Energy Conversion and Management: X 16 (2022) 100302

Contents lists available at ScienceDirect

Energy Conversion and Management: X

journal homepage: www.sciencedirect.com/journal/energy-conversion-and-management-x

Coal fly ash supported ZnO catalyzed transesterification of Jatropha curcas

oil: Optimization by response surface methodology
Adeyinka Sikiru Yusuff a, *, Lekan Taofeek Popoola a, David Olalekan Adeniyi b,
Moses Aderemi Olutoye b
a
Department of Chemical and Petroleum Engineering, College of Engineering, Afe Babalola University, Ado-Ekiti, Nigeria
b
Department of Chemical Engineering, School of Infrastructure, Process Engineering and Technology, Federal University of Technology, Minna, Niger State, Nigeria

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

Keywords: Herein, a zinc oxide based coal fly ash (ZnO-CFA) composite as heterogeneous catalyst was formulated and
Biodiesel explored for transesterification of non-edible oil (Jatropha curcas oil, JCO). The as-synthesized catalyst was
ZnO analyzed to gain insights into its properties using various characterization techniques (EDX, FTIR, TEM, XRD and
Fly ash
BET). The solid catalyst’s ability to catalyze the methanolysis reaction was investigated and optimized at varying
Transesterification
reaction temperature, methanol/JCO molar ratio and catalyst dosage using Box-Behnken design. Predicted
Catalyst (
Optimization values of biodiesel yield were found to be in agreement with the experimental values obtained R2 =
2
)
0.9751 and Adj.R = 0.9431 . It was determined that optimum transesterification process conditions of 60.4 ◦ C
reaction temperature, 11.8:1 methanol/JCO molar ratio and 1.63 wt% catalyst loading resulted in biodiesel yield
and FAME content of 91.08 ± 0.06 % and 97.22 %, respectively. Additionally, the produced biodiesel at the
optimum conditions was shown to be conformed to ASTM standard. The ZnO-CFA catalyst was simple to syn­
thesize and handle, eliminating the need for costly aluminosilicate-based catalysts in biofuel synthesis. It was
also simple to separate from the product stream and could be recycled up to four times.

1. Introduction initiate the saponification reaction, which results in the formation of

soap [4,6]. Thus, to avoid soap formation and enhance product yield, a
The quest for sources of renewable fuel to meet energy consumptions feedstock with a high level of FFA, such as Jatropha curcas oil (JCO), that
has become increasingly more important due to the global warming cannot be directly converted to biodiesel via a homogeneously or het­
concern. The crude oil reserve is depleting at an alarming rate, neces­ erogeneously catalyzed transesterification process must be pretreated.
sitating significant effort to identify suitable and sustainable energy An extensive information regarding the merits and demerits of homo­
sources to replace fossil fuel. Biodiesel is recognized as one of the po­ geneously and heterogeneously catalyzed transesterification of feed­
tential fuel substitutes derived from renewable feedstocks (plant oils and stock with high FFA content has been compiled by Elgarbawy et al. [7].
animal fat), with significant potential for use as fuel in diesel engines The use of solid (heterogeneous) catalysts in the esterification of
[1,2]. Biodiesels have numerous advantages, including the ability to organic acid (fatty acid) and transesterification of triglycerides with
reduce greenhouse gas emissions, biodegradability, eco-friendliness and alcohols has dominated the global scene in recent years. These reactions
improved lubricity [3,4]. are frequently catalyzed by basic or acidic catalysts. For example, an
For the alcoholysis process, two types of catalysts are commonly eggshell-based heterogeneous catalyst was used in the methanolysis of
used: homogeneous and heterogeneous [5]. Homogeneous catalyst is canola oil [5], a barium-modified montmorillonite K10 (BMK10) cata­
more advantageous than other catalysts because it is less expensive, lyst was used in the transesterification of waste cooking oil (WCO) [1], a
provides a higher yield and allows for a faster reaction rate [1]. How­ pristine organoclay catalyst was used in the production of biodiesel from
ever, liquid catalyst has the disadvantage of being sensitive to moisture JCO [8] and CaO-modified zeolite catalysts were used in catalyzing
and free fatty acid (FFA) in feedstock. For example, if the FFA content of waste lard transesterification [9]. Furthermore, it has been reported that
the oil or fat exceeds 3.0 wt%, the homogeneous base catalyst will biodiesel can be produced by transesterifying JCO with methanol using

* Corresponding author.
E-mail address: [email protected] (A.S. Yusuff).

https://doi.org/10.1016/j.ecmx.2022.100302
Received 26 July 2022; Received in revised form 18 September 2022; Accepted 25 September 2022
Available online 27 September 2022
2590-1745/© 2022 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-
nc-nd/4.0/).
A.S. Yusuff et al. Energy Conversion and Management: X 16 (2022) 100302

an aluminosilicate-based catalyst (CaO/fly ash). At 60 ◦ C for 1 h with a Table 1

methanol/JCO molar ratio of 12:1 and 0.5 wt% catalyst loading, the Studied process factors and their levels.
highest biodiesel yield (94.72 %) and FAME conversion (97.33 %) were Level
obtained [10]. Under optimum esterification conditions, oleic acid
Symbol Parameter low middle high
esterification over the sulfonated eucalyptus bark biochar catalyst T Temperature (◦ C) 60 65 70
resulted in high oleic acid conversion (97.05 %) [11]. Palm oil mill fly M Methanol/JCO molar ratio 8:1 10:1 12:1
ash supported CaO was utilized as a heterogeneous base catalyst for the C Catalyst loading (wt.%) 1.0 1.5 2.0
transesterification of crude palm oil (CPO) with methanol to produce
FAME, and 97.09 % CPO conversion was achieved [12].
2.2. ZnO-CFA catalyst preparation
Fly ash is an industrial byproduct produced by the combustion of
coal in thermal power plants to generate electricity. The use of fly ash as
Fly ash was activated via chemical activation in order to remove
a catalyst to catalyze reaction processes is one approach to solving a
carbon residue and 5.5 g of activated CFA was mixed with 4.5 g of Zn
solid waste disposal problem. Furthermore, fly ash is nontoxic, inex­
(NO3)2⋅6H2O The resulting solid mixture was then suspended in 50 mL
pensive, chemically active, and reusable, as well as suitable for the in­
of distilled water, stirred on a magnetic stirrer at 70 ◦ C for 12 h and dried
dustrial synthesis of silica gel [10,12,13]. To modify fly ash with basic
at 110 ◦ C overnight. Finally, zinc nitrate loaded CFA was calcined in a
oxides, coal fly ash supported ZnO was synthesized for biodiesel pro­
muffle furnace at different temperatures (300, 400 and 500 ◦ C) for 4.5 h.
duction. Previous research found that unmodified fly ash has poor
The as-prepared catalysts were henceforth denoted as ZnO-CFAT, where
textural properties, is low in activity and produces passive sites [14].
T denoted calcination temperature value.
Thus, improving fly ash activity through modification is critical to
enrich textural characteristics and avoid the drawbacks of industrial by-
products due to their low surface area. 2.3. ZnO-CFA catalyst characterization
When compared to the one-factor-at-a-time technique, which results
in an unreliable number of experiments, the use of statistical experi­ Fourier transform infrared (FTIR) spectrometer (Perkin Elmer-
mental design techniques in the development of the transesterification Spectrum TWO ATR 2015, USA), Micromeritics surface area and
process can result in reduced process variability as well as the require­ porosity analyzer (Model ASAP 2010, USA), X-ray diffractometer (A8
ment of fewer resources (time, raw reagents and experimental work) Advance, Bruker AXS Gmbh, Germany 2010), high-resolution trans­
[15,16]. The response surface methodology (RSM) is an example of mission electron microscope (HR-TEM, JEOL-JEM 2010) were used to
experimental design methodology. RSM is a set of statistical and evaluate the surface functional groups, textural properties, phases and
mathematical techniques that can be applied to process analysis, opti­ crystallographic structures and surface morphology and elemental
mization and modeling [17]. With the use of RSM, estimation of linear, compositions of the activated CFA and ZnO-CFA catalyst, respectively.
interaction and quadratic effects of the process variables and prediction
model for the process output (response) are possible. The experimental
data required are dependent on the selected design: central composite or 2.4. Catalytic activity during methanolysis of JCO
Box-Behnken design [18]. These are different in the number of runs
required and in the combinations of the levels used in the experiment. Due to the high FFA content in the feedstock, a two-step trans­
The Box-Behnken design (BBD) is thought to be more capable and esterification process, i.e., acid esterification followed by base trans­
powerful than the central composite design (CCD) because it requires esterification, was used in this current work to synthesize biodiesel. The
fewer experiments and has been shown to be adequate in describing the acid esterification was carried out in a two-neck round bottom flask
majority of steady-state process responses [15,19]. equipped with a condenser and thermometer for temperature mea­
This work is particularly significant because it adds value to the surement, using 50 mL of JCO, 25 mL of methanol and 0.1 mL of
waste stream of coal fly ash by using it as a catalyst support, and it re­ concentrated H2SO4 for 1.5 h at 60 ◦ C. Excess methanol, H2SO4 and
duces raw material costs by using JCO (non-edible oil) to synthesize water were removed from the product mixtures after the reaction was
biodiesel. Because of the easy availability of both feedstock and catalyst completed. The acid value of the JCO after esterification reaction was
support, the current study focused on the development and character­ determined immediately using the titration method, and it was found to
ization of ZnO loaded coal fly ash (ZnO-CFA) as a heterogeneous base have decreased from 11.16 mg KOH/g to 1.09 mg KOH/g. Following
catalyst for the methanolysis of JCO. The Box-Behken design of response that, the esterified product was transesterified over a ZnO-CFA catalyst
surface methodology was used to optimize transesterification process in the same experimental setup.
parameters such as temperature, methanol/oil molar ratio and catalyst For all the reactions conducted, the temperature was monitored by
dosage. inserting the flask in silicon oil bowl. The required amounts of methanol,
esterified JCO and ZnO-CFA catalyst were all poured into the flask and
2. Materials and methods the reaction process started as soon as the temperature of the mixture
attained a desired value. In addition, the reaction mixtures were stirred
2.1. Materials at 400 rpm to prevent mass transfer limitation. The different parameters
of the methanolysis process were optimized in the given level for tem­
Coal fly ash (CFA) was collected from a coal power plant station, perature, methanol/JCO molar ratio and catalyst loading using response
Lucknow, India. The chemical composition of the CFA, as determined surface methodology (see Table 1). After the completion of reaction, the
via X-ray fluorescence analysis, were 30.2 % Al2O3, 56.8 % SiO2, 4.9 % reaction broth was centrifuged to remove spent catalyst and then left for
Fe2O3, 0.2 % SO3, 2.1 % other metal oxides and 5.1 % loss on ignition a while for further separation. The unused methanol was removed from
[10]. All chemical compounds used, including zinc nitrate hexahydrate synthesized biodiesel using rotary evaporator. Each experiment was
(Zn(NO3)2⋅6H2O), n-hexane (99 %), methanol (99.8 %), concentrated duplicated under the same conditions. The biodiesel yield (Y) was then
sulfuric acid (H2SO4, 98 %), potassium hydroxide (KOH), ethanol (95 calculated using Eq. (1).
%), phenolphathalein and diethyl ether were purchased from Sigma- Mbiodiesel
Aldrich Chemicals (private) limited, Mumbai, India. Information Y= × 100% (1)
MJCO
regarding the extraction and characteristics of JCO used herein has
already been reported [20]. where Mbiodiesel and MJCO are masses of biodiesel and JCO used,
respectively.

2
A.S. Yusuff et al. Energy Conversion and Management: X 16 (2022) 100302

Table 2 the composite catalyst.

Percentage composition of CFA and calcined ZnO-CFA catalyst at different
temperatures. 3.1.1. EDX analysis
Sample Composition (%) Calcination temperature is an important factor to consider when
Si Ca Mg O Al Fe C Ti Zn
formulating heterogeneous (solid) catalysts. Table 2 depicts the
elemental composition of ZnO-CFA at various calcination temperatures
ZnO-CFA300 34.6 0.4 45.8 11.8 1.1 1.1 0.8 1.7
–
and activated CFA. The EDX results revealed that the ZnO-CFA catalysts
ZnO-CFA400 25.4 1.0 0.5 43.4 14.7 4.5 – 1.3 7.7
ZnO-CFA500 23.8 – – 46.2 17.4 6.4 – 1.1 4.1 were dominated by elements such as O, Si, Al, Zn and Fe, with lesser
Activated CFA 22.8 1.1 1.3 48.1 18.4 5.7 1.9 0.2 – amounts of other elements. When compared to other calcination tem­
peratures studied, ZnO-CFA showed the highest percent mass fraction of
ZnO at 400 ◦ C, indicating complete degradation of Zn(NO3)2. In addi­
2.4.1. Analysis of produced JCO biodiesel tion, calcined ZnO-CFA sample at 400 ◦ C contained significant amounts
The fuel properties of the synthesized biodiesel such as specific of Mg and Ca. The higher levels of Mg and Ca in ZnO-CFA400 samples
gravity, kinematic viscosity, flash point, cloud and pour point were compared to ZnO-CFA300 and ZnO-CFA500 samples suggested an
evaluated in accordance with the American Society for Testing and important role of alkaline earth metals as active components, such that
Materials (ASTM) procedures. FTIR spectrometer (Perkin Elmer, USA) the catalytic activity was improved in the methanolysis process.
was used to examine the functional groups in the synthesized biodiesel, As shown in Table 2, the carbonaceous material in the composite
with spectra recorded from 4000 to 500 cm− 1. In addition, the fatty acid catalyst completely decomposed at temperatures above 300 ◦ C, with the
methyl ester content of JCO biodiesel was determined using a GC carbon content decreasing to 0.0 % in the ZnO-CFA400 and ZnO-
(Agilent 7890A, USA) equipped with a flame ionization detector and a CFA500 samples. Based on these findings, it could be assumed that the
capillary column (J & W DB-5HT), as previously reported [11]. calcined Zn(NO3)2 loaded fly ash catalyst was rich in basic oxides such as
SiO2, Al2O3, ZnO, Fe2O3, TiO2 and CaO, indicating that calcination at
2.5. Optimization studies higher temperature transformed the raw catalyst sample to a mixed
oxides, as reported by Olutoye et al. [1]. The obtained biodiesel yields
For the methanolysis of esterified JCO, the Box-Behnken design for activated CFA, ZnO-CFA300, ZnO-CFA400 and ZnO-CFA500 cata­
(BBD), which is embedded in RSM, was applied. A total number of 17 lysts were 5.7 %, 46.2 %, 87.5 % and 68.1 %, respectively, using acid
experimental runs have been designed using Design-Expert 12.0.0 to esterified JCO with a methanol/oil molar ratio of 10:1, catalyst dosage
optimize the operating parameters. The transesterification process var­ of 1.0 wt.% and reaction time of 2 h at 60 ◦ C. As a result, an optimum
iables such as temperature (T), methanol/JCO molar ratio (M) and value of 400 ◦ C was chosen as the best calcination temperature, and a
catalyst loading (C) have been studied in the present work and the significant amount of Zn(NO3)2 loaded coal fly ash composite was syn­
biodiesel yield (Y) was considered as a dependent parameter. The range thesized and calcined at 400 ◦ C.
and levels of operating parameters investigated in the current work are
presented in Table 1. 3.1.2. TEM analysis
The internal morphologies of the CFA and ZnO-CFA400 are depicted
3. Results and discussion in Fig. 1. The results (Fig. 1a) revealed that the CFA sample was
composed of spherical particles, which was consistent with previous
3.1. Characterization of the ZnO-CFA catalyst research [12,21]. However, after loading Zn ions on the catalyst support,
a change in the morphological structure was observed due to active
The fly ash and Zn-containing fly ash catalyst were analyzed for their ingredient dispersion on the fly ash, as revealed by the TEM pattern of
efficacy by energy dispersive X-ray (EDX) spectroscopy, transmission the composite catalyst. Some particles seen around the spherical struc­
electron microscopy (TEM), Fourier transform infrared (FTIR) spec­ ture (see Fig. 1b) were thought to be ZnO deposits on the CFA surface.
troscopy, Brunauer-Emmett-Teller (BET) surface area and X-ray
diffraction (XRD). The analysis data were in line with the hypothesis
that the ZnO-CFA was more active as a result of the presence of ZnO in

Fig. 1. TEM images of (a) CFA and ZnO-CFA400 catalyst.

3
A.S. Yusuff et al. Energy Conversion and Management: X 16 (2022) 100302

Table 3
Textural properties of CFA and ZnO-CFA400 catalyst.
Sample specific surface area Pore volume Average pore
(m2/g) (cm3/g) diameter (Å)

CFA 1.7 0.004 82.0

ZnO- 58.8 0.230 80.0
CFA400

Table 4
Box-Behnken Design Matrix and the values of response function.
Run Transesterification process parameters Biodiesel yield

Temperature Methanol/ Catalyst Experimental Predicted

(◦ C) JCO molar loading value (%) value (%)
ratio (wt.%)

1 65 10:1 1.5 75.1 ± 0.28 73.40

2 65 10:1 1.5 72.77 ± 0.47 73.40
3 60 12:1 1.5 83.36 ± 1.03 83.39
4 60 10:1 1.0 74.57 ± 1.88 73.77
5 60 10:1 2.0 82.62 ± 0.86 82.26
6 65 12:1 1.0 68.56 ± 0.81 69.33
7 60 8:1 1.5 80.71 ± 0.70 81.84
8 70 10:1 2.0 71.77 ± 0.37 72.57
Fig. 2. FTIR spectra of (a) activated CFA (b) fresh ZnO-CFA400 catalyst and (c)
9 65 10:1 1.5 73.62 ± 0.17 73.40
reused ZnO-CFA400 catalyst. 10 65 10:1 1.5 73.81 ± 1.27 73.40
11 65 8:1 2.0 69.64 ± 1.60 68.87
12 70 8:1 1.5 69.14 ± 1.32 69.11
13 65 8:1 1.0 66.34 ± 0.79 66.01
14 65 12:1 2.0 78.74 ± 1.72 79.08
15 70 12:1 1.5 82.23 ± 1.31 81.10
16 70 10:1 1.0 68.08 ± 0.26 68.44
17 65 10:1 1.5 71.72 ± 1.92 73.40

However, upon the utilization of the catalyst in transesterification re­

action, its FTIR spectrum (Fig. 2c) changed due to interaction with re­
actants. The peaks detected at 2926–2966 cm− 1 and 1717–1788 cm− 1
were corresponded to –C–H stretching of methyl group and C– –O
stretching vibration of ester group, respectively, thus indicating the
coverage of the surface of the spent catalyst by methyl esters. It should
be noted that these peaks were quite intense, which can be attributed to
the loss of active sites [10].

3.1.4. XRD analysis

The X-ray diffractograms of the CFA and ZnO-CFA400 catalyst,
shown in Fig. 3, revealed that the coal fly ash contained SiO2 (at 2θ =
20.6◦ , 26.4◦ , 39.2◦ , 50.0◦ , 59.7◦ , and 67.6◦ ), CaSiO2 (at 2 θ = 33.4◦ ) and
Ca2AlSiO2 (at 2 θ = 42.5◦ ). These XRD results were consistent with the
XRD pattern of fly ash from a palm oil mill [12]. However, after
impregnation of ZnO on CFA and calcination at elevated temperature,
Fig. 3. XRD patterns of (a) CFA and (b) ZnO-CFA400 catalyst. The corre­
new peaks were observed. Furthermore, the XRD pattern of ZnO-
sponding crystalline phases are labeled as: w – SiO2; x – CaSiO2; y- Ca2AlSiO2; z
CFA400 (see Fig. 3b) revealed ZnO peaks at 31.6◦ , 34.3◦ , 36.1◦ and
– ZnO; k - ZnFe2O4.
47.6◦ , indicating that the incorporation of Zn ions into the fly ash
structure improved crystallization. A peak (56.8◦ ) attributed to ZnFe2O4
3.1.3. FTIR analysis
was also detected, possibly due to the combination of ZnO and fly ash.
The FTIR spectra of CFA, calcined ZnO-CFA400 and reused ZnO-
CFA400 catalysts are displayed in Fig. 2. Broad bands at 3509 cm− 1
3.1.5. Surface area analysis
detected in the three samples were attributed to the hydroxyl (O–H)
The textural features of the fly ash and the composite catalyst are
stretching vibration, indicating the adsorption of moisture onto the
listed in Table 3. The obtained surface area and pore size in this study
surfaces of the samples [22]. The spectrum of the fresh ZnO-CFA400
showed an improvement in the textural properties of the resulting
catalyst revealed a unique peak at 2408 cm− 1 which could be assigned
catalyst. This could be due to the formation of pores following zinc
to asymmetric stretching of C–H. The CFA and fresh ZnO-CFA400
insertion, which resulted in an increase in the surface area of the catalyst
catalyst exhibited series of similar bands which included 1669–1676
[10]. As shown in Table 3, the surface area increased from 1.7 m2/g to
cm− 1 (H–O–H deformation vibration), 1121–1123 cm− 1 (Si-O-Si
58.8 m2/g, indicating that loading Zn ions onto CFA had a significant
stretching vibration), 847–868 cm− 1 (Al-O out of plane vibration) and
impact on catalyst performance.
487 cm− 1 (metal–oxygen bond linked with Al-O in the alumina tetra­
hedral sheet and Si-O in the silica octahedral plate). The FTIR data ob­
tained indicated that the as-synthesized composite catalyst was made up
of various metal oxides as revealed by the XRD analysis (Fig. 3).

4
A.S. Yusuff et al. Energy Conversion and Management: X 16 (2022) 100302

Table 5 not significant relative to the pure error. Non-significant lack of fit is
Analysis of variance (ANOVA) for response surface quadratic model for biodiesel good as it confirms the adequacy of the model fit. Also, an adequate
yield. precision (a parameter that measures ratio of signal to noise) value
Source Sum of Degree of Mean F- P-value Remark greater than 4 is desirable as it indicates an adequate signal. As evident
squares freedom square value in the presented results (Table 5), the model for the yields of biodiesel
Model 463.30 9 51.48 30.44 <0.0001 Significant obtained via transesterification of JCO with methanol over ZnO-CFA400
T 112.80 1 112.80 66.70 <0.0001 catalyst satisfied those above-mentioned criteria, thus confirming the
M 91.53 1 91.53 54.13 0.0002 adequacy of the model fit [23]. These findings were supported by the
C 79.51 1 79.51 47.02 0.0002
correspondence plot of the predicted and experimental JCO biodiesel
TM 27.25 1 27.25 16.11 0.0051
TC 4.75 1 4.75 2.81 0.1376 yields, as seen in Fig. 4.
MC 11.83 1 11.83 7.00 0.0332 As can be seen in Table 5, the ANOVA revealed that T, M, C, TM, MC,
T2 83.30 1 83.30 49.26 0.0002 T 2 and C2 were the significant model terms as their p-values were less
M2 4.28 1 4.28 2.53 0.1557 than 0.05 which suggested the significant parameters at 95 % confi­
C2 54.33 1 54.33 32.13 0.0008
dence level. However, TC and M2 were insignificant to the biodiesel
Residual 11.84 7 1.69 –
Lack of 5.51 3 1.84 1.16 0.4276 Non- yield. It is worthy of note that all the three parameters studied were all
fit significant significant to the response in the ascending order of catalyst loading,
Pure 6.33 4 1.58 methanol/JCO molar ratio and reaction temperature.
error
2 2
R = 0.9751; Adj − R = 0.9431; Adeq − precision = 17.43; std.dev =
1.30
3.2.1. Effect of parameters as response surface plot
The surface plot in Fig. 5 (a-c) depicts the interaction effect of the
transesterification process parameters (temperature, methanol/JCO
molar ratio and catalyst dosage) on the yield of JCO biodiesel. The in­
fluence of methanol/JCO molar ratio and temperature on JCO biodiesel
yield is depicted in Fig. 5a. This plot demonstrated that increasing the
methanol/oil molar ratio increased biodiesel yield while decreasing
with increasing temperature. The reason for this observation was
thought to be that a higher methanol/JCO molar ratio would enhance
the formation of the biodiesel as per Le Chatelier’s principle, as the
excess methanol in transesterification reaction process reduces the vis­
cosity of the reaction medium, promotes the mass transfer and eventu­
ally improves the performance of the catalyst [2]. However, as the
reaction temperature increased, the biodiesel yield decreased (see
Fig. 5a and 5b). The assumed reason was that at much higher reaction
temperatures greater than 60 ◦ C, the biodiesel yield steadily decreased
because the methanol solvent vapourized at its boiling point (around
65 ◦ C), reducing the net amount of solvent available for trans­
esterification and lowering the biodiesel yield [1]. Furthermore, the
reaction by-product (glycerol) obtained at temperatures above 65 ◦ C
was much more viscous than normal because the intermediate products
(esters of mono and diglycerides) appeared to dissolve in the glycerol
phase, inhibiting the desired product yield [4,10]. As shown in Fig. 5b
Fig. 4. Predicted versus experimental biodiesel yield. and 5c, the biodiesel yield increased as catalyst dosage increased from
1.0 to 1.6 wt% and then decreased as the catalyst concentration
3.2. Box-Behnken design (BBD) model and statistical analysis increased above 1.6 wt%. This could be attributed to the fact that higher
catalyst loading resulted in the formation of an emulsion (a more viscous
The 3-factor BBD matrix as generated by the design expert software phase) as a result of a side reaction (saponification reaction) that
and the experimental data obtained in the JCO conversion to biodiesel occurred between the remaining FFA in the esterified JCO and excessive
are presented in Table 4. The quadratic response equation for trans­ catalyst, which inhibited the diffusion of reactant to the catalyst active
esterification of JCO to produce biodiesel is given in Eq. (2). sites, lowering the biodiesel yield [24,25].

Y = + 73.40 − 3.76T + 3.38M + 3.15C + 2.61TM − 1.09TC + 1.72MC

+ 4.45T 2 + 1.01M 2 − 3.59C2 3.3. Optimal conditions of transesterification of JCO over ZnO-CFA400
(2) catalyst

As contained in Eq. (2), positive coefficient signified synergetic effect The optimum values for the three process parameters investigated
on biodiesel yield, while negative coefficient suggested antagonistic were numerically predicted to be 60.4 ◦ C reaction temperature, 11.8:1
influence on the response. The model quality was examined based on methanol/JCO molar ratio and 1.63 wt% catalyst loading with 86.75 %
some criteria which included coefficient of determination (R2 ), adjusted- for JCO biodiesel yield. The prediction was validated by using the best
R2 (Adj-R2 ), lack of fit, standard deviation (std. dev), model F-value, conditions in three replicates, yielding an average biodiesel yield of
probability distribution function (p-value) and adequate precision. A 91.08 ± 0.06 %. Thus, the model accurately represented the trans­
model fit is said to be adequate if the R2 value is closer to unity and the esterification process for JCO biodiesel production. The biodiesel yield
Adj–R2 value is more comparable to the corresponding R2 value [18]. obtained in this study was greater than the maximum biodiesel yield
The larger the model F-value and the smaller the standard deviation, the (79.76 %) reported by Ho et al. [12], who evaluated a palm oil mill fly
more accurate the response could be predicted by the model. In addition ash supported calcium oxide (PFA/CaO) catalyst in biodiesel production
to this, the p-value less than 0.0500 suggests that the model term is from crude palm oil (CPO) at 45 ◦ C for 3 h with a methanol/oil molar
significant, whereas the lack of fit F-value of 1.16 implies the lack of fit is ratio of 12:1 and a catalyst loading of 6.0 wt%. It is worth noting that the

5
A.S. Yusuff et al. Energy Conversion and Management: X 16 (2022) 100302

Fig. 5. 3-D response surface plots of biodiesel yield (%) as the function of (a) methanol/JCO molar ratio, M and reaction temperature, T (◦ C); (b) catalyst loading, C
(wt.%) and reaction temperature, T (◦ C) and (c) catalyst loading, C (wt.%) and methanol/JCO molar ratio, M.

amount of ZnO-CFA400 (1.63 wt% of the loaded JCO) used in the cur­ power a diesel engine. Because of the significant reduction in FFA
rent study was less than the amount of PFA/CaO (6.0 wt% of the loaded content in the feedstock after acid esterification, no soap formation was
CPO) used in transesterification reaction, making the ZnO-CFA400 observed during the base transesterification process [26]. In addition to
superior. this, the kinematic viscosity of the produced biodiesel was significantly
lower than that of its associated oil, confirming a successful two-step
transesterification process. The kinematic viscosity value within
3.4. Analysis of JCO biodiesel produced under optimum conditions 1.9–6.0 mm2/s was preferred because viscous fuel caused engine per­
formance issues such as injector coking, wear and tear, and carbon
3.4.1. Physicochemical properties deposition [4,27]. According to ASTM D664, the maximum permissible
Table 6 shows the properties of biodiesel produced by trans­ biodiesel acid value is 0.8 mg KOH/g [6,28], indicating that the JCO
esterification of JCO over a ZnO-CFA400 catalyst under optimal reaction biodiesel produced herein with an acid value of 0.57 ± 0.02 mg KOH/g
conditions. As shown in the table, the majority of the values of the could be used in diesel engines without causing engine corrosion.
produced biodiesel’s properties were in accordance with the EN 14,214
and ASTM D6757 standards, confirming the biofuel’s suitability to

6
A.S. Yusuff et al. Energy Conversion and Management: X 16 (2022) 100302

Table 6 Table 8
Physicochemical and fuel properties of JCO and JCO-B. EDX analysis for fresh ZnO-CFA400 and spent ZnO-CFA400 used for last cycle.
Property ASTM Value ASTM biodiesel Element Composition (%)
test specification
Fresh ZnO-CFA400 Spent ZnO-CFA400
JCOa JCO Si 25.4 23.9
Biodiesel Ca 1.0 0.6
Mg 0.5 0.1
Specific gravity @ D4052 920 886 ± 860–900
O 43.4 53.2
25 ◦ C 0.002
Al 14.7 8.6
Kinematic viscosity @ D445 38.1 4.23 ± 1.9–6.0
Fe 4.5 3.9
40 ◦ C (mm2/s) 0.17
C – 4.5
Acid value (mg KOH/g) D664 11.2 0.57 ± 0.8 max
Ti 1.3 0.9
0.02
Zn 7.7 4.3
FFA content (wt.%) – 5.6 0.285 ± –
0.01
Flash point (◦ C) D93 256 171 ≥ 130
Cloud point (◦ C) D2500 +24 +11 − 3 to 12
suggesting transformation of triglycerides to methyl esters [30]. The
Pour point (◦ C) D97 +17 +6 − 15 to 10 characteristic peaks at 2917–2934 cm− 1 confirmed asymmetric and
a symmetric stretching vibration of –C–H in methylene and methyl
Yusuff [20].
groups. The sharp peaks observed at around 1750 cm− 1 and 1174 cm− 1
were corresponded to C– – O stretching vibration of ester and C–O
stretching of ester, respectively [31]. It is important to mention that the
absorption band at 1469 cm− 1 in JCO appeared as 1469 and 1433 cm− 1
in synthesized biodiesel, which might be due to deformation vibration of
O-CH3 group [32].

3.4.3. FAME content analysis

The data obtained from the holistic analysis of the JCO biodiesel
indicated that the composition of the biodiesel obtained under the op­
timum operating variables included the FAME profile as presented in
Table 7 which signified methyl oleate (37.41 %), methyl linoleate
(30.53 %), methyl palmitate (14.08 %) and methyl stearate (11.98 %) as
major components in the produced JCO biodiesel. Table 7 revealed a
total of 26.15 % saturated and 71.07 % unsaturated methyl ester con­
tents. The qualitative analysis results confirmed that the FAME content
(97.22 %) in the synthesized JCO biodiesel was higher than the ASTM
and EN specifications [12]. Maftuchah et al. [33] produced biodiesel
from several genotype crossbred Jatropha curcas seeds and obtained a
similar FAME profile. It should be noted that the FAME content in bio­
diesel produced under optimal conditions in this study was significantly
higher than many others reported in the literature for various catalysts
developed (see Table 9). For example, when BMK10 was used as catalyst
in the transesterification of WCO at 150 ◦ C, methanol/oil molar ratio of
12:1, catalyst loading of 3.5 wt% and reaction time of 5 h, a maximum
Fig. 6. FTIR spectra of (a) JCO and (b) JCO biodiesel. FAME content of 83.38 % was obtained [1].

Table 7 3.5. Catalyst reusability study

FAME content of the JCO biodiesel.
S/N FAME content Composition (wt.%) Following each experiment, the spent catalyst was centrifuged from
the reaction product, washed with n-hexane to remove the adsorbed
1 C16:0 Methyl palmitate 14.08
2 C16:1 Methyl palmitoleate 0.72 triglycerides and esters, dried at 100 ◦ C for 4 h and recalcined at 350 ◦ C
3 C18:0 Methyl stearate 11.98 for 2 h to restore its catalytic performance. The regenerated ZnO-
4 C18:1 Methyl oleate 37.41 CFA400 catalyst was then reused for transesterification of JCO to bio­
5 C18:2 Methyl linoleate 30.53
diesel under optimum conditions (60.4 ◦ C reaction temperature, 11.8:1
6 C18:3 Methyl linolenate 1.19
7 C20:0 Methyl arachidate 0.09
methanol/JCO molar ratio and 1.63 wt% catalyst loading). The bio­
8 C20:1 Methyl eicosenoate 0.38 diesel yields of 78.04, 65.31, 53.20 and 39.52 % were obtained for four
9 C20:2 Methyl 11, 14-eicosadienoate 0.41 consecutive cycles of JCO methanolysis using regenerated ZnO-CFA400
10 C20:3 Methyl 5, 8, 11-eicosatrienoate 0.08 catalyst, implying that the composite catalyst could be reused at least
11 C20:4 Methyl arachidonate 0.31
four times with a significant drop in biodiesel yield compared to the
12 C20:5 Methyl eicosapentenoate 0.04
13 others 2.78 91.08 ± 0.06 % conversion achieved with fresh catalyst. The decreased
activity of the ZnO-CFA400 catalyst after each cycle was caused by re­
action product blocking active sites on the catalyst surface, preventing
3.4.2. FTIR analysis reactants from accessing the active basic sites [34]. This was consistent
The biodiesel and associated oil from Jatropha curcas seed were with the functional groups observed in the FTIR spectrum of the
analyzed using FTIR technique. As seen in Fig. 6, the spectra of JCO and recovered catalyst after the last (fourth) cycle (see Fig. 2c). The peaks
its biodiesel seemed to be similar probably due to the chemical simi­ associated with C– – O (ester) stretching at 1717 and 1788 cm− 1 were
larities that exist among triglycerides and esters [29]. However, more observed on the spectrum of the spent catalyst, as shown in the FTIR
peaks were observed in the JCO biodiesel than in the JCO spectrum, result (Fig. 2c). Furthermore, the intensity of the peaks at 2926 and

7
A.S. Yusuff et al. Energy Conversion and Management: X 16 (2022) 100302

Table 9
Comparison of FAME production condition and yield for the current study and previously reported works.
Catalyst Feedstock Experimental Conditions FAME content (%) Reference

T ( C)
◦
M C (wt.%) t (h)

BMK10 WCO 150 12:1 3.5 5.0 83.38 [1]

Ba-ZEL700 WFO 65.38 12:1 3.0 2.0 90.89 [36]
Li/CaO Nahor oil 65 10:1 5.0 4.0 94.0 [37]
Li-Chicken bone UCO 60 18:1 4.0 3.0 94.90 [5]
KOH/JS WFO 60 16:1 – 2.0 86.7 [35]
KSF clay JCO 160 1:1 4.8 6.0 70 [38]
AENICO WFO 70 12:1 3.0 2.0 90.2 [39]
SO2-
4 /ZrO2/Al2O3 Palm oil 127 8.1 6.0 3.0 83.3 [40]
ZnO-CFA400 JCO 60.4 11.8:1 1.63 3.0 97.22 Current study

WFO: waste frying oil; JS: Jatropha curcas fruit shell; AENICO: anthill-eggshell-Ni-Co.

2966 cm− 1 (C–H stretching vibration of the methyl group) appeared to Declaration of Competing Interest
increase after the last time reuse, confirming the poisoning of catalyst
surface sites by methyl esters [10]. This is supported by EDX results The authors declare that they have no known competing financial
(Table 8), which showed that the spent catalyst contained a lower mass interests or personal relationships that could have appeared to influence
fraction of zinc and other elements than the fresh catalyst, indicating the work reported in this paper.
leaching of catalyst active ingredient. It is worth noting that the
increased oxygen content and presence of carbon in the spent catalyst Data availability
after regeneration suggested that its treatment with hexane favored the
formation of oxygenated and carboxylic groups, as confirmed by the The authors do not have permission to share data.
FTIR results. This result was consistent with the findings of Olutoye and
Hameed [22], who synthesized FAME from waste cooking palm oil using Acknowledgment
a highly active clay-based catalyst.
The first author (Dr. Adeyinka Yusuff) gratefully acknowledges the
postdoctoral fellowship awarded him by TWAS-CSIR (FR number:
3.6. Comparison between ZnO-CFA400 and other catalysts 3240306317).

Table 9 compares the performance of ZnO-CFA400 catalyst to that of References

other previously used heterogeneous catalysts for biodiesel synthesis
from different feedstock varieties and at different optimum reaction [1] Olutoye MA, Wong SW, Chin LH, Amani SW, Asif M, Hameed BH. Synthesis of fatty
acid methyl esters via the transesterification of waste cooking oil by methanol with
conditions. It was demonstrated that the prepared ZnO-CFA400 catalyst
a barium-modified montmorillonite K10 catalyst. Renewable Energy 2016;86:
performed better, providing a FAME yield of 97.22 % under optimum 392–8.
reaction conditions of 60.4 ◦ C reaction temperature (T), 11.8 methanol/ [2] Ajala EO, Ajala MA, Ajao AO, Saka HB, Oladipo AC. Calcium-carbide residue: A
precursor for the synthesis of CaO-Al2O3-SiO2-CaSO4 solid acid catalyst for
JCO molar ratio (M), 1.63 wt% catalyst loading (C), and a fixed reaction
biodiesel production using waste lard. Chem Eng J Adv 2020;4:100033.
time (t) of 3 h, implying that ZnO-CFA400 could be considered as an [3] Dong S, Zhu M, Dai B. Synthesis, characterization, and application of ZS/HMS
efficient catalyst for FAME synthesis from JCO, especially when catalyst in the esterification of gossypol. Green Sustain Chem 2012;2:8–13.
compared with KOH-supported Jatropha curcas fruit shell (KOH/JS) [4] Tan TH, Abdullah MO, Nolasco-Hipolito C, Taufiq-Yap YH. Waste ostrich- and
chicken-eggshells as heterogeneous base catalyst for biodiesel production from
[35] and BMK10 [1]. used cooking oil: Catalyst characterization and biodiesel yield performance. Appl
Energy 2015;160:58–70.
[5] AlSharifi M, Znad H. M, Znad H, Development of a lithium based chicken bone (Li-
4. Conclusions Cb) composite as an efficient catalyst for biodiesel production. Renewable Energy
2019;136:856–64.
This paper investigated the performance of a zinc oxide-based coal [6] Singh D, Sharma D, Soni SL, Inda CS, Sharma S, Sharma PK, et al. A comprehensive
review of physicochemical properties, production process, performance and
fly ash (ZnO-CFA) composite as a heterogeneous catalyst for biodiesel emissions characteristics of 2nd generation biodiesel feedstock: Jatropha curcas.
synthesis from JCO. Zn ions were impregnated into coal fly ash and the Fuel 2021;285:119110.
activity of the composite catalyst was evaluated under mild process [7] Elgharbawy A, Sadik WA, Sadek OM, Kasaby M. A review on biodiesel feedstocks
and production technologies. J Chil Chem Soc 2021;66(1):5098–109.
conditions. Calcination of Zn(NO3)2 loaded fly ash at 400 ◦ C was
[8] Ajaaro AS, Ojo MS, Lawal BS. Pristine organoclay as catalyst for biodiesel synthesis
discovered to be critical for obtaining an efficient catalyst for JCO from Jatropha curcas seed oil. J Pure Appl Sci 2018;28(3):561–8.
conversion to biodiesel. The response surface plot was used to assess the [9] Lawan I, Garba ZN, Zhou W, Zhang M, Yuan Z. Synergies between the microwave
impact of operational parameters on JCO biodiesel yield. The optimum reactor and CaO/zeolite catalyst in waste lard biodiesel production. Renewable
Energy 2020;145:2550–60.
values of the reaction temperature, methanol/JCO molar ratio and [10] Yusuff AS, Kumar M, Obe BO, Mudashiru LA. Calcium oxide supported on coal fly
catalyst dosage were 60.4 ◦ C, 11.8:1 and 1.63 wt%, respectively, and the ash (CaO/CFA) as an efficient catalyst for biodiesel production from Jatropha
physicochemical properties of biodiesel produced at these conditions curcas oil. Top Catal 2021. https://doi.org/10.1007/s11244-021-01478.
[11] Yusuff AS, Thompson-Yusuff AK, Porwal J. Sulfonated biochar catalyst derived
conformed with ASTM standard. Based on the promising results ob­ from eucalyptus tree shed bark: synthesis, characterization and its evaluation in
tained herein, the ZnO loaded coal fly ash catalyst had good activity and oleic acid esterification. RSC Adv 2022;12:10237–48.
was relatively stable during reuse. [12] Ho WWS, Ng HK, Gan S, Tan SH. Evaluation of palm oil mill fly ash supported
calcium oxide as a heterogeneous base catalyst in biodiesel synthesis from crude
palm oil. Energy Convers Manage 2014;8:1167–78.
CRediT authorship contribution statement [13] Franus W, Wdowin M, Franus M. Synthesis and characterization of zeolites
prepared from industrial fly ash. Environ Monit Assess 2014;186:5721–9.
[14] Ho WWW, Ng HK, Gan S. Development and characterization of novel
Adeyinka Sikiru Yusuff: Conceptualization, Methodology, Soft­ heterogeneous palm oil mill boiler ash-based catalysts for biodiesel production.
ware, Validation. Lekan Taofeek Popoola: Writing – review & editing. Bioresour Technol 2012;1255:158–64.
David Olalekan Adeniyi: Visualization, Investigation. Moses Aderemi [15] Giwa A, Akpan UG, Hameed BH. Optimization of photocatalytic degradation of an
anthraquinone dye using design of experiment. J Eng Res 2012;17:20–31.
Olutoye: Writing – review & editing, Visualization.

8
A.S. Yusuff et al. Energy Conversion and Management: X 16 (2022) 100302

[16] Yusuff AS, Popoola LT, Igbafe AI. Response surface modeling and optimization of [29] Taufiq-Yap YH, Abdullah NF, Basri M. Biodiesel production via transesterification
hexavalent chromium adsorption onto eucalyptus tree bark-derived pristine and of palm oil using NaOH/Al2O3 catalysts. Sains Malaysiana 2011;40:587–94.
chemically-modified biochar. Chem Eng Res Des 2022;182:592–603. [30] Niju S, Raj FR, Anushya C, Balajii M. Optimization of acid catalyzed esterification
[17] Bezerra MA, Santelli RE, Oliveira EP, Villar LS, Escaleira LA. Response surface and mixed metal oxide catalyzed transesterification for biodiesel production from
methodology (RSM) as a tool for optimization in analytical Chemistry. Talanta Moringa Oleifera oil. Green Process Synth, 2019;8:756–75.
2008;76(5):965–77. [31] Siatis NG, Kimbaris AC, Pappas CS, Tarantilis PA, Polissiou MG. Improvement of
[18] Montogomery DC. Design and Analysis of experiments. New York, USA: John bioidiesel production based on the application of ultrasound monitoring of the
Wiley and Sons; 2005. procedure by FTIR spectroscopy. J Am Oil Chemists’ Soc 2011;83(1).
[19] Obeng DP, Morrell S, Napier-Munu TJ. Application of central composite rotatable [32] Betiku E, Okeleye AA, Ishola NB, Osunleke AS, Ojumu TV. Development of a novel
design to modelling the effect of some operating variables on the perforamnce of mesoporous biocatalyst derived from kola nut pod husk for conversion of Kariya
the three-product cyclone. Int J Miner Process 2015;76:181–92. seed oil to methyl esters: A case of synthesis, modeling and optimization studies.
[20] Yusuff AS. Parametric optimization of solvent extraction of Jatropha curcas seed oil Catalysis Letter 2019;149(7):1772–87.
using design of experiment and its quality characterization. S Afr J Chem Eng [33] Maftuchah, Zainudin A, Winaya A, Rahmadesi Y. Biodiesel generated from
2021;35:60–8. Jatropha (Jatropha curcas Linn.) seeds selected based on various genotypes
[21] Izidoro JDC, Fungaro DA, dos Santos FS, Wang S. Characteristics of Brazilian coal crossbred. Energy Rep 2020;6:345–50.
fly ashes and their synthesized zeolites. Fuel Process Technol 2012;97:38–44. [34] Georgogianni K, Kontominas M, Tegru E, Avlonitis D, Gergis V. Biodiesel
[22] Olutoye MA, Hameed BH. A highly active clay-based catalyst for the synthesis of production: reaction and process parameters of alkali-catalyzed tarnsesterification
fatty acid methyl ester from waste cooking palm oil. Appl Catal A 2013;450:57–62. of waste frying oils. Energy and Fuel 2011;21(5):3023–7.
[23] Adesina OA, Abdulkareem F, Yusuff AS, Lala M, Okewale A. Response surface [35] Buasri A, Chaikwan T, Lorychenyong V, Rodklum C, Chaikwan T, Kumphan N.
methodology approach to optimization of process parameter for coagulation Continuous process for biodiesel production in packed bed reactor from waste
process of surface water using Moringa oleifera seed, South African Journal of. frying oil using potassium hydroxide supported on Jatropha curcas fruit as solid
Chem Eng 2019;28:46–51. catalyst, Applied. Science 2012;2:647–53.
[24] Cao LC, Zhang SC. Production and characterization of biodiesel derived from [36] Yusuff AS, Bhonsle AK, Bangwal DP, Atray N. Development of a barium-modified
Hodgsomia macrocarpa seed oil. Appl Energy 2010;146:135–40. zeolite catalyst for biodiesel production from waste frying oil: Process optimization
[25] Kafuku G, Mbarawa M. Alkaline catalyzed biodiesel production from Moringa by design of experiment. Renewable Energy 2021;177:1253–64.
oleifera with optimized production parameter. Appl Energy 2010;87:2561–5. [37] Boro J, Konwar LJ, Deka D. Transesterification of nonedible feedstock with lithium
[26] Yusuff AS, Bangwal DP, Gbadamosi AO, Atray N. Kinetic and thermodynamic incorporated eggshell derived CaO for biodiesel production. Fuel Process Technol
analysis of biodiesel and associated oil from Jatropha curcas L. during thermal 2014;122:72–8.
degradation. Biomass Convers Biorefin 2021. https://doi.org/10.1007/s13399- [38] Calgaroto C, Calgaroto S, Mazutti MA, de Oliveir D, Pergher S, de Oliveira J.
021-01545-3. Production of biodiesel from soybean and Jatropha curcas oils with KSF and
[27] Zheng S, Kates M, Dube MA, Mclean DD. Acid-catalyzed production of biodiesel amberlyst 15 catalysts in the presence of co-solvents. Sustainable Chem Processes
from waste frying oil. Biomass Bioenergy 2006;30:267–72. 2013;1:1–6.
[28] Hossain M, Goni LKMO, Muntaha N, Jamal MS, Sujan SMA, Ahmed S, Islam D, [39] Yusuff AS, Adeniyi OD, Azeez OS, Olutoye MA, Akpan UG. Synthesis and
Bhuiyan RH, Fakhruddin ANM. Box-Behnken design-based optimization for characterization of anthill-eggshell-Ni-Co mixed oxides composite catalyst for
biodiesel production from waste cooking oil using Mahogany (Swietenia biodiesel production from waste frying oil. Biofuels, Bioproduct & Biorefinery
macrophylla) fruit shell derived activated carbon as a heterogeneous base catalyst. 2019;13:37–47.
Reaction kinetics, Mechanisms and Catalysis, doi.org/10.1007/s11144-021-01995- [40] K.F. Yee, K.T. Lee, Palm oil as feedstocks for biodiesel production via
w. heterogeneous transesterification: optimization study, International Conference on
Environment 2008 (ICENV 2008).