a low clearance. For
distributed writing, comprising of a few thousand distributed multiple impeller systems, the most favourable clearance at
articles and a few as of now accessible books, and pamphlets which there is minimal intrusion among the flow generated by
from supplies sellers [3]. For the blending process, a couple of the top and lower impellers is the same as the tank diameter.
challengers must to think about on the grounds that this Well-spaced impellers generate smooth and high fluid flows,
methodology obliged long living arrangement time, high which are characterized by mean velocity of the fluid and
working expense, high vitality utilization and low of turbulence concentration [7,8].

*Corresponding author: [email protected]

2019 FAZ Publishing. All right reserved.
 Journal of Complex Flow, Vol. 1 No. 2 (2019) p. 10-13

The homogeneous mixture is a mixture that has mixing shape of mixer that is designed in order to get better efficiency
element uniformly. There are several factors that influence the to mix the fluid homogeneously. At the same time, each of the
homogeneous mixture which are mechanisms of mixing, design requires a high cost of manufacturing and need to
characteristics of materials to be mixed and characteristics of spend a lot of time in manufacturing and installation.
the mixer. To analyse a mixture, samples are taken from the This research study will come out with a simple design
mixture at random which represent the state of the mixture. of mixing tank and at the same time having standard
The level of acceptable homogeneity of the mixture depends efficiency of a mixer in order to reduce the current cost of
on each application. It can be specified in terms variation of manufacturing but still producing same results of mixing fluid
coefficient, CoV. The standard CoV value equal to 0 as another mixing tank. In order to design optimal mixer
represents a complete distributive mixing, while CoV value geometries, appropriate tools and methods are needed to
more than or equal to 1 represents total segregation [9]. characterize the flow conditions and their influence on the
Actually they have different between, methanol and mixing process. In this study, CoV will be applied in order to
ethanol, the ethanol is reacted with the triglycerides molecules measure for presenting the uniformity of concentration at a
in the oil, producing glycerine and alkyl esters of fatty acids cross section of the mixing tank. The simulation of mixing
(biodiesel) Worldwide, methanol is the most commonly used fluid can be simulated by using computational fluid dynamic
alcohol for biodiesel production, but the large-scale (CFD) software. The simulation will predict the behaviour of
production of sugarcane in Brazil makes ethanol cheaper and fluid circulation and mixing within the tank. This study will
easier to use for biodiesel production. Since ethanol is focus on the simulation of fluid circulation and mixing within
obtained from plant sugars and methanol is commonly the tank at a specific distance of interest.
produced from natural gas (or petroleum), the production of Mostly, main focused of this study is to propose a new
biodiesel by mixing of oils with ethanol (ethanolysis) approach of fractal concept (square grid fractal) for baffle in
represents a more sustainable pathway for this befoul stirred tank. At the same time, the capability of fractal pattern
production than that employing methanol (methanolysis). in mixing process by determining the CoV will be determine.
Methanolysis is a multiphase reaction, which occurs only at By the results, a recommendation for new concept of mixing
the boundaries of the methanol droplets dispersed in the in tank by using a fractal concept based on square grids fractal
vegetable oil phase [10,11]. will be provide.

1.1 Fractal Concept 2. Simulation Model and Methodology

Fractal is fundamentally a repeated shape and similarity For the first step of the geometry drawing, the reactor
toward oneself to a limitlessly little scale [12,13]. drawing is done by using SolidWorks. Based on the actual
Homogeneous isotropic turbulence has been widely studied geometry, 3D modelling using SolidWorks is constructed. The
both experimentally in wind or water tunnels and numerically role of analysis in the reactor with the rotating 4-blades 90°
by Direct Numerical Simulations (DNS). Recently, used design is to simulate the flow over a range of the operation. In
various multiscale (fractal) grids to generate turbulence in this study, the fluid simulates at full load condition only with
wind tunnels and found that complex multi-scale H=398 mm and T=300mm. The condition is due to the Ruston
boundary/initial conditions can drastically influence the turbine and also 100 mm from the bottom was impeller
behaviour of a turbulent flow, especially when a fractal square position and there are two types of baffles, normal and fractal.
grid is placed at the entry of a wind tunnel test section. After the drawing of the reactor is done drew in the Solid
In these simulations, the fractal stirrers were mainly Works Software, the file is saved as IGES (*, igs) file then it
based on a square pattern with three multi-scale iterations and is converted to the ANSYS Workbench.
with a relatively small computational domain in the stream
wise direction. There were also successfully performed DNS
of turbulence generated by strategy [14]. The square grid
fractal is as in Fig. 1 below.

Fig. 1 - Construction of a panel fractal square grid based

on three fractal iterations [10]. Fig. 2 - Full vessel geometry with normal and fractal
baffles
1.2 Mixing Tank
Numerical simulation results could be performed in
A mixing tank is one of the devices in engineering
graphical form. The estimation of a variable is represented to
industries that is used for the continuous mixing of fluid
in color. Volume fraction of fluid flow through the selected
materials. Generally, mixing tank is used to mix liquid, but it
surfaces can be calculated as well. According to certain
can also be used to mix gas streams, disperse gas into a liquid
variables, fluid mixing is usually represented utilizing
or blend immiscible liquids.
pathways color. An important part of simulation result is
In industry, there are many types of mixing tank have
corroboration of the solution to check its rightness and to
been designed and it is used widely in industry [15]. However,
check the soundness of selected models. The system was first
there are many mixing tanks that have been proposed in the
represented by the reactor entirely with palm oil and
industry having complex and complicated in design. The type
methanol. The rotational velocity of the impeller was set to
of mixing tank that usually used in industry is CSTR, PFR and
150 rpm.
BR tank. Each of mixing tank has its own arrangement and

11
Published by FAZ Publishing
http://www.fazpublishing.com/jcf
 Asmayou, H. A. et al., Journal of Complex Flow, Vol. 1 No. 2 (2019) p. 10-13

The transient analysis took the results of the steady state

analysis as an initial guess, and then a mixture containing
methanol was injected into the palm oil vessel through the
inlet pipe, as in Figure 2. Mesh deformation was applied to
account for the increase in volume of liquid in the tank caused
by the addition of the methanol mixture. The k-epsilon
turbulence model was selected, since it is deemed suitable for
a wide range of simulations, with relative low computational
demand. In simulation of fluid mixing, there are several
simulation results that want to be achieved by the novelist.
The types of the result are volume fraction contours. This
result will give details the behavior of homogeneous fluid
mixing in the mixing tank. The visualization of contours was
used to describe the magnitude of an erratic that reacts on the
surface of the model. Basically, the various colors represent
the amount and behavior of the model in visualization. Fig. 4 - Distribution of the local volume fraction for
fractal baffles
2.1 Coefficient of Variance (COV)
The COV is defined as the standard deviation of Table 1 show that the homogeneity level of the local
concentration, σ over the mean concentration, x for a given set volume fraction for normal and fractal baffles. It is clearly
of data points. The standard deviation of concentration is; showed that the CoV for normal baffles is 0.0632 while for the
fractal baffles, the CoV is 0.0424. According to the CoV
∑ ( ) value, the implementation of the fractal baffles was giving
√ (1)
better mixing compare to the normal baffles. The CoV for the
where σ = standard deviation, n = number of data points, x = fractal baffles which is lower than 0.05 showed that the
mean of the xi and xi = values of the data. An alternative mixing tank equipped with a fractal baffles give a reasonable
measure of homogeneity is provided by the coefficient of target for most applications.
variation;
Table 1 - Homogeneity level of the local volume fraction
(2) for normal and fractal baffles
Basically, the lower the value of CoV shows the better Standard Average of Coefficient
mixture quality. The required level of mixture quality is deviation of volume of variation ,
usually process specific. However, the CoV of between 0.01 volume Fraction for (CoV)
and 0.05 is a reasonable target for most applications. This fraction for methanol
means that 95% of all concentration measurements to be taken methanol
from the tank cross section will be within ±2% of the mean Normal 0.00436 0.0691 0.0632
concentration for CoV=0.01 and ±10% for CoV=0.05. baffles
Fractal 0.002379 0.0561 0.0424
3. Results and Discussion baffles
The results discussed on the distribution of local volume
fraction for the dispersed phase which is methanol in palm oil Fig. 5 showed the rate of mixing for methanol in the
for normal and fractal baffles in mixing tank. Fig. 3 shows the tank for several time. From the graph, it showed that the time
distribution of the local volume fraction for the normal baffles also become another factor in fluid mixing. The results for the
while Fig. 4 is the distribution of the local volume fraction for normal baffles look quite slow in mixing performance in the
the fractal baffles after 3600 s. From the simulation, the beginning. However, the results after 3600 s look nearly close
volume fraction of distributed phase (methanol) and the COV to the results for the fractal baffles. Therefore, it can be said
could be resolved in order to define the mixing homogeneity that if we let the mixing process for a long time, there are
level. maybe no effect on the implementation of the fractal baffles.

0.2
CoV for
normal…
0.15
CoV

0.1

0.05

0
500 1500 2500 3500
Time (s)
Fig. 3 - Distribution of the local volume fraction for Fig. 5 - Comparison of CoV between normal and fractal
normal baffles baffles.

12
Published by FAZ Publishing
http://www.fazpublishing.com/jcf
 Journal of Complex Flow, Vol. 1 No. 2 (2019) p. 10-13

4. Conclusion [6] R.Zadghaffari, J.S. Moghaddas and J.Revstedt. A Study

The aims of this study is to present the capability of the of Liquid-liquid Mixing in a Stirred Tank with 6- Blade
fractal pattern as a baffles in the mixing tank that can be Rushton Turbine.” Iranian Journal of Chemical
enhance the mixing performance between ethanol and crude Engineering 5(4) (2008): 12-22.
palm oil in bio-diesel process. The performance of the fractal [7] A. Ochienga, M. Onyangob and K. Kiriamitic.
baffles was proved by comparing the coefficient of variation “Experimental Measurement and CFD Simulation of
for both type of baffles tested. As the lower the value of CoV, Mixing in a Stirred Tank.” South African Journal of
the better the mixture quality, the CoV for the mixing by using Science 105 (2009): 421-426.
the fractal baffles give a value of 0.0424 which is lover that [8] Manshoor, B., Jaat, M., Izzuddin, Z., and Khalid, A.
0.5. Basically, the CoV of between 0.01 and 0.05 is a “CFD Analysis of Thin Film Lubricated Journal
reasonable target for most applications. As a conclusion, Bearing.”, Procedia Engineering 68 (2013): 56-62.
implementation of the fractal baffles in the mixing tank is one DOI: 10.1016/j.proeng.2013.12.147
of the best alternatives to enhance the mixing process if the
time becomes one of the constraints in mixing process. [9] Guillard F. and Tragardh C. “Mixing in Industrial
Rushton Turbine Agitated Reactors under Aerated
Acknowledgement Conditions.” Chem. Eng. Process 421 (2009): 373–386.
This research was funded by a grant from Ministry of Higher [10] Demirbas, Ayhan. “Biodiesel Production from
Education of Malaysia and (ERGS: vot E033), Universiti Vegetable Oils via Catalytic and Non-Catalytic
Teknologi Malaysia and Universiti Muhammadiyah Supercritical Methanol Transesterification Methods.
Surakarta, Indonesia, under the mobility staff/student Progress in Energy and Combustion Science 31(5-6)
exchange programme. (2005): 466-487
[11] S.Furuta, H.Matsuhashi and K. Arata. “Biodiesel Fuel
References Production with Solid Superacid Catalysis in Fixed Bed
Reactor under Atmospheric Pressure.” Catalysis
[1] Fangrui Ma and Milford A. Hanna. “Biodiesel Communications 5(12) (2005): 721-723
production: A review.” Bioresource Technology 70(1),
[12] Manshoor B, Nicolleau FCGA and Beck SBM. “The
(1999): 1-15.
Fractal Flow Conditioner for Orifice Plate Flow
[2] Thomas Hörmann, Daniele Suzzi and Johannes G. Meters.” Flow Measurement and Instrumentation 22
Khinast. “Mixing and Dissolution Processes of (2011):208–214
Pharmaceutical Bulk Materials in Stirred Tanks:
[13] Khalid, Amir., Mustaffa, Norrizal., Manshoor, Bukhari.,
Experimental and Numerical Investigations.” Industrial
Zakaria, Hanis., Alimin, Ahmad Jais., Leman, Mutalib.,
& Engineering Chemistry Research 50 (21) (2011),
and Sadikin, Azmahani. “The Comparison of Preheat
[3] Norman, Mohd Azim, and Muhamad Najib Hassan. Fuel Characteristics of Biodiesel and Straight Vegetable
“Flow Visualization of Perforated Baffles and Impellers Oil.” Applied Mechanics and Materials 465 (2014):
for Stirred Tank Reactor with Single Stage Rushton 161-166
Turbine.” Journal of Complex Flow 1 (1) (2019): 21-25.
[14] Sylvain, Laizet., and John, Christos Vassilicos. “DNS
[4] M.Nikbakht, Ali & Hosseini, Mehdi. “Biodiesel of Fractal-Generated Turbulence.” Flow Turbulence
Production in Batch Tank Reactor Equipped to Helical Combustion 87 (2011):673–705.
Ribbon-like Agitator.” Modern Applied Science Vol. 6
[15] A. Khalid, A.S.A. Tajuddin, N. Jaat, B. Manshoor, I.
(2012): 40-45
Zaman, S.A.A. Hadi and R.S. Nursa., “Performance and
[5] Ihsak, Norasyida, Muhammad Elyas Muiz Ahmad Emissions of Diesel Engine Fuelled with Preheated
Shaffie, Muhammad Amirul Arif Mohd Azam, and Biodiesel Fuel Derived from Crude Palm, Jatropha and
Mohd Azahari Razali. “Experimental Study of Mixing waste Cooking Oils.” International Journal of
Efficiency in Laminar Flow for Circular Pipe with Automotive and Mechanical Engineering 14(2) (2017):
Perforated Plate Mixer”. Journal of Complex Flow 1 (1) 4273-4284
(2019): 16-20

13
Published by FAZ Publishing
http://www.fazpublishing.com/jcf