Accepted Manuscript

Process intensification of the transesterification of palm oil to biodiesel in a batch

agitated vessel provided with mesh screen extended baffles

Rodaina Metawea, Taghreed Zewail, El-Sayed El-Ashtoukhy, Iman El Gheriany,

Hesham Hamad

PII: S0360-5442(18)31060-0
DOI: 10.1016/j.energy.2018.06.007
Reference: EGY 13049

To appear in: Energy

Received Date: 29 January 2018

Revised Date: 31 May 2018
Accepted Date: 2 June 2018

Please cite this article as: Metawea R, Zewail T, El-Ashtoukhy E-S, El Gheriany I, Hamad H, Process
intensification of the transesterification of palm oil to biodiesel in a batch agitated vessel provided with
mesh screen extended baffles, Energy (2018), doi: 10.1016/j.energy.2018.06.007.

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

Process intensification of the transesterification of palm oil to biodiesel

in a batch agitated vessel provided with mesh screen extended baffles.
Rodaina Metawea a,*, Taghreed Zewail a, El-Sayed El-Ashtoukhy a, Iman El Gheriany a, Hesham
Hamad b,*
a
Chemical Engineering Department, Faculty of Engineering, Alexandria University, Alexandria 21544,

PT
Egypt.
b
Fabrication Technology Research Department, Advanced Technology and New Materials Research

RI
Institute (ATNMRI), City of Scientific Research and Technological Applications (SRTA-City), New
Borg El-Arab City, Alexandria, 21934, Egypt.

SC
*Corresponding author: Tel. + (203)5914475; Fax. + (203)5921853
E-mail: [email protected] (Rodaina Metawea), [email protected]

U
(Hesham Hamad) AN
Abstract:
One of the major challenges in the production of biodiesel is the immiscibility of oil and methanol. The
M

extent of dispersion between the two phases controls the mass and heat transfer rates and consequently
the rate of the transesterification reaction. In this study, we report a novel design of a batch agitated
D

vessel used for the production of biodiesel from palm oil. The new reactor is provided with static
stainless steel mesh screen baffles to promote dispersion between the two immiscible reactants. The
TE

effect of the key parameters was investigated. The reaction yield was expressed in terms of FAME
(fatty acid methyl ester) concentration. The optimum transesterification reaction conditions that yielded
EP

the highest biodiesel yield (97%) were as follows: methanol to oil molar ratio of 6:1, a reaction
temperature of 60 °C, a one percent (wt.%) NaOH solution, a mesh screen size of 12″ and an agitation
C

speed of 250 rpm. In order to test whether the produced biodiesel can replace diesel oil in combustion
engines we evaluated the biodiesel quality, the engine performance and the emission characteristics of
AC

a B20 (20% biodiesel and 80% petroleum diesel) blend. A reduction in carbon monoxide and a
marginal increase in nitrogen oxides emissions was observed.
Keywords: Biodiesel, Palm oil methyl ester, Base-catalyzed transesterification, Baffled reactors,
Performance, Emission.

1
 ACCEPTED MANUSCRIPT

1. Introduction.

Rising global energy demand and growing environmental concern about vehicle exhausts are major
triggers for scientists all over the world to find green renewable alternative fuels [1,2]. Biodiesel is a
sustainable biofuel that is expected to play a growing role in the future. It can replace or supplement
petrol-based diesel fuels without any modifications to diesel engines [3,4]. Biodiesel is derived from

PT
renewable feedstocks such as; edible and nonedible vegetable oils, animal fats, or recycled greases
from the food industry [5, 6]. The variability of the feedstocks used around the world mirrors their

RI
abundance. Rapeseed and sunflower oil are predominantly used in Europe; palm oil predominates in
tropical countries, and soybean oil and animal fats are the common feedstocks used in the USA [6].

SC
Despite the fact that the aforementioned lipid feedstocks possess high heating values their high
viscosity is a major disadvantage [4, 6]. Biodiesel is produced by four basic methods that include

U
blending with diesel [7], micro-emulsion [8], thermal cracking [9], and transesterification [6]. It is an
esterification reaction between an alcohol and the fatty acids abundant in the lipid feedstock. If
AN
methanol is used the product is FAME (fatty acid methyl esters)[6].

Transesterification has traditionally been catalyzed by an enzyme, an acid or a base to improve the
M

reaction rate and yield. Enzymatic transesterification is expensive due to the high cost of the enzyme.
The acid-catalyzed transesterification reaction is relatively slow and requires a high methanol to oil
D

molar ratio. The corrosive nature of the acids used (sulfuric acid, sulfonic acid and hydrochloric acid) is
TE

another disadvantage [10]. The base-catalyzed transesterification process is the most preferred method
because of the faster reaction rates, the higher yields of FAME, the moderate temperatures and hence
the better economy of the process [11]. The most commonly used alkali catalysts are NaOH, KOH and
EP

the corresponding sodium and potassium alkoxides. The main drawback of this process is the
instability of the base catalyst due to its sensitivity to both moisture and free fatty acid (FFA) content of
C

the used feedstock. The possible saponification reaction between the alkali catalyst, the FFA and the
AC

triglyceride is detrimental to the whole process. It leads to the loss of catalyst activity and decreased
ester yield. In addition, the presence of soap complicates the downstream processes of biodiesel
recovery and purification [6]. Thus, alkali-catalyzed transesterification must be performed using
dehydrated vegetable oil containing less than 0.5 wt.% FFA. Anhydrous alkali catalysts are required for
commercially viable alkali–catalyzed systems [12].

The price of biodiesel (relative to petroleum-based fuels) is a major obstacle to its

commercialization. Biodiesel costs over $3.06 per gallon compared to $2.58 per gallon for petroleum-
2
 ACCEPTED MANUSCRIPT

based diesel (April 2017). The high production cost of biodiesel can be reduced either by using cheaper
raw materials (e.g. waste vegetable oil or non-edible oil) or by the intensification of the production
process. The latter can be accomplished by reducing the process steps or by lowering the energy
requirements [13].

The majority of commercial batch transesterification processes use stirred tank reactors. The process

PT
is limited by the slow kinetics of the ester exchange reaction. In addition, the rate of reaction is reduced
by the immiscibility of methanol and oil. Recent research has focused both on finding efficient

RI
catalysts and on increasing the interfacial area i.e. area of mass and heat transfer in the reactor. Various
process intensification technologies such as microwave [14], ultrasonic cavitation [15], hydrodynamic

SC
cavitation [16], static mixer [17], micro-channel reactor [18], and spinning reactor [19] have been
studied. The target was to introduce new mixing technologies that improve the rate of mass transfer i.e.
reduce the reaction time at minimal energy input (compared to the conventional processes).

U
AN
The presentwork is a novel contribution to the intensification of the base-catalyzed
transesterification of palm oil. It introduces a novel batch reactor with extended stainless steel mesh
screen baffles that are expected to enhance the rate of heat and mass transfer between palm oil and
M

methanol. The effects of reaction temperature, alcohol to oil ratio, mesh number and NaOH
concentration are investigated to find the optimum reaction conditions for biodiesel production. The
D

quality of the produced biodiesel is evaluated and the combustion performance of a B20 blend of the
TE

produced biodiesel is tested. Finally, the emission characteristics of B20 blend are determined and
compared to the baseline case of petroleum diesel.
EP

2. Materials and methods.

2.1. Materials and design of reactor.

C

A schematic diagram of the bench scale batch agitated vessel used in the present study is shown in
AC

Fig. 1. It is a 3 liter stainless steel vessel provided with a temperature controlled water bath and a
digital D.C. electrical motor. The cylindrical reactor is 14 cm in diameter and 25 cm long. It is fitted
with four vertical baffles spaced 90° apart. A 45° pitched stainless steel turbine is connected to the
electrical motor by an insulated steel shaft and is adjusted at the center of the vessel 5 cm from the
bottom. The vessel is covered by a stainless steel lid. Previous research related to the effect of impeller
design on mixing [20-22] has shown that in baffled tanks the power efficiency decreased slightly with
pitch angle up to 45° and increased substantially thereafter. In unbaffled tanks the flow circulation

3
 ACCEPTED MANUSCRIPT

efficiency is not affected by impeller parameters i.e. by pitch angle or the number of blades. Therefore,
the speed of rotation of the impeller is likely to be the main variable that will enhance methanol/oil
mixing and biodiesel production.

2.2. Physico-chemical properties of refined palm oil.

In this study biodiesel was produced from refined olein palm oil provided by the Egyptian Company

PT
of Extracted Oils (Alexandria, Egypt). Palm oil specifications provided by the company are shown in
Table 1. A complete Gas Chromatography- Mass Spectroscopy (GC-MS) analysis of the palm oil is

RI
shown in Table 2. An Agilent 7890 A chromatograph (Agilent Technologies, Santa Clara, C.A.,
U.S.A.) with a 7697 A headspace sampler was used. The other chemicals were reagent grade and were

SC
used without further purification.
2.3. Transesterification reaction and product analysis.
Two liters of heated refined palm oil were fed into the vessel. In a separate beaker, 1 wt.%

U
(expressed as a percentage of the oil mass) of NaOH was completely dissolved in warmed methanol,
AN
then the NaOH/methanol solution was added to the heated oil stepwise. The temperature of the
reactants was adjusted to the predetermined temperature (50 °C, 55 °C, 60 °C and 65 °C) using a
M

thermostated water bath. The transesterification reactions started once the motor was turned on at the
predetermined rpm. The impeller speeds of 200, 250, 300, 400 and 600 rpm were used to provide
D

different degrees of dispersion. The transesterification was conducted for 30 min, after which agitation
was stopped. Phase separation between the biodiesel and glycerol took place immediately. The reaction
TE

products were separated using gravity settling in a funnel separator. After phase separation, water
washing and refining of the crude biodiesel were performed. Warm water at 50-60 °C was used for
EP

biodiesel washing. The volume of water needed in each washing step was about twice the volume of
the crude biodiesel. Washing was repeated to assure the complete removal of the excess reactants and
of the reaction products. The transparency of the waste water was the indicator of the completeness of
C

the washing process. Sulfuric acid was added to the washing water to neutralize the excess amount of
AC

NaOH and keep the pH of biodiesel at 7.

A summary of the investigated parameters and their ranges is presented in Table 3. The yield of the
produced biodiesel is expressed in terms of %FAME and was calculated according to Eq. (1):

 
 .∗
% FAME = 100(

) (1)

where As is the saponification value, Aa is the acid value and Gt is the total glycerol.

4
 ACCEPTED MANUSCRIPT

Volumetric analysis was used to determine the acid value (Aa), the saponification value (As) and the
percent of total glycerol in the refined biodiesel (Gt) [13]. A complete GC-MS of the biodiesel
produced under the optimum conditions is provided in Table 2. Biodiesel quality is evaluated according
to ASTM standard D6751.
2.4. Engine test.

PT
The produced biodiesel was tested in a standard combustion engine at the Department of Mechanical
Engineering at the Faculty of Engineering in Alexandria University. The specifications of the standard
combustion engine are presented in Table 4. The engine experiments were performed on a B20 blend

RI
of the produced biodiesel. Petrol-based diesel was used as a standard fuel. The fuel properties of the
diesel used are shown in Table 5.

SC
2.5. Experimental setup for the emission test.
The combustion engine was coupled to an exhaust gas analyzer (IMR Environmental Equipment, Inc.,

U
St. Petersburg, F.L., USA). The analyzer was provided by the occupational health and air pollution
AN
center at Alexandria University. A calibration was carried out on the gas analyzer before all tests.
Initially, the engine was operated with standard diesel to acquire the reference data.
3. Results and discussion.
M

3.1. Optimization of reaction conditions.

The effect of the novel extended mesh screen baffle reactor compared to the conventional baffled
D

reactor was studied at five agitation speeds. The performance of the extended mesh screen baffled
TE

reactor was evaluated at four different alcohol to oil molar ratios. The effect of catalyst loading and
temperature was investigated. The range of parameters is shown in Table 3.
3.1.1. Effect of agitation speed.
EP

The extended baffled reactor was used for the production of biodiesel by the base-catalyzed
methanolysis of refined palm oil. Biodiesel was produced in the presence and absence of the extended
C

baffles (mesh number 12) in the reactor at 60 °C at a methanol to oil molar ratio of 6:1. The catalyst
AC

loading was 1 wt.% NaOH. The effect of the extended baffles was tested at different agitation speeds.
Interestingly, the yield of biodiesel using the new reactor configuration was higher than the yield of the
conventional baffled reactor at all agitation rates as shown in the Fig. 2a. High conversion (almost
90%) was obtained at 200 rpm compared to 79% in the conventional baffled reactor at 200 rpm. It is
obvious that the new proposed baffled reactor enhanced oil dispersion in the alcoholic phase and hence
the yield of FAME without any increase in the speed of agitation. This could be significant on the

5
 ACCEPTED MANUSCRIPT

industrial scale since it would lead to an appreciable decrease of the operating (mixing) cost. The
power consumption is less at low rpm according to Eq. (2):
P=Np. ρ.N3.D5 (2)
where Np is the power number, ρ is the density of fluid, N is the rotation speed and D is the mixer
diameter

PT
The percent enhancement of the FAME yield due to the use of mesh screen extended baffles can be
calculated according to Eq. (3):
%      %  !""  

RI
% Enhancement = ∗ 100 (3)
%  !""  

SC
Table 6 shows the percent enhancement at different rpm values. The percent enhancement is
particularly evident at low agitation speeds. As expected, mixing plays an important role since the
transesterification reaction is limited by the degree of oil dispersion in the alcoholic phase containing

U
the alkali catalyst. This is confirmed by the increase in yield from 79% at 200 rpm to 94% at 600 rpm
AN
in the standard baffled reactor and from 90% at 200 rpm to 97% at 600 rpm in the mesh screen baffled
reactor as shown in Fig. 2a. It is important to point out that the increase in % FAME with rpm becomes
less significant at rpm values greater than 300 rpm. Stamenkovic´ et al. [23] reported a similar trend
M

when they studied the effect of agitation speed on the alkali-catalyzed methanolysis of sunflower oil. In
the oscillatory baffled reactor studied by Aubin et al. [24] no improvement in yield could be achieved
D

beyond 4.5 Hz despite the significant dependence of yield/turbulence on the oscillation frequency
TE

(below 4.5 Hz). It is likely that equilibrium conversion is attained i.e. no increase in conversion can be
reached at the current process parameters.
3.1.2. Effect of alcohol to oil molar ratio.
EP

One of the most important parameters affecting the yield of ester is the molar ratio of alcohol to
vegetable oil [6]. According to the overall transesterfication equation shown in Eq. (4), the
C

stoichiometric molar ratio of methanol to the triglyceride is 3 to 1.

AC

The amount of methanol used is higher than the stoichiometric ratio in order to shift the reversible
reaction toward the formation of biodiesel. In this study the methanol to oil ratio was varied in the
6
 ACCEPTED MANUSCRIPT

range of 3:1 to 12:1. The yield of biodiesel increased sharply from 62% to 89% as the molar ratio
increased from 3:1 to 6:1. This increase is expected since the transesterification reaction requires the
use of excessive methanol to get reasonable conversion as stated earlier. A further increase of the ratio
to 9:1 and 12:1 resulted in little improvement of yield as shown in Fig. 2b. This observation was
previously reported in the literature [25-28] and was explained by the research effort performed by

PT
Arzamendi et al. [29-31] to monitor the methanolysis reaction. These aforementioned studies revealed
that at short reaction times, the rate of the base-catalyzed methanolysis of sunflower oil decreased as
the methanol excess increased. Little improvement of yield is observed beyond the 6:1 molar ratio due

RI
to the biphasic nature of the esterification reaction and the presence of the base catalyst (NaOH) in the
alcoholic phase. As the alcohol to oil molar ratio increases the concentration of the catalyst or rather the

SC
methoxide anion (active species) decrease due to the dilution effect. In fact Arzamendi et al. [30] have
shown that it is possible to restore the transesterification rate if the catalyst is increased to compensate

U
for the dilution of the catalyst at higher alcohol to oil ratios. The maximum % FAME was 95.5% at the
AN
12:1 molar ratio. However, these alcohol to oil ratios are avoided. As the methanol concentration
increases, glycerol separation becomes more difficult due to the formation of an emulsion of glycerol
with the produced FAME resulting in a decrease in yield [26, 27]. Furthermore, the use of higher molar
M

ratios increases the cost of excess methanol separation and recovery. Therefore, the reactant ratio of 6:1
is used in subsequent experiments. This optimum alcohol to oil ratio was optimum for sunflower,
D

rapeseed and canola oil [6]

TE

3.1.3. Effect of temperature.

FAME yield increases from 87% to 95% with the increase of temperature from 50 °C to 60 °C.
EP

However, %FAME yield declines at 65 °C as shown in Fig. 2c. This has been frequently observed in
the literature and is primarily linked to the partial vaporization of methanol when the reaction
temperature exceeds 60 °C since the normal boiling point of methanol is 64.7 °C. The loss of methanol
C

by vaporization decreases yield. Furthermore, at temperatures above 60 °C the saponification reaction

AC

of the glycerides becomes faster than the esterification reaction as observed for karanja oil [27].
Therefore, the base-catalyzed transesterification reaction using methanol is carried out at 60 °C at
atmospheric pressure [6].
3.1.4. Effect of NaOH catalyst concentration.
The effect of varying the NaOH concentration on methyl ester conversion was studied at four
catalyst concentrations. NaOH is by far the most commonly used homogeneous base catalyst [6]. It was
found more efficient than potassium hydroxide (KOH) and sodium methoxide (CH3ONa) giving the
7
 ACCEPTED MANUSCRIPT

highest FAME conversion at short reaction times [28]. Biodiesel yield increased linearly with alkali
loading as shown in Fig. 2d in the range of NaOH concentrations studied (0.25, 0.8, 1 and 1.5 wt.%). It
was interesting to see the effect of small catalyst loading since NaOH concentrations as low as 0.2
wt.% were reported [31]. Increasing the % NaOH concentration from 0.25 to 1.5 wt.% led to an
enhancement of yield from 80% to 96% FAME after 30 min. Jain and Sharma [26] also observe a

PT
significant increase in FAME yield when the NaOH loading is increased from 0.5 to 1 wt.%. Higher
catalyst concentration increases the tendency of soap formation and results in decreased yield. The
presence of soap makes the separation of ester difficult in the washing step. The optimum catalyst

RI
loading used in subsequent experiments is 1 wt.%. It seems to agree with the optimum NaOH
concentration used in the literature [6, 25, 26].

SC
3.1.5. Effect of mesh number of the extended screen baffles.
The hydrodynamics of batch and continuous oscillatory baffled reactor has been extensively studied

U
and optimized [24, 32-34]. Mazubert et al. [33] studied the effect of mesh number on the performance
AN
of continuous oscillatory baffled reactor using computational fluid dynamics and found it to be
irrelevant to the hydrodynamics of the reactor. In the present study the effect of the mesh number of the
extended baffle was investigated using 5 mesh numbers (6, 8, 10, 12 and 14). The % FAME increases
M

gradually from 86% to 95% as the mesh number increase from 6 to 12 as shown in Fig. 2e. As the
mesh number increases the pore size of the extended screen baffles decreases. This is in turn leads to
D

the formation of more vortices due to increased flow blockage and flow reversal near the baffles i.e. the
TE

higher the mesh number the higher the degree of turbulence. However, the increase of % FAME is
relatively small with the increase of the mesh number beyond 12. It is likely that the equilibrium
conversion is attained i.e. no increase in conversion can be reached at the current process parameters.
EP

Finally, the yield of the mesh screen baffled reactor was compared to yield of other reactor designs.
Results in Table 7 show that the present study gives a higher yield with a simple design compared to an
C

ultrasonic reactor [35], microreactors [36-38] and the microwave technique [39].
AC

3.2. Characteristics and quality of the produced biodiesel.

The GC-MS analysis of the fatty acid methyl ester of palm oil produced in the new baffled reactor at
the optimum conditions is given in Table 2. The fuel quality of the biodiesel produced was compared to
the fuel properties of standard diesel and biodiesel shown in Table 5. The density of the biofuel is
0.8754 g/cc. It lies within the range of standard diesel and biodiesel of 0.82-0.85 g/cc and 0.86-0.89
g/cc, respectively. The kinematic viscosity lies within the range of ASTM biodiesel standard D6715.

8
 ACCEPTED MANUSCRIPT

The pour point (16 °C) and cloud point (9 °C) values agree well with published data of POME (palm
oil methyl ester) [40, 41]. The cloud and pour point of biodiesel are known to be higher than those of
petroleum based diesel and largely depend on the fatty acid composition of the used feed stock. The
two cold flow indicators are higher for palm oil methyl esters due to the high percent of saturated
palmitic acid found in this triglyceride mixture. The poor cold flow characteristics of biodiesel can be

PT
reduced either by the addition of low temperature additives or by branching of the straight carbon
chains. The biodiesel flash point (186 °C) is quite high compared to the biodiesel ASTM limit (93 °C)
and the EN limit (120 °C) but is typical of POME. Ali et al.[40] report a flashpoint of palm oil methyl

RI
esters equal to 170° C. The gross and net calorific values of the biodiesel produced are 45.17 MJ/kg
and 42.45 MJ/kg, respectively. The produced biodiesel is almost free of sulfur and moisture. So, its

SC
combustion will not produce acidic sulfur oxides which typically lead to engine corrosion and may
cause a reversion of FAME to FA as well as filter plugging. In addition, the biodiesel is ash free and

U
hence will not cause engine abrasion. After verifying that the palm oil based biodiesel meets the
AN
biodiesel standard it was important to evaluate its engine performance.

3.3. Engine performance characteristics.

M

A standard combustion engine was used to study the fuel consumption rates (Q) of the B20 biodiesel
blend. The relation between fuel consumption and torque is shown in Fig. 3a. The rate of consumption
of B20 is slightly higher than that of diesel fuel. Due to the higher density and viscosity of the biodiesel
D

a higher fuel mass flow rate is needed to give the same energy output. The brake thermal efficiency at
TE

varying torques for both diesel and B20 is shown in Fig. 3b. The brake thermal efficiency is defined as
the relation between the power delivered and the power that could be obtained if the engine operates
EP

without loss of power. It was calculated based on Eq. (5).

BP
BTE =
mf * C.V (5)
C

where BP is the brake power (kW), mf is the fuel mass flow rate and C.V is the calorific value of the
AC

fuel.
The variation of the brake specific fuel consumption (BSFC) with torque is shown in Fig. 3c. The
brake specific fuel consumption is the measure of fuel efficiency. It was obtained from Eq. (6):
%
BSFC =
&'
(6)
The relatively low BTE of the blend is explained by the high viscosity of the biodiesel. The higher
viscosity results in poor atomization. Higher fuel mass flow rates (mf) are needed to give the same

9
 ACCEPTED MANUSCRIPT

energy and hence the BTE value given by Eq. (5) is reduced [7]. The higher viscosity of biodiesel also
explains the high BSFC of the B20 blend at high torque values. Due to the poor atomization of the
biofuel some fuel remains unburned during the main combustion phase. This is often referred to as
ignition delay and results in a delayed combustion and a low thermal efficiency. The high exhaust gas
temperatures observed (Fig. 3d) confirm that there is an ignition delay and explain the low thermal

PT
efficiency. The engine performance test results of the FAME produced in the present study showed that
the biodiesel blend could be used as a fuel in combustion engines without further modification. The
ignition delay and the low thermal efficiency could be improved by increasing the cetane number. In a

RI
recent study [42] the brake thermal efficiency and the brake specific fuel consumption were improved
by the addition of 1% DTBP (di-tert-butyl peroxide), a cetane improver. According to this study the

SC
thermal efficiency was improved by 2-3.5%.
3.4. Emission characteristics.

U
The exhaust of the standard combustion engine fueled with diesel and the B20 blend was analyzed for
AN
CO, SO2 and NOX. The results of the gas analyzer are shown in Table 8. CO emissions are related to
the combustion quality and amount of oxygen available during the combustion reaction. When an
adequate amount of oxygen is present most of the carbon atoms are converted to CO2 (lean
M

combustion)[43]. Otherwise, most of the carbon atoms are converted to CO (rich combustion). The
emission test results (Table 8) revealed that the presence of biodiesel in the blended fuel decreased CO
D

emissions compared to diesel fuel. This observation was reported in the literature [41, 44]. The addition
TE

of biodiesel to diesel provided more oxygen to the combustion reaction and promoted complete
combustion due to the chemical structure of ester group resulting in higher combustion temperatures
and lower carbon monoxide emissions [45-47]. Oxides of nitrogen (NO and NO2) are produced by the
EP

reaction of nitrogen atoms with oxygen at high temperature and pressure. The amount of NOx
emissions depends on the flame temperature, oxygen availability and combustion duration. When the
C

combustion engine was run with the B20 blend more NOx emissions were produced compared to diesel
fuel. This is likely due to high combustion temperature mirrored by the high temperature of the flue gas
AC

(124 °C compared to 75 °C) caused by ignition delay [7, 41, 44]. NOx emissions could be reduced by
the use of cetane improvers [42]. The exhaust gas temperature of the B20 blend supplemented with 1%
DTBP was lower than biodiesel. NOx emissions were reduced by 11 to 14% and CO emission was
reduced by 12 to 15%. The SO2 emission of diesel is 9.16 mg/m3 and decreases to zero for the B20
sample confirming that the vegetable derived fuel is free from sulfur [2].

10
 ACCEPTED MANUSCRIPT

Acknowledgments

The authors are deeply grateful to both Alexandria University and the City of Scientific Research

and Technological Applications (SRTA-City) in Egypt for supporting and facilitating this study.

PT
4. Conclusions.

RI
In this study a new stirred tank reactor configuration for the production of biodiesel is reported. The
performance of the mesh screen baffled reactor was optimized. Biodiesel was produced by

SC
homogeneous base-catalyzed transesterification of palm oil. The engine performance and emission
characteristics of a B20 fuel blend are investigated.

U
The highest FAME yield was achieved at 60 °C using a methanol to oil ratio of 6:1 at 1 wt.% NaOH
AN
and an agitation speed of 250 rpm. The quality of the biodiesel produced in the reactor was found to be
within the range of palm oil methyl esters. The difference in the engine performance of biodiesel and
the B20 blend was not significant. Emission data agreed fairly well with the emission characteristics of
M

biodiesel. Lower CO emissions and higher NOx emissions were measured.

Based on the present results, we conclude that the B20 blend produced in the baffled mesh screen
D

reactor can be considered as an alternative engine fuel without any considerable modifications of the
TE

engine. The scale up of the mesh screen baffled reactor system needs to be studied in depth. According
to a recent study [48] the energy required for agitating a 50 L transesterification stirred reactor was
estimated to be 27.6% of the processing cost. The installation of baffles is expected to reduce the cost
EP

and time required to obtain high FAME yield. The capital cost of installing the baffles needs to be
compared with saving in reaction time and mixing cost.
C
AC

5. References.
[1] Conti J, Holtberg P, Diefenderfer J, LaRose A, Turnure JT, Westfall L. International Energy
Outlook 2016 With Projections to 2040. USDOE Energy Information Administration (EIA),
Washington, DC (United States). Office of Energy Analysis; 2016.
[2] Coronado CR, de Carvalho Jr JoA, Silveira JL. Biodiesel CO2 emissions: A comparison with the
main fuels in the Brazilian market. Fuel Process Technol 2009;90(2):204-11.
[3] Li B, Li Y, Liu H, Liu F, Wang Z, Wang J. Combustion and emission characteristics of diesel
engine fueled with biodiesel/PODE blends. Appl Energ 2017;206:425-31.
[4] Alptekin E, Canakci M. Determination of the density and the viscosities of biodiesel–diesel fuel
blends. Renew Energy 2008;33(12):2623-30.
11
 ACCEPTED MANUSCRIPT

[5] Tran D-T, Chang J-S, Lee D-J. Recent insights into continuous-flow biodiesel production via
catalytic and non-catalytic transesterification processes. Appl Energ. 2017;185:376-409.
[6] Issariyakul T, Dalai AK. Biodiesel from vegetable oils. Renew Sust Energ Rev 2014;31:446-71.
[7] Ndayishimiye P, Tazerout M. Use of palm oil-based biofuel in the internal combustion engines:
performance and emissions characteristics. Energy. 2011;36(3):1790-6.
[8] Tong K, Zhao C, Sun Z, Sun D. Formation of concentrated nanoemulsion by w/o microemulsion
dilution method: biodiesel, tween 80, and water system. ACS Sustainable Chem Eng 2015;3(12):3299-
306.

PT
[9] Parvizsedghy R, Sadrameli SM, Towfighi Darian J. Upgraded biofuel diesel production by thermal
cracking of castor biodiesel. Energy Fuels 2015;30(1):326-33.
[10] Demirbas A. Comparison of transesterification methods for production of biodiesel from vegetable

RI
oils and fats. Energy Convers Manage 2008;49(1):125-30.
[11] Stamenković OS, Todorović ZB, Lazić ML, Veljković VB, Skala DU. Kinetics of sunflower oil
methanolysis at low temperatures. Bioresour Technol 2008;99(5):1131-40.

SC
[12] Van Gerpen J. Biodiesel from vegetable oils. Biomass to Biofuels, John Wiley and Sons Ltd,
Chichester, West Sussex, UK. 2006:141-63.
[13] Pierucci S, Ferraris GB. 20th European Symposium of Computer Aided Process Engineering:
ESCAPE-20: Elsevier, 2010.

U
[14] Cui Y, Liang Y. Direct transesterification of wet Cryptococcus curvatus cells to biodiesel through
use of microwave irradiation. Appl Energ 2014;119:438-44.
AN
[15] Kumar G, Singh V, Kumar D. Ultrasonic-assisted continuous methanolysis of Jatropha curcas oil
in the appearance of biodiesel used as an intermediate solvent. Ultrason Sonochem 2017;39:384-91.
[16] Chuah LF, Klemeš JJ, Yusup S, Bokhari A, Akbar MM, Chong ZK. Kinetic studies on waste
cooking oil into biodiesel via hydrodynamic cavitation. J Clean Prod 2017;146:47-56.
M

[17] Santana HS, Tortola DS, Silva Jr JL, Taranto OP. Biodiesel synthesis in micromixer with static
elements. Energy Convers Manage 2017;141:28-39.
[18] Budžaki S, Miljić G, Tišma M, Sundaram S, Hessel V. Is there a future for enzymatic biodiesel
D

industrial production in microreactors? Appl Energ 2017;201:124-34.

[19] Shivaprasad P, Jones MD, Patterson DA, Emanuelsson EAC. Process intensification of catalysed
TE

henry reaction using copper-wool catalyst in a spinning mesh disc reactor. Chem Eng Process
2017;122:550-9.
[20] Wang S, Boger DV, Wu J. Energy efficient solids suspension in an agitated vessel-water slurry.
Chem Eng Sci 2012;74:233-43.
EP

[21] Wu J, Graham LJ, Nguyen B, Mehidi MNN. Energy efficiency study on axial flow impellers.
Chem Eng Process 2006;45(8):625-32.
[22] Kumaresan T, Joshi JB. Effect of impeller design on the flow pattern and mixing in stirred tanks.
C

Chem Eng J 2006;115(3):173-93.

[23] Stamenković OS, Lazić ML, Todorović ZB, Veljković VB, Skala DU. The effect of agitation
AC

intensity on alkali-catalyzed methanolysis of sunflower oil. Bioresour Technol 2007;98(14):2688-99.

[24] Soufi MD, Ghobadian B, Najafi G, Mousavi SM, Aubin J. Optimization of methyl ester
production from waste cooking oil in a batch tri-orifice oscillatory baffled reactor. Fuel Process
Technol 2017;167:641-7.
[25] Gupta J, Agarwal M, Dalai AK. Optimization of biodiesel production from mixture of edible and
nonedible vegetable oils. Biocatal Agric Biotechnol 2016;8:112-20.
[26] Jain S, Sharma MP. Kinetics of acid base catalyzed transesterification of Jatropha curcas oil.
Bioresour Technol 2010;101(20):7701-6.
[27] Patil PD, Deng S. Optimization of biodiesel production from edible and non-edible vegetable oils.
Fuel 2009;88(7):1302-6.

12
 ACCEPTED MANUSCRIPT

[28] Pullen J, Saeed K. Investigation of the factors affecting the progress of base-catalyzed
transesterification of rapeseed oil to biodiesel FAME. Fuel Process Technol 2015;130:127-35.
[29] Arzamendi G, Arguiñarena E, Campo I, Gandía LM. Monitoring of biodiesel production:
Simultaneous analysis of the transesterification products using size-exclusion chromatography. Chem
Eng J 2006;122(1-2):31-40.
[30] Arzamendi G, Campo I, Arguinarena E, Sánchez M, Montes M, Gandía LM. Synthesis of
biodiesel with heterogeneous NaOH/alumina catalysts: comparison with homogeneous NaOH. Chem
Eng J 2007;134(1-3):123-30.

PT
[31] Zabala S, Arzamendi G, Reyero I, Gandía LM. Monitoring of the methanolysis reaction for
biodiesel production by off-line and on-line refractive index and speed of sound measurements. Fuel
2014;121:157-64.

RI
[32] Mazubert A, Poux M, Aubin J. Intensified processes for FAME production from waste cooking
oil: a technological review. Chem Eng J. 2013;233:201-23.
[33] Mazubert A, Fletcher DF, Poux M, Aubin J. Hydrodynamics and mixing in continuous oscillatory

SC
flow reactors-Part II: Characterisation methods. Chem Eng Process 2016;102:102-16.
[34] Mazubert A, Fletcher DF, Poux M, Aubin J. Hydrodynamics and mixing in continuous oscillatory
flow reactors-Part I: Effect of baffle geometry. Chem Eng Process 2016;108:78-92.
[35] Choedkiatsakul I, Ngaosuwan K, Cravotto G, Assabumrungrat S. Biodiesel production from palm

U
oil using combined mechanical stirred and ultrasonic reactor. Ultrason Sonochem 2014;21(4):1585-91.
[36] Guan G, Kusakabe K, Moriyama K, Sakurai N. Transesterification of sunflower oil with methanol
AN
in a microtube reactor. Ind Eng Chem Res 2009;48(3):1357-63.
[37] Rahimi M, Aghel B, Alitabar M, Sepahvand A, Ghasempour HR. Optimization of biodiesel
production from soybean oil in a microreactor. Energy Convers Manage 2014;79:599-605.
[38] Yeh SI, Huang YC, Cheng CH, Cheng CM, Yang JT. Development of a millimetrically scaled
M

biodiesel transesterification device that relies on droplet-based co-axial fluidics. Sci Rep 2016;6:29288.
[39] Lertsathapornsuk V, Pairintra R, Aryusuk K, Krisnangkura K. Microwave assisted in continuous
biodiesel production from waste frying palm oil and its performance in a 100 kW diesel generator. Fuel
D

Process Technol 2008;89(12):1330-6.

[40] Ali EN, Tay CI. Characterization of biodiesel produced from palm oil via base catalyzed
TE

transesterification. Procedia Eng 2013;53:7-12.

[41] Mosarof MH, Kalam MA, Masjuki HH, Ashraful AM, Rashed MM, Imdadul HK, et al.
Implementation of palm biodiesel based on economic aspects, performance, emission, and wear
characteristics. Energy Convers Manage 2015;105:617-29.
EP

[42] Musthafa MM, Kumar TA, Mohanraj T, Chandramouli R. A comparative study on performance,
combustion and emission characteristics of diesel engine fuelled by biodiesel blends with and without
an additive. Fuel 2018;225:343-8.
C

[43] Serin H, Yıldızhan Ş. Influence of the compression ratio on the performance and emission
characteristics of a vcr diesel engine fuelled with alcohol blended fuels. Eur Mech Sci (EMS).
AC

2017;1:39-46.
[44] Sharon H, Karuppasamy K, Kumar DRS, Sundaresan A. A test on DI diesel engine fueled with
methyl esters of used palm oil. Renew Energy 2012;47:160-6.
[45] Kathirvelu B, Subramanian S, Govindan N, Santhanam S. Emission characteristics of biodiesel
obtained from jatropha seeds and fish wastes in a diesel engine. Sust Environ Res 2017;27(6):283-90.
[46] Szabados G, Bereczky Á. Experimental investigation of physicochemical properties of diesel,
biodiesel and TBK-biodiesel fuels and combustion and emission analysis in CI internal combustion
engine. Renew Energy 2018.
[47] Nabi MN, Rahman MM, Akhter MS. Biodiesel from cotton seed oil and its effect on engine
performance and exhaust emissions. Appl Therm Eng 2009;29(11-12):2265-70.

13
 ACCEPTED MANUSCRIPT

[48] Anyanwu CN, Mbajiorgu CC, Oparaku OU, Odigboh EU, Emmanuel UN. A design algorithm for
batch stirred tank transesterification reactors. Energy Convers Manage 2012;62:40-6.

PT
RI
U SC
AN
M
D
TE
C EP
AC

14
 ACCEPTED MANUSCRIPT
Table caption
Table 1: Refined palm oil specifications.

Properties Value
Density (40 °C) , (g/cm3) 0.897
Total glycerol % 1.03
Acid value (mg KOH/g oil ) 0.75
Saponification value , (mg KOH/g oil) 196.35

PT
Table 2: GC-MS analysis of palm oil and the biodiesel obtained at optimum

RI
conditions.

Fatty Acid concentration % FA

Concentrations % FA

SC
(FA) of FA in of FA in FAME
mg/100ml of
mg/100ml of
Oil
FAME

U
AN
C8:0 1.752 0.276
8.063 1.5
C10:0 2.9 0.456 4.726 0.9
M

C11:0 2.076 0.327 0.277 0.05

C12:0 6.268 0.987 61.850 11.60
D

C13:0 0.637 0.1 0.447 0.08

TE

C14:1 0.829 0.131 1.525 0.29

C14:0 26.253 4.134
224.546 42.11
EP

C15:0 2.056 0.323

10.043 1.88
C16:1 3.048 0.48 79.408 14.89
C

C16:0 190.455 29.99

30.778 5.77
AC

C17:1 2.169 0.34

1.676 0.31
C17:0 4.716 0.743
18.641 3.50
C18:3γ ω6 4.213 0.663 2.150 0.40
C18:3α ω3 173.384 27.3
1.421 0.27
C18:1c ω9 -- --
8.244 1.55
C18:1t ω9 -- -- 1.027 0.19

3
 ACCEPTED MANUSCRIPT
C18:0 146.644 23.1
6.714 1.26
C20:4 ω6 1.564 0.246
1.683 0.32
C20:5 ω3 2.049 0.323
3.078 0.58
C20:3 ω3 -- -- 1.661 0.31
C20:3ω6 4.096 0.645

PT
- 0
C20:2 2.413 0.38
1.618 0.30

RI
C20:1 8.632 1.359
16.132 3.03
C20:3 ω3 3.858 0.61

SC
1.713 0.32
C20:0 12.286 1.934 15.913 2.98

U
C21:0 4.657 0.733 1.252 0.24
AN
C22:6 ω3 3.004 0.473
2.870 0.54
C22:0 8.58 1.351
8.134 1.53
M

C22:2 2.964 0.467

1.719 0.32
D

C22:1 ω9 -- --
1.538 0.29
TE

C24:1 2.512 0.395 1.692 0.32

C24:0 11.068 1.743
12.651 2.37
C EP
AC

3
 ACCEPTED MANUSCRIPT
Table 3: Summary of the investigated parameters and their ranges:

Parameters Parameter varied Parameter kept constant during the experiment

studied during the experiment
Agitation 200-600 6:1 alcohol to oil ratio, 60 °C, 30 min, 1% wt.
speed catalyst conc., mesh number 12
(rpm)
Alcohol to oil 3-12 55 °C , 30 min, 1% wt. catalyst con., mesh
ratio number 8, 200 rpm

PT
Temp. 50-65 6:1 alcohol to oil ratio, 30 min, 1% wt. catalyst
(°C) conc., mesh number 8, 200 rpm

RI
Catalyst 0.25-1.5 6:1 alcohol to oil ratio, 60 °C, 30 min, mesh
(%wt) number 8, 200 rpm
Mesh number 6-14

SC
size 6:1 alcohol to oil ratio, 60 °C, 30 min, 1 wt. %
catalyst conc., 200 rpm

U
AN
Table 4: Specifications of the standard combustion engine.

Item Value
M

Model TecQuipment (TD212)

No. of cylinders 1
Rated power,kW 3.1 at 3000rpm
D

Compression ratio 22:1

Engine capacity,cm3 232
TE

Injection type Direct injection

Bore/Stroke,mm 69/62
C EP
AC

3
 ACCEPTED MANUSCRIPT
Table 5: Fuel properties of diesel and biodiesel diesel produced in the present
study.

Test Method Biodiesel Standard diesel Standard Diesel

biodiesel
Density @ 15 °C ASTM D1298 0.8754 0.82- 0.86-0.89 0.845
0.85
Ash content ASTM D482 0.001 0.01 - -
% wt max.

PT
Water content ASTM D 95 0.01 0.05 0.05 max -
% wt max.
Sulphur content ASTM D2622 0.003 0.05 0.05 max 0.243

RI
% wt max.
Gross calorific value ASTM D4868 45.17 44.8 42 45.53
MJ/kg

SC
Net calorific value ASTM D4868 42.45 43.4 39.28 42.72
MJ/kg
Flash point ASTM D93 186 °C 52 < 130 < 64
(°C)

U
Kinematic vicosity ASTM D445 4.643 1.3-4.1 1.9-6 3.42
(mm2/s)
AN
Pour point - 16 °C - -10- 12 -
(°C)
Cloud point - 9 °C - - -
M

(°C)
D

Table 6: The percentage enhancement at different rpm values.

TE

rpm Degree of enhancement (%)

200 19.74
EP

250 16.13

300 10.08
C

400 6.91
AC

600 4.21

3
 ACCEPTED MANUSCRIPT
Table 7: Comparison between different enhancement techniques for biodiesel
production

Type of reactor Optimum conditions %FAME Reference

yield
Batch vessel with Reaction time : 30 min 97% The
extended mesh Molar ratio of oil to methanol : 1:6 present
screen baffles Catalyst concentration : 1 wt.% study
NaOH

PT
Reaction temp : 60 °C
Rotational speed : 250 rpm
Combined Reaction time : 1 hr 76% [35]

RI
mechanical stirred Molar ratio of oil to methanol 1:6
and ultrasonic Catalyst concentration : 0.15 wt.%
reactor NaOH

SC
Rotational speed : 25 rpm
Micro reactor with Reaction time : 180 s 89% [36]
T-mixer Molar ratio of oil to methanol :
1:9

U
Reaction temp : 60 °C
Catalyst concentration : 1.2 wt %.
AN
Droplet-based co- Reaction time : 180 s 92.4 [37]
axial Molar ratio of oil to methanol
fluidic system 1:23.9
M

Reaction temp : 20 °C
Catalyst concentration : 2.2 wt %.
Microtube reactor Reaction time : 4 min. 80 [38]
D

Molar ratio of oil to methanol

1:23.9
TE

Reaction temp : 20 °C
Catalyst concentration : 4.5 wt %
Microwave Reaction time : 30 sec. 97 [39]
irradiation Molar ratio of oil to methanol :
EP

1:12
Catalyst concentration : 3 wt.%
NaOH
C

A household microwave (800W)

AC

3
 ACCEPTED MANUSCRIPT
Table 8: Concentrations of air pollutants (mg/m3) in exhaust emissions.

Fuel CO NO2 SO2 Temp. Oxidation of Oxidation of NO2

used mg/ mg/m3 mg/m3 of gas CO
m3 (°C) O2 CO2 O2 NO NOx
% % % mg/ mg/m3
m3

B20 30 64 0 124 20.9 4.68 20.9 312 376

PT
Diesel 213 6 9.16 75 14.6 zero 14.6 146 152
Max. 250 400 350 -- --- 500 --
allowed

RI
conc.*
* Annex (6) table (2) from the Prime Minister's Decree No 1095 of 2011and 710 for
2012 Issuing the Executive Regulations of the Environment Law Promulgated by Law

SC
No 4-1994 and Amended by Law No 9-2009 in Egypt.

U
AN
M
D
TE
C EP
AC

3
 ACCEPTED MANUSCRIPT

Figure caption

PT
RI
U SC
AN
M

1) DC digital Motor 2) Stainless steel lid

3) Three liters Stainless steel vessel 4) Extended baffles

D

5) 45 ° pitch turbine impeller 6) Thermometer

TE

7) Liquid level 8) Heating coil

9) Thermostated water bath

EP

(a)
C
AC

1
 ACCEPTED MANUSCRIPT

PT
RI
SC
Fig. 1: The bench-scale batch agitated reactor, b) Stainless-steel vessel with four extended mesh
screen baffles.

100

U
AN
90
FAME yeild (%)

80

70
D
TE

60

3:1 6:1 9:1 12:1

EP

Alcohol to oil ratio

(a)
C
AC

2
 ACCEPTED MANUSCRIPT

100

95
FAME yeild (%)
90

PT
85

RI
80

75

SC
50 55 60 65
Temperature (oC)

(b)

100
U
AN
95
FAME yeild (%)

90

85
D
TE

80

75
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6
EP

Catalyst Conc. (%)

(c)
C
AC

3
 ACCEPTED MANUSCRIPT

100

FAME yeild (%)

95

PT
90

RI
85

SC
6 8 10 12 14
Mesh number

(d)

100
U
AN
95
FAME yeild (%)

90

85
D
TE

80
12 ' mesh screen baffles
standard baffles
75
200 300 400 500 600
EP

Agitation speed (rpm)

(e)
C

Fig. 2: Effect of a) alcohol to molar ratio, b) temperature , c) catalyst concentration , d) mesh

number of the extended screen baffles and e) agitation speed on the % FAME.
AC

4
 ACCEPTED MANUSCRIPT

7
(a)
6

5
Q (ml/min.)

PT
4

RI
2 B20
Diesel

SC
0 1 2 3 4 5 6 7 8 9
Torque (N.m.)

U
AN
40
35 (b)
Brake Thermal Efficiency

30
M

25
20
D

15
TE

10
5 B20
0 Diesel
EP

0 1 2 3 4 5 6 7 8 9
Torque (N.m.)
C
AC

5
 ACCEPTED MANUSCRIPT

1.2
(c)
1.0

BSFC (kg/kw.hr) 0.8

0.6

PT
0.4

RI
0.2
B20
Diesel
0.0

SC
0 1 2 3 4 5 6 7 8 9
Torque (N.m.)

U
250
AN
(d)
200
Exhaust Temp. ( oC)

150
M

100
D

50
TE

B20
Diesel
0
0 1 2 3 4 5 6 7 8 9
Torque (N.m.)
EP

Fig. 3: Variation of a) volumetric flow rate of fuel; b) brake thermal efficiency; c) brake specific
fuel consumption and d) exhaust temperature with torque (N=1500 ± 2).
C
AC

6
 ACCEPTED MANUSCRIPT
Highlights

1. Biodiesel is produced in a mesh screen baffled batch reactor.

2. The optimal condition led to high FAME yield of 97% at 200 rpm.

3. B20 of the produced palm oil methyl ester can be used as a fuel without
engine modification.

PT
4. The cost of biodiesel and glycerol are of 1.05 and 0.68 €/L, respectively.

RI
U SC
AN
M
D
TE
C EP
AC