Energy Conversion and Management 158 (2018) 168–175

Contents lists available at ScienceDirect

Energy Conversion and Management

journal homepage: www.elsevier.com/locate/enconman

Enzymatic production of biodiesel from insect fat using methyl acetate as an T

acyl acceptor: Optimization by using response surface methodology
⁎
Hoang Chinh Nguyena, Shih-Hsiang Liangb, Shang-Sian Chenc, Chia-Hung Suc, , Jhih-Huei Linc,
Chien-Chung Chienc
a
Faculty of Applied Sciences, Ton Duc Thang University, Ho Chi Minh City 700000, Vietnam
b
Hsinchu Branch, Livestock Research Institute, Council of Agriculture, Miaoli County 36841, Taiwan
c
Graduate School of Biochemical Engineering, Ming Chi University of Technology, New Taipei City 24301, Taiwan

A R T I C L E I N F O A B S T R A C T

Keywords: Black soldier fly larvae (BSFL) are oleaginous insects that can assimilate organic waste for fat accumulation, and
Insect fat thus serve as an alternative feedstock for biodiesel production. In lipase-catalyzed transesterification, enzymes
Interesterification are deactivated by excess methanol. To address this obstacle, methyl acetate is suggested as an alternative acyl
Lipase acceptor to methanol. In this study, methyl acetate was first used in the enzymatic production of biodiesel with
Methyl acetate
BSFL as a triglyceride source. The interesterification of BSFL fat with methyl acetate was catalyzed using
Response surface methodology
Novozym 435 as an efficient immobilized lipase. Response surface methodology was used to optimize the re-
action and establish a reliable mathematical model for prediction. A maximum biodiesel yield of 96.97% was
reached at a reaction time of 12 h, molar ratio of methyl acetate to fat of 14.64: 1, enzyme loading of 17.58%,
and temperature of 39.5 °C. Under these optimal reaction conditions, Novozym 435 could be reused for up to 20
cycles without loss in enzyme activity. The properties of BSFL biodiesel were also investigated and all met the
European standard EN 14214. This study indicates that the enzymatic interesterification of BSFL fat with methyl
acetate is a promising and ecofriendly method for green fuel production.

1. Introduction shown to meet the specifications of the European biodiesel standard EN

With rapidly increasing demands for energy and environmental
protection, biodiesel has been increasingly developed worldwide as a
green fuel to replace petroleum [1,2] because of its combustion effi-
ciency, renewability, reduced environmental footprint, and compat-
ibility with diesel engines without modification [3–5]. However, a
major drawback of biodiesel compared with petrodiesel is its high price
due to the high cost of feedstocks, which account for 75% of the pro-
duction cost [6,7]. Therefore, developing a novel and inexpensive
feedstock for biodiesel production is an urgent requirement.
Insects have attracted much attention as a feedstock source for
producing biodiesel because of their high fat content, short life cycles,
and high reproduction rates [8–10]. In particular, black soldier fly
larvae (BSFL; Hermetia illucens) have been reported as a promising
biodiesel feedstock [10,11]. These insects can degrade various organic
wastes derived from animals and plants, such as animal manure
[10,12], restaurant waste [13], and lignocellulosic biomass [14] to
accumulate high fat content, which is subsequently used for biodiesel
production [11,15]. The fuel properties of the resulting biodiesel were
14214 [11]. In addition, the cell debris after fat extraction can be used
as a high-protein feed for cultivating aquatic animals, poultry, and li-
vestock [10,15]. Consequently, this insect has been increasingly used as
feedstock for biodiesel production to reduce production costs and fa-
cilitate the green conversion of waste into energy.
In the conventional process of using BSFL for biodiesel production,
larval fat is reacted with an acyl acceptor using sulfuric acid and sodium
hydroxide as catalysts [12,15]. The most common acyl acceptors for
biodiesel production are methanol and ethanol because of their avail-
ability and low price [12,16]. Although biodiesel can be successfully
produced using a chemically catalyzed process, several associated
problems exist, including corrosion damage to equipment and the
complicated separation of saponified products and catalysts from bio-
diesel [11,17,18]. The use of supercritical alcohols in the transester-
ification of oil for biodiesel production addresses these problems;
however, this process operates at extreme pressures (20–43 Mpa) and
temperatures (350–400 °C), resulting in product degradation [19–21].
Lipase-catalyzed reactions were proven to be a promising method for
biodiesel production to resolve the aforementioned limitations [11].

⁎
Corresponding author.
E-mail addresses: nguyenhoangchinh@tdt.edu.vn (H.C. Nguyen), chsu@mail.mcut.edu.tw (C.-H. Su).

https://doi.org/10.1016/j.enconman.2017.12.068
Received 20 October 2017; Received in revised form 13 December 2017; Accepted 20 December 2017
Available online 09 January 2018
0196-8904/ © 2017 Elsevier Ltd. All rights reserved.
H.C. Nguyen et al.
Nguyen et al. [11] successfully produced biodiesel through the trans-
esterification of BSFL fat with methanol using immobilized lipase under
mild reaction conditions. The lipase-catalyzed process is energy saving
and ecofriendly [22–24], and thus decreases environmental damage.
Nevertheless, lipases are deactivated by high dosages of methanol or
ethanol [11]. To overcome this obstacle, several solutions have been
proposed, such as the use of solvents as diluents or the stepwise addi-
tion of alcohol [25,26]. However, these methods decrease the reaction
rate because they maintain low concentrations of alcohol in the reac-
tion mixture [27,28] and thus limit the industrial application of enzy-
matic processes for biodiesel production.
Methyl acetate, a novel acyl acceptor, has been developed for bio-
diesel production to overcome the problems associated with enzymatic
transesterification using methanol [29,30]. The transesterification of
triglyceride using methyl acetate is also known as interesterification or
an ester exchange reaction, which involves three consecutive reversible
reactions to convert triglycerides into biodiesel and triacetin [31,32].
High dosages of methyl acetate in the reaction have been shown to have
no adverse effects on enzyme activity and stability, which are the key
concerns in the enzymatic process, thus enhancing the reaction rate
[33–35]. Du et al. [36] obtained a higher biodiesel yield by using
methyl acetate compared with methanol, and lipase could be used
continually for 100 cycles without loss of enzyme activity. In addition,
triacetin, a byproduct of this process, is used as an additive in the to-
bacco, cosmetic, and pharmaceutical industries [31]. Studies have
shown that triacetin can be added to the formulation of biodiesel at up
to 10% to improve certain biodiesel properties [31,37,38]. Because of
these merits, this method has been employed in biodiesel production
using olive oil [38], soybean oil [36], sunflower oil [39], and waste
cooking oil [35]. However, no study has reported the use of methyl
acetate for biodiesel production from insect fat.
This study examined the lipase-catalyzed interesterification of BSFL
fat using methyl acetate as an acyl acceptor for biodiesel production.
Response surface methodology (RSM) was used to optimize the reaction
conditions by analyzing the effects of reaction factors (reaction time,
molar ratio of methyl acetate to fat, enzyme loading, and temperature)
on the biodiesel yield. The reusability of the enzyme was also in-
vestigated under optimal reaction conditions. Finally, the properties of
BSFL biodiesel were determined according to the American Society for
Testing and Materials (ASTM) method.
2. Materials and methods
2.1. Materials
Novozym 435 (Candida antarctica lipase) was purchased from
Novozymes A/S (Bagsvaerd, Denmark). Methanol (HPLC grade, Tedia,
USA), n-hexane (HPLC grade, Tedia, USA), methyl acetate (HPLC
grade), and other reagents were purchased from ECHO Chemical Co.
Ltd. (Miaoli, Taiwan).
2.2. Insect species and growth conditions
BSFL (Hermetia illucens) were obtained from the Livestock Research
Institute (Hsinchu Branch, Miaoli County, Taiwan) and maintained for
more than 10 generations before being used in this study. To produce
biomass, the larvae were inoculated into fermented wheat bran at a
ratio of 1200 larvae per kilogram of substrate and incubated at 30 °C
with 65% moisture. After 20 days, the larvae were harvested from re-
sidue and inactivated at 105 °C for 10 min. They were subsequently
dried at 60 °C for 2 days and stored at 4 °C until use.
2.3. Fat extraction from BSFL
The BSFL were ground with a RT-02B micromill (Rong-Tsong
Precision Technology Co., Taiwan) to produce BSFL powder. The BSFL
169
Energy Conversion and Management 158 (2018) 168–175
powder was then immersed in n-hexane at a ratio of 1:5 (w/v) for 48 h
at room temperature. The hexane phase containing crude fat was se-
parated from cell debris through filtration. Finally, the crude fat was
obtained by evaporating hexane using a rotary evaporator (N-1200,
Eyela, Tokyo, Japan). The acid, iodine, and saponification values of
crude fat were determined according to the standard method [40,41].
2.4. Effect of various acyl acceptors on biodiesel production
A comparative study was conducted on the enzyme-catalyzed
transesterification of BSFL fat with methanol and interesterification
with methyl acetate to investigate the effects of the acyl acceptor types
(methanol and methyl acetate) on biodiesel yield. The reaction was
initiated by adding 4% Novozym 435 into reaction mixtures containing
the acyl acceptors and BSFL fat at various molar ratios (from 1:1 to
12:1). The reaction was then maintained at 40 °C with shaking for 12 h.
After the reaction was completed, the sample was collected, washed 3
times with deionized water, and subsequently centrifuged to remove
the aqueous layer [42]. The upper layer was collected and subjected to
rotary evaporation (N-1200, Eyela, Tokyo, Japan) to recover the re-
sidual methyl acetate, after which biodiesel was obtained and used to
determine the biodiesel yield.
2.5. Optimization of interesterification using RSM
A three-level and four-factorial Box–Behnken design was employed
to investigate the effects of reaction factors on the biodiesel yield.
Interesterification reactions with various reaction times (4–12 h), molar
ratios of methyl acetate to fat (9:1–15:1), enzyme loadings (10%–20%),
and reaction temperatures (30–50 °C) were performed in 150-mL
Erlenmeyer flasks with vigorous shaking. The biodiesel yield was de-
termined from the sample withdrawn from the reaction mixtures. The
relationship between the determined biodiesel yield and reaction fac-
tors was established using the following quadratic equation:
Y = β0 + β1 X1 + β2 X2 + β3 X3 + β4 X 4 + β11 X12 + β22 + β33 X32 + β44 X42
+ β12 X1 X2 + β13 X1 X3 + β14 X1 X 4 + β23 X2 X3 + β24 X2 X 4 + β34 X3 X 4
(1)
where Y is the biodiesel yield; X1 is the reaction time; X2 is the molar
ratio of methyl acetate to fat; X3 is the enzyme loading; X4 is the re-
action temperature; β0 is the regression coefficient for the intercept
term; β1-β4 are linear parameters; β12, β13, β14, β23, β24, and β34 are
interaction parameters; and β11, β22, β33, and β44 are quadratic para-
meters. The model parameters were determined using the least-squares
method [43]. A mathematical model was then used to determine the
optimal reaction conditions for obtaining the maximal biodiesel yield
using a canonical method [43]. Minitab 16 (Minitab Inc., State College,
PA, USA) was used to establish the empirical model, conduct an ana-
lysis of variance (ANOVA), and determine the optimal reaction condi-
tions.
2.6. Reusability of the enzyme
Novozym 435 was reused in interesterification with methyl acetate
and its stability was compared with that in the transesterification with
methanol reported in our previous study [11]. The reaction was carried
out under the optimal conditions determined using RSM. After the re-
action was completed, Novozym 435 was removed from the reaction
mixture through filtration. The recovered catalyst was then remixed
with fresh reactants to initiate a new reaction. In each reaction cycle,
the sample was withdrawn to determine the biodiesel yield.
2.7. Analysis
The biodiesel composition was determined using a gas
H.C. Nguyen et al. Energy Conversion and Management 158 (2018) 168–175

chromatograph (GC-2014, Shimadzu, Japan) equipped with a flame 80

ionization detector (FID) and capillary column (Stabilwax, Restek,
70
USA). Nitrogen at a flow rate of 29 mL/min was used as the mobile
phase. The temperatures of the injector and detector were set at 220 60

Biodiesel yield (%)

and 250 °C, respectively. The column was initially operated at 140 °C
for 5 min; the temperature was then increased to 240 °C at a rate of
4 °C/min and maintained at 240 °C for 15 min. Methyl pentadecanoate
was used as the internal standard with which the peak areas of fatty
acid methyl esters were compared to quantify the content of biodiesel.
The fatty acid profiles of the BSFL biodiesel were identified from re-
ference standards (Supelco 37, Sigma-Aldrich, USA). The BSFL biodiesel
yield (%) was calculated using the following formula:
Biodiesel yield (%)
peak area of biodiesel sample weight of internal standard
= x x
peak area of internal standard weight of biodiesel sample
weight of total biodiesel product
x100
weight of fat used
The biodiesel properties such as density, viscosity, sulfur content, ester
content, water content, acid value, flash point, and cetane index were
also measured according to the ASTM standard methods [44].
3. Results and discussion
3.1. Characteristics of BSFL fat
Table 1 presents the properties of BSFL fat. The fat yield extracted
from BSFL was 30.2%, indicating a high fat content. The acid value of
the BSFL fat was 11.1 mg KOH/g. This is because of the presence of free
fatty acids in BSFL fat, which was found to be higher than those in
rapeseed oil (1.14 mg KOH/g) [6]. The iodine value of BSFL fat was
76.5 g I/100 g, which is lower than that of rapeseed oil (115.5 g I/
100 g) [6], indicating that BSFL fat is more saturated. The saponifica-
tion value of BSFL was determined as 232.4 mg KOH/g. This value in-
dicated that the average molecular weight of the BSFL fat was 760.5 g/
mol. Different oil feedstocks produce biodiesel with different compo-
sitions and properties, thus affecting the production process [6]. Raw
materials that contain a smaller proportion of free fatty acids are pre-
ferred because large amounts of free fatty acids are difficult to be
converted to biodiesel [6]. These results are in agreement with previous
studies [6,10], demonstrating that BSFL fat has suitable properties for
biodiesel production.
3.2. Effect of different acyl acceptors on biodiesel production
This study investigated the effects of two acyl acceptors, methanol
and methyl acetate, on lipase-catalyzed biodiesel synthesis from BSFL
fat. As shown in Fig. 1, excess methanol caused a substantial decrease in
biodiesel yield. This might be because the lipase was markedly in-
activated by the absorption of methanol and the byproduct glycerol on
the surface of the immobilized lipase [36,45]. To overcome this ob-
stacle, methyl acetate was chosen as an alternative acyl receptor for the
reaction. The results indicated that the biodiesel yield increased sig-
nificantly when the molar ratio of methyl acetate to fat increased
(Fig. 1). This result corresponds to other studies [33–36], which have
indicated that excess methyl acetate had no negative effect on enzyme
activity and that methyl acetate is a suitable acyl acceptor for biodiesel
Table 1
Properties of crude fat extracted from BSFL.
50
40
30
20
10
0
0 2 4 6 8 10 12
Molar ratio of acyl acceptor to oil
Fig. 1. Effects of various acyl acceptors on biodiesel production from BSFL fat. The re-
action was conducted under the following conditions: a temperature of 40 °C, enzyme
loading of 4%, reaction time of 12 h, and various molar ratios of methanol to fat (■) or
methyl acetate to fat (●).
production. Therefore, methyl acetate was selected for further experi-
ments.
3.3. Development of the RSM model
In this study, the relationship between biodiesel yield (measured
response) and four reaction factors (input variables), the reaction time,
molar ratio of methyl acetate to fat, enzyme loading, and reaction
temperature, was evaluated using a Box–Behnken RSM model involving
three central replicates. The input variables and their coded and un-
coded values are presented in Table 2. To obtain the optimal reaction
conditions, the experiments were performed according to the experi-
mental design (Table 3). Table 3 shows the experimental results based
on the RSM model. The measured responses were modeled using an
empirical model. A quadratic polynomial equation in terms of the
coded variables was generated to fit the experimental responses by
using the pseudo-inverse technique, as follows:
Y = 84.83 + 15.35X1 + 8.33X2 −0.62X3−0.4X 4 −4.35X12 −7.09X22 −4.27X32
−9.49X42 + 1.51X1 X2 + 1.13X1 X3 + 0.74X1 X 4 + 4.71X2 X3−2.96X2 X 4
+ 2.46X3 X 4 (2)
where X1, X2, X1X2, X1X3, X1X4, X2X3, and X3X4 have positive influences
on the response, whereas the other parameters have adverse effects.
A relatively low coefficient of variance (1.55%) based on the central
runs (25–27) of repeated experiments indicates high precision and re-
producibility of the experiments. The statistical significance of the
model was evaluated using the F test for ANOVA (Table 4). A very low P
value (< 0.0001) of the model in the F test indicates that the regression
is statistically significant at the 95% confidence level. The quality of the
developed model was evaluated using the coefficient of determination
(R2). A high R2 value (0.97) was obtained, indicating high reliability of
the model in predicting the BSFL biodiesel yield. As shown in Fig. 2, the
model predictions and experimental values showed strong agreement,
Table 2
Coded values of the variables for the Box–Behnken design.
Variables Symbols Variable levels
−1 0 1

Fat yield Acid value Iodine value Saponification value Reaction time (h) X1 4 8 12
(%) (mg KOH/g) (gI/100 g) (mg KOH/g) Molar ratio of methyl acetate to fat X2 9 12 15
Enzyme loading (%) X3 10 15 20
30.2 ± 1.3 11.1 ± 0.5 76.5 ± 2.8 232.4 ± 7.6 Temperature (°C) X4 30 40 50

170
H.C. Nguyen et al. Energy Conversion and Management 158 (2018) 168–175

Table 3 Table 5
Box–Behnken design matrix for the influence of the four independent variables on the Significance of the coefficients in the empirical model.
reaction yield in coded values and experimental results.
Model term Parameter estimate Standard error t valuea P value
Run Variable Response, Y
β0 84.83 1.97 43.15 0.000b
X1 X2 X3 X4 β1 15.35 0.98 15.61 0.000b
β2 8.33 0.98 8.47 0.000b
1 0 −1 −1 0 66.33 β3 −0.62 0.98 −0.63 0.538
2 1 0 0 −1 86.68 β4 −0.4 0.98 −0.40 0.693
3 0 1 −1 0 78.35 β11 −4.35 1.47 −2.95 0.012b
4 0 −1 0 1 58.69 β22 −7.09 1.47 −4.81 0.000b
5 0 0 −1 −1 71.26 β33 −4.27 1.47 −2.90 0.013b
6 1 0 1 0 88.8 β44 −9.49 1.47 −6.44 0.000b
7 1 1 0 0 93.67 β12 1.51 1.70 0.89 0.393
8 −1 0 0 −1 54.49 β13 1.13 1.70 0.67 0.519
9 0 1 0 1 71.28 β14 0.74 1.70 0.43 0.673
10 −1 0 0 1 50.48 β23 4.71 1.70 2.77 0.017b
11 0 1 0 −1 79.99 β24 −2.96 1.70 −1.74 0.108
12 0 0 1 1 70.9 β34 2.46 1.70 1.45 0.174
13 1 −1 0 0 80.63
14 −1 1 0 0 58.22 a
tα/2,n-p = t0.025,12 = 2.18.
15 −1 0 1 0 60.55 b
P < .05 indicates that the model terms are significant.
16 −1 −1 0 0 51.22
17 1 0 −1 0 85.87
18 0 −1 1 0 55.82
19 −1 0 −1 0 62.15
100
20 0 1 1 0 86.67 95
21 0 −1 0 −1 55.57
22 0 0 1 −1 63.03 90

)
Biodiesel yield (%
23 1 0 0 1 85.62
24 0 0 −1 1 69.28 85
25 0 0 0 0 82.83
80
26 0 0 0 0 81.55
27 0 0 0 0 84.12 75
70
Table 4 65 20
Analysis of variance for the empirical model.
60 18

)
16

(%
Source DFb SSb MSb F value Probability (P) > F 9
Mola 10 11

ng
14

di
r rati 12 12

oa
Model a
14 4402.32 314.45 20.06 < 0.0001 o of 13
meth

el
yl ac 14 10

m
Residual (error) 12 139.17 11.6
etate 15

zy
to fa

En
Total 26 4541.49 t
a
Coefficient of determination (R2) = 0.97; adjusted R2 = 0.93. Fig. 3. Response surface plot of the combined effects of the reactant molar ratio and
b
DF, degree of freedom; SS, sum of squares; MS, mean square. enzyme loading on the biodiesel yield at a constant temperature (40 °C) and reaction time
(12 h).
100
be used to plot response surface curves and predict the optimal con-
90 ditions for maximizing responses.
Predicted yield (%)

80 3.4. Effect of reaction factors on biodiesel yield

Fig. 3 illustrates the effects of the molar ratio of methyl acetate to fat
70 and enzyme loading on the biodiesel yield, showing a significant in-
teraction. At a low molar ratio of methyl acetate to fat, increasing the
enzyme loading adversely affected biodiesel synthesis and resulted in
60
low biodiesel yield. This could be because the excess enzyme caused the
polymer beads to form aggregates where the enzymes were im-
50 mobilized and thus hindered mass transfer, limiting the flexibility of the
50 60 70 80 90 100 enzyme to react with the oil–water interface, and subsequently low-
Actual yield (%) ering the conversion yield [35,46,47]. However, at a high molar ratio of
methyl acetate to fat, an increase in enzyme loading caused a marked
Fig. 2. Correlation between experimental and fitted conversions of reaction.
increase in biodiesel yield. This is because an increase in the amount of
the enzyme increases contact between the reactants and the active
signifying that the established model provided accurate and satisfactory surface area of the enzyme, thus enhancing the reaction [47,48]. The
results. The overall effects of the input variables on the biodiesel yield maximum biodiesel yields were obtained at a high enzyme loading and
were analyzed using t tests (Table 5). Low P values (< 0.05) were high molar ratio of methyl acetate to fat.
obtained for the intercept term, two linear terms (X1 and X2), all Fig. 4 shows the effects of the temperature and molar ratio of methyl
quadratic terms, and an interaction term (X2X3), indicating that they acetate to fat on the biodiesel yield; no interaction occurred. At any
were significant factors in the reaction. The developed model can thus given temperature, the biodiesel yield significantly increased when the

171
H.C. Nguyen et al. Energy Conversion and Management 158 (2018) 168–175

method was employed to predict the optimal reaction conditions for

100 obtaining the maximum biodiesel yield. The maximal biodiesel yield
was predicted to be 98.28% at a reaction time of 12 h, a molar ratio of
95
methyl acetate to fat of 14.64:1, an enzyme loading of 17.58%, and a
)

90 reaction temperature of 39.50 °C. To verify the prediction, an experi-

Biodiesel yield (%

ment was conducted under the optimal conditions. A biodiesel yield of

85 96.97% ± 1.04% was obtained, indicating agreement with the em-
pirical model prediction. Thus, the developed RSM model can be used
80
to adequately represent the relationship between the variables and the
75 response in the enzymatic synthesis of BSFL biodiesel. In addition, the
yield is comparable with that of the conventional process [12,14,15],
70 50
suggesting that the enzymatic interesterification of BSFL fat with me-
45
thyl acetate is a promising process for producing BSFL biodiesel.

)
65

( Co
9 40
Mola 10 11

re
35

tu
r rati
o of 12 13 3.6. Reusability of Novozym 435

ra
m eth

yl ac 14
e ta t e
15
to fa
t alyst reusability is one of the most crucial factors. However, lipase ac-
Fig. 4. Response surface plot of the combined effects of the temperature and reactant
molar ratio on the biodiesel yield at a constant enzyme loading (15%) and reaction time
(12 h).
molar ratio of methyl acetate to fat was increased from 9:1 to 14:1.
However, the biodiesel yield leveled off with a further increase in the
molar ratio of methyl acetate to fat. This might be because excess
methyl acetate causes excessive dilution of the reaction medium
[31,35,36]. The results of this study differ slightly from those of other
studies, in which the optimal ratio of methyl acetate to oil was 12:1
[31,33,35]. This could be because of different oil substrates being used.
Fig. 5 presents a response surface curve obtained by plotting the
biodiesel yield against the enzyme loading and temperature. At any
enzyme loading level, the biodiesel yield increased as temperature in-
creased. However, a further increase in temperature led to a decrease in
the biodiesel yield. This is because the enzyme is deactivated by high
temperature; thus, the biodiesel yield is reduced. Several studies have
indicated that lipases are sensitive to temperature and their activity
levels are substantially decreased at low or elevated temperatures
[11,49]. In this study, Novozym 435 exhibited its highest activity at
39.5 °C.
3.5. Obtaining optimal reaction conditions
The root of the empirical model [Eq. (2)] for the reaction yield re-
presents the maximization of the response function. The canonical
100
pe
30
m
Te
To make enzymatic synthesis of BSFL biodiesel cost effective, cat-
tivity declines because of the inactivation effect of methanol, which
limits the reusability of this enzyme in transesterification reactions. In
this study, to investigate the reusability of Novozym 435, the inter-
esterification of BSFL fat with methyl acetate was conducted under the
optimal reaction conditions and its yield was compared with that of the
methanolysis reaction. As shown in Fig. 6, the biodiesel yield drama-
tically decreased when methanol was used in the reaction. This could be
because of the inactivation effect caused by the absorption of methanol
and the byproduct glycerol on the surface of the immobilized Novozym
435 [36,45]. To overcome this obstacle, the enzyme regeneration
process was used to recover the catalytic activity of Novozym 435 [11];
however, this complicated regenerative step resulted in additional
production costs. Therefore, methyl acetate appears to be an alternative
acyl acceptor that overcomes the problems associated with methanol.
As shown in Fig. 5, Novozym 435 could be effectively reused up to 20
times to drive the reaction with a biodiesel yield of 95.97%, indicating
that methyl acetate and triacetin have no negative effect on enzyme
activity. This result is similar to that of a study on biodiesel production
from soybean oil conducted by Du et al. [36], suggesting that methyl
acetate is a suitable acyl acceptor for enzyme-catalyzed biodiesel pro-
duction and that Novozym 435 can be used repeatedly to improve the
economic viability.
3.7. Fatty acid methyl ester profiles of BSFL biodiesel
The fatty acid methyl ester composition of BSFL biodiesel was
analyzed using a gas chromatograph and compared with that of rape-
seed-oil-based biodiesel. As shown in Table 6, nine fatty acid methyl
100

95
)
Biodiesel yield (%

95
Biodiesel yield (%)

90

90
85

50
80 45 85
)
(C
o

10 40
12
re

14 35
tu

Enzy
ra

me l 16
pe

oadi 18 30 80
ng (%
m

20 0 5 10 15 20
Te

)
Cycle number
Fig. 5. Response surface plot of the combined effects of the enzyme loading and tem-
perature on the reaction yield at a constant reactant molar ratio (12:1) and reaction time Fig. 6. Reusability of the enzyme in biodiesel production from BSFL fat with the use of
(12 h). methanol (●) or methyl acetate (■) as an acyl acceptor.

172
H.C. Nguyen et al. Energy Conversion and Management 158 (2018) 168–175

Table 6 kg), ester content (99.5%), acid value (< 0.8 mg KOH/g), viscosity
Fatty acid methyl ester profiles of BSFL biodiesel and rapeseed biodiesel. (5.4 mm2/s), flash point (122 °C), and cetane index (49), met the Eur-
opean standard EN 14214 [53], and most met ASTM specifications
Composition Rapeseed biodiesela BSFL biodieselb (%)
(%) [44]. Among these fuel properties, sulfur content, ester content, and
acid value are strongly related to the reaction type [54,55]. Because
Capric acid methyl ester (C10:0) nac 1.62 sulfur is absent, an additional process for removing sulfur from the
Lauric acid methyl ester (C12:0) nac 43.22
biodiesel is unnecessary in the enzyme-catalyzed process. Moreover,
Myristic acid methyl ester (C14:0) nac 6.99
Palmitic acid methyl ester (C16:0) 3.5 19.67 the higher ester content and lower acid value in the biodiesel indicate
Palmitoleic acid methyl ester nac 2.88 that the enzyme-catalyzed process is an alternative to the conventional
(C16:1) process for driving the reaction toward completion. By contrast, the
Stearic acid methyl ester (18:0) 0.8 5.52 remaining fuel properties are mainly influenced by the structure of the
Oleic acid methyl ester (C18:1) 64.4 13.03
individual fatty acid methyl esters, such as their degree of saturation
Linoleic acid methyl ester (C18:2) 22.3 2.33
Linolenic acid methyl ester 8.2 ndd and fatty acid profile. The viscosity of our BSFL biodiesel is lower than
(C18:3) of rapeseed-oil-based biodiesel [12] because this property is adversely
Nonadecanoic acid methyl ester nac 4.74 affected by a decrease in the carbon chain length of the feedstock oil
(C19:0)
[56,57]. These results suggest that BSFL biodiesel has suitable proper-
a
Data obtained from Ref. [12].
ties and can be a substitute for petrodiesel.
b
This study.
c
na = not reported. 3.9. The economic feasibility of BSFL biodiesel
d
nd = not detected.
In biodiesel production, the selection of catalyst for the reaction is
esters were detected, with lauric acid methyl ester (43.22%), palmitic critical for designing an efficient process that minimizes both waste
acid methyl ester (19.67%), oleic acid methyl ester (13.03%), and generation and energy consumption and maximizes the value of ma-
myristic acid methyl ester (8.8%) being the main ones. The composition terials [23]. Nonedible and waste oils, including insect fat, contain high
of medium-chain fatty acid methyl esters in our BSFL biodiesel was levels of free fatty acids, which lowers biodiesel production efficiency
44.84%, which is higher than those in rapeseed-oil-based, palm-oil- [23]. To enhance the biodiesel yield, biodiesel production from these
based, and soybean-oil-based biodiesels [6,7]. The relatively high level oils includes acid-catalyzed esterification to reduce the free fatty acid
of medium chain fatty acid methyl esters can reduce the viscosity of the level prior to tranesterification [10,12]. However, this acid esterifica-
BSFL biodiesel [7]. The fatty acid composition of the oil also sig- tion step is not required in enzymatic tranesterification because en-
nificantly affects other properties of the produced biodiesel, such as the zymes can simultaneously convert free fatty acids and oils into biodiesel
cetane index [8]. The more saturated the constituent fatty acid mole- [11]. Additionally, the biocatalyst used in the reaction is advantageous
cules, the higher the cetane index of the fuel [50]. The saturated fatty compared with chemical catalysts because the reaction occurs under
acid methyl ester content in the BSFL fat (81.46%) examined in this mild conditions without a negative effect on the equipment and en-
study is higher than that in rapeseed oil (4.3%), which could account vironment [11,22]. Therefore, a lipase-catalyzed process was selected
for the higher cetane index of the BSFL biodiesel. In addition, because for biodiesel production from BSFL fat.
saturated fatty acid methyl esters are more stable against oxidation than Methanol and ethanol are the most widely used alcohols as acyl
unsaturated fatty acid methyl esters [51,52], our BSFL biodiesel can be acceptors for transesterification because of their availability and low
expected to have higher oxidative stability than rapeseed-oil-based price [12,16]. However, these alcohols deactivate lipase activity, thus
biodiesel. Moreover, the number of carbons in the BSFL biodiesel limiting the industrial application of the enzymatic process [11]. To
ranged from 12 to 18, which was similar to that of fossil diesel, in- solve this obstacle, the interesterification of BSFL fat with methyl
dicating that BSFL biodiesel is a feasible substitute for petrodiesel and acetate was performed in our study. Under optimal conditions, a
BSFL is a suitable biodiesel feedstock. maximum biodiesel yield of 96.97% was obtained, which was higher
than the yield of soybean-oil-based biodiesel (92%) and olive-oil-based
3.8. Properties of BSFL biodiesel biodiesel (80%) synthesized using the same method by Du et al. [36]
and Usai et al. [38], respectively. Additionally, the Novozym 435
Table 7 shows the properties of the BSFL biodiesel determined using loading required in our study (17.58%) was lower than that of the study
ASTM standard methods compared with rapeseed-oil-based biodiesel by Du et al. (30%) [36]. Several studies have used a lower enzyme
[12]. As indicated in Table 7, most properties of our biodiesel are si- loading (6%–8%, w/w), but a longer reaction time is required (68–96 h)
milar to those of rapeseed-oil-based biodiesel. Moreover, all properties compared with the 12-h reaction time in our study [23,38]. To reduce
of our biodiesel, namely density (875 kg/m3), water content (0.03 mg/ production costs, enzyme reuse is critical in the enzymatic process.

Table 7
Fuel properties of BSFL biodiesel in comparison with rapeseed-oil-based biodiesel, the European standard EN 14214, and ASTM D6751.

Properties ASTM method ASTM D6751a EN 14214 Rapeseed-oil-basedb biodiesela This study

3 c
Density (kg/m ) D1480 na 860–900 880 875
Water content (mg/kg) D95 nac < 0.03 0.03 0.03
Sulfur content (wt.%) D5453 < 0.05 < 0.05 < 0.01 ndd
Ester content (%) D7371 nac 96.5 nac 99.5
Acid value (mg KOH/g) D664 < 0.5 < 0.8 0.31 < 0.8
Viscosity at 40 °C (mm2/s) D445 1.9–6.0 1.9–6.0 6.35 5.4
Flash point (closed cup) (°C) D93 > 130 > 120 nac 122
Cetane index D613 > 47 48–60 45 49

a
Data obtained from Ref. [43].
b
Data were obtained from Ref. [12].
c
na = not reported.
d
nd = not detected.

173
H.C. Nguyen et al.
Novozym 435 was reused up to 20 times without loss of enzyme activity
in our study. This result corresponds to that presented for the inter-
esterification reaction of vegetable oils with methyl acetate [36,39],
indicating that the lipase-interesterification of BSFL fat with methyl
acetate is a promising process for producing biodiesel.
The economic feasibility of using immobilized enzymes for produ-
cing fine chemicals has been evaluated in previous studies. The product
yield when using immobilized enzymes as a catalyst should be in the
range of 2–10,000 kg/kg of biocatalyst [58,59]. In this study, after 20
reuses of Novozym 435, 110 kg of biodiesel was produced using 1 kg of
immobilized enzyme. The achieved yield, which is in the range re-
quired, indicated that the process is economically feasible. The enzy-
matic process for biodiesel production using methanol as the acyl ac-
ceptor has been demonstrated to be economically feasible for industrial
applications [11]. Compared with that process, biodiesel production in
the present study was considered cheaper because less immobilized
enzyme was used and no enzyme regeneration step was required. In
addition, the high price of biodiesel is due to the cost of feedstock,
which comprises 75% of the total production cost [6,7]. Edible oils,
such as rapeseed, soybean, and palm oil, are commonly used for pro-
ducing biodiesel in Europe, the United States, and Southeast Asia, re-
spectively, resulting in an approximately 1.5-fold increase in the pro-
duction cost compared with petrodiesel [6,60]. BSFL fat is a nonedible
oil that was demonstrated to be a promising and cheap biodiesel
feedstock without requiring arable land or competition with human
food [14]. In addition, BSFL fat can quickly be produced from various
types of organic waste with a high yield (30.2%), and BSFL use can
solve challenges encountered in organic waste management [11].
Therefore, BSFL fat use can reduce biodiesel production costs, and en-
zyme-catalyzed interesterification of BSFL fat is an economically fea-
sible process for biodiesel production.
4. Conclusions
In this paper, we report the enzymatic interesterification of BSFL fat
with methyl acetate as an acyl acceptor for biodiesel production.
Compared with methanol, methyl acetate had no negative effect on
lipase activity and increased the biodiesel yield. The reaction conditions
were subsequently optimized for maximizing the biodiesel yield using
RSM. A maximal biodiesel yield of 96.97% was obtained and verified.
Under optimal conditions, Novozym 435 could be reused for 20 cycles,
yielding a 95.97% biodiesel yield. The properties of the BSFL biodiesel
were also studied and met the specifications of European standard EN
14214. The results suggest that the enzymatic interesterification of
BSFL fat with methyl acetate is a promising process for biodiesel pro-
duction and that the synthesized BSFL biodiesel can be a substitute for
petrodiesel.
References
[1] Tran DT, Chang JS, Lee DJ. Recent insights into continuous-flow biodiesel pro-
duction via catalytic and non-catalytic transesterification processes. Appl Energy
2017;185:376–409.
[2] Tran DT, Chen CL, Chang JS. Continuous biodiesel conversion via enzymatic
transesterification catalyzed by immobilized Burkholderia lipase in a packed-bed
bioreactor. Appl Energy 2016;168:340–50.
[3] Guan Q, Shang H, Liu J, Gu J, Li B, Miao R, et al. Biodiesel from transesterification
at low temperature by AlCl3 catalysis in ethanol and carbon dioxide as cosolvent:
process, mechanism and application. Appl Energy 2016;164:380–6.
[4] Mosarof MH, Kalam MA, Masjuki HH, Alabdulkarem A, Ashraful AM, Arslan A, et al.
Optimization of performance, emission, friction and wear characteristics of palm
and Calophyllum inophyllum biodiesel blends. Energy Convers Manage
2016;118:119–34.
[5] Kakati J, Gogoi TK. Biodiesel production from Kutkura (Meyna spinosa Roxb. Ex.)
fruit seed oil: its characterization and engine performance evaluation with 10% and
20% blends. Energy Convers Manage 2016;121:152–61.
[6] Zheng L, Hou Y, Li W, Yang S, Li Q, Yu Z. Exploring the potential of grease from
yellow mealworm beetle (Tenebrio molitor) as a novel biodiesel feedstock. Appl
Energy 2013;101:618–21.
[7] Canakci M, Sanli H. Biodiesel production from various feedstocks and their effects
174
Energy Conversion and Management 158 (2018) 168–175
on the fuel properties. J Ind Microbiol Biotechnol 2008;35:431–41.
[8] Surendra K, Olivier R, Tomberlin JK, Jha R, Khanal SK. Bioconversion of organic
wastes into biodiesel and animal feed via insect farming. Renew Energy
2016;98:197–202.
[9] Bowling JJ, Anderson JB, Armbrust KL, Hamann MT. Evaluation of potential bio-
diesel feedstock production from oleaginous insect Solenopsis sp. Fuel 2014;117:5–7.
[10] Li Q, Zheng L, Qiu N, Cai H, Tomberlin JK, Yu Z. Bioconversion of dairy manure by
black soldier fly (Diptera: Stratiomyidae) for biodiesel and sugar production. Waste
Manage 2011;31:1316–20.
[11] Nguyen HC, Liang SH, Doan TT, Su CH, Yang PC. Lipase-catalyzed synthesis of
biodiesel from black soldier fly (Hermetica illucens): optimization by using response
surface methodology. Energy Convers Manage 2017;145:335–42.
[12] Li Q, Zheng L, Cai H, Garza E, Yu Z, Zhou S. From organic waste to biodiesel: black
soldier fly, Hermetia illucens, makes it feasible. Fuel 2011;90:1545–8.
[13] Zheng L, Hou Y, Li W, Yang S, Li Q, Yu Z. Biodiesel production from rice straw and
restaurant waste employing black soldier fly assisted by microbes. Energy
2012;47:225–9.
[14] Li W, Li Q, Zheng L, Wang Y, Zhang J, Yu Z, et al. Potential biodiesel and biogas
production from corncob by anaerobic fermentation and black soldier fly. Biores
Technol 2015;194:276–82.
[15] Zheng L, Li Q, Zhang J, Yu Z. Double the biodiesel yield: rearing black soldier fly
larvae, Hermetia illucens, on solid residual fraction of restaurant waste after grease
extraction for biodiesel production. Renew Energy 2012;41:75–9.
[16] Alavijeh RS, Tabandeh F, Tavakoli O, Karkhane A, Shariati P. Enzymatic production
of biodiesel from microalgal oil using ethyl acetate as an acyl acceptor. J Oleo Sci
2015;64:69–74.
[17] Campanelli P, Banchero M, Manna L. Synthesis of biodiesel from edible, non-edible
and waste cooking oils via supercritical methyl acetate transesterification. Fuel
2010;89:3675–82.
[18] Minami E, Saka S. Kinetics of hydrolysis and methyl esterification for biodiesel
production in two-step supercritical methanol process. Fuel 2006;85:2479–83.
[19] Saka S, Isayama Y. A new process for catalyst-free production of biodiesel using
supercritical methyl acetate. Fuel 2009;88:1307–13.
[20] Imahara H, Minami E, Hari S, Saka S. Thermal stability of biodiesel in supercritical
methanol. Fuel 2008;87:1–6.
[21] García-Martínez N, Andreo-Martínez P, Quesada-Medina J, de los Ríos AP, Chica A,
Beneito-Ruiz R, et al. Optimization of non-catalytic transesterification of tobacco
(Nicotiana tabacum) seed oil using supercritical methanol to biodiesel production.
Energy Convers Manage 2017;131:99–108.
[22] Amini Z, Ilham Z, Ong HC, Mazaheri H, Chen WH. State of the art and prospective
of lipase-catalyzed transesterification reaction for biodiesel production. Energy
Convers Manage 2017;141:339–53.
[23] Amini Z, Ong HC, Harrison MD, Kusumo F, Mazaheri H, Ilham Z. Biodiesel pro-
duction by lipase-catalyzed transesterification of Ocimum basilicum L. (sweet basil)
seed oil. Energy Convers Manage 2017;132:82–90.
[24] Ungcharoenwiwat P, Canyuk B, Aran H. Synthesis of jatropha oil based wax esters
using an immobilized lipase from Burkholderia sp. EQ3 and Lipozyme RM IM.
Process Biochem 2016;51:392–8.
[25] Royon D, Daz M, Ellenrieder G, Locatelli S. Enzymatic production of biodiesel from
cotton seed oil using t-butanol as a solvent. Biores Technol 2007;98:648–53.
[26] Shimada Y, Watanabe Y, Sugihara A, Tominaga Y. Enzymatic alcoholysis for bio-
diesel fuel production and application of the reaction to oil processing. J Mol Catal
B Enzym 2002;17:133–42.
[27] Modi MK, Reddy JR, Rao BV, Prasad RB. Lipase-mediated conversion of vegetable
oils into biodiesel using ethyl acetate as acyl acceptor. Biores Technol
2007;98:1260–4.
[28] Gog A, Roman M, Toşa M, Paizs C, Irimie FD. Biodiesel production using enzymatic
transesterification–current state and perspectives. Renew Energy 2012;39:10–6.
[29] Tan KT, Lee KT, Mohamed AR. A glycerol-free process to produce biodiesel by
supercritical methyl acetate technology: an optimization study via response surface
methodology. Biores Technol 2010;101:965–9.
[30] Casas A, Ramos MJ, Perez A. New trends in biodiesel production: chemical inter-
esterification of sunflower oil with methyl acetate. Biomass Bioenergy
2011;35:1702–9.
[31] Maddikeri GL, Pandit AB, Gogate PR. Ultrasound assisted interesterification of
waste cooking oil and methyl acetate for biodiesel and triacetin production. Fuel
Process Technol 2013;116:241–9.
[32] Casas A, Ramos MJ, Pérez Á. Kinetics of chemical interesterification of sunflower oil
with methyl acetate for biodiesel and triacetin production. Chem Eng J
2011;171:1324–32.
[33] Subhedar PB, Gogate PR. Ultrasound assisted intensification of biodiesel production
using enzymatic interesterification. Ultrason Sonochem 2016;29:67–75.
[34] Xu Y, Du W, Liu D. Study on the kinetics of enzymatic interesterification of tri-
glycerides for biodiesel production with methyl acetate as the acyl acceptor. J Mol
Catal B Enzym 2005;32:241–5.
[35] Razack SA, Duraiarasan S. Response surface methodology assisted biodiesel pro-
duction from waste cooking oil using encapsulated mixed enzyme. Waste Manage
2016;47:98–104.
[36] Du W, Xu Y, Liu D, Zeng J. Comparative study on lipase-catalyzed transformation of
soybean oil for biodiesel production with different acyl acceptors. J Mol Catal B
Enzym 2004;30:125–9.
[37] Goembira F, Saka S. Optimization of biodiesel production by supercritical methyl
acetate. Biores Technol 2013;131:47–52.
[38] Usai EM, Gualdi E, Solinas V, Battistel E. Simultaneous enzymatic synthesis of
FAME and triacetyl glycerol from triglycerides and methyl acetate. Biores Technol
2010;101:7707–12.
H.C. Nguyen et al.
[39] Ognjanovic N, Bezbradica D, Knezevic-Jugovic Z. Enzymatic conversion of sun-
flower oil to biodiesel in a solvent-free system: process optimization and the im-
mobilized system stability. Biores Technol 2009;100:5146–54.
[40] Su CH, Fu CC, Gomes J, Chu I, Wu WT. A heterogeneous acid-catalyzed process for
biodiesel production from enzyme hydrolyzed fatty acids. AIChE J 2008;54:327–36.
[41] Vicente G, Martınez M, Aracil J. Integrated biodiesel production: a comparison of
different homogeneous catalysts systems. Biores Technol 2004;92:297–305.
[42] Casas A, Ruiz JR, Ramos MJ, Pérez A. Effects of triacetin on biodiesel quality.
Energy Fuels 2010;24:4481–9.
[43] Arteaga G, Li-Chan E, Vazquez-Arteaga M, Nakai S. Systematic experimental de-
signs for product formula optimization. Trends Food Sci Technol 1994;5:243–54.
[44] American standards for testing of materials (ASTM). D1480, D95, D5453, D7371,
D664, D445, D93, D613, D6751. St. Joseph, MI, ASAE; 2003.
[45] Chen JW, Wu WT. Regeneration of immobilized Candida antarctica lipase for
transesterification. J Biosci Bioeng 2003;95:466–9.
[46] Calero J, Verdugo C, Luna D, Sancho ED, Luna C, Posadillo A, et al. Selective
ethanolysis of sunflower oil with Lipozyme RM IM, an immobilized Rhizomucor
miehei lipase, to obtain a biodiesel-like biofuel, which avoids glycerol production
through the monoglyceride formation. New Biotechnol 2014;31:596–601.
[47] Waghmare GV, Rathod VK. Ultrasound assisted enzyme catalyzed hydrolysis of
waste cooking oil under solvent free condition. Ultrason Sonochem 2016;32:60–7.
[48] Wu H, Liu Y, Zhang J, Li G. In situ reactive extraction of cottonseeds with methyl
acetate for biodiesel production using magnetic solid acid catalysts. Biores Technol
2014;174:182–9.
[49] Köse Ö, Tüter M, Aksoy HA. Immobilized Candida antarctica lipase-catalyzed al-
coholysis of cotton seed oil in a solvent-free medium. Biores Technol
2002;83:125–9.
175
Energy Conversion and Management 158 (2018) 168–175
[50] Yang S, Li Q, Gao Y, Zheng L, Liu Z. Biodiesel production from swine manure via
housefly larvae (Musca domestica L.). Renew Energy 2014;66:222–7.
[51] Knothe G. ‘‘Designer” biodiesel: optimizing fatty ester composition to improve fuel
properties. Energy Fuel 2008;22:1358–64.
[52] Ramos MJ, Fernández CM, Casas A, Rodríguez L, Pérez Á. Influence of fatty acid
composition of raw materials on biodiesel properties. Biores Technol
2009;100:261–8.
[53] European Standard, EN 14214. Automotive fuels-fatty acid methyl esters (FAME)
for diesel engines-requirements and test methods; 2008.
[54] Feng Y, Zhang A, Li J, He B. A continuous process for biodiesel production in a fixed
bed reactor packed with cation-exchange resin as heterogeneous catalyst. Biores
Technol 2011;102:3607–9.
[55] Nielsen PM, Rancke-Madsen A, Holm HC, Burton R. Production of biodiesel using
liquid lipase formulations. J Am Oil Chem Soc 2016;93:905–10.
[56] Shu Q, Yang B, Yang J, Qing S. Predicting the viscosity of biodiesel fuels based on
the mixture topological index method. Fuel 2007;86:1849–54.
[57] Yang S, Li Q, Zeng Q, Zhang J, Yu Z, Liu Z. Conversion of solid organic wastes into
oil via Boettcherisca peregrine (Diptera: Sarcophagidae) larvae and optimization of
parameters for biodiesel production. PLoS One 2012;7:e45940.
[58] Nordblad M, Pedersen AK, Rancke-Madsen A, Woodley JM. Enzymatic pretreatment
of low-grade oils for biodiesel production. Biotechnol Bioeng 2016;113:754–60.
[59] Tufvesson P, Lima-Ramos J, Nordblad M, Woodley JM. Guidelines and cost analysis
for catalyst production in biocatalytic processes. Org Process Res Dev
2010;15:266–74.
[60] Mardhiah HH, Ong HC, Masjuki HH, Lim S, Lee HV. A review on latest develop-
ments and future prospects of heterogeneous catalyst in biodiesel production from
non-edible oils. Renew Sustain Energy Rev 2017;67:1225–36.