ASIA-PACIFIC JOURNAL OF CHEMICAL ENGINEERING

Asia-Pac. J. Chem. Eng. (2015)

Published online in Wiley Online Library
(wileyonlinelibrary.com) DOI: 10.1002/apj.1930

Research article
Fly ash supported Ni―Fe solid acid catalyst for efficient
production of diesel additive: intensification through
far-infrared radiation
Piasy Pradhan and Rajat Chakraborty*
Chemical Engineering Department, Jadavpur University, Kolkata 700032, India

Received 11 January 2015; Revised 15 July 2015; Accepted 27 July 2015

ABSTRACT: Novel Ni―Fe solid acid catalyst was prepared through co-wet impregnation of Fe2 (SO4)3·H2O and Ni (NO3)
2·6H2O on fly ash (FA) support. XRD, BET-BJH, SEM-EDX, XPS and FTIR methods were employed to characterize the
catalyst. The excellent catalyst activity was assessed and optimized in the production of methyl oleate (biodiesel/diesel-
additive). The response surface methodology (RSM) computed optimal parametric values pertaining to maximum oleic acid
(OA) conversion (98.37 ± 0.01%) were 15:1 methanol/OA molar ratio, 1:15:2 nickel nitrate: ferric sulphate:FA (wt. ratio) and
4 wt.% (of OA) catalyst concentration in 1 h under conventional heating at mild reaction temperature (60 °C). The optimal
catalyst possessed 27.64 m2/g specific surface area and 0.0233 cc/g pore volume with predominantly mesoporous framework.
XPS bands confirmed the presence of Fe3+ and mixed Ni2+/Ni3+species rendering an acidity of 5.5 mmol KOH/g. The
catalyst comprised of crystalline Fe2O3, Ni2O3 and NiS phases as revealed from XRD analysis. This study also reports the
remarkable intensification of esterification reaction through an energy-efficient application of far-infrared radiation protocol
(1/3rd of conventional power requirement) resulting 99.20% OA conversion in only 10 min at the derived optimal conditions.
© 2015 Curtin University of Technology and John Wiley & Sons, Ltd.

KEYWORDS: fly ash; Ni―Fe; far-infrared radiation; ester; RSM optimization

INTRODUCTION of air pollution, greenhouse gas emission and global

warming; on the contrary, biodiesel is nontoxic,
Fly ash (FA) is a typical waste of thermal power plants, renewable, biodegradable and thus, environment
which characteristically contains SiO2, Al2O3, Fe2O3, friendly. Esterification of methanol (M) and oleic acid
Na2O, CaO, MgO, TiO2, BaO and K2O.[1] In recent (OA) produces methyl oleate (MO), which can be used
days many technologies have been reported for as biodiesel and also as additive to the fossil diesel.
utilization of FA for development of value added Notably, the technical problems in biodiesel, e.g.
products. Solid heterogeneous catalysts were prepared oxidative stability, cold flow and increased NOX
using FA as a support by several research groups. In exhaust emission can be overcome by using methyl
Knoevenagel condensation reaction, FA supported oleate (MO) as additive.[9] Kusmiyati and Sugiharto[10]
calcium oxide base catalyst was employed[2]; while, reported esterification of M and OA using hazardous
other FA supported metallic acid and base catalysts sulphuric acid as a catalyst employing reactive
were synthesized and characterized for different distillation. Besides, sulfuric acid was employed to
reactions.[3–6] In transesterification of soybean oil[7] catalyse esterification reaction between ethanol and
and sunflower oil,[8] FA based catalysts were used. acetic acid.[11]
However, to the best of our knowledge, all the reported Nevertheless, owing to the difficulty in separation of
catalysts used a single active metal precursor for homogeneous sulphuric acid catalyst, more recently,
catalyst preparation. catalysts were effectively employed to produce biofuel
Nowadays, biodiesel has been considered as the components over various heterogeneous supported
potential alternative to replace fossil diesel to overcome noble metal and bimetallic catalysts[12] owing to ease
environmental problems. Fossil fuel is the major source of product purification;[13] and reduce waste stream
generation to meet the global stringent pollution
regulations.[14] Application of solid heterogeneous
*Correspondence to: Rajat Chakraborty, Chemical Engineering
Department, Jadavpur University, Kolkata-700032, India. E-mail: acid catalyst, e.g. cation exchange resin could
rajat_chakraborty25@yahoo.com successfully produce MO through esterification
© 2015 Curtin University of Technology and John Wiley & Sons, Ltd.
Curtin University is a trademark of Curtin University of Technology
P. PRADHAN AND R. CHAKRABORTY Asia-Pacific Journal of Chemical Engineering

reaction.[15,16] Commercial heterogeneous catalyst viz. 1:15:2 (NI-15). The mixture was vigorously agitated
Amberlyst-16 was used for esterification of acetic acid with as per wet-impregnation protocol at 60 °C, for 2 h under
M to produce methyl acetate.[17] However, appreciable total reflux in a three-necked flask. Subsequently,
ester yield was obtained at the expense of moderately high ammonium hydroxide solution was added to maintain
catalyst concentration (5%) and temperature (80.15 °C). pH at 10 for completion of precipitation. It was aged
Yan et al. developed mesoporous H4SiW12O40–SiO2 for 48 h and consequently dried for 12 h, at 105 °C in
catalysts which demonstrated high efficacy for synthesis hot air oven. The dried mass was grinded to fine powder
of methyl (73% yield) and ethyl (67% yield) levulinate (150 mesh). Finally, at 700 °C, the dried mass was
biodiesel.[18] calcined for 4 h to prepare the Ni―Fe/FA catalyst.
The present work reports the development of a FA
supported low-cost, highly efficient Ni―Fe solid acid Catalyst characterization
catalyst (Ni―Fe/FA catalyst)for performing esterification
of OA to synthesize MO (biodiesel). Ferric sulphate [Fe2 The XRD (X-ray diffraction) patterns (RigakuMiniflex
(SO4)3·H2O] was used in different ratio keeping the Co., Japan) of the three different ratios of Ni―Fe/FA
amount of Ni(NO3)2·6H2O and calcined-FA constant to catalyst were detected using a Cu Kα source equipped
prepare catalysts through wet impregnation method. with an Inel CPS 120 hemispherical detector. Analysis
Ni―Fe/FA catalyst characterization has been performed was performed at 2θ ranging from 0° to 80° at a
through XRD line-broadening, SEM-EDX, FTIR, BET- scanning speed of 1° min
1. The specific surface area
BJH and XPS analyses. M/OA molar ratio (QMR),Fe2 of the catalyst sample was measured by BET (Brunner–
(SO4)3·H2O loading (QL) and catalyst concentration Emmett–Teller) method (Quantachrome make NOVA
(QC) have been selected as process independent factors 4000e). The pore volume and pore size distribution of
through response surface methodology (RSM) to the catalyst were investigated using the BJH (Barret,
determine maximum OA conversion. The efficacy of Joyner, Halenda) method. The surface morphology of
the catalyst was also assessed in presence of novel, the prepared catalyst was determined using scanning
energy efficient far-infrared radiation for intensification electron microscope (SEM) at 15 kV (JEOL Ltd., JSM-
of OA conversion. 6360). FTIR (Fourier Transform Infrared) spectra of the
catalyst were detected with a FTIR, SHIMADZU
(Alpha), from 400 to 4000 cm
1. The acidity of the
MATERIAL AND METHODS catalyst was determined as per the method of titration.[19]
PHI 5000 Versaprobe II scanning XPS (X-ray
Materials photoelectron spectroscopy) microprobe instrument
(manufactured by UL-VAC PHI, USA) was used to
Hexahydrate nickel nitrate, methanol (>99% purity), measure the binding energy of Fe 2p and Ni 2p of NI-
oleic acid (OA), ammonium hydroxide, silica gel, 15 catalyst. For monochromatic scanning, Al Kα X-ray
molecular sieve etc. were purchased from Merck and source and 180° hemispherical electron analyser, model
Ferric sulphate monohydrate was procured from Loba PHI 10-371 was used.
Chemicals(India). All the chemicals used were of Statistical analysis was performed using Design
analytical reagent (AR) grade. Fly ash (FA) was collected Expert 8.0 software. Face centred central composite
from Kolaghat Thermal Power Plant, West Bengal, India. design (FCCD) was used with three factors viz. M to
OA molar ratio (QMR), Fe2 (SO4)3·H2O loading (QL)
Catalyst preparation and catalyst concentration (QC) (Table 1). Experimental
runs were performed based on 15 different combinations
FA was calcined at 900 °C in a muffle furnace for 2 h to of the coded variables (Table 2).
remove carbonaceous impurities. Aqueous solutions of
Ni(NO3)2·6H2O and Fe2(SO4)3·H2O were separately Reaction procedure
prepared and finally added to a measured amount of
calcined-FA in three weight ratios: Ni(NO3)2·6H2O: Esterification of OA was primarily performed in a
Fe2(SO4)3·H2O:FA of 1:5:2 (NI-5), 1:10:2 (NI-10) and conventionally heated reactor (CHR) (500-mL three-necked

Table 1. Experimental ranges and levels of the process factors used in RSM.

Factors Name Units
1 level 0 level +1 level

QMR Methanol/OA molar ratio — 5.00 10.00 15.00
QL Fe2(SO4)3·H2O loading wt.% 5.00 10.00 15.00
QC Catalyst concentration wt.% of OA 2.00 3.00 4.00
QC
3a
qMR ¼ QMR5
10 qL ¼ QL
10
5 qC ¼ 1

© 2015 Curtin University of Technology and John Wiley & Sons, Ltd. Asia-Pac. J. Chem. Eng. (2015)
DOI: 10.1002/apj
Asia-Pacific Journal of Chemical Engineering NI―FE/FLY ASH CATALYST: DIESEL ADDITIVE
Table 2. Face-centred central composite design (FCCD) layout for esterification of OA using Ni―Fe/FA catalyst.

Factor (qMR) Factor (qL) Factor (qC) Response (γOA)

Methanol/oleic Fe2(SO4)3·H2O Catalyst concentration OA conversion
Run acid molar ratio loading (wt.%) (wt.% of OA) (%)
1 0.00 0.00
1.00 91.09
2
1.00 1.00
1.00 97.24
3
1.00
1.00 1.00 87.77
4
1.00
1.00
1.00 90.11
5 0.00
1.00 0.00 88.78
6 1.00
1.00
1.00 86.50
7 1.00
1.00 1.00 87.55
8
1.00 1.00 1.00 94.90
9 1.00 0.00 0.00 92.02
10 0.00 1.00 0.00 97.21
11 0.00 0.00 1.00 92.34
12 0.00 0.00 0.00 92.77
13 1.00 1.00 1.00 98.37
14
1.00 0.00 0.00 92.09
15 1.00 1.00
1.00 97.49
16 0.00 0.00 0.00 92.53

flask, equipped with two reflux condensers and a PID the acid value, the OA conversion (Eqn (2)) was
temperature controller provided with a 300-W evaluated and the details are given below:
conductive heating mantle) at a controlled stirrer speed
of 600 rpm. Initially fixed amount of OA (10 mL, 9.1 g) 56:11NV
and measured amount of catalyst were stirred for few COA ¼ ; (1)
m
minutes at 60 °C to make OA-catalyst slurry.
Afterwards, required quantity of M was added to the  
reactor, and the reaction took place for 1 h. Experiments AV OA
C OA
γOA ¼ 100 : (2)
were conducted according to parametric values AV OA
(Table 1) and OA conversions were measured as the
response values in accordance with the FCCD matrix
(Table 2) to evaluate optimal factors for maximizing
OA conversion. At the end of the reaction, the mixture
was vacuum filtrated to separate the solid catalyst. After RESULTS AND DISCUSSION
filtration, hot distilled water was added to the filtrate to
remove the excess M. To separate the ester phase from Statistical analyses
the aqueous phase, it is centrifuged for 10 min at about
5000 rpm, to obtain MO as the top layer. In the present study, to evaluate the optimal process
After assessing the optimal factorial values factors for maximum conversion of OA to MO in
corresponding to maximum OA conversion (from esterification reaction, the RSM model based on FCCD
Table 2) in CHR; an attempt was made to enhance and has been developed with three factors viz. QMR, QL and
accelerate the OA conversion by incorporating 40 wt.% QC. The experimental results showing OA conversion
of silica gel (SG) or molecular sieve (MS) desiccants to (γOA) as a function of process factors are tabulated
facilitate higher OA conversion by adsorbing the by- according to FCCD (Table 2).
product water. Moreover, at optimal values of the process 2FI model has been the best fitted model to represent
factors, in presence of desiccants (SG or MS), efficacy of the parametric effects on γOA compared to the linear
a far-infrared radiated reactor (FIRRR) system (50 Hz, and quadratic models as assessed from statistical
100 W, 220 V, frequency 3 THz–300 GHz, wavelength ‘Model Fit Summary’ (Table 3). Here, 2FI model
100 µm–1 mm, energy 12.4 meV–1.24 meV) on OA exhibits superior statistical criteria than the linear or
conversion was also investigated. quadratic models in terms of the standard deviation,
R2, predicted-R2 and adjusted-R2. The 2FI model
equation that correlates γOA and process factors in
Conversion analysis coded form is given by Eqn (3) and that in terms of
actual factors is given by Eqn (4).
Through the titration process,[20] the concentration of Analysis of variance (ANOVA) depicting F-Test
OA (Eqn (1)) in the product MO was analysed. From (Table 4) suggests that a two-factor interaction (2FI)
© 2015 Curtin University of Technology and John Wiley & Sons, Ltd. Asia-Pac. J. Chem. Eng. (2015)
DOI: 10.1002/apj
P. PRADHAN AND R. CHAKRABORTY Asia-Pacific Journal of Chemical Engineering
Table 3. Model summary statistics of linear, 2FI and quadratic model formulated for predicting OA conversion to
methyl oleate using different Ni―Fe/FA catalysts.

Source Prob > F Stander deviation R2 Adjusted R2 Predicted R2 Press value

Linear <0.0001 1.17 0.9239 0.9049 0.8292 36.66
2FI 0.0031 0.64 0.9826 0.9709 0.9528 10.13
Quadratic 0.2451 0.57 0.9909 0.9771 0.9300 15.01

Table 4. ANOVA results of OA conversion corresponding to Eqn (3).

Source Sum of squares df Mean square F value p-Value Probability p > F

Model 210.84 6 35.14 84.54 <0.0001
qMR 0.00324 1 0.00324 0.00795 0.9316
qL 198.03 1 198.03 476.43 <0.0001
qC 0.22 1 0.22 0.54 0.4806
qMRqL 7.13 1 7.13 17.14 0.0025
qMRqC 5.46 1 5.46 13.14 0.0055
qLqC 0.003613 1 0.003613 0.008691 0.9278
Residual 3.74 9 0.42
Lack of fit 3.71 8 0.46 16.11 0.1905
Pure error 0.029 1 0.029
Corrected total 214.58 15

model with a p-value (probability of error value) of process conditions viz. M/OA molar ratio (QMR) of
<0.0001 was significant to predict γOA. 15:1, NI-15 Fe2 (SO4)3·H2O loading (QL) and 4 wt.%

γOA ¼ 92:42
0:018qMR þ 4:45qL
0:15qC þ 0:94qMR qL þ 0:83qMR qC
0:021qL qC

(3)
R2 ¼ 0:9826 ð
1 < qMR ; qL ; qC < þ1Þ

γOA ¼ 92:42250
0:018000QMR þ 4:45000QL of catalyst concentration (QC). MO synthesis was

0:15000QC þ 0:94375QMR QL performed in triplicate at the predicted optimum
conditions for an experimental 98.37 ± 0.01% OA
þ 0:82625QMR QC
0:021250QL QC (4)

Influence of individual parametric effect on OA

conversion

In perturbation plot (Fig. 1), γOA is plotted as a function

of one factor over its range keeping other factors
unchanged. The perturbation plot revealed that qL have
positive influence on γOA; on the other hand qMR and qC
have negative effects; these inferences corroborate with
the model equation (Eqn (3)). The increase in ferric
sulphate loading resulted greater acidity of the catalyst,
thus, augmenting the conversion of OA to MO.

Optimal process conditions

Figure 1. Influence of main (individual) effects on Oleic
Maximum predicted OA conversion (98.43%) was acid conversion: (A) methanol/OA molar ratio; (B) Fe2
determined through RSM corresponding to the optimal (SO4)3·H2O loading; (C) catalyst concentration.
© 2015 Curtin University of Technology and John Wiley & Sons, Ltd. Asia-Pac. J. Chem. Eng. (2015)
DOI: 10.1002/apj
Asia-Pacific Journal of Chemical Engineering NI―FE/FLY ASH CATALYST: DIESEL ADDITIVE

conversion. It shows approximately 0.06% differences Characterization of Ni―Fe/FA catalyst

between predicted and experimental OA conversion.
The optimum OA conversion in the present study is Acidity analyses
greater than that obtained (97.71%) through Titration method[19,21,22] was used for the acidity
homogeneous H2SO4 catalyst.[10] Besides, the analysis of Ni―Fe/FA catalyst samples, i.e. NI-5, NI-
maximum conversion of OA achieved in the present 10 and NI-15. With increase in the ferric sulphate
study is remarkably higher than that reported (93.0%) loading, the acidity was found to increase gradually
in our previous work[16] while using biological- from 2.4 to 5.5 mmol KOH/g catalyst corresponding
hydroxyapatite supported Fe catalyst. to NI-5 to NI-15.
Careful observations of the contour and 3-D plots
(Fig. 2(A–B)) indicate that γOA increases monotonically
with increase in QL. At lower QL, QMR has positive XRD analysis
effect, while a relatively weaker effect of QMR on γOA The intense difractrogram enunciates the presence of
is observed at higher QMR corresponding to higher QL Fe2O3 (card no. 33-0664) crystalline phase at scanning
keeping QC at optimal level. Excess of M (high QMR) angles 2θ = 24.29°, 33.26°, 35.78°, 43.52°, 49.61°,
provides negative effect on γOA at fixed acidity of the 54.17°, 57.47°, 62.51°, 72.20°, 71.78° and 72.20°. It
catalyst(i.e. fixed QL) as second order esterification can be well depicted from the Fig. 3c that because of
mechanism changes into pseudo-first order reaction increase in ferric sulphate loading, prominence of
mechanism. At high QMR, M molecule may obstruct Fe2O3 peaks gradually increases from NI-5 to NI-15.
the surface chemical adsorption of OA on the Presence of less intense peaks of Al2SiO5(card no.
Ni―Fe/FA catalyst and thus, rendering reduced 11-0046) (2θ = 28.16°, 30.38°, 37.88°, 48.35°, 55.79°
conversion of OA. At both the lowest and highest and 64.13°) (Fig. 3a–c) in NI-15 compared to NI-5
M/OA molar ratios, an increase in catalyst attributes to the fact that Al2SiO5 renders basic
concentration could enhance the OA conversion characteristics to the catalyst sample. In other words,
because of the presence of augmented catalyst active decrease in Al2SiO5 crystalline phase increases the
sites for the esterification reaction. acidity of NI-15; thus, rendering higher yield of methyl
oleate. Figure 3 also represents the presence of Ni2O3
(card no. 14-0481) and NiS (card no. 02-1273)
crystalline phases highlighted at 27.65°, 39.35°,
44.99°, 51.86°, 66.59° and at 29.96°, 34.22°, 53.03°,
78.32° respectively, which corroborate the fact that
gradual increase in intensity of Ni2O3 and NiS peaks

Figure 3. X-ray diffraction (XRD) patterns of fly ash supported

Figure 2. Catalyst loading versus methanol/OA molar Ni–Fe catalyst prepared at different iron loading ratio (a) NI-5,
ratio at optimal catalyst concentration for predicted C (b) NI-10 and (c) NI-15 [Characteristic peaks due to Fe2O3
(A) Contour plot and (B) 3-D plot. This figure is available ( ), Al2SiO5 ( ), Ni2O3 ( ) and NiS ( )]. This figure is
in colour online at www.apjChemEng.com. available in colour online at www.apjChemEng.com.
© 2015 Curtin University of Technology and John Wiley & Sons, Ltd. Asia-Pac. J. Chem. Eng. (2015)
DOI: 10.1002/apj
P. PRADHAN AND R. CHAKRABORTY Asia-Pacific Journal of Chemical Engineering

from NI-5 to NI-15 signified higher acidity of the NI-

15 catalyst sample.
With increase in ferric sulphate loading from 5% to
15% (Fig. 3(a–c)); crystallite size of the corresponding
catalyst was found to increase progressively. Using
Bragg’s law and Scherrer’s equation, lattice parameter
and crystallite size of the specific peaks of Fe2O3,
Ni2O3 and NiS (Table 5) were evaluated in concurrence
with the XRD analysis.[23]

SEM and EDX analyses

The morphology of the unevenly porous and granular
NI-15 uncalcined catalyst sample has been exhibited
through SEM image (Fig. 4A). Upon calcination, the
catalyst particles became agglomerated and flattened
(Fig. 4B) with non-uniform pores distributed over catalyst
surfaces. Existing elements and the corresponding weight
% and atomic % are presented in the EDX analyses for
the optimal catalyst, i.e. NI-15 (Table 6).

BET and BJH analyses Figure 4. SEM images of the NI-15 catalyst: (A)
The BET specific surface area of the NI-15 optimal before calcination and (B) after calcination at
catalyst has been determined as 27.64 m2/g indicating a 700 °C.
substantial increase in BET area compared to the FA
support (0.60 m2/g). The BET plot of the NI-15 catalyst Table 6. EDX analyses of NI-15 catalyst.
characterizes type II isotherm (IUPAC) (Fig. 5A) implying
unrestricted monolayer-multilayer adsorption.[24] The Elements present in NI-15 Weight% Atomic%
catalyst pore volume of 0.0233 cc/g (0.005 cc/g for FA) O 46.29 68.83
and a modal pore diameter of 1.5 nm (1.8 nm for FA) were Al 1.43 1.26
determined using BJH method (inset of Fig. 5A). The Si 2.37 2.01
penetration curve of the NI-15 catalyst sample (Fig. 5B) S 21.26 15.77
indicates presence of 11.21% micropores, 77.27% Fe 25.40 10.82
Ni 3.25 1.32
mesopores and 11.52% macropores in the NI-15
catalyst. Thus, together 88.79% of pore volume was
contributed by meso and macro-pores rendering high
conversion (98.37%) of OA to MO.
peaks at 1098.25 cm
1 representing Si―O―Si detected
FTIR analyses of FA and NI-15 catalyst for FA corroborate well with previous findings.[32,33]
FTIR analyses of the calcined-FA (Fig. 6a) and the The peak at 1643.83 cm
1 is ascribed to the vibration
prepared NI-15 (Fig. 6b) solid catalyst detect the peaks at of O―H stretching (δO
H) present in FA,[34] and weak
537.99, 554, 570.01, 602.02, 646.44 and 1108.92 cm
1 band of OH
ions is located around 2343.66 cm
1.[17]
of the octahedral and tetrahedral sites of Fe―O
deformation.[25–27] The peak located at 2000 cm
1 XPS analysis of NI-15 catalyst
represents the DMD (double-metal) Fe―Ni vibration[28] The XPS diagram of the NI-15 catalyst that demonstrated
and stretching vibration at 463.29 and 479.64 cm
1 the optimal catalytic performance for synthesis of methyl
represent Ni―O.[29,30] Asymmetric stretching vibration oleate is displayed in Fig. 7 (A and B). The XPS analysis
at 996.87 cm
1 of X―O (X = Si, Al) bonds,[27,31] and of NI-15 catalyst (Fig. 7A) detects the binding energies

Table 5. Crystallite size and lattice parameter of Fe2O3, Ni2O3 and NiS determined through XRD analysis.

Fe2O3 Ni2O3 NiS

Fig. Crystallite Lattice Crystallite Lattice Crystallite Lattice
no. 2θ (°) size (nm) parameter (nm) 2θ (°) size (nm) parameter (nm) 2θ (°) size (nm) parameter (nm)
3(c) 33.26 2.61 0.4779 27.65 3.92 0.3833 34.22 2.52 0.4645
3(b) 33.37 2.35 0.5305 28.25 3.36 0.4371 33.26 2.14 0.5831
3(a) 33.23 2.14 0.5836 27.68 3.23 0.5629 33.23 2.09 0.5912

© 2015 Curtin University of Technology and John Wiley & Sons, Ltd. Asia-Pac. J. Chem. Eng. (2015)
DOI: 10.1002/apj
Asia-Pacific Journal of Chemical Engineering NI―FE/FLY ASH CATALYST: DIESEL ADDITIVE

Figure 5. (A) BET and BJH (inset) plot and (B) penetration curve of NI-15
catalyst. This figure is available in colour online at www.apjChemEng.com.

binding energy is attributed to Fe 2p3/2 spin, while that at

higher binding energy is ascribed to Fe 2p1/2. In
concurrence with these observations, it might be inferred
that Fe3+species existing in the prepared NI-15 catalyst
rendered appreciable acidity (5.5 mol KOH/g) vis-à-vis
catalytic activity.
In Fig. 7B, the 854.8-eV XPS peak represents a mixed
Ni2+/Ni3+ phase of Ni compounds which corroborates
well with the findings of Grosvenor et al.[36] It may be
noted that the absolute binding energy values were
allowed to vary by ±0.1 eV to allow for error associated
with charge referencing to adventitious C 1s.
From the experimental result of MO synthesis (Table 2)
it may be inferred that an increase in Fe precursor, i.e.
Fe2 (SO4)3·H2O loading could improve the catalytic
performance. This signifies that Fe3+ detected in XPS
analysis was the major active phase in the OA
conversion to yield MO.

Effect of dessicant and heating protocol on

esterification
Figure 6. FTIR analyses of: (a) calcined FA and (b) NI-15 catalyst.
This figure is available in colour online at www.apjChemEng.com. SG and MS were used to improve OA conversion for
enhancing MO yield at the derived optimal reaction
conditions through removal of by-product water
pertaining to Fe2O3 through the characteristic XPS bands thereby reducing the extent of backward reaction
at 710.3 eV corresponding to Fe3+ 2p3/2 and at 724.1 eV significantly. In absence of any desiccant, 98.37% MO
in relation to Fe3+ 2p1/2.[35] The peak pertaining to lower conversion was achieved in 1h using CHR, whereas,

Figure 7. XPS spectra of (A) Fe 2p and (B) Ni 2p of NI-15 catalyst. This figure is
available in colour online at www.apjChemEng.com.
© 2015 Curtin University of Technology and John Wiley & Sons, Ltd. Asia-Pac. J. Chem. Eng. (2015)
DOI: 10.1002/apj
P. PRADHAN AND R. CHAKRABORTY Asia-Pacific Journal of Chemical Engineering

application of 40 wt.% of SG and MS helped to gain of unsaturates (cis-double bond stretching)in the ester
99.1% and 99.2% conversions (Fig. 8A) respectively product.[39]
at otherwise identical conditions. From Fig. 8 A–B, it
is evident that MS rendered better performance for both Properties of B10 fuel derived from product MO
CHR and FIRRR systems in comparison with SG.
Notably, at optimal condition, changing the reactor The product methyl oleate obtained at optimal condition
system from CHR (97.36%) to FIRRR (99.20%) in (using FIRRR) was blended with commercial petro-
the presence of MS (Table 7), resulted significant diesel in 1:9 volumetric ratio to prepare B10 diesel fuel
enhancement in OA conversion within only 10 min. and tested to evaluate the pertinent fuel properties
Table 7 presents some interesting experimental results (Table 8). The results indicate that fuel properties of
demonstrating that FIRRR system incorporating MS B10 satisfied ASTM D7467-10 specifications.
desiccants could provide the highest conversion in
minimum time and hence, rendered best performance Catalyst reusability
while consuming lower energy (because FIRRR had a
power rating of 100 W, i.e. 1/3rd power of the CHR). The reusability of the NI-15 catalyst was tested by
carrying out the esterification reactions using the
recycled catalyst several times at the optimal process
FTIR analyses of optimally produced methyl conditions. At optimal conditions, 1h esterification
oleate resulted same conversion for eight successive batches,
however, in ninth run OA conversion was found to
Infrared spectra of optimally synthesized MO are decrease from 98.37% to 97.76%. The decrease in
depicted in Fig. 9. The peaks at 1285,1381.05 and conversion in the ninth batch might be ascribed to
1413.06 cm
1 show the presence of CH3 group which negligible leaching of the active metal species into the
corroborated well with the findings of Chen Man reaction medium as confirmed form ICP analyses.
et al.[37] Vibrations of methylene (―CH2) group are The NI-15 catalyst could be regenerated by methanol
detected at 1466.42, 2855.77 and 2925.31 cm
1. The washing followed by air drying for 2 h at 105 °C in
intense peaks of ester carbonyl group(C¼O vibrations) the hot air oven. Afterwards, the dried sample was
at 1711.84 cm
1 and that of C―O vibrations around calcined at 700 °C for 4 h, and the regenerated catalyst
1124.93 and 1247.65 cm
1 are evident.[38] A weak could render approximately 97.5% OA conversion for
band at 3005.53 cm
1 signify insignificant presence additional five batches.

Figure 8. Effect of heating system on OA conversion: (A) conductive heated

reactor (CHR), (B) far-infrared radiated reactor (FIRRR). This figure is available
in colour online at www.apjChemEng.com.

Table 7. Effect of different types of desiccants (SG and MS) used in CHR and FIRRR on OA esterificaton.

Reaction time (min) OA conversion (%) using CHR OA conversion (%) using FIRRR
In the presence of SG In the presence of MS In the presence of SG In the presence of MS
10 96.49 97.36 98.56 99.20
20 97.83 98.43 99.11 99.83
30 98.71 98.82 99.56 99.83
40 98.90 99.02 99.74 99.83
50 99.04 99.11 99.74 99.83
60 99.10 99.20 99.74 99.83

© 2015 Curtin University of Technology and John Wiley & Sons, Ltd. Asia-Pac. J. Chem. Eng. (2015)
DOI: 10.1002/apj
Asia-Pacific Journal of Chemical Engineering NI―FE/FLY ASH CATALYST: DIESEL ADDITIVE

BJH Barret, Joyner, Halenda

SEM Scanning electron microscope
FTIR Fourier Transform Infrared
XPS X-ray photoelectron spectroscopy
EDX Energy-dispersiveX-rayspectroscopy
OA Oleic acid
M Methanol
MO Methyl oleate
RSM Response surface methodology
CHR Conductive heated reactor
FIRRR Far-infrared radiated reactor
MS Molecular sieve
SG Silica gel
Figure 9. FTIR analyses of product methyl oleate.
QMR M to OA molar ratio
QL Fe2 (SO4)3·H2O loading
QC Catalyst concentration
qMR Coded value of methanol to
Table 8. Properties of B10 fuel (10% methyl oleate oleic acid molar ratio
and 90% petro-diesel). qL Coded value of catalyst loading
Properties Specification Result qC Coded value of catalyst concentration
COA Concentration of OA
B6-B20 (ASTM B10 (ASTM
Standard number D7467-10) D7467-10) 56.11 Molecular weight of potassium
hydroxide
Specific gravity 0.88 0.85 N Normality of the potassium
(g/mL) hydroxide solution
Flash point (K) 52 (min) 105
Cetane number 40 (min) 50 V Volume of the KOH solution
Cloud point (K) Reporta
1 consumed from the burette
Pour point (K)
3
13 m Weight of the ester layer (oleic
Kinematic viscosity 1.9–4.1 2.41 acid taken for titration)
(40 °C, mm2/s)
γOA Conversion of oleic acid (%)
a
Cloud point is dependent on several factors, viz. place and time of AVOA Acid value of oleic acid (200)
year, and is not specified in ASTM D7467-10.

REFERENCES
CONCLUSION [1] Ojha K, Pradhan NC. Treated fly ash: a potential catalyst
for catalytic cracking. Indian J. Eng. Mater. Sci., 2001; 8,
Fly ash, a waste generated from coal based thermal power 100–103.
[2] Jain D, Khatri C, Rani A. Fly ash supported calcium oxide as
plant has been efficiently employed as a support for recyclable solid base catalyst for Knoevenagel condensation
development of Ni―Fe solid acid catalyst. The catalyst reaction. Fuel Process. Technol., 2010; 91, 1015–1021.
demonstrated superior performance in OA-methanol [3] Khatri C, Rani A. Synthesis of nano-crystalline solid acid
catalyst from fly ash and its catalytic performance. Fuel,
esterification in comparison with homogeneous H2SO4 2008; 87, 2886–2892.
catalyst and hydroxyapatite supported Fe solid acid [4] Khatri C, Jain D, Rani A. Fly ash-supported cerium triflate as
catalyst. The prepared FA supported Ni―Fe catalyst an active recyclable solid acid catalyst for Friedel–Crafts
acylation reaction. Fuel, 2010; 89, 3853–3859.
displayed appreciable reusability attributes. The [5] Khatri C, Mishra MK, Rani A. Synthesis and characterization
application of far-infrared radiation exhibited energy- of fly ash supported sulfated zirconia catalyst for benzylation
efficient, appreciable enhancement of OA conversion reactions. Fuel Process. Technol., 2010; 91(10), 1288–1295.
[6] Jain D, Khatri C, Rani A. Synthesis and characterization of
compared to conventional heating protocol. Thus, an novel solid base catalyst from fly ash. Fuel, 2011; 90(6),
efficient, cost-effective and reusable solid catalyst could 2083–2088.
be developed using FA, paving a greener pathway for [7] Chakraborty R, Bepari S., Banerjee A. Transesterification of
soybean oil catalyzed by fly ash and egg shell derived solid
effective management of such abundant industrial waste. catalysts. Chem. Eng. J., 2010; 165, 798–805.
[8] Kotwal MS, Niphadkar PS, Deshpande SS, Bokade VV,
Joshi PN. Transesterification of sunflower oil catalyzed by
NOMENCLATURE: fly ashbased solid catalysts. Fuel, 2009; 88, 1773–17789.
[9] Knothe G. “Designer” biodiesel: optimizing fatty ester
FA Fly ash composition to improve fuel properties. Energ. Fuel., 2008;
22, 1358–1364.
XRD X-ray diffraction [10] Kusmiyati SA. Production of biodiesel from oleic acid and
BET Brunner–Emmett–Teller methanol by reactive distillation. BCREC, 2010; 5, 1–6.
© 2015 Curtin University of Technology and John Wiley & Sons, Ltd. Asia-Pac. J. Chem. Eng. (2015)
DOI: 10.1002/apj
P. PRADHAN AND R. CHAKRABORTY Asia-Pacific Journal of Chemical Engineering
[11] Atalay FS. Kinetics of the esterification reaction between determination of surface area and porosity. Pure Appl. Chem.,
ethanol and acetic acid. Asia-Pac. J. Chem. Eng., 1994; 2, 1985; 57, 603–619.
181–184. [25] Mahmoudi M, Simchi A, Imani M, Häfeli UO.
[12] Yan K, Wu G, Lafleur T, Jarvis C. Production, properties and Superparamagnetic iron oxide nanoparticles with rigid
catalytic hydrogenation of furfural to fuel additives and cross-linked polyethylene glycol fumarate coating for
value-added chemicals. Renew. Sust. Energ. Rev., 2014; 38, application in imaging and drug delivery. J. Phys. Chem. C,
663–676. 2009; 113, 8124–8131.
[13] Chakraborty R, RoyChowdhury D. Fish bone derived natural [26] Fan Y, Zhang FS, Feng Y. An effective adsorbent developed
hydroxyapatite-supported copper acid catalyst: Taguchi from municipal solid waste and coal co-combustion ash for
optimization of semibatcholeic acid esterification. Chem. As (V) removal from aqueous solution. J. Hazard. Mater.,
Eng. J., 2013; 215–216, 491–499. 2008; 159, 313–318.
[14] Lopez DE, Suwannakarn K, Bruce DA, Goodwin Jr JG. [27] Li Y, Zhang FS, Xiu FR. Arsenic (V) removal from aqueous
Esterification and transesterification on tungstated zirconia: system using adsorbent developed from a high iron-containing
effect of calcination temperature. J. Catal., 2007; 247, 43–50. fly ash. Sci. Total Environ., 2009; 407, 5780–5786.
[15] Chakraborty R, RoyChowdhury D. Optimization of [28] Sreeprasanth PS, Srivastava R, Srinivas D, Ratnasamy P.
biological-hydroxyapatite supported iron catalyzed methyl Hydrophobic, solid acid catalysts for production of biofuels
oleate synthesis using response surface methodology. and lubricants. Appl. Catal., A: Gen., 2006; 314, 148–159.
J. Taiwan Inst. Chem. Eng., 2014; 45, 92–100. [29] Biju V, Khadar MA. Fourier transform infrared spectroscopy
[16] Chen F, Kawasaki J , Naka Y. Computational studies of the study of nanostructured nickel oxide. Spectrochim. Acta. Part
countercurrent multistage extraction-coupled esterification A, 2003; 59, 121–134.
process of oleic acid with methanol using excess methanol [30] Chakraborty R, Das SK. Optimization of biodiesel synthesis
as extractant. Chem. Eng. Res. Des., 2004; 82, 599–604. from waste frying soybean oil using fish scale-supported Ni
[17] Mekala M, Goli VR . Comparative kinetics of esterification catalyst. Ind. Eng. Chem. Res., 2012; 51, 8404–8414.
of methanol–acetic acid in the presence of liquid and solid [31] El-Naggar MR, El-Kamash AM, El-Dessouky MI, Ghonaim
catalysts. Asia-Pac. J. Chem. Eng., 2014; 9, 791–799. AK. Two-step method for preparation of NaA–X zeolite
[18] Yan K, Wu G, Wen J, Chen A.. One-step synthesis of blend from fly ash for removal of cesium ions. J. Hazard.
mesoporous H4SiW12O40-SiO2 catalysts for the production Mater., 2008; 154, 963–72.
of methyl and ethyl levulinate biodiesel. Catal. Commun., [32] Jain D, Khatri C, Rani A. Synthesis and characterization of
2013; 34, 58–63. novel solid base catalyst from fly ash. Fuel, 2011; 90(6),
[19] Singh AK, Fernando SD. Transesterification of soyabean 2083–2088.
oil using heterogeneous catalysts. Energ. Fuel., 2008; 22, [33] Liping Y, Hanning X, Yin C. Influence of magnesia on the
2067–2069. structure and properties of MgO–Al2O3–SiO2–F–glass
[20] Tesser R, Casale L, Verde D, Serio MD, Santacesaria E. ceramics. Ceram. Intern., 2008; 34, 63–68.
Kinetics of free fatty acids esterification: batch and loop [34] Jain D, Rani A. MgO enriched coal fly ash as highly active
reactor modeling. Chem. Eng. J., 2009; 154, 25–33. heterogeneous base catalyst for Claisen–Schmidt
[21] Chakraborty R, Das S, Pradhan P, Mukhopadhyay P. condensation reaction. ACSj, 2011; 1(2), 37–49.
Prediction of optimal conditions in the methanolysis of [35] Liu F, He H, Lian Z, Shan W, Xie L, Asakura K, Yang W,
mustard oil for biodiesel production using cost-effective Deng H. Highly dispersed iron vanadate catalyst supported
Mg-solid catalysts. Ind. Eng. Chem. Res., 2014; 53(51), on TiO2 for the selective catalytic reduction of NOx with
19681–19689. NH3. J. Catal., 2013; 307, 340–351.
[22] Chakraborty R, Das S, Bhattacharjee SK. Optimization of [36] Grosvenor AP, Biesinger MC, Smart RStC, McIntyre NS.
biodiesel production from Indian mustard oil by biological New interpretations of XPS spectra of nickel metal and
tri-calcium phosphate catalyst derived from turkey bone oxides. Surf. Sci., 2006; 600, 1771–1779.
ash. Clean Techn. Environ. Policy., 2015; 17, 455–463. [37] Che Man YB, Setiowaty G. Application of Fourier transform
[23] Lan L, Chen S, Zhao M, Gong M, Chen Y. The effect of infrared spectroscopy to determine free fatty acid contents in
synthesis method on the properties and catalytic performance palm olein. Food Chem., 1999; 66, 109–114.
of Pd/Ce0.5Zr0.5O2-AI2O3 three-way catalyst. J. Mol. Catal. [38] Knothe G. Analyzing biodiesel: standers and others methods:
A-Chem., 2014; 394, 10–21. a review. J. Am. Oil Chem. Soc., 2006; 83, 823–833.
[24] Sing KSW, Everett DH, Haul RAW, Moscou L, Pierotti RA, [39] Rohman A, Che Man YB. Fourier transform infrared (FTIR)
Rouquerol J, Siemieniewska T. Reporting physisorption data spectroscopy for analysis of extra virgin olive oil adulterated
for gas/solid systems with special reference to the with palm oil. Food Res. Int., 2010; 43, 886–892.

© 2015 Curtin University of Technology and John Wiley & Sons, Ltd. Asia-Pac. J. Chem. Eng. (2015)
DOI: 10.1002/apj