Chapter 4

Chapter 4

4B – Demonstrating photocatalytic esterification as

a potential strategy to improve the properties of
feedstock oil derived from dairy waste scum for
biodiesel production
Abstract

The availability of quality feedstock for biodiesel production is challenging; and therefore,
techniques to improve the quality of feedstock are highly desired, as they could be utilized to
modify any feedstock and enhance the biodiesel production. Dairy waste scum oil (DWSO)
is recently known to be a valuable feedstock for producing biodiesel. However, DWSO needs
to go through an esterification process to reduce its high free fatty acid (FFA) content. To do
this, herein, we have treated the DWSO using photocatalytic tin oxide (SnO2) particles under
UV light irradiation and reduced the FFA content significantly compared to the conventional
cumbersome acid-mediated esterification process. The synthesized catalyst is characterized
for its structural, morphological, optical and surface properties using various characterization
techniques. Parameters such as catalysts dosage, methanol to oil ratio, reaction time, and
temperature involved in the photo-assisted esterification processes are optimized, and a
mechanism for the photocatalytic esterification reaction is also proposed. Importantly, SnO2
particles also showed excellent reusability in the photo-assisted esterification process. Overall,
the physiochemical and fuel properties of the photo-esterified DWSO-derived biodiesel are
found to be improved compared to the biodiesel synthesized via conventional acid-esterified
oil.
Keywords: Biodiesel; Feedstock; Free fatty acids; Esterification; Transesterification;
Photocatalysis.

1
 Chapter 4

5.1. Introduction

Biodiesel is a renewable resource-derived fuel and can be a potential substitute for fossil fuels.
However, the availability of ‘quality’ feedstock is challenging, and the price of such quality
feedstock has a greater impact on the overall cost of biodiesel production [1, 2]. Enormous
research on finding quality and cheaper alternative feedstock is underway at many levels. One
of them is the use of waste oils, which can reduce the overall production cost and also improve
the yield of biodiesel, showing prospects of long-term viability as a quality feedstock [3].
These waste oils include used cooking oil and brown grease (collected from waste water
treatment plants), which typically contain more than 15% free fatty acid (FFA) contents [4-
6]. On the other hand, the homogeneous catalysts namely KOH and NaOH are largely used in
biodiesel production[7], but they are not often effectively suitable for reducing FFA in
feedstock, leading to saponification reactions rather than biodiesel formation [8, 9]. Therefore,
a pre-treatment step (i.e., esterification process) is commonly performed to reduce the FFA
content of the oil using the homogeneous acid catalysts such as H2SO4 and HCl [10]. However,
this process requires high pressure (~1.7 to 2 bar), temperature (> 100 oC), higher methanol
to oil molar ratios, strong acids, and longer reaction times due to its slower reaction rate [11,
12]. Also, it is essential to use expensive and sophisticated equipment to accelerate the
reaction and mitigate other associated issues such as corrosion, stability, etc. [13, 14]. In this
direction, the supercritical transesterification and enzyme catalyzed processes have been
reported as novel one-step techniques to make biodiesel from high FFA oils. However,
supercritical processes also require high temperatures (~300-400 oC), higher methanol to oil
ratio (~40:1), and high pressure (> 100 bar) to produce biodiesel [15]. Similarly, enzyme
catalyzed processes require longer reaction time (above 10 h), and its reusability is also not
possible. Hence, it is crucial to develop cost-effective processes for converting of high FFA
waste oils to biodiesel [16].
The production of biodiesel by catalytic esterification or transesterification using
heterogeneous catalysts is a very active area of research. Therefore, many heterogeneous
catalysts have been developed for biodiesel production using high FFA oils through different
methods. Recently, cation exchange resin (CER) was used in the esterification of high FFA
oleic acid and achieved around 91.63% oleic acid conversion by adopting the Taguchi
algorithm. The catalyst showed high activity over six consecutive catalytic cycles, indicating
its satisfactory catalytic stability [17].

2
 Chapter 4

The PEG/MgO/ZSM-5@Fe3O4 magnetic catalyst was synthesized using polyethylene glycol

(PEG) as a mediator and was used in microwave-assisted biodiesel [Link] PEG
content beneficial to the catalyst's higher basicity, larger pore size, and greater surface area
resulting in a 95.8% biodiesel yield [18]. Furthermore, a fixed-bed coupled pervaporation
(PV) technology was utilized in esterification reaction of oleic acid (OA) and achieved a
99.32% conversion yield at optimum conditions using response surface methodology. The
conversion rate of OA increased by 12.69% compared to the esterification reaction without
PV, and the cation exchange resin used in the fixed bed showed greater reusability [19].
Similarly, solid acid catalyst Zr4+-CER was used in the esterification reaction. The
esterification of tall oil fatty acids (TOFA) with ethanol dehydration was studied by fixed-bed
coupled PV and achieved around 99.51% conversion. After six esterification cycles, the
TOFA conversion of the catalyst reached around 90.03%, indicating that the catalyst had high
stability [20]. In addition, MgO/ZSM-5 catalysts [21] and mixed metal oxide of CaO-Al2O3
[22] were used as potential catalysts in biodiesel production.
In recent times, photocatalysts have gained more importance in the esterification of
oils under light radiations, and this motivated us to envision that this technique can evolve a
potential method to reduce FFA content in the feedstock. Eventually, it can be designated as
a feedstock modification technique. A sustainable method for producing biodiesel that uses
renewable resources and has less of an impact on the environment is photocatalytic
esterification. With this process, fatty acids and triglycerides are converted into biodiesel more
quickly by using photocatalysts in mild esterification reactions [23]. Photocatalysis is mainly
mediated through reduction-oxidation reactions, initiated by excited electron-hole pairs when
light of suitable wavelength/energy [24-26] is irradiated on the photocatalyst. Also, this
process can be conducted at atmospheric pressure and room temperature using direct sunlight
or other artificial light sources. It can hence be a viable technique to reduce FFA in the
feedstock and can be reused for several cycles as well, eventually reducing the overall cost of
biodiesel production [27-29].
Currently, very limited heterogeneous photocatalysts have been reported for the
esterification of FFA to produce biodiesel. The ZnO/SiO2 heterogeneous photocatalyst
showed high activity in the esterification of FFA with methanol under UV irradiation, and the
catalyst’s activity remained unchanged after being used 10 times [30]. Furthermore, the
Cr/SiO2 photocatalyst calcined between 700-1000 oC showed greater photocatalytic activity
in the esterification of high FFA oil. The coexistence of Cr6+ and Cr3+ moieties in the
photocatalyst facilitated a wide range absorption of light energies and made it extremely

3
 Chapter 4

efficient for photocatalytic esterification of FFAs with methanol [31]. In 2021, Guo et al. [32]
prepared La3+/ZnO-TiO2 photocatalyst for the esterification of FFA with waste cooking oil
and reported around 96.14% FFA conversion at 4 wt.% catalyst dosage, 12:1 methanol to oil
ratio, 3 h UV irradiation at 35 oC temperature. This catalyst also showed good reusability up
to 5 cycles with an FFA conversion of 87%. Similarly, a TiO2-based composite with graphitic
carbon [33] [34] and CuO/ZnO composites have also been reported for photocatalytic
esterification reactions [35]. These reports demonstrated that the photocatalytic esterification
process is efficient, with high selectivity and reduced energy consumption compared to
traditional methods. Photocatalytic esterification shows promise for sustainable biodiesel
production, contributing to the development of greener and more efficient processes in the
renewable energy sector.
In this context, herein, SnO2 has been used as a potential photocatalyst for photo-
assisted esterification reactions. SnO2 possesses a band gap energy of ~3.6 eV with excitation
binding energy of 130 eV [36]. The UV light absorption property of this material is found to
be promising and similar to that of TiO2, where SnO2 also has outstanding electrical, optical,
mechanical, and photocatalytic properties [37]. In this direction, this study demonstrates the
synthesis of highly active, stable, and recyclable tin oxide (SnO2) photocatalyst through
solution combustion route and their photocatalytic esterification reaction towards the
degradation of FFA in the dairy waste scum-derived oil (DWSO) to produce biodiesel. The
properties of the prepared photocatalyst are characterized to understand their various
physiochemical properties. Then, a conventional transesterification reaction using KOH
catalyst is performed to produce high-quality biodiesel from the obtained photocatalysis-
assisted FFA-degraded DWSO feedstock. Parametric studies have also been conducted to
understand the influence of various parameters in the photocatalysis-assisted FFA degradation
in the DWSO feedstock.

5.2. Experimental
5.2.1 Materials
The dairy waste scum is collected from Karnataka Milk Federation (KMF), Tumakuru Milk
Union (TMU), Mallasandra, Karnataka, India. The collected dairy waste scum was heated to
100 oC to convert it into a liquid form. Then, the unwanted materials and semi-solid particles
present in the liquid were filtered and centrifuged to obtain pure dairy waste scum oil
(DWSO). The fatty acid composition of DWSO includes myristic acid (32.60%), capric acid
(1.60), lauric acid (3.39%), stearic acid (5%), pentadecyclic acid (1.20%), palmitic acid

4
 Chapter 4

(13.56%), elaidic acid (11.42%), oleic acid (29.61%), and oleic acid (29.61%) [38]. The
chemicals such as tin chloride, urea, calcium oxide, and methanol used in this study were
purchased from Fisher Scientifics, Mumbai, India.

5.2.2. Synthesis of SnO2

In the process of combustion synthesis, 2 g of SnCl2.2H2O was added to 20 mL of distilled
water to form a transparent solution. A magnetic stirrer was used to enhance reactivity and
homogeneity of the mixture and mixed for 30 min. While stirring, 2 mL of HNO3 was added
dropwise to form the tin nitrate solution, acting as an oxidizer. Subsequently, 4 g of urea and
3.3 g of CTAB were added to the solution and stirred for 30 min. The resulting solution was
then transferred into a crucible and heated in a muffle furnace at 650 ˚C for 2 h for calcination,
yielding the final product as depicted in Fig. 1. Based on the reactants, the formation of
Sn(OH)₂ and its transformation to SnO2 during the solution combustion process can be
represented as given in Eq. (1)-(3).
SnCl2 + 2HNO3 → Sn(NO3)2 + 2HCl (Eq. 1)
Sn(NO3)2 + 2CO(NH2)2 → Sn(OH)2 + 2NH4NO3 + CO2 (Eq. 2)
2Sn(OH)2 → SnO2 + 2H2O (Eq. 3)

Figure 5.1. Schematic representation of the synthesis of SnO2 via solution combustion
method.

5.3. Characterizations
The phase identification of SnO2 was conducted using Rigaku Ultima IV X-ray diffraction
(XRD) analysis with Cu-Kα (1.5418 Å) radiation. Morphology analysis of the sample was
carried out using field emission scanning electron microscopy (FE-SEM) with a JEOL JSM-
7100F instrument. Optical absorption properties of the prepared SnO2 were determined using
a UV-Vis-spectrometer (Shimadzu UV-1800), and Tauc’s plot was used to estimate the band
gap energy.
5
 Chapter 4

Specific surface area of the samples was estimated using the Brunner-Emmer-Teller (BET)
with as Smart Instruments, SMART SORB 92/93, SS 93/11/11-12 instrument, based on the
obtained N2 adsorption and desorption isotherms. FTIR analysis of the samples was conducted
using a Brucker Alpha–T FTIR spectrometer.

5.4. Determination of FFA in DWSO

Estimating the free fatty acid (FFA) content in the oil is crucial as it can lead to soap formation
instead of biodiesel [39]. To determine the FFA content of dairy waste scum oil (DWSO), the
acid value of DWSO was initially assessed. In this procedure, 10 g of DWSO was added to
50 mL of isopropyl alcohol in a conical flask and gently heated to approximately 35°C while
shaking. The solution was then titrated using a phenolphthalein indicator with 0.1 N KOH,
and the appearance of a light pink color indicated the endpoint. The acid value of DWSO was
calculated using Eq. 4, the % of FFA in the feedstock was estimated using Eq. 5, and the %
FFA degradation was calculated using Eq. 6.
V x N x 56.1
Acid value (KOH g-1) = (Eq. 4)
W
Acid value
%FFA = x 100 (Eq. 5)
2
Initial FFA−Final FFA
FFA degradation (%) = X 100 (Eq. 6)
Initial FFA

Here, N, V and W represent the normality of KOH, volume of alkali added (in mL), and
weight of the oil sample (in g), respectively. The constant 56.1 represents the equivalent mass
of KOH.

5.5. Photocatalysis-assisted feedstock modification through esterification reaction

In this procedure, 100 g of DWSO feedstock was taken in a three-neck flask equipped with a
reflux condenser, thermometer, and a magnetic condenser. Different oil to methanol molar
ratios (ranging from 1:6 to 1:14) and SnO2 (2.5 to 5 wt.%) were added and stirred at ~35 oC.
The setup was enclosed in a closed black box with a 30 W ultraviolet (UV) light (~254 nm,
0.5 mW/cm2). The reaction proceeded for up to 4 h at a constant stirring speed of 400 rpm.
After the reaction, the resulting mixture was filtered through a separating funnel, yielding two
layers: an upper layer containing methanol and water, and a lower layer consisting of catalyst,
unreacted triglycerides, and methyl ester. The lower layer, containing the catalyst, was
separated. Methanol was collected through distillation, and the water was removed by heating
the reaction mixture in a hot-air oven at 100 °C.

6
 Chapter 4

This resulting esterified DWSO, obtained through photocatalysis, was then sued for biodiesel
production, as discussed in the following section.

5.6. Transesterification of DWSO catalyzed by KOH

The transesterification was conducted after the esterification of DWSO (with acid value less
than 2% or FFA less than 4%). The esterified DWSO was taken in a three-neck flask with a
condenser and placed on magnetic stirrer, and the fixed transesterification reaction parameters
such as 4.5:1 molar ratio of methanol to oil, 75 minutes reaction time, 1.20 wt% KOH, 62 oC
temperature and 650 rpm were selected based on our previous work [40]. After the completion
of reaction, the formed biodiesel and glycerin were separated. The excess methanol was
recollected from the reaction mixture through distillation, and the resulting biodiesel was
treated with water to eliminate the soap and other residues, and then heated at 100 oC to
eliminate the water molecules present in the biodiesel. The biodiesel yield was calculated
using Eq. 7.
Mass of biodiesel produced
Biodiesel yield (%) = X 100 (Eq. 7)
Mass of oil taken

5.7. Results and discussion

5.7.1 Characterizations of the synthesized SnO2
5.7.2 Structural studies
The XRD pattern of the synthesized SnO2 is presented in Figure 5.2(a). The obtained XRD
pattern matched with the JCPDS card No. 41-1445, which indicates that the obtained tin oxide
is crystalized into a tetragonal rutile phase of SnO2. The observed shift in the XRD peaks with
respect to the standard could be attributed to the presence of strain in the system due to the
smaller crystallite size of the particles, which can also be correlated with the observed peak
broadening in the XRD peaks. Accordingly, the average crystallite size of the synthesized
SnO2 estimated using Scherrer’s formula [41] is found to be ~9.63 nm, which is in agreement
with the values reported by Kumar et al.[42]. Clear and distinct peaks indicate good
crystallinity, and the absence of any other peaks suggests that the prepared SnO2 is free from
impurities and secondary phases. The Williamson and Hall (W-H) plot was obtained to
understand the association between grain size and micro-strain [35] in the system. The
obtained plot, as given in Figure 5.2(b), shows a low value of 1.98 × 10-2 corresponding to a
compressive micro-strain in the system with a respective crystallite size of around 10.6 nm,
consistent with the crystallite size estimated using the Scherrer's formula.

7
 Chapter 4

A similar observation was reported by Abdel-latief et al [43], where they reported the
synthesis of SnO2 via the hydrothermal route in the presence of cationic surfactant CTAB.
Consequently, a crystallite size of around 2.54 and 2.12 nm was estimated through the
Scherrer and Williamson-Hall plot, respectively. Furthermore, the micro-strain reported was
0.041. In another study, Amruta et al., [44] used the W-H equation to determine the lattice
parameters, strain, and effective particle size of SnO2 nanoparticles synthesized through the
co-precipitation method. They reported that intercept of the slop for pure tin oxide sample
exhibits the mean particle size and the micro strain in the systems.

(a) (b)

Figure 5.2. (a) XRD pattern and (b) W-H plot of the synthesized SnO2 particles.
The FTIR spectrum of the synthesized SnO2 NPs (Figure 5.3) displays bands at 542
and 660 cm-1 corresponding to the stretching and bending vibrations of Sn-O and O-Sn-O,
respectively. The presence of bands at these frequencies indicates that the synthesized SnO 2
is free from defects such as tin and oxygen vacancies. This ensures that the obtained SnO2 is
stable in terms of its structure and chemical composition, which can ultimately support durable
catalytic activity during successive reactions. The bands at 1425 and 1097 cm-1 can be
assigned to stretching O-H and C-H bonds, respectively. Typically, the bands in the range of
3200-3500 cm-1 are associated with O-H bending, indicating the presence of OH groups on
the SnO2 surface [45, 46]. However, in the current study, the band related to OH group is not
observed. This absence could be due to combustion and subsequent annealing processes that
removed the hydroxyl group and water molecules from the catalyst surface.

8
 Chapter 4

Figure 5.3. FTIR spectrum of the synthesized SnO2 system.

5.7.3. SEM, EDX and TEM analysis

The morphological features and elemental analysis of SnO2 particles are presented in Figure
5.4 (a-c). The obtained SEM and TEM images obtained show irregularly shaped micron-sized
particles with rough surface along with uneven distribution. The abrupt and uncontrolled
exothermic reactions during the combustion process leads to the formation of particles with
uneven distributions since there is no strict control over the size and shape of the particles.
However, due to the nucleation and subsequent particle growth occurring rapidly during the
combustion process, the particles tend be highly crystalline and robust in nature, as observed
from their structural analysis using XRD and IR techniques.
This a robust formation of particles also ensures the proper chemical composition of sample,
as observed in the EDS spectrum shown in Figure 5.4 (d). The presence of respective
elements such as Sn and O, along with their estimated atomic percentages (78.8% (Sn) and
21.2% (O)) (Figure 5.4 (d)), suggested that the formed particles are stoichiometric SnO2
without any impurities and secondary phases.

9
 Chapter 4

(a) (b)

50 µm 10 µm

(c) (d)

Figure 5.4. (a-b) SEM images at different scales, (c) TEM image and (d) EDS elemental
spectrum of the synthesized SnO2 particles.

5.7.4. BET surface property analysis

The specific surface area and pore size distribution of the synthesized SnO2 were determined

using BET N2 gas adsorption-desorption analysis, and the obtained isotherm and pore-size

distribution curves are shown in Figure 5.5 (a)-(b), respectively. The observed isotherm

indicates the existence of type IV hysteresis, suggesting the mesoporous nature of the

particles. Furthermore, the obtained Barrette-Joynere-Halenda (BJH) plot indicates that the

particles possess a multimodal pore structure with pore size ranging from 2 to 16 nm.

The surface area of SnO2 is estimated to be around 23.45 m2 g-1, which is relatively low due

to their larger particle size [38, 47]. Accordingly, the formation mechanism of mesoporous

SnO2 during calcination is schematically shown in Figure 5.6. This signifies that the particles

can establish a smooth adsorption-desorption process on their surface, which is essential for

the robust catalytic reactions with prolonged stability and activity.

10
 Chapter 4

Fig. 5.5. (a) BET N2 adsorption-desorption isotherm and (b) BJH pore size distribution plot

of the synthesized SnO2 particles.

Fig. 5.6. Formation mechanism of mesoporous structures in the synthesized SnO2 particles.

5.7.5 Optical analysis

The UV-visible absorption spectrum and Tauc’s plot for the band gap estimation of the
synthesized SnO2 particles are displayed in Figure 5.7(a)-(b), respectively. An absorption
maximum is observed around 313 nm, indicating the typical UV-light driven properties of the
SnO2 particles. However, the spectrum extends across the visible light region up to 700 nm,
suggesting that the SnO2 is active over a broad range of solar spectrum, which is consistent
with previous literature [48]. The observed visible light activity of SnO2 could be originated
due to the induced strain in the material owing to their smaller crystallites, which gives rise to
the localized active optical transitions in the system. The optical bandgap energy of SnO 2 is
calculated using Tauc’s plot as shown in Figure 5.7(b) and found to be ~3.42 eV, which is
less than the band gap energy of the typical bulk-SnO2 (3.6 eV). This decreased bandgap
energy could be attributed to the disorder-induced modification in the direct-bandgap structure
and anisotropic nano-crystallite size of the synthesized SnO2 particles [49, 50].

11
 Chapter 4

(a) (b)

3.42 eV

Figure 5.7. (a) UV-Visible absorption spectrum and (b) Tauc’s plot of the synthesized SnO2

particles.

5.8 Photocatalytic esterification-mediated degradation of FFA in feedstock: Effect of

reaction parameters
As discussed, the feedstock modification (i.e., the degradation of free fatty acid in the obtained
dairy waste-scum derived oil) was carried out through photocatalytic esterification using the
developed SnO2 particles under UV light irradiation. This investigation includes a parametric
study, where key parameters such as methanol-to-oil ratio, catalyst dosage, and UV irradiation
time are varied. These results are presented in the following sections.

5.8.1 Effect of methanol/oil ratio and catalyst dosage

The photocatalytic efficiency of SnO2 particles in degrading free fatty acids (FFA) was studied
by varying the methanol-to-oil (M/O) ratios from 6:1 to 14:1, while keeping the other
parameters constant at 100 mL of oil, 3.5 wt% of catalyst, 3 h of UV irradiation. The results
are presented in Figure 5.8(a).

(a) (b)

12
 Chapter 4

Figure 5.8. Parametric studies showing the (a) methanol-to-oil ratio and (b) catalyst
concentration dependent FFA degradation efficiency of the synthesized photocatalyst under
UV irradiation.
The FFA conversion rate increases with increasing M/O ratio, reaching a maximum
value of ~94.4% (1.2% FFA) at 12:1. However, beyond this molar ration, the conversion rate
decreases, which may be due to the reversible esterification reactions. Here, increasing the
M/O ratio improves reactant concentration, leading to faster reactions and higher conversion
rate [51, 52]. However, excessively high methanol volumes inhibit the reaction in the positive
direction, resulting decreased conversion rates. Therefore, the optimal M/O molar ratio is
determined to be 12:1.
Similarly, by increasing the dosage of SnO2 particles from 2.5 to 4 wt% (as shown in
Figure 5.8 (b)), while keeping the other parameters constant at 100 mL of oil, 12:1 M/O ratio,
3 h of UV irradiation, the FFA degradation increases from around 57.1 (9.3% FFA) to 96.1%
(0.81% FFA). However, with a further increase to 5 wt%, the FFA degradation decreases to
77.8% (4.8% FFA). It is known that the surface-active sites play crucial role in effective
catalytic reactions over heterogeneous catalysts. Accordingly, the observed decrease in FAA
degradation at higher catalyst concentrations may be attributed to the inhibition of active sites
due to increased particle-particle interactions rather than particle-reactant interactions in the
reaction medium. Also, photocatalytic reactions largely depend on effective redox reactions
mediated by radicals produced by photo-excited carriers. Their formation and transfer to
reactant molecules may be limited by particle agglomeration via particle-particle interactions
[53, 54]. Therefore, 4 wt% of catalyst was determined as the optimized concentration for this
reaction. Further, the catalytic activity of SnO2 (5 wt%) was conducted without UV
irradiation, resulting in around 63% degradation of FFA.

5.8.2 Effect of UV irradiation time

The impact of UV irradiation time on FFA degradation efficiency of SnO 2 was studied by
varying the irradiation time from 1 to 5 h, maintaining other parameters constant at 100 mL
of oil, a 12:1 M/O ratio, and 4 wt% of photocatalyst. In Figure 5.9(a), it is evident that without
UV irradiation, photocatalytic esterification of FFA at 35°C does not occur, with FFA
conversion remaining nearly zero even after a three-hour reaction without illumination. The
FFA degradation shows an increasing trend with irradiation time up to 3.5 h, reaching around
96.3% (0.8% FFA), but it starts to decline with further increases in irradiation time up to 5 h.

13
 Chapter 4

During photocatalytic reactions, activating the photocatalyst is crucial, and this activation is
time-dependent, closely linked to the response of the surface-active sites. The participation of
these sites increases with irradiation time from 1 to 3 h, peaking at 3.5 hours. Beyond this,
additional irradiation time does not enhance reactions as all surface-active sites are already
engaged, leading to a saturation point in FFA degradation beyond 3.5 h of irradiation.
Moreover, a conventional esterification reaction conducted at 65 °C also resulted in FFA
degradation, reaching a maximum of approximately 74.6% (5.5% FFA) after 5 h, as shown in
Figure 5.9(b). This highlights the effectiveness of the prepared SnO2 photocatalyst under UV
irradiation and its potential as a heterogeneous catalyst for reducing FFA.

(a) (b)

Figure 5.9. FFA degradation over SnO2 photocatalyst (a) with and without UV light
irradiation and (b) without light irradiation under heating at 65 °C.

5.8.3 Effect of reaction time

The effect of reaction time on FFA degradation was investigated by varying the reaction time
from 2 to 4.5 h while maintaining other reaction conditions at 4 wt% catalyst loading, 3.5 h
UV irradiation, and 12:1 methanol/oil ratio. The increase in FFA degradation was observed
with longer reaction times, as shown in Figure 5.10. Degradation of FFA increased and
reached 94.7% at 3.5 h, with no further increase in degradation observed with longer reaction
time. This could be due to the balance between reverse and forwards reactions [55, 56].
Therefore, the optimum reaction time was determined to be 3.5 h.

14
 Chapter 4

Figure 5.10 Effect of reaction time on FFA degradation by the synthesized SnO2 photocatalyst

5.9 Photo-esterification mechanism of SnO2 in FFA degradation

Photo-esterification for FFA degradation is mediated via the active radicals generated by the

reaction of excited charge carriers with the surrounding molecules. In a typical reaction, upon

the incident photons, the electrons are excited to the conduction band (CB) and holes are

generated in the valence band (VB) [Eq. (8)]. The electrons (e ̄) react with carboxylic

molecules (ROOH) and form their corresponding anions (RCOO• ̄) [Eq. (9)], while the holes

(h+) react with methanol (CH3OH) and produce anionic CH3O ̄ radicals and protons (H+) [Eq.

(10)]. Then, these formed RCOO• ̄ species react with CH3OH and produce RHCOO• [Eq.

(11)], which further react with CH3O• radicals and H+, forming the methyl ester (RCOOCH3)

and water (H2O) as shown in Eq. (12). The similar mechanism is demonstrated in Cr/SiO2

[31] and La3+/ZnO-TiO2 [32] photocatalysts in esterification. The overall mechanism is

depicted in Figure 5.11(a)-(b).

SnO2 + hν  SnO2 (e + h+) (Eq. 8)

SnO2 + (e-CB) + ROOHads  SnO2 + RCOO. (Eq. 9)

SnO2 (h+VB) + CH3OH  SnO2 + [Link] + H+ (Eq. 10)

RCOO. + CH3OH  RHCOO. + CH3O. (Eq. 11)

RHCOO. + CH3O. + H+  RCOOCH3 + H2O (Eq. 12)

15
 Chapter 4
RCOOH RCOO.

UV irradiation e- e - e-
CB

SnO2
photocatalyst

VB

h+ h+ h+

CH30H CH30-

Figure 5.11. (a) Mechanism of photocatalytic esterification over SnO2 and (b) reaction steps

in FFA degradation.

5.10 Conventional esterification of DWSO using H2SO4 and reusability of SnO2

photocatalyst

For comparison, the esterification reaction was conducted using the homogeneous acid

catalyst H2SO4. The reaction parameters were selected according to the literature [10]. The

obtained results (Figure 5.12(a)) showed that at the end of 1st (1 h) and 4th cycle (1 h), the

FFA content is decreased to ~16.25% (i.e. 25.34% degradation of FFA) and ~1.6% (92.62%

degradation of FFA), respectively, from the initial FFA content of 21.7% (i.e. 100% FAA). It

should be noted that the FFA content is decreased to ~1.02% (96.31% degradation of FFA) in

the first step (i.e., 3.5 h) of the photocatalytic reaction itself under the optimized conditions of

100 mL of oil, M/O ratio 12:1, 4 wt% catalyst, and 3.5 h of UV irradiation. Whilst, it took 4

repeated processes to decrease the FFA content to ~1.6% in the conventional acid-based

homogeneous catalytic process. Moreover, in each step, the obtained H2SO4-treated feedstock

should be thoroughly washed several times using water in order to avoid the sulfur residues

in the final biodiesel, which otherwise will impair the quality of biodiesel.

It is evident that the conventional acid-based homogeneous catalytic esterification

process can be relatively far expensive, time-consuming, and can adversely affect the quality

16
 Chapter 4

of biodiesel compared to the photocatalyst-assisted esterification process. Importantly, one of

the key advantages of heterogenous catalysts is their reusability. Accordingly, the synthesized
SnO2 photocatalyst was reused up to 5 cycles for FFA degradation, and the obtained results
are shown in Figure 5.12(b). Under the optimized conditions, the FFA degradation efficiency
of the photocatalyst remained consistent up to 3 cycles with an average of ~90.5% degradation
efficiency. However, it started decreasing to 86 and 84% in 4th and 5th cycle, respectively.
Even though the catalyst was washed with n-hexane and dried at 80 °C before starting the next
cycle, complete regeneration of catalyst is difficult during the recycles. This observed
decreased efficiency of the photocatalyst with cycles could be attributed to the catalyst loss,
strong interaction of triglycerides and by-products formed during the esterification reaction
with the surface of photocatalyst, significantly affect the reactive sites thus poisoning the
catalyst surface.
Furthermore, the recovered SnO2 photocatalyst was characterized using XRD and SEM
techniques. As depicted in Figure 5.12(c), the crystalline phase and structure of the recovered
photocatalyst remained unaltered compared to the fresh photocatalyst. Similarly, the SEM
image of the recycled photocatalyst (Fig. 12(d)) shows irregularly shaped micron-sized
particles with a relatively rough surface and uneven distribution of particles, emphasizing their
morphology and particulate structure are relatively stable compared to the fresh SnO2
photocatalyst. However, the observed slight decrease in catalytic activity can be attributed to
the deposition of organic materials on the catalyst surface, leading to the suppression of active
sites on the surface of the photocatalyst.

17
 Chapter 4

Figure 5.12. (a) FFA degradation of feedstock oil using the conventional homogenous acid-

based catalyst (H2SO4), (b) reusable efficiency of the photocatalyst toward FFA degradation,

(c) XRD and (d) SEM image of the recovered SnO2 photocatalyst.

5.11 Kinetic study

Typically, a pseudo-first-order kinetic model is utilized to describe the kinetics of

heterogeneously catalyzed esterification when the mole ratio of methanol to oil exceeds 5

[57]. To investigate the reaction kinetics of the SnO2 photocatalyst in the esterification

process, experiments were performed under optimized conditions. The kinetic equation for

the esterification reaction can be expressed as shown in Eq. 13.

RCOOH + CH3OH ↔ RCOOCH3 + H2O (Eq. 12)

The apparent reaction rate for the esterification reaction is given by Eq. 14.

𝑑𝐶𝐹𝐹𝐴
− = 𝑘1,𝑒𝑠𝑡 𝐶𝐹𝐹𝐴 𝐶𝑚𝑒𝑡ℎ𝑎𝑛𝑜𝑙 (Eq. 14)
𝑑𝑡

Where k1,est is the kinetic constant, C represents the concentration, and t represents the time.

18
 Chapter 4

[𝐶𝐹𝐹𝐴 (1−𝑥)]
−𝑑 = 𝑘1,𝑒𝑠𝑡 𝐶𝐹𝐹𝐴 (1 − 𝑥)𝐶𝑚𝑒𝑡ℎ𝑎𝑛𝑜𝑙 (Eq. 15)
𝑑𝑡

On cancelling the terms in Eq. 15, the following simplified differential Eq. 16 can be obtained.
𝑑𝑥
= 𝑘1,𝑒𝑠𝑡 𝐶𝑚𝑒𝑡ℎ𝑎𝑛𝑜𝑙 (1 − 𝑥) (Eq. 16)
𝑑𝑡

The concentration of the methanol is unchanged in the esterification reaction, and

therefore, the Cmethanol is known and the Eq. 17 is an analytical equation, which is obtained by
solving the differential equation of Eq 16.
− ln(1 − 𝑥) = 𝑘1,𝑒𝑠𝑡 𝐶𝑚𝑒𝑡ℎ𝑎𝑛𝑜𝑙 𝑡 (Eq. 17)
Since the molar ratio of methanol to oil exceeds 5 (i.e., 12:1), it can be assumed to
follow a pseudo-first-order reaction kinetic given the fixed concentration of catalyst. A plot
was created for –ln(1-x) vs ‘t’ at various reaction temperatures using the 4 wt.% catalyst.
Figure 5.13(a) shows the plot of –ln(1-x) vs ‘t’, where the slope of the linear plot curve is
utilized to determine the rate constant (k). The rate constant was determined to be 0.00296 k
(min-1), 0.00305 k (min-1), 0.00314 k (min-1), and 0.00325 k (min-1) at different temperatures
of 35 oC, 45 oC, 55 oC, and 65 oC, respectively.
The activation energy can be calculated by using the Arrhenius equation as given in Eq. 18.
𝐸𝑎
𝐾 = 𝐴𝑒𝑥𝑝 [− 𝑅𝑇] (Eq. 18)

By performing the logarithm of Eq. 18 on both sides, Eq. 19 can be obtained as,
𝐸𝑎
𝑙𝑛𝐾 = − 𝑅𝑇 + 𝑙𝑛𝐴 (Eq. 19)

The activation energy (Ea) can be determined by plotting lnK against 1/T, as shown
in Figure 5.13(b). Based on this plot, the activation energy for the SnO2 photocatalyst was
found to be ~33.82 kJ/mol for catalyzing the esterification of feedstock oil. This indicates that
the SnO2 photocatalyst relatively requires relatively lower activation energy for the
esterification, enabling the degradation of FFA content in the feedstock oil under mild reaction
conditions.

19
 Chapter 4

(a) (b)

Figure 5.13. (a) –ln(1-x) vs time at different temperatures and (b) Arrhenius plot of lnK vs

1/T.

5.12 Synthesis of biodiesel

The FFA-reduced esterified oil obtained through (i) photocatalysis-assisted and (ii)
conventional acid-based homogeneous catalysis-assisted process was further utilized to
produce biodiesel via conventional homogeneous transesterification using KOH, as described
in section 2.4.2. At the end of the reaction, biodiesel yields of around 94.6 and 93.7% were
achieved using the esterified oil from photocatalysis and conventional catalysis process,
respectively. It is important to note that the biodiesel yield of 93.7% corresponds to the oil
obtained after the 4th round of esterification via the conventional homogeneous catalysis
process, while the yield of 94.6% is achieved using the esterified oil obtained in a single step
photo-esterification reaction. Consequently, this photo-esterified-oil derived biodiesel was
further taken for its structural and physiochemical analysis using various technique as
discussed in the following sections.

5.13 Characterization of the biodiesel produced

The FTIR spectra of DWSO (oil) and DSME (biodiesel) are displayed in Figure 5.14. The

peaks at 3256, 2359, and 1914 cm-1 indicate the presence of alcohol, alkyne, and aromatic

compound, respectively in the esterified oil (DWSO). Conversely, the observed peak shift

from 1700 to 1740 cm-1, as well as the disappearance of the C-O stretch corresponding to

DWSO at 1700 cm-1 in the spectrum of DSME, clearly indicates the effective conversion of

oil into biodiesel.

20
 Chapter 4

The absence of carboxylic acid peaks between 2500-3300 cm-1 represents the absence of

moisture in the obtained biodiesel (DSME) [58, 59].

Figure 5.14. FTIR spectra of the photo-esterified dairy waste scum oil and its methyl ester.

5.14 Physicochemical properties of biodiesel

Various physicochemical features associated with fuel properties of the biodiesel produced
from photo- and conventional-esterified oil are analyzed and provided in Table 1 in
comparison with the ASTM standards. Notably, density and kinematic viscosity are important
fuel properties, where lower density and viscosity of biodiesel improve spray atomization and
flow ability in diesel engines [60, 61]. Accordingly, the density and viscosity of biodiesel
synthesized from photo- and conventional esterified feedstock were found to be comparable
with respect to the ASTM standards. The copper corrosion of biodiesel synthesized from both
photo- and conventionally esterified feedstock was found to be 1a, indicating that the fuel is
less corrosive. The calorific value of the biodiesel produced from photo- and conventionally
esterified feedstock is found to be relatively similar, showing the good combustibility of the
fuel. In addition, the cloud (28 oC) and pour point (21 oC) of DWSO are greater due to the
presence of saturated fatty acids in it [62]. After the transesterification of photo- and
conventional esterified feedstock, the cloud (7 oC) and pour point (3 oC) are reduced, and these
results are similar in both photo- and conventionally esterified feedstock biodiesel. The acid
value of photo- and conventional esterified feedstock biodiesel was determined to be ~0.09
mg KOH/g and ~0.11 mg KOH/g, respectively, which lies within the limit of the ASTM
standard of 0.5 mg KOH/g.

21
 Chapter 4

The flash point of biodiesel produced from photo-esterified feedstock (144 oC) was found to
be higher compared to the biodiesel produced from conventionally esterified feedstock (138
o
C). This decreased reduced flashpoint temperature can be attributed to the high quantity of
methanol used in the conventional esterification process to reduce FFA [63]. However,
properties of both the biodiesels are in the range of ASTM standards, which essentially means
that they are safe for storage and transportation. The carbon residue of photo-esterified
feedstock biodiesel was found to be less compared to conventionally esterified-feedstock
biodiesel and within the limits of ASTM standards, suggesting the effective burning of the
fuel without making any deposits in the engine [64]. The sulfur content in conventionally
esterified feedstock biodiesel (0.051% m/m) is higher compared to the photo-esterified
feedstock biodiesel (0.018% m/m). This is because a large quantity of sulfuric acid was used
in the esterification reaction to reduce FFA in the DWSO. This indicates that photo-esterified
biodiesel is less corrosive in nature and hence it can be safely stored. The acid value of photo-
esterified feedstock biodiesel (0.09 mg KOH/g) is lower than the conventionally esterified
feedstock (0.11 mg KOH/g). The higher acid value of the biodiesel results in the corrosion of
fuel injectors in the diesel engines [65]. Overall, the fuel properties of biodiesel from photo-
esterified feedstock, such as viscosity, flash point, sulfur content, ash content, and acid value,
are superior to those of conventionally esterified feedstock.
Table 1. Physicochemical properties of photo-esterified oil (DWSO) and the derived

biodiesel.

Biodiesel
Biodiesel
Bare from Biodiesel
from photo-
Properties Diesel Feedstock conventional standard
esterified
(DWSO) esterified (ASTM-6751)
feedstock
feedstock
Density at
834 900 880 880 870-900
15oC (kg/m3)
Viscosity at
2.6 32.1 4.41 4.48 1.9-6.0
40 oC (cSt)
Copperstrip
corrosion, 1a 1a 1a 1a no. 3 max
50 oC, 3h

22
 Chapter 4
Flash point
56 228 144 138 >130
(oC)
Ash content
Nd Nd 0.004 0.023 0.050
(% mass)
Sulphur Nd Nd 0.018 0.051 0.05 max
content
(% m/m)

Calorific
42800 37969 39970 39844 -----
value (kJ/Kg)

Carbon
residue (% Nd Nd Nil 0.011 0.020
mass)

Pour point
-10 21 3 3 ----
(oC)

Cloud point
-6 28 7 7 ----
(oC)

Acid value
Nd 10.85 0.09 0.11 0.5 max
(mg KOH/g)

Nd = Not determined, DWSO = dairy waste scum oil, FAME = Fatty acid methyl ester

23
 Chapter 4

5.15 Conclusion

In this study, tin oxide (SnO2) particles were synthesized via solution combustion method and
employed as a photocatalyst to degrade free fatty acid (FFA) in diary waste scum-derived oil
(DWSO), rendering it as a suitable high-quality feedstock for biodiesel production through
conventional homogeneous catalytic transesterification. The developed SnO2 particle
exhibited a wide band gap energy, indicating their UV light driven photocatalytic properties.
Through parametric studies by varying the catalyst concentration, methanol-to-oil ratio, light
irradiation time, and reaction time an optimized condition of 12:1, 4 wt%, 3.5 h, and 3.5 h
respectively was determined. This condition resulted in the degradation of around 96.31% of
FFA, leading to a final FFA concentration of 0.8% in the treated DWSO, compared to ~21.7%
in the untreated DWSO. In contrast, the conventional acid-treatment process using H2SO4 as
an acid catalyst degraded around 92.62% of FFA but required for consecutive esterification
processes. Furthermore, the stability of SnO2 allows for its recyclable usage in FFA
degradation in DWSO, maintaining over 80% efficiency up to 4 cycles. Finally, the biodiesel
produced using the SnO2 photocatalysts-treated DWSO met ASTM 6751 standards with
improved physiochemical and fuel properties. This study underscores that compared to the
conventional acid-assisted esterification process; the photocatalytic esterification can be an
effective alternative for degrading FFA in feedstock oil. It offers advantages such as reduced
process time, cost-effectiveness, and scalability for large-scale biodiesel production.

24