G Model

CATTOD-9661; No. of Pages 8 ARTICLE IN PRESS

Catalysis Today xxx (2015) xxxxxx

Contents lists available at ScienceDirect

Catalysis Today
journal homepage: www.elsevier.com/locate/cattod

Kinetic studies on biodiesel production using a trace acid catalyst

Yingying Liu a,f , Houfang Lu a,b , Kwesi Ampong-Nyarko c , Tanya MacDonald d ,
Lawrence L. Tavlarides e , Shijie Liu f, , Bin Liang a,b,
a
Institute of New Energy and Low-Carbon Technology, Sichuan University, Chengdu 610065, PR China
b
College of Chemical Engineering, Sichuan University, Chengdu 610065, PR China
c
Alberta Agriculture and Rural Development, Edmonton, AB, Canada
d
School of Innovation, Olds College, Olds, AB, Canada
e
Department of Biomedical and Chemical Engineering, Syracuse University, Syracuse, NY, USA
f
Department of Paper and Bioprocess Engineering, State University of New YorkCollege of Environmental Science and Forestry, Syracuse, NY 13210, USA

a r t i c l e i n f o a b s t r a c t

Article history: Biodiesel was produced by transesterication in the presence of trace sulfuric acid (0.020.1 wt% of
Received 13 April 2015 oil mass). The kinetics for the transesterication of corn oil with ethanol was investigated between
Received in revised form 7 July 2015 155195 C. The transesterication in the pseudo-homogeneous system consisted of three consecu-
Accepted 10 July 2015
tive steps. All steps were found to be second order reactions, with the rst and the second steps being
Available online xxx
irreversible, while the third step was reversible. It was observed that the apparent rate constants of the
forward reactions increased linearly with increasing acid concentration. Compared to the initial two steps
Keywords:
of the ethanolysis reaction, the effect of the reversible transformation in the third step is signicant, as the
Biodiesel
Transesterication conversion of monoglyceride to glycerol is difcult. In addition, the presence of free fatty acids (<30 wt%),
Trace acid catalyst water (<3 wt%), and the use of stirring had little effect on the nal content of biodiesel. Finally, it was
Kinetics observed that methanol exhibited a higher reactivity than ethanol in transesterication, while corn oil
Corn oil exhibited higher reactivity than Pennycress oil.
Pennycress oil 2015 Elsevier B.V. All rights reserved.

1. Introduction can be used to catalyze esterication and transesterication

Over the past few decades, alternative and renewable energy
sources have attracted a growing amount of attention. Biodiesel,
for example, is a renewable fuel produced from the trans-
esterication of renewable resources such as alcohols and
vegetable oils/animal fats [14]. As these oils normally con-
tain free fatty acids (FFAs), which can induce saponication in
a traditional base-catalyzed transesterication process, a pre-
esterication process is required. This process involves the
esterication of such oils with alcohol to convert the FFAs to
esters under acidic conditions, thus reducing the acid value of
the oil below 1 mg KOH/g. Many literatures have reviewed dif-
ferent methods of reducing FFAs in the raw oil [1,57]. Both
acid including homogeneous and heterogeneous acid and enzyme
Corresponding author at: College of Chemical Engineering, Sichuan University,
Chengdu 610065, China. Tel.: +86 28 85990278.
Corresponding authors at: Department of Paper and Bioprocess Engineering,
State University of New York-College of Environmental Science and Forestry, Syra-
cuse, NY 13210, USA. Tel.: +1 315 470 6885.
E-mail addresses: sliu@esf.edu (S. Liu), liangbin@scu.edu.cn (B. Liang).
simultaneously and avoid soap formation. But in commercial pro-
cesses, approximately 0.55 wt% (based on oil weight) of a liquid
acid is employed as catalyst, and the spent catalyst is removed
after pre-esterication to prevent interference in the subsequent
alkali-catalyzed transesterication. This process results in not only
the corrosion of pipelines, but also in the production of signicant
quantities of wastewater. Although many solid acids have been
studied to avoid soap formation and waste water, they are easy to
deactivate and expensive [5,6]. The high price, slow reaction rate
and deactivation of enzyme also limits its commercialization [6].
It has previously been demonstrated that liquid acids such as
sulfuric acid can slowly catalyze transesterication reactions at
room temperature, with accelerated rates achieved by increasing
the reaction temperature [8,9]. We previously reported a subcrit-
ical production process for biodiesel using high acid value oils
using trace acid catalysts (00.1% weight of oil) at an elevated
temperature below the supercritical temperature of the alcohol.
Experimental results showed that the conversion rate was reason-
able at signicantly lower pressures than those required for the
supercritical method [10].
In the acid catalyzed subcritical process, the trace acid catalyst
dissolves in the glycerol product following evaporation of excess

http://dx.doi.org/10.1016/j.cattod.2015.07.004
0920-5861/ 2015 Elsevier B.V. All rights reserved.

Please cite this article in press as: Y. Liu, et al., Kinetic studies on biodiesel production using a trace acid catalyst, Catal. Today (2015),
http://dx.doi.org/10.1016/j.cattod.2015.07.004
G Model
CATTOD-9661; No. of Pages 8 ARTICLE IN PRESS
2 Y. Liu et al. / Catalysis Today xxx (2015) xxxxxx
alcohol, giving a residual amount of acid catalyst in the biodiesel
phase which can be within acceptable biodiesel standards [11].
Transesterication between methanol and Jatropha curcas L. oil in
the presence of trace amounts of acid was previously investigated in
our group [10], where we found that both the esterication and the
transesterication reactions could be carried out simultaneously.
With regard to renewable resources for such processes, ethanol
can be produced by the fermentation of grain and sugarcane, and
has additional advantages over methanol in that it is less toxic.
Recently, the production of ethanol from lignocellulosic biomass
has been investigated, to increase ethanol production from abun-
dant renewable feedstocks [12]. In addition, it has been found that
the biodiesel produced from ethanol has improved low temper-
ature ow properties compared to that produced from methanol
[13].
We herein report the use of H2 SO4 (00.1 wt% based on the
weight of oil) for the catalysis of the simultaneous esterication
and transesterication of alcohol under subcritical conditions. Lit-
eratures relating to kinetics studies of biodiesel production with
acid catalyst have been published with different oils at different
reaction conditions in the past decade. Marchetti et al. investigated
the kinetics of esterication of FFA in acid oils between 3555 C
[9]. Two kinetic models were set up with and without considering
the effect of transesterication. Nautiyal et al. studied the kinetics
and thermodynamics on biodiesel production from algae [14]. Hex-
ane, sulfuric acid and methanol was added to dried algae biomass to
carry out extraction and transesterication in one step. It showed
that transesterication followed the rst order kinetics and the
rate of reaction varied linearly with yield of methyl ester from 30
to 60 C. Shuit et al. studied the kinetics of esterication of palm
fatty acid distillate with solid acid catalyst [15]. But few studies
have been carried out into kinetic studies of the transesterication
under trace acid catalyst conditions. We therefore report inves-
tigations into the measurement of kinetic properties relating to
the intermeidates of monoglyceride (MG) and diglyceride (DG) in
the trace H2 SO4 -catalyzed transesterication process. In addition,
a range of oils (corn oil and Pennycress oil) and alcohols (methanol
and ethanol) were compared. The impact of water and FFA in the
reaction system was also investigated.
2. Experimental
2.1. Materials
Corn oil was purchased from a local supermarket (Syracuse,
New York, USA). Pennycress oil was supplied by Alberta Agriculture
and Rural Division (Edmonton, Alberta, Canada) (see Table 1 for
composition details). Methanol (99.9%; Fisher Scientic), ethanol
(absolute 99%, extra dry, Fisher Scientic), and n-heptane (anhy-
drous, 99%, Sigma Aldrich), were used without further purication.
Standard solutions and internal standards of triglyceride (TG),
diglyceride (DG) and monoglyceride (MG) for analysis according to
the ASTM method D6584 were purchased from Sigma Aldrich. N-
methyl-N-(trimethylsilyl) triuoroacetamide (MSTFA) (97%) was
purchased from Fisher Scientic (Agawam, MA, USA). Unless oth-
erwise stated, all other chemicals were purchased from Fisher
Scientic (Agawam, MA, USA) and were of analytical grade. The
components of the fatty acids in corn oil and Pennycress oil are
listed in Table 1.
2.2. Biodiesel preparation
Transesterication was carried out in a 1 L stainless steel auto-
clave (Parr 4048), equipped with a heating jacket, an internal
Please cite this article in press as: Y. Liu, et al., Kinetic studies on biodiesel production using a trace acid catalyst, Catal. Today (2015),
http://dx.doi.org/10.1016/j.cattod.2015.07.004
cooling coil, an adjustable two-blade propeller agitator, a sampling
port, and a reux condenser.
A mixture of the desired alcohol and oil with the sulfuric acid
catalyst were charged to the autoclave (total volume 400 mL). After
heating the closed autoclave to the desired temperature (155 C,
165 C, 175 C, 185 C, or 195 C) over 35 min (5 min), the agita-
tor was switched on (at 300 rpm, unless specied otherwise), and
the zero point of the reaction time taken. The reaction mixture was
sampled at a range of time intervals. Each sample was washed with
distilled water to give a pH of approximately 7. After drying over
MgSO4 overnight, the transparent golden product was analyzed
by GC (see details below) to give the biodiesel components, MG,
DG, and TG. All experiments were conducted in triplicate to ensure
reproducibility. And the errors were within 3.5%.
2.3. Analytical methods
Analysis of the esters present in biodiesel samples was
by GC, using a Thermo Scientic Focus GC system equipped
with a Triplus automatic sampler and Trace TR-WaxMS
(30 m 0.25 mm 0.25 m) GC column. During analysis, the
temperature was increased from 120 C to 260 C at a rate of
5 C/min. The injector temperature was set at 250 C. Methyl
salicylate was used as internal standard.
Quantitative analysis of TG, DG, and MG was carried out using
an HP 5890 Series II GC system equipped with an FID detector, in
combination with a previously reported GC method [16].
For GC analysis, each sample was analyzed twice. Concentra-
tions of ethanol and glycerol were determined using the mass
balance of the reaction. The equations are as following,
[EtOH] = [EtOH]0 [FAEE] (1)
[GL] = [TG]0 [TG] [DG] [MG] (2)
where [TG], [DG], [MG], [EtOH], and [GL] are the concentrations
of triglyceride, diglyceride, monoglyceride, ethanol, and glycerol,
respectively. [EtOH]0 and [TG]0 are the initial concentrations of
ethanol and triglyceride when they were added into the autoclave.
3. Results and discussion
3.1. Transesterication process
The three successive reactions, which take place during the
transesterication of TG, are shown in Scheme 1.
As can be seen in Scheme 1, during the transesterication, TG
initially reacts with ethanol (EtOH) to form DG and FAEE (1). DG
then proceeds to react with an additional equivalent of ethanol to
form MG and FAEE (2). Finally, MG reacts with ethanol to produce
glycerol and FAEE (3).
Fig. 1 shows a typical concentration prole during the transes-
terication of corn oil with ethanol. As expected, the concentrations
of TG and ethanol decreased, whereas that of FAEE and glycerol
increased over time. The concentrations of the intermediates, DG
and MG, increased initially, and then decreased as the reaction pro-
gressed. The presence of small quantities of FAEE, DG, and MG at
t = 0 is due to the reaction beginning to take place during the heat
ramp to the reaction temperature. It should be noted that these
concentrations were taken as the initial concentrations for kinetic
analysis. It can also be seen from Fig. 1 that after 141 min, the con-
version of TG was complete, but after 240 min DG and MG were
still present, and the FAEE yield according to the fatty acid in TG
was < 100%. These factors suggest that the yield of FAEE is limited
by the three equilibria.
G Model
CATTOD-9661; No. of Pages 8 ARTICLE IN PRESS
Y. Liu et al. / Catalysis Today xxx (2015) xxxxxx 3

Table 1
Fatty acid compositions (wt%) of corn oil and Pennycress oil.

Fatty acid C16:0a C18:1 C18:2 C18:3 C20:1 C20:2 C22:1 C24:1 Others

Corn oil (wt%) 11.2 27.8 58.3 2.7

Pennycress oil (wt%) 1.61 8.8 24.2 12.6 8.6 1.9 38.9 2.3 1.2
a
For Cx:y, x is the number of carbon atoms and y is the number of double bonds.

Scheme 1. Three-step transesterication of triglyceride. k1 , k2 , and k3 are the forward reaction rate constants and k1 , k2 , and k3 are the reverse reaction rate constants.

11 respect to the mass of oil used) and temperatures (155 C, 165 C,

10 175 C, 185 C, and 195 C).
9 In order to build the kinetic models, it was assumed that:

(1) In the initial stage, mass transfer effects were negligible.

Concentration(mol/L)

1.0 (2) FFA content in rened corn oil is negligible. The corn oil contains
only triglycerides.
0.8 (3) The forward and reverse reactions of the three-step reversible
reactions were rst order reactions with respect to each reac-
0.6 tant.

0.4 The reaction rate for each step can therefore be written as:

r1 = k1 [TG] [EtOH] k1 [DG] [FAEE] (3)

0.2
r2 = k2 [DG] [EtOH] k2 [MG] [FAEE] (4)
0.0
0 50 100 150 200 250 r3 = k3 [MG] [EtOH] k3 [GL] [FAEE] (5)
Time(min)
therefore,
Fig. 1. Concentration proles of TG, DG, MG, FAEE, GL and EtOH during corn oil d [TG]
ethanolysis. Reaction conditions: 175 C, 0.1 wt% sulfuric acid, ethanol to corn oil = r1 (6)
dt
molar ratio 24:1. Symbols: () TG, () EtOH, () DG, () FAEE, () MG, and () GL.
d [DG]
= r1 r2 (7)
dt
3.2. Kinetics of the transesterication using trace acid catalyst
d [MG]
= r2 r3 (8)
Studies relating to kinetic measurements are important for the dt
understanding of the reaction behavior, to predict the reaction d [FAEE]
= r1 + r2 + r3 (9)
extent, and to determine reaction rate parameters. Although kinet- dt
ics study for biodiesel production with acid catalyst have received a d [GL]
signicant amount of attention over the past few decades [9,14,15], = r3 (10)
dt
kinetics study of ethanolysis for biodiesel production with small
amount of sulfuric acid as catalyst are rarely discussed. We there- d [EtOH]
= (r1 + r2 + r3 ) (11)
fore carried out kinetic studies into the transesterication between dt
corn oil and ethanol, based on consecutive reaction models and Substituting Eqs. (3)(5) into Eqs. (6)(11) gives the following:
experimental measurements. d [TG]
Reactions were carried out using a range of catalyst concen- = k1 [TG] [EtOH] + k1 [DG] [FAEE] (12)
dt
trations (0.02 wt%, 0.04 wt%, 0.06 wt%, 0.08 wt%, and 0.1 wt%, with

Please cite this article in press as: Y. Liu, et al., Kinetic studies on biodiesel production using a trace acid catalyst, Catal. Today (2015),
http://dx.doi.org/10.1016/j.cattod.2015.07.004
G Model
CATTOD-9661; No. of Pages 8 ARTICLE IN PRESS
4 Y. Liu et al. / Catalysis Today xxx (2015) xxxxxx

Table 2
Apparent rate constants.

H2 SO4 , wt% H2 SO4 , 102 mol L1 T, C k, 102 L mol1 min1

k1 k1 k2 k2 k3 k3

0.02 0.0765 175 0.094 0 0.372 0 0.395 0.972

0.04 0.1531 175 0.166 0 0.503 0 0.416 0.471
0.06 0.2296 175 0.219 0 0.663 0 0.477 0.403
0.08 0.3060 175 0.317 0 1.005 0 0.644 0.551
0.1 0.3827 175 0.350 0 1.150 0 0.670 0.400
0.06 0.2296 155 0.118 0 0.325 0 0.266 0.172
0.06 0.2296 165 0.158 0 0.357 0 0.294 0.210
0.06 0.2296 175 0.219 0 0.663 0 0.477 0.403
0.06 0.2296 185 0.243 0 0.858 0.391 0.653 0.566
0.06 0.2296 195 0.396 0 1.12 0.883 0.708 0.597

11 11
(a) 10
(b)
10

9 9

Concentration(mol/L)
1.0
Concentration(mol/L)

1.0

0.8 0.8

0.6 0.6

0.4 0.4

0.2 0.2

0.0 0.0
0 50 100 150 200 250 0 50 100 150 200 250

Time(min) Time(min)

11 11

10
(c) 10 (d)
9 9

1.2
Concentration(mol/L)

1.0
Concentration(mol/L)

1.0
0.8
0.8

0.6
0.6

0.4
0.4

0.2
0.2

0.0 0.0
0 50 100 150 200 250 0 50 100 150 200 250

Time Time(min)

11 8.8
8.4 (f)
(e)
10 8.0
9 7.6

1.2
Concentration(mol/L)
Concentration(mol/L)

1.0

1.0
0.8
0.8
0.6
0.6

0.4
0.4

0.2 0.2

0.0 0.0
0 50 100 150 200 250 0 50 100 150 200 250

Time(min) Time(min)

Fig. 2. Comparison of concentrations between the predicted results and experimental data under a range of reaction conditions: (a) 155 C, 0.02 wt% sulfuric acid, ethanol to
corn oil molar ratio = 24:1; (b) 195 C, 0.01 wt% sulfuric acid, ethanol to corn oil ratio = 24; (c) 155 C, 0.1 wt% sulfuric acid, ethanol to corn oil molar ratio = 24:1; (d) 195 C,
0.1 wt% sulfuric acid, ethanol to corn oil molar ratio = 24:1; (e) 165 C, 0.1 wt% sulfuric acid, ethanol to corn oil molar ratio = 24:1; (f) 175 C, sulfuric acid 2.296 103 mol L1 ,
ethanol to corn oil molar ratio = 18:1. Key: () TG, () EtOH, () DG, () FAEE, () MG, () GL. Solid line = model prediction.

Please cite this article in press as: Y. Liu, et al., Kinetic studies on biodiesel production using a trace acid catalyst, Catal. Today (2015),
http://dx.doi.org/10.1016/j.cattod.2015.07.004
G Model
CATTOD-9661; No. of Pages 8 ARTICLE IN PRESS
Y. Liu et al. / Catalysis Today xxx (2015) xxxxxx 5

Table 3 2
K1 , K2 and K3 for the transesterication of corn oil, conversion of TG (xTG ), and 0.012 k1, k=0.827[H2SO4], R =0.9838
selectivities of DG (SDG ), MG (SMG ), and GL (SGL ) with 0.1 wt% H2 SO4 at a range of 2
k2, k=2.535[H2SO4]+0.001, R =0.9718
temperatures.
0.010 2
k3, k=0.946[H2SO4]+0.002, R =0.9148
Temperature ( C) 155 165 175 195

k (L mol min )
-1
K1 a 0.02
0.008
K2 a 1.84 12.70

-1
K3 a 1.07 1.30 1.99 1.19 0.006
xTG (%)a 95.13 100 100 100
SDG (%)a 0.66 0.10 0 0
0.004
SMG (%)a 9.77 9.21 6.31 9.96
SGL (%)a 85.28 90.50 91.67 90.27
a
0.002
Reaction conditions: ethanol to corn oil ratio = 24:1, 0.1 wt% H2 SO4 , 240 min.

0.000
0.000 0.001 0.002 0.003 0.004
d [DG] [H2SO4] (mol/L)
= k1 [TG] [EtOH] k1 [DG] [FAEE] k2 [DG] [EtOH]
dt
Fig. 3. Apparent rate constants as a function of catalyst concentration.
+ k2 [MG] [FAEE] (13)
From Table 2, it could be seen that k1 is the lowest rate constant
of the three forward reactions, which conrms that the conversion
d [MG]
= k2 [DG] [EtOH] k2 [MG] [FAEE] k3 [MG] [EtOH] of TG to DG is the rate-determining step in the overall reaction.
dt
All forward reaction rate constants were found to increase with an
+ k3 [GL] [FAEE] (14) increase in both sulfuric acid concentration and reaction temper-
ature. In addition, a k1 value of 0 means that the rst step in the
transesterication process is irreversible. This was veried by the
d [FAEE]
= k1 [TG] [EtOH] k1 [DG] [FAEE] + k2 [DG] [EtOH] low value of K1 in Table 3 at 155 C. At temperatures greater than
dt (15) 165 C, the conversion of TG reached 100% after 240 min.
k2 [MG] [FAEE] + k3 [MG] [EtOH] k3 [GL] [FAEE] The progress of the second step was also examined. At 155 C,
165 C, and 175 C, it could also be seen that k2 was equal to zero,
d [GL] thus conrming that the second step of the transesterication reac-
= k3 [MG] [EtOH] k3 [GL] [FAEE] (16)
dt tion is also irreversible. As the values for K2 were larger than 1, and
k2 s are the highest apparent rate constants, it appears that DG con-
d [EtOH] verts rapidly into MG. This result is similar to the alkali-catalyzed
= (k1 [TG] [EtOH] k1 [DG] [FAEE] + k2 [DG] [EtOH]
dt transesterication reaction at low temperatures [18].
k2 [MG] [FAEE] + k3 [MG] [EtOH] k3 [GL] [FAEE]) (17) In contrast to the rst and second steps, the third step is indeed
reversible, and so MG cannot be completely converted to glycerol.
Even after 240 min at 195 C, the selectivity of MG was still 9.96%.
The rate constants can be calculated by integration of the differ- In the NaOH-catalyzed ethanolysis of sunower oil, Reyero et al.
ential equations. The ODexLims function in Microsoft Excel 2003 [19] found similar results. But this is in contrast to results reported
was used to solve differential equations by the least square method by Vicente et al. [18], where the third step of the transesterica-
[17]. Table 2 gives the apparent rate constants for each reaction. tion was irreversible. They attributed this effect to the immiscibility
To identify the model, F-test was done to test the models and of methyl ester and glycerol. In such studies, we observed that in
parameters. The results showed that the model and parameters the nal stages of the reaction, reactants in our experiments were
were t to describe the experiments. homogeneous. As all values for K3 were found to be close to 1, it
Models using the reaction rate constants given in Table 2 were is clear that the removal of glycerol from the system is optimal to
used to simulate the reaction procedure, and the results were com- drive equilibrium.
pared with the experimental data shown in Fig. 2. The effects of catalyst concentration were investigated and were
Fig. 2 displays an excellent match between the kinetic mod- found to signicantly affect the transesterication rate. In the
els and the experimental data, allowing us to conclude that the absence of catalyst, after 240 min at 175 C, the FAEE content was
selected kinetic models are suitable for prediction of the reaction found to be only 1.95 wt%, indicating an extremely slow transes-
path. Using the experimental results, the equilibrium of transes- terication rate. However, with the addition of a trace amount of
terication can also be calculated. K1 , K2 , and K3 represent the sulfuric acid (0.02 wt%), an FAEE content of 75.2 wt% was mea-
equilibrium constants for the three steps when at equilibrium and sured under the same conditions. Fig. 3 shows the plots of the
can be dened as: forward reaction rate constants versus the catalyst concentration.
[FAEE] [DG] In all cases, the points tted to a straight line, demonstrating that
K1 = (18)
[EtOH] [TG] all forward reaction rates were proportional to the acidic catalyst
concentration.
[FAEE] [MG]
K2 = (19) According to Arrheniuss law (Eq. (21)) the temperature depend-
[EtOH] [DG]
ency of the reaction rate constant was studied (Fig. 4). In addition,
[FAEE] [GL] the activation energies for each step of the reaction were calculated
K3 = (20)
[EtOH] [MG] (Table 4). From these data, it can be concluded that the appar-
ent reaction rate constants are in agreement with the Arrhenius
Table 3 gives the values for K1 , K2 , and K3 under a range of
equation.
reaction conditions. In addition, the conversions of TG (xTG ), and  E
selectivities of DG (SDG ), MG (SMG ), and glycerol (SGL ) are also shown k = k0 exp (21)
in Table 3. RT

Please cite this article in press as: Y. Liu, et al., Kinetic studies on biodiesel production using a trace acid catalyst, Catal. Today (2015),
http://dx.doi.org/10.1016/j.cattod.2015.07.004
G Model
CATTOD-9661; No. of Pages 8 ARTICLE IN PRESS
6 Y. Liu et al. / Catalysis Today xxx (2015) xxxxxx

-4.0 100
-4.5 90

FAEE concentration (wt%)

-5.0
80
70 with agitation
-5.5
60 without agitation
lnk

-6.0
2
k1, R =0.9665
50
-6.5 2 40
k2, R =0.9576
2
-7.0 k-2, R =0.9401 30
2
k3, R =0.9465 20
-7.5
2
k-3, R =0.9371
10
-8.0
0.00210 0.00215 0.00220 0.00225 0.00230 0.00235 0
-1 0 30 60 90 120 150 180 210 240
1/T (K )
time (min)
Fig. 4. Arrhenius plot of reaction rate versus temperature.
Fig. 5. Effects of agitation on the FAEE content. Reaction conditions: 175 C, ethanol
to corn oil molar ratio = 24:1, 0.06 wt% sulfuric acid.
Table 4
Activation energies and pre-exponential factors.

Reaction TG DG DG MG MG GL MG DG GL MG comparison. In Fig. 5, the FAEE contents both under agitation and
in the static state are shown.
E (kJ/mol) 47.4 55.8 46.0 264.5 58.1
k0 (L mol1 min1 ) 702.7 19,262.2 1028.6 3.57 1027 20,685.7
As can be seen, the curves relating to the reaction with and
without agitation are comparable. In addition to the FAEE concen-
trations, the TG, DG, and MG contents were measured and were
Table 5 also found to be comparable. This indicates that at elevated tem-
Conversion of TG and selectivities of DG, MG, and glycerol at a range of ethanol to peratures the reactions were conducted in a miscible system.
corn oil molar ratios.
In the case of the agitation experiments, the samples withdrawn
Ethanol:corn oil (mol/mol) 3:1 6:1 9:1 12:1 18:1 24:1 from the system contained two phases in the initial stages of the
xTG (%) 91.67 97.71 100 100 100 100
reaction, with the upper phase being ethanol-rich, and the lower
SDG (%) 18.74 5.09 2.14 0.80 0 0 phase being oil-rich. As the reaction progressed, the lower phase
SMG (%) 26.08 27.55 20.26 15.93 11.49 7.98 was found to decrease in quantity, with a homogenous system
SGL (%) 64.53 78.80 85.74 87.47 90.56 91.91 being present nally. As these samples were observed at room tem-
Reaction conditions: 175 C, 0.06 wt% H2 SO4 , 240 min. perature, it would be expected that a homogenous system was also
present at the higher reaction temperatures.
In addition, as previously reported, a homogenous phase is
3.3. Effect of ethanol/corn oil molar ratio formed between soybean oil and ethanol at 70 C [20]. Follegatti-
Romero et al. [21] also investigated the solubility of soybean oil,
We then chose to investigate the effects of ethanol dosage on sunower oil, rice bran oil, cottonseed oil, palm olein oil, and palm
the transesterication reaction, and found that it had a signicant oil in anhydrous ethanol between 25 C and 60 C. This therefore
effect on the conversion. Using a constant total volume of 400 mL, conrmed that these oils can form homogeneous phases at approx-
the molar ratio of ethanol to corn oil was varied between 3:1 and imately 70 C. We therefore expect that at 175 C, corn oil and
24:1 at a temperature of 175 C. During these experiments, the con- ethanol form a homogeneous phase.
centration of sulfuric acid used was 2.296 103 mol L1 . Table 5
below shows the conversion of TG, and selectivities of DG and MG
after 240 min.
3.5. Effect of water content
The results in Table 5 support previous data, showing that the
transesterication reaction follows a consecutive mechanism with
As water content is a critical factor in conventional base- or
a nal reversible third step. It was found that increasing the molar
acid-catalyzed transesterication reactions [22], we subsequently
ratio of ethanol to corn oil pushed the equilibrium to the right, as
investigated the water tolerance of the system by the addition of
can be seen from an increase in the conversion of TG and the selec-
1 wt% and 3 wt% of water (Fig. 6).
tivity of glycerol, while a decrease in the selectivity of both DG and
Fig. 6 shows how the reaction curves were altered with a vari-
MG was observed. In addition, after 240 min, with a molar ratio of
ation in water content. It can clearly be seen that lower reaction
24:1, the selectivity of MG was still 7.98%, while DG was completely
rates were observed where 1 wt% and 3 wt% of water were added.
converted when the molar ratio exceeded 18:1. This demonstrates
It has been previously reported that water inuences reactivity in
that the rst and the second steps of transesterication process are
transesterication [22], while also decreasing the mutual-solubility
able to reach completion under these conditions, while the conver-
of the ethanol-oil system [23,24]. As this reaction is catalyzed by
sion of MG to glycerol reaches equilibrium due to the reversibility
trace amounts of acid, the presence of moisture may dilute the acid
of the third step.
in the reaction system, thus reducing the rate. However, in the nal
stage of the reaction, although the presence of water aids in the
3.4. Effect of mixing separation of glycerol from the oil phase, it also leaches the acid
catalyst from the oil phase where the transesterication reaction
As corn oil and ethanol are not completely miscible at room takes place.
temperature, it is important to enhance interphase mass transfer. However, it is expected that the process using trace acid cata-
The effects of mixing were therefore investigated by experimental lyst can tolerate greater quantities of water due to the high reaction

Please cite this article in press as: Y. Liu, et al., Kinetic studies on biodiesel production using a trace acid catalyst, Catal. Today (2015),
http://dx.doi.org/10.1016/j.cattod.2015.07.004
G Model
CATTOD-9661; No. of Pages 8 ARTICLE IN PRESS
Y. Liu et al. / Catalysis Today xxx (2015) xxxxxx 7

100 100
90 90

FAEE concentration (wt%)

80 80
FAEE concentration (wt%)

70 without water 70
60 1 wt% water
60
3 wt% water
50
50
40 o
40 Pennycress oil, 175 C, 0.06 wt% H2SO4
30 o
30 Pennycress oil, 175 C, 0.1 wt% H2SO4
20 o
20 corn oil,175 C, 0.06 wt% H2SO4
10 o
10 corn oil,175 C, 0.1 wt% H2SO4
0
0 30 60 90 120 150 180 210 240 0
time (min) 0 30 60 90 120 150 180 210 240
time (min)
Fig. 6. Effects of water on FAEE content. Reaction conditions: 175 C, ethanol to corn
oil molar ratio = 24:1, 0.06 wt% sulfuric acid. Fig. 8. Comparison of the transesterication process using corn oil and Pennycress
oil. Reaction conditions: 175 C, ethanol to corn oil molar ratio = 24.

100
90
FAEE concentration (wt%)

80 three reaction curves, it appears that the esterication of FFA is

70 faster than the transesterication of TG. FFA esteried rapidly with
ethanol as the reaction mixture was heated to the desired temper-
60 without FFA ature. However, the presence of FFA lowered the nal yield of FAEE,
50 20% FFA likely due to the formation of water during the esterication of FFA
30% FFA with ethanol.
40
If we consider that the 30% of the added FFA is converted com-
30
pletely, approximately 1.9 wt% of water will be produced (based
20 on the mass of oil), which has been shown previously to be com-
10 patible with the system reported herein. The results from these
experiments therefore demonstrate our system is suitable for oils
0
0 30 60 90 120 150 180 210 240 containing a high FFA content, unlike commercial processes, where
the FFA content is limited to <0.5%.
time (min)

Fig. 7. Effects of FFA on the FAEE concentration. Reaction conditions: 175 C, ethanol
to corn oil molar ratio = 24:1, 0.06 wt% sulfuric acid. 3.7. Transesterication of Pennycress oil

Pennycress oil, which has a signicantly different fatty acid com-

temperatures used. Signicant effects of water on the transesteri-
cation reactions have been reported, with the water content in
conventional processes being limited to <0.06% [22] or <0.5% [8].
However, in our experiments, it was observed that in the presence
of 1 wt% and 3 wt% water, the nal FAEE content reached 94.2% in
240 min, which is comparable with 96.0% FAEE content achieved in
the absence of water. In this case, the reaction rate was inuenced
only slightly, giving similar results to those previously reported by
our group in relation to the transesterication of Jatropha curcas L.
oil and methanol [10].
3.6. Effects of free fatty acids (FFA)
As a number of commercially recovered oils and unrened oils
contain free fatty acids (FFA), the tolerance of the system to FFA
was investigated by the addition of oleic acid to simulate feedstocks
containing a range of acid values. Under acidic conditions, FFA can
react with alcohol to form biodiesel. The esterication of FFA with
ethanol is shown in the following equation:
RCOOH + CH3 CH2 OH  RCOOCH2 CH3 + H2 O
Fig. 7 shows how the reaction curves varied in both the pres-
ence and absence of FFA. It can clearly be seen that where FFA was
present, higher initial FAEE concentrations and lower nal FAEE
concentrations were recorded. Upon further examination of the
position compared to corn oil, is widespread in North America.
Pennycress oil was found to contain a wide range of fatty acids ran-
ging from C16 to C24, while corn oil contained mainly C16 to C18
fatty acids. The carbon chain length in corn oil is therefore shorter
than that in Pennycress oil. Corn oil has an unsaturated fatty acid
content of >86%, whereas the content in Pennycress oil is appro-
aching 97%, resulting in biodiesel from such oils having good cold
ow property.
Comparisons of the transesterication of Pennycress oil with
corn oil are shown in Fig. 8.
As can be seen from the FAEE concentration plots in Fig. 8,
differences in reaction rate between the two oils were apparent.
Although the conversion rates using 0.1 wt% H2 SO4 were compara-
ble, the FAEE content from the corn oil reaction was slightly higher
than that of Pennycress oil. In addition, the presence of TG from
the corn oil reaction was not detected in the nal product after
140 min when 0.1 wt% H2 SO4 was used as catalyst. In contrast, after
the same reaction time, the TG content in the Pennycress oil reac-
tion was found to be 1.80%. In the presence of 0.06 wt% H2 SO4 , the
initial conversion rate of corn oil was slower than that of Penny-
(22)
cress oil, whereas the nal conversion was higher. It indicates that
corn oil has lower reactivity than Pennycress oil at the initial time
of reaction, but a higher conversion at the end time of reaction. The
difference in reactivity between the two oils is likely to be due to
the difference of carbon chain length in two oils.

Please cite this article in press as: Y. Liu, et al., Kinetic studies on biodiesel production using a trace acid catalyst, Catal. Today (2015),
http://dx.doi.org/10.1016/j.cattod.2015.07.004
G Model
CATTOD-9661; No. of Pages 8 ARTICLE IN PRESS
8 Y. Liu et al. / Catalysis Today xxx (2015) xxxxxx
100
FAME/FAEE concentration (wt%)
90
80
70
60
50
40 ethanol, 0.06 wt% H 2SO4
30 ethanol, 0.1 wt% H 2SO4
methanol, 0.06 wt% H 2SO4
20
methanol, 0.1 wt% H 2SO4
10
0 [7] E. Lotero, Y.J. Liu, D.E. Lopez, K. Suwannakarn, D.A. Bruce, J.G. Goodwin,
0 30 60 90 120 150
time (min)
Fig. 9. Comparison between the transesterication reactions of methanol
and ethanol based on FAME/FAEE concentration. Reaction conditions: 175 C,
ethanol/methanol to Pennycress oil molar ratio = 24:1.
3.8. Effect of alcohol chain length on transesterication
Fig. 9 shows the transesterication curves of Pennycress oil
using both methanol and ethanol, where it can be seen that
the transesterication reaction of methanol is faster than that
of ethanol. Kwon et al. [25] reported similar results for the
uncatalyzed transesterication using methanol and ethanol below
400 C. Both results suggest that methanol is more active than
ethanol in the transesterication reaction, as signicantly higher
FAME concentrations were observed at both 0.06 wt% and 0.1 wt%
H2 SO4 .
4. Conclusions
Biodiesel production using a trace acid catalyst was systemat-
ically investigated. It was found that both the esterication and
transesterication processes could be carried out in a single batch
reactor. Based on experimental data, the reaction rate equations
were used to predict the reaction pathway, which was shown to be
a 3-step successive transesterication reaction, in which the rst
step is the rate-determining step. The rst and the second steps
were found to be irreversible, while the third step was reversible.
Calculations and experimental data showed that all steps in the
overall reaction were second order reactions, with the apparent
rate constants being in agreement with the Arrhenius equation.
In addition, it was found that the apparent rate constants for the
three forward reactions were proportional to the sulfuric acid
content. Water (<3 wt%), FFA(<30 wt%), and stirring were found to
have minimal effects on the nal compositions of the biodiesel.
Finally, methanol appeared to display greater reactivity than
ethanol, while corn oil displayed higher reactivity than Pennycress
oil. Oils contained FFA and/or water can be carried out to produce
biodiesel using trace acid catalyst.
Please cite this article in press as: Y. Liu, et al., Kinetic studies on biodiesel production using a trace acid catalyst, Catal. Today (2015),
http://dx.doi.org/10.1016/j.cattod.2015.07.004
References
[1] M.G. Kulkarni, A.K. Dalai, Waste cooking oilan economical source for
biodiesel: a review, Ind. Eng. Chem. Res. 45 (2006) 29012913.
[2] Y. Wang, J. Nie, M. Zhao, S. Ma, L. Kuang, X. Han, S. Tang, Production of
biodiesel from waste cooking oil via a two-step catalyzed process and
molecular distillation, Energy Fuels 24 (2010) 21042108.
[3] L.H. Chin, B.H. Hameed, A.L. Ahmad, Process optimization for biodiesel
production from waste cooking palm oil (Elaeis guineensis) using response
surface methodology, Energy Fuels 23 (2009) 10401044.
[4] G.L. Maddikeri, A.B. Pandit, P.R. Gogate, Intensication approaches for
biodiesel synthesis from waste cooking oil: a review, Ind. Eng. Chem. Res. 51
(2012) 1461014628.
[5] D.Y.C. Leung, X. Wu, M.K.H. Leung, A review on biodiesel production using
catalyzed transesterication, Appl. Energy 87 (2010) 10831095.
[6] M.K. Lam, K.T. Lee, A.R. Mohamed, Homogeneous, heterogeneous and
enzymatic catalysis for transesterication of high free fatty acid oil (waste
cooking oil) to biodiesel: a review, Biotechnol. Adv. 28 (2010) 500518.
180 210 240 Synthesis of biodiesel via acid catalysis, Ind. Eng. Chem. Res. 44 (2005)
53535363.
[8] M. Canakci, J. Van Gerpen, Biodiesel production via acid catalysis, Trans. ASAE
42 (1999) 12031210.
[9] J.M. Marchetti, M.N. Pedernera, N.S. Schbib, Production of biodiesel from acid
oil using sulfuric acid as catalyst: kinetics study, Int. J. Low Carbon Technol. 6
(2011) 3843.
[10] Y.Y. Liu, H.F. Lu, W. Jiang, D.S. Li, S.J. Liu, B. Liang, Biodiesel production from
crude Jatropha curcas L. oil with trace acid catalyst, Chin. J. Chem. Eng. 20
(2012) 740746.
[11] C.W. Chiu, M.J. Goff, G.J. Suppes, Distribution of methanol and catalysts
between biodiesel and glycerin phases, AIChE J. 51 (2005) 12741278.
[12] G. Sivakumar, D.R. Vail, J.F. Xu, D.M. Burner, J.O. Lay, X.M. Ge, P.J. Weathers,
Bioethanol and biodiesel: alternative liquid fuels for future generations, Eng.
Life Sci. 10 (2010) 818.
[13] M.H. Su, J.X. Wang, T.J. Li, Improvement of low temperature uidity of
soybean oil biodiesel with ethanol, China Oils Fats 33 (2008) 5961.
[14] P. Nautiyal, K.A. Subramanian, M.G. Dastidar, Kinetic and thermodynamic
studies on biodiesel production from Spirulina platensis algae biomass using
single stage extractiontransesterication process, Fuel 135 (2014)
228234.
[15] S. Shuit, S. Tan, Biodiesel production via dsterication of palm fatty acid
distillate using sulphonated multi-walled carbon nanotubes as a solid acid
catalyst: process study, catalyst reusability and kinetic study, Bioenergy Res. 8
(2015) 605617.
[16] V.F. Marulanda, G. Anitescu, L.L. Tavlarides, Investigations on supercritical
transesterication of chicken fat for biodiesel production from low-cost lipid
feedstocks, J. Supercrit. Fluids 54 (2010) 5360.
[17] S. Liu, Bioprocess Engineering: Kinetics, Biosystems, Sustainability and
Reactor Design, Elsevier, Amsterdam, 2013.
[18] G. Vicente, M. Martnez, J. Aracil, A. Esteban, Kinetics of sunower oil
methanolysis, Ind. Eng. Chem. Res. 44 (2005) 54475454.
[19] I. Reyero, G. Arzamendi, S. Zabala, L.M. Ganda, Kinetics of the
NaOH-catalyzed transesterication of sunower oil with ethanol to produce
biodiesel, Fuel Process. Technol. 129 (2015) 147155.
[20] X.J. Liu, X.L. Piao, Y.J. Wang, S.L. Zhu, Liquid-liquid equilibrium for systems of
(fatty acid ethyl esters plus ethanol plus soybean oil and fatty acid ethyl
esters plus ethanol plus glycerol), J. Chem. Eng. Data 53 (2008) 359362.
[21] L.A. Follegatti-Romero, M. Lanza, C.A.S. da Silva, E.A.C. Batista, A.J.A. Meirelles,
Mutual solubility of pseudobinary systems containing vegetable oils and
anhydrous ethanol from (298.15 to 333.15) K, J. Chem. Eng. Data 55 (2010)
27502756.
[22] F. Ma, L.D. Clements, M.A. Hanna, The effects of catalyst, free fatty acids, and
water on transesterication of beef tallow, Trans. ASAE 41 (1998)
12611264.
[23] C.s.A.S. da Silva, G. Sanaiotti, M. Lanza, L.A. Follegatti-Romero, A.J.A. Meirelles,
E.A.C. Batista, Mutual solubility for systems composed of vegetable
oil + ethanol + water at different temperatures, J. Chem. Eng. Data 55 (2009)
440447.
[24] C.E.C. Rodrigues, R. Antoniassi, A.J.A. Meirelles, Equilibrium data for the
system rice bran oil + fatty acids + ethanol + water at 298.2 K, J. Chem. Eng.
Data 48 (2003) 367373.
[25] E.E. Kwon, H. Yi, Y.J. Jeon, Transforming rapeseed oil into fatty acid ethyl Ester
(FAEE) via the noncatalytic transesterication reaction, AIChE J. 59 (2013)
14681471.