Chapter 3 : Hydrolysis of Vegetable Oils and Animal Fat

Chapter 3
Hydrolysis of Vegetable Oils and
and Animal Fat

Ph.D. Thesis, University of Pune Page 77

 Chapter 3 : Hydrolysis of Vegetable Oils and Animal Fat

3.1. Introduction
Biodiesel is getting more and more attention due to its renewability,
biodegradability, better lubricity and better emission properties. Depleting fossil fuels
and their escalating prices are also favouring the use of biodiesel. It is generally
prepared by transesterification of vegetable oils and fats in the presence of alkali
catalysts. For an alkali-based process, feedstock should be free from water and fatty
acids (<0.5 wt%) [1- 4]. This makes the biodiesel production costly and non-
competitive to petrodiesel as 70-80% of the biodiesel production cost depends upon
the feedstock used. An alternate, more generic process for the production of biodiesel
is the hydrolysis of waste oils/fats to fatty acids followed by their esterification to
fatty acid methyl esters, FAME (Fig. 3.1) [5]. Fatty acids produced from non-edible
oils and fats are more advantageous as they cannot only make biodiesel more
competitive but also avoid the food vs fuel issue. The hydrolysis of oils, prior to
saponification, provides possibilities of overcoming many problems associated with
the recovery of byproduct glycerol. Fatty acids and glycerol are valuable chemical
intermediates with a variety of end uses. Fatty acids (FA) are major components used
in the preparation of a wide variety of products, such as soaps, surfactants, lubricants,
plasticizers, paints, coatings, pharmaceuticals, foods, agricultural, industrial and
personal care products [6]. Yet quantitative information on the kinetics,
thermodynamics, and engineering aspects of hydrolysis of vegetable oils and fats is
limited.

Fig. 3.1. Two routes for biodiesel production

In the widely used Colgate-Emery steam hydrolysis process, the reaction is
conducted in continuous, counter-currents of fat and water (40–50 wt% of fat), in the
absence of a catalyst, at high temperatures (523–603 K) and pressures (50–60 bar) [7,

Ph.D. Thesis, University of Pune Page 78

 Chapter 3 : Hydrolysis of Vegetable Oils and Animal Fat

8]. This pressure splitting process is efficient and vegetable oil/fat splitting yields of
95% and above are obtained. However, its applicability to low-cost, waste vegetable
oils and animal fats especially those containing multiple/conjugated double bonds
and hydroxyl groups (castor oil and fish oil, for example) is limited. At high
temperatures, these triglycerides and the fatty acids derived from them undergo
undesired thermal decomposition leading to deterioration in colour/odour and to a
reduced yield of fatty acids [9]. Additional major drawbacks of the high temperature–
pressure fat splitting process include: (a) it is an energy intensive process, (b) it uses
considerable amount of superheated steam as a reagent, (c) it requires the use of large
reactors made of expensive corrosion-resistant material of construction, (d) the
quality of the product is poor and necessitates additional process steps to purify the
fatty acids and glycerol escalating the cost of the overall manufacturing process.
The Twitchell fat splitting process operates at milder conditions (373 K and 1
bar) but needs longer contact times (12–24 h) and fat splitting is only 80–85% [10].
Fat splitting via saponification produces significant amount of acid sludge and hard
water in the down-stream processing [11, 12]. The quality of glycerin by-product is
also poor. Enzymatic splitting of fat (308 K) requires long contact times (16 h to
several days) [13–15]. It is, therefore, desirable to have a more efficient hydrolysis
process of vegetable oils which operates at moderate temperatures/pressures and
produces higher yields of fatty acids and glycerin with minimum number of process
steps.
Solid acid catalysts are more beneficial than homogenous catalysts due to
engineering advantages and easy catalyst separation from the product stream. There
are not many reports in the journal or patent literature wherein solid catalysts have
been used successfully for the hydrolysis of oils and fats to fatty acids. Ngaosuwan et
al. [16] reported tungstated zirconia and Nafion resin nanoparticles supported on
mesoporous silica (SAC-13) catalysts for the hydrolysis of tricaprylin, a model
compound. This chapter describes application of different solid acid catalysts for the
hydrolysis of vegetable oils and fats as a means to ultimately lower the capital and
operating costs by conducting the reaction under moderately low reaction conditions.
Such a process could be applied in a novel two-step (hydrolysis-cum-esterification)
biodiesel synthesis route using low-cost feedstocks containing 5-15% FFAs. Since
the esterification of FFAs is faster than transesterification of triglycerides on acid
catalysts, there could be a possibility to construct a more efficient biodiesel synthesis

Ph.D. Thesis, University of Pune Page 79

 Chapter 3 : Hydrolysis of Vegetable Oils and Animal Fat

process around the use of two-step hydrolysis-cum-esterification route using solid

acid catalyst rather than what is currently done with homogeneous catalysts [17].

3.2. Experimental Section

3.2.1. Material Preparation

The Fe-Zn double-metal cyanide (DMC) complex, Co-Zn DMC complex and
ZnO catalyst were prepared as reported in Chapter 2. Hβ (SiO2/Al2O3 = 28) and HY
(SiO2/Al2O3 = 5) zeolites were supplied by the Catalyst Pilot Plant, National
Chemical Laboratory, Pune. SAPO-11 (1.5 mm TRX) and MoOx supported on
phosphate-modified alumina (MoOx/Al2O3; MoOx content = 15 wt%, P content = 1.2
wt%) were obtained from Sud-Chemie India Ltd. Sulfated zirconia (1.5 mm pellets)
was procured from Saint-Gobain, Norpro. High temperature, strongly acidic,
macroporous polymer – AmberlystTM70 was obtained from Rohm and Haas, USA.
Detailed characterization of the catalysts is reported in Chapter 2.

3.2.2. Characterization Techniques

X-ray diffractograms of the catalysts were recorded on an X’Pert Pro (Philips)
diffractometer using Cu Kα radiation and a proportional counter as detector.
Transmission electron micrographs (TEM) of the samples were scanned on a JEOL
(model 1200 EX) microscope operating at 100 kV. The C, H, N & S composition was
determined by a Carlo-Erba 1106 analyzer. The specific surface area (BET) of the
samples was determined using a NOVA 1200 Quanta Chrome [Link] acidic
properties of the catalysts were determined using pyridine and NH3 as probe
molecules. The type and density of the acid sites were determined by diffuse
reflectance infrared Fourier transform (DRIFT) spectroscopy of adsorbed pyridine
and temperature-programmed desorption of ammonia (NH3-TPD) techniques. In
DRIFT measurements, the samples were initially activated at 473 K (in the case of
DMC) or at 723 K (in the case of sulfated zirconia, Hβ, H-Y MoOx/Al2O3 and ZnO).
The temperature was brought down to 323 K and pyridine was adsorbed. Then, the
temperature of the sample was raised and held at a desired value (373 K) for 30 min
before starting the measurements (spectral resolution = 2 cm-1; number of scans =
200). Difference FTIR spectra were obtained by subtracting the spectrum of the
catalyst from that of the sample with adsorbed pyridine. NH3-TPD measurements
were performed on a Micromeritics AutoChem 2910 instrument. In a typical
experiment, 0.1 g of catalyst was taken in a U-shaped, flow-thru, quartz sample tube.

Ph.D. Thesis, University of Pune Page 80

 Chapter 3 : Hydrolysis of Vegetable Oils and Animal Fat

Prior to measurements, the catalyst was pretreated in He (30 cm3/min) at 473 K (in
the case of DMC) or at 773 K (in the case of sulfated zirconia, Hβ, MoOx/Al2O3 and
ZnO) for 1 h. A mixture of NH3 in He (10%) was passed (30 cm3/min) at 323 K for
0.5 h. Then, the sample was, subsequently flushed with He (30 cm3/min) for 1 h. The
TPD measurements were carried out in the range 323 – 723 K at a heating rate of 10
K/min. Ammonia concentration in the effluent was monitored with a gold-plated,
filament thermal conductivity detector. The amount of desorbed ammonia was
determined based on the area under the peak.

3.2.3. Reaction Procedure – Hydrolysis of Oils / Fat

Catalytic activity of various solid acid catalysts was evaluated for the hydrolysis
of vegetable oils or fat. A detailed reaction procedure is given in Chapter 2. The
product was analyzed by high performance liquid chromatography (HPLC), titration
and 1H NMR spectroscopy. In some cases, petroleum ether (5 – 10 mL) was added
after the completion of reaction to facilitate separation of the solid catalyst. A few
reactions were conducted in a 300 mL Paar stainless steel stirred tank reactor. Several
edible (coconut, palmolein, and soybean) and non-edible (castor, jatropha, karanja,
palm and rubber seed) oils and chicken fat have been hydrolyzed. With a view to
improve the miscibility of oil/fat and water, some experiments were carried out in the
presence of solvents (N, N-dimethylformamide (DMF), tetrahydrofuran (THF),
acetone), phase transfer agents (tetrapropylammonium bromide, TPAB) or
surfactants (Triton X-100). In other studies, prior to conducting the reactions, a
portion of the product mixture or oleic acid (1 – 20 wt%) was added to reactants. At
the end of the hydrolysis reaction, the solid catalyst was separated from the product
mixture and, without any further treatment was used in another fresh batch of
hydrolysis reaction. The catalyst was reused in five consecutive recycles.

3.2.4. Product Analysis

The fatty acid products of the reaction were analyzed using a Perkin-Elmer
(Series 200) HPLC equipped with a quaternary pump, auto sampler, degasser,
thermostated column compartment and TotalChrom navigator-PE-HPLC software.
An ELSD detector from Gilson and a reverse-phase, Perkin-Elmer Brownlee column
(C-18 Spheri-5, 250 x 4.6 mm with a 5 µm particle size) were used in the analysis. 1H
NMR (Bruker Avance 200 MHz) [23] and Fourier transform infrared (Shimadzu
FTIR 8300) spectroscopic techniques were also used to identify and estimate the

Ph.D. Thesis, University of Pune Page 81

 Chapter 3 : Hydrolysis of Vegetable Oils and Animal Fat

products. The acid values of the product mixture and of vegetable oils and fats were
determined by titration with 0.1 N NaOH solution and using phenolphthalein as the
indicator.
Quantification of fatty acids (FA) and glycerides (TG, DG and MG) in vegetable
oil/fat and the hydrolyzed products by 1H NMR is based on the fact that α-CH2 peaks
of FA appear at higher δ values than those of glycerides. The difference in chemical
shift (between the FA and glycerides) is due to the greater de-shielding effect of the
carboxylic group compared to the ester group. Due to this shift, one of the peaks of
the triplet of FA (at 2.38 ppm) shifts out of the α-CH2 region of the ester while the
other two peaks (2.34 and 2.30 ppm, respectively) are merged with those due to the
glycerides at 2.35 and 2.31 ppm, respectively. In other words, a sample containing
both FA and glyceride (i.e., partially hydrolyzed product) shows a quartet spectral
pattern in the α-CH2 region of the 1H NMR spectrum with the intensity of the peaks
depending on the relative FA and glyceride concentrations. The unmerged peak of
the FA triplet can be used to determine the FA content in vegetable oil, animal fat
and hydrolyzed product. The area (AFA) of the unmerged peak of FA triplet
(appearing around 2.38 ppm, out of the glycerides triplet) can, thus, be determined by
integration of the spectral region 2.37 – 2.41 ppm. The triplet appears with an
intensity ratio of [Link]. Total area corresponding to the α-CH2 groups of the FA, will
thus, be four times the area of the single unmerged FA peak around 2.38 ppm. The
total area corresponding to α-CH2 of both FA and glycerides can be determined by
integrating the spectral region 2.20 – 2.41 ppm. The concentration of FA (wt%) in oil
or hydrolyzed product is, thus,

Alternatively, the peaks due to the glycerides and FA can be deconvoluted and
the FA content can be calculated using the following equation:

Ph.D. Thesis, University of Pune Page 82

 Chapter 3 : Hydrolysis of Vegetable Oils and Animal Fat

For validation of this method, calibration curves for the standard solutions of
oleic acid (as the representative FA) and soybean oil were constructed based on both
the 1H NMR and titrimetric methods. A very good correlation was observed.

3.3. Results and Discussion

3.3.1. Physicochemical Properties

The textural and acidic properties of different solid catalysts used for the
hydrolysis are listed in Table 3.1. Fe-Zn DMC has a molecular formula :
K4Zn4[Fe(CN)6]3.6H2O.2(tert.-BuOH) with two molecules of tert.-butanol included
in its structure [18]. The complex crystallizes with a cubic unit cell (unit cell
parameter = 0.904 nm). N2 adsorption-desorption isotherms and high resolution
transmission electron micrographs reveal that DMC is mostly microporous with some
mesoporosity in its structure (average pore diameter = 2.1 nm and pore volume =
0.03 cc/g). It has a specific surface area; 52 m2/g (Table 3.1). DMC is a Lewis acid
catalyst. Tetracoordinated Zn ions are the acidic, active sites.

Sulfated zirconia possesses a tetragonal crystalline phase (a = 0.362 nm and c =

0.511 nm) and a long-range mesopore ordering (average pore diameter = 4.1 nm and
pore volume = 0.14 cc/g, specific surface = 136 m2/g). Sulfated zirconia, Hβ, HY and
SAPO-11 contain both Lewis (IR bands at 1607, 1596 and 1443 cm-1 due to pyridine
molecules adsorbed on Lewis acid sites) and Brönsted (IR bands at 1639 and 1542
cm-1 due to pyridinium ions) acid sites. The total amount of ammonia desorbed from
sulfated zirconia, SAPO-11 and Hβ is lower (0.48, 0.42 and 0.68 mmol/g,
respectively) than that from Fe-Zn DMC (0.84 mmol/g). However, the former group
contains relatively stronger Brönsted acid sites. The Fe-Zn DMC contains only Lewis
acid sites (absence of IR bands due to pyridinium ions).

Number of acid sites (1.28 mmol NH3/g) in zeolite HY is higher but has smaller
average pore diameter than Fe-Zn DMC. Sulfated zirconia has a Na+ ion exchange
capacity of 0.26 mmol/g [24]. AmberlystTM70 is a high temperature macroporous
polymer with strong Brönsted acidic sites (0.26 mmol/g). It has a specific surface
area of 36 m2/g and a pore diameter of 22 nm. MoOx/Al2O3 contains both Lewis and
Brönsted acid sites (0.61 mmol/g). It has a specific surface area of 205 m2/g and a
pore volume of 0.28 cc/g.

Ph.D. Thesis, University of Pune Page 83

 Chapter 3 : Hydrolysis of Vegetable Oils and Animal Fat

Table 3.1. Surface and acidic properties of catalysts

Catalyst Surface properties Acidic properties

SBET Av. pore Pore Nature Typea Acid site densityb
(m2/g) diameter volume mmol/g mol/m2
(BJH, nm) (cc/g)

Fe-Zn DMC 52 2.1 0.03 Hydrophobic L 0.84 16.2

AmberlystTM70 36 22 - Hydrophilic B 0.26 7.2
Sulfated zirconia 136 4.1 0.14 Hydrophilic L, B 0.48 3.5
MoOx/Al2O3 205 5.5 0.28 Hydrophilic L, B 0.61 3.0
a
SAPO-11 206 3.6 0.19 Hydrophilic L, B 0.42 2.0
HY (SiO2/Al2O3= 5) 550 1.8a 0.50 Hydrophilic L, B 1.28 2.3
H-β 577 2.7a 0.39 Hydrophilic- L, B 0.68 1.2
(SiO2/Al2O3= 28) hydrophobic
a
L = Lewis, B = Brönsted. bEstimated from NH3-TPD measurements. cReaction conditions: Oil/fat = 10 g, water: oil/fat = 20, Fe-Zn
DMC catalyst = 5 wt%, temperature = 463 K, reaction time = 8 h. dTG = triglycerides, FA = fatty acids.
a
Average pore diameter of alumina extrudates and not the zeolites per se.

Ph.D. Thesis, University of Pune Page 84

 Chapter 3 : Hydrolysis of Vegetable Oils and Animal Fat

Scheme 3.1. Vegetable oil or fat hydrolysis

3.3.2. Catalytic Activity

The hydrolysis of triglycerides is a three-step, consecutive, reversible reaction. In

the first step, the triglyceride (TG) is converted into diglycerides (DG), which is then
in the second and third steps converted sequentially into monoglycerides (MG) and
glycerol (G), respectively, producing one mole of fatty acid (FA) at each step
(Scheme 3.1). Stoichiometrically, one mole of triglyceride requires three moles of
water (oil : water molar ratio = 1 : 3) to produce three moles of fatty acids and one
mole of glycerol. Excess water would drive the equilibrium towards the end-products
– fatty acids and glycerol.

Scheme 3.2. Enthalpy of hydrolysis of glyceryl triacetate

It is an endothermic reaction as can be inferred from the calculated enthalpy for

each step of hydrolysis of glyceryl triacetate (Scheme 3.2) [19]. The third step of the
reaction i.e., conversion of MG into glycerin and FA is relatively more endothermic
than the first two steps. Similar thermodynamics is expected for the vegetable oils
and fats. Energy of activation for these transformations has been reported to be in
range of 45 – 75 kJ/mol depending on the type of the catalyst and chain length of the
fatty acid [16, 20]. Hydrolysis is accelerated by mineral acids, a number of metallic
oxides or hydroxides, and various organic catalysts besides certain lypolytic enzymes
of plant and animal origin. In practice the effectiveness of a catalyst is dependent not

Ph.D. Thesis, University of Pune Page 85

 Chapter 3 : Hydrolysis of Vegetable Oils and Animal Fat

only upon its fat-splitting action per se, but also to a great extent upon its ability to
promote emulsion between the oily and aqueous phase undergoing reaction. The
establishment of large water-oil interface is very necessary, because of the limited
mutual solubility of the two phases [21].

Table 3.2. Hydrolysis of soybean oil over different catalystsa

Catalyst TG TONb Products distribution (wt%) Acid value

conv. FA MG DG TG of product
(mgKOH/g)
(wt%)

No catalyst 34.5 - 14.4 0.8 19.3 65.5 28.4

HY 22.0 4 8.6 0.4 13.0 78.0 16.9

SAPO-11 27.0 14 10.9 0.8 15.3 73.0 21.5

AmberlystTM70 38.5 33 18.7 1.8 18.1 61.5 36.9

Sulfated zirconia 47.8 22 22.3 2.0 23.5 52.2 44.1

MoOx/Al2O3 64.7 24 31.9 4.0 28.8 35.3 63.0

Hβ 67.2 22 34.2 7.3 25.8 32.8 67.5

Co-Zn DMC 96.7 (7) 67.5 9.3 19.9 3.3 133.8

Fe-Zn DMC 93.9 25 (7) 71.8 4.5 17.7 6.1 141.7

ZnO 99.4 (2) 81.3 7.7 10.4 0.6 160.6

Conc. H2SO4 98.5 1 82.2 2.1 14.2 1.5 162.3

a
Reaction conditions: Soybean oil = 10 g, soybean oil : water (molar ratio) = 1 : 0,
catalyst = 5 wt% of oil, temperature = 463 K, reaction time = 8 h,
TG = triglycerides, DG = diglycerides, MG = monoglycerides, FA = fatty acids.
b
Turnover number (TON) = moles of TG converted per mole of acid sites (NH3-TPD)
on the catalyst. Values in parentheses indicate TON calculated per mole of Zn2+ ions
in the catalyst.

[Link]. Hydrolysis over Various Solid Acid Catalysts

Table 3.2 lists the catalytic activity data of different solid acid catalyst for the
hydrolysis of soybean oil. Blank experiments under the reaction condition (463 K,
reaction time = 8 h and oil : water molar ratio = 1 : 20, no catalyst), revealed that the

Ph.D. Thesis, University of Pune Page 86

 Chapter 3 : Hydrolysis of Vegetable Oils and Animal Fat

hydrolysis of soybean oil takes place even in the absence of any catalyst but to a
small extent only (Table 3.2, TG conversion = 34.5% and fatty acids selectivity =
41.7%). However, the catalysts (except HY and SAPO-11) increase the TG
conversion significantly. Near-quantitative conversion of TG was obtained over
DMC and ZnO catalysts. TON (TON = moles of TG converted per mole of acid sites
on the catalyst) values are significantly greater than one for most of the catalysts
indicate that the reaction is catalytic and not stoichiometric in nature. Catalytic
activity expressed in terms of turnover number indicates that Amberlyst TM70 (TON =
33) is the most active catalyst for the conversion of the triglyceride among those
investigated in this study. But the fatty acids yield over this catalyst is very low
(18.7%), the mono- and diglycerides constituting the main products. DMC shows
high activity (TON = 25) together with a high yield of fatty acids. Fe-Zn DMC was
superior (fatty acids yield = 71.8%) to Co-Zn DMC (fatty acids yield = 67.5%).
(Table 3.2). ZnO was efficient and comparable to the homogeneous concentrated
sulfuric acid catalyst but its separation from the product mixture was difficult. A
paste-like substance formed which required washing with ether or hexane several
times before the catalyst could be separated. It is due to the fact that ZnO and MgO
increases the solubility of water in oil phase by forming metal soaps [22].

[Link]. Kinetics of Hydrolysis of Soybean Oil

Hydrolysis of soybean oil is conducted in a batch reactor and product yield

monitored by acid-base titration, HPLC, FTIR and 1H NMR spectroscopy. Fig. 3.2
shows the product composition in the hydrolysis of soybean oil as a function of
reaction time over the Fe-Zn DMC catalyst analyzed by HPLC. Conversion of oil and
yield of fatty acids increased with time and remained nearly constant after 8 h.
Kinetic plot for the hydrolysis clearly indicates that it is a three-step, consecutive and
reversible reaction (Scheme 3.1). The concentration of DG in the products increased
with time, reaching a maximum at 7 h and then it decreased with further increase in
reaction time (due to conversion into monoglycerides). The concentration of MG was
negligible during the initial 5 h and then increased and remained at 7 – 9% beyond 8
h. This plot, suggesting that DG and MG are the intermediates while the fatty acids
are the final products in the reaction sequence, is typical of sequential reactions
(Scheme 3.1). In contrast to DG, no clear maximum was observed for MG in the
kinetics plot because of the attainment of equilibrium between MG and FA at higher

Ph.D. Thesis, University of Pune Page 87

 Chapter 3 : Hydrolysis of Vegetable Oils and Animal Fat

conversion levels of triglycerides. Minami and Saka [23] reported a similar

observation of equilibrium at around 90 wt% in yield of FA in the hydrolysis of
rapeseed oil in subcritial water (reaction conditions: 593 K, 200 bar, oil to water
molar ratio ~1: 55). DMC is a reusable catalyst. Hence, this behavior of MG variation
in the kinetics plot is not due to deactivation of the catalyst but to attainment of
equilibrium between MG and FA.

100

80 FA%
MG%
DG%
TG%
60
Wt.%

40

20

0
0 2 4 6 8 10 12
Reaction time (h)

Fig. 3.2. Kinetics of hydrolysis of soybean oil over Fe-Zn DMC

FA - fatty acids, MG - monoglycerides, DG - diglycerides, TG - triglycerides

The kinetics of hydrolysis of soybean oil was also monitored by 1H NMR (Fig.
3.3) and FTIR (Fig. 3.4) spectroscopic techniques. In 1H NMR, the glycerides show
characteristic signals at 4 – 4.5 ppm. As the reaction proceeded, intensity of these
peaks decreased indicating their conversion into fatty acids. A significant increase in
fatty acids yield was observed only after an induction period of 4 h (Figs. 3.3 and
3.4). No decrease in the intensity of peaks due to C=C groups ( = 5.5 ppm) was
observed indicating that there is no thermal polymerization or oxidation occurring as
side reactions. Based on 1H NMR spectroscopy, it can be concluded that the
hydrolysis product at the end of 8 h contains mainly FA with negligible amounts of
glycerides (Fig. 3.3). Fatty acid glycerides show a characteristic IR peak for the
carboxylate group at 1750 cm-1. The corresponding peak for FA appears at 1710 cm-1
(Fig. 3.4). .

Ph.D. Thesis, University of Pune Page 88

 Chapter 3 : Hydrolysis of Vegetable Oils and Animal Fat

1h
Glycerides 2h

3h 4h
Intensity (a.u.)

5h 6h

7h 8h

10h
12h

6 5 4 3 2 1 6 5 4 3 2 1

 (ppm)
Fig. 3.3. Kinetic study of hydrolysis of soybean oil by 1H-NMR

Ph.D. Thesis, University of Pune Page 89

 Chapter 3 : Hydrolysis of Vegetable Oils and Animal Fat

Table 3.3. Hydrolysis of soybean oil over Fe-Zn DMC catalyst : influence of amount of catalyst

Amount of Reaction time = 5 h Reaction time = 8 h

catalyst (wt% TG Products distribution (wt%) Acid value TG Products distribution (wt%) Acid value
of oil) conv. of product conv. of product
FA MG DG TG FA MG DG TG
(wt%) (mgKOH/g) (wt%) (mgKOH/g)

1 16.0 6.6 0.1 9.3 84.0 13.0 66.1 30.3 13.6 22.3 33.9 59.8
2 24.2 11.3 0.3 12.6 75.8 22.4 80.9 44.2 10.1 26.6 19.1 87.4

3 28.8 12.4 0.5 15.9 71.2 24.6 86.7 51.6 11.2 23.9 13.3 101.9
5 45.7 22.1 2.0 21.7 54.3 43.6 93.4 65.7 9.4 18.2 6.6 129.8
7 59.7 31.1 2.7 25.9 40.3 61.4 97.2 72.1 10.5 14.6 2.8 142.4
10 67.9 39.7 4.2 24.1 32.1 78.4 97.6 75.4 8.6 13.6 2.4 149.0
Reaction conditions: Soybean oil = 10 g, soybean oil : water (molar ratio) = 1 : 20, temperature = 463 K, reaction mode = batch
process in a Teflon-lined stainless steel reactor (100 mL). Product distribution was analyzed by HPLC. FA content was also determined
by titration method.

Ph.D. Thesis, University of Pune Page 90

 Chapter 3 : Hydrolysis of Vegetable Oils and Animal Fat

While the intensity of the former peak decreased, the intensity of the latter
increased with reaction time in the FTIR spectrum. While 1H NMR revealed
complete conversion of glycerides to FA beyond 8 h of reaction, FTIR and HPLC do
reveal the presence of some residual amounts of glycerides in the product. This
disparity is due to differences in the sensitivity of the techniques for different
products.

1h 2h

3h 4h
Absorbance

Glycerides
5h FA 6h

7h 8h

10 h 12 h

1800 1780 1760 1740 1720 1700 1680 1660 1800 1780 1760 1740 1720 1700 1680 1660

Wave number(cm-1)
Fig. 3.4. Hydrolysis of soybean oil monitored by FTIR spectroscopy

[Link]. Influence of Amount of Catalyst

TG conversion and FA selectivity increased markedly with an increase in the
amount of catalyst up to 5 wt%. Beyond that, the change is only marginal (Table 3.3).
According to the Madon and Boudart test [24], a modified form of the Koros and
Nowak test [25], in the absence of any mass transfer limitations, the TG conversion
should increase proportionately with the amount of the catalyst. We observed this
increase in the range of 0 – 5 wt% catalyst. The deviation from this linearity beyond
5 wt% of the catalyst indicates that, at higher catalyst concentrations, there exists
mass transfer limitation on the reaction rate. At longer reaction time (8 h),

Ph.D. Thesis, University of Pune Page 91

 Chapter 3 : Hydrolysis of Vegetable Oils and Animal Fat

conversions are also influenced by equilibrium attainment. The non-homogeneity of

the slurry at such high concentrations of the solid catalyst may also contribute to such
a mass transfer limitation.

[Link]. Influence of Reaction Temperature

Reaction temperature is another important parameter that influences the rate of
hydrolysis. As expected, both TG conversion and FA yield increased with reaction
temperature from 433 to 463 K. Above 463 K, however, the selectivity for FA
declined due to equilibrium limitations (Table 3.4). At higher temperatures, the
reverse esterification reaction of FA with glycerol becomes significant.

Table 3.4. Hydrolysis of soybean oil over Fe-Zn DMC catalyst: effect of reaction
temperaturea

Reaction TG TONb Product distribution (wt%) Acid value of

temperature conv. the product
FA MG DG TG
(K) (wt%) (mgKOH/g)

433 24.3 7 8.6 1.2 16.5 73.7 16.9

443 90.3 24 73.0 6.3 11.0 9.7 144.2

463 84.1 23 72.3 5.8 6.0 15.9 121.5

473 89.7 24 69.4 11.8 8.6 10.3 137.1

483 93.2 25 66.3 20.1 6.8 6.8 131.0

a
Reaction conditions: Soybean oil = 20 g, soybean oil : water (molar ratio) = 1 : 30,
reaction mode = batch process in a Ti-line stainless steel Paar reactor (300 mL, 300
revolutions per minute).
b
Turnover number (TON) = moles of TG converted per mole of acid sites (NH3-
TPD) on the catalyst.

[Link]. Influence of Water to Oil Molar Ratio

Table 3.5 shows the influence of water to oil molar ratio on the hydrolysis of
soybean oil over DMC catalyst. The yield of FA increased with water/oil molar ratio
up to 20. Maximum FA yields could be obtained using 5 wt% of catalyst, an oil to
water molar ratio of 1 : 20, a reaction temperature of 463 K and a reaction time of 8
h.

Ph.D. Thesis, University of Pune Page 92

 Chapter 3 : Hydrolysis of Vegetable Oils and Animal Fat

Table 3.5. Hydrolysis of soybean oil over Fe-Zn DMC catalyst : influence of water
to oil molar ratio

Water : oil TG TONb Products distribution (wt%) Acid value of

molar ratio the product
conv. FA MG DG TG
(mgKOH/g)
(wt%)

5 92.7 25 49.4 14.3 29.0 7.3 97.7

10 92.6 25 53.4 7.5 31.7 7.5 105.5

15 96.2 26 66.0 8.8 21.4 3.8 130.3

20 93.4 25 65.7 9.4 18.2 6.6 129.8

30 97.0 26 69.1 7.3 20.5 3.0 136.6

40 96.4 26 68.9 7.2 20.4 3.6 136.1

a
Reaction conditions: Soybean oil = 10 g, catalyst = 5 wt% of oil, temperature = 463
K, reaction time = 8 h.
b
Turnover number (TON) = moles of TG converted per mole of acid sites (NH3-
TPD) on the catalyst.

[Link]. Hydrolysis of Different Oils and Fat

Fe-Zn DMC was found to be very active in the hydrolysis of a range of edible
(coconut, palmolein and soybean) and non-edible (castor, jatropha, karanja, palm and
rubber seed) oils and chicken fat (Table 3.6). Non-edible oils (used soybean oil and
fat) were converted with equal, if not higher efficiency than the edible oils under
optimized conditions. Triglyceride conversion greater than 97% and FA selectivity
greater than 76% were obtained. Non-edible oils and fats are low-quality feedstock
compared to edible (refined) vegetable oils because of their free fatty acid content
that may increase up to 15 wt% in contrast with edible oils which typically contain
less than 0.5 wt% fatty acid. In addition, non-edible oils contain phospholipids,
sterols, water, odorants and other impurities in small amounts (~2 wt%). These
impurities deactivate conventional, homogeneous and heterogeneous catalysts.
However, no such deactivation of DMC was observed when using non-edible oils.

Ph.D. Thesis, University of Pune Page 93

 Chapter 3 : Hydrolysis of Vegetable Oils and Animal Fat

Table 3.6. Hydrolysis of edible and non-edible oils and fat over Fe-Zn DMC
catalysts
Oil/fat FA content (wt%) TG Product selectivity (wt%)
Before the After the Conv. FA MG DG TG
reaction reaction (wt%)
Edible oils:
Coconut oil 0.2 88.4 99.1 89.4 9.3 1.3 1.0
Palmolein oil 0.3 80.5 98.9 80.5 11.1 7.3 1.1
Soybean oil 0.3 77.2 98.5 77.2 12.0 9.3 1.5
Used 0.5 83.2 99.2 83.2 9.6 6.3 0.8
soybean oil
Non-edible oils:
Castor Oil 3.1 51.9
Jatropha Oil 3.7 82.7 99.3 82.7 9.8 6.8 0.7
Karanja Oil 5.8 76.7 99.9 76.7 20.4 2.7 0.1
Palm oil 6.6 80.2 99.1 80.2 11.1 7.8 0.9
Rubber Seed 17.0 84.4 94.3 84.4 5.6 4.3 5.7
Oil
Animal fat:
Chicken Fat 10.5 78.5 97.2 78.5 11.5 7.2 2.8
Reaction conditions: Oil or fat = 10 g, oil or fat : water molar ratio = 1 : 30, catalyst
= 5 wt% of oil, temperature = 463 K, reaction time = 12 h.
100 100

95 95
TG Conv.
TG Conv. FA Sel.
Wt. %

FA sel. 90
Wt %

90

85 85

80 80

(a) (b)
75 75
0 1 2 3 4 5 6 13 14 15 16 17 18
Unsaturation Index Average No. of Carbon

Fig. 3.5. Influence of unsaturation and chain length of fatty acids on hydrolysis
activity

Ph.D. Thesis, University of Pune Page 94

 Chapter 3 : Hydrolysis of Vegetable Oils and Animal Fat

Unsaturation in vegetable oils and fatty acid chain length influenced the
hydrolysis activity. Average fatty acid chain lengths of different oils were estimated
based on their compositions projected in Table 1.2 in Chapter 1. Fig. 3.5(a) shows the
influence of unsaturation index (ratio of unsaturated fatty acids to saturated fatty
acids) on TG conversion. As the unsaturation of oils increases a slight decrease in TG
conversion and FA yield was detected. A similar decrease was also observed with
increasing fatty acid chain length (Fig. 3.5 (b)). The reason for this may be the
electron rich double bond containing fats adsorb more strongly on the Lewis acidic
catalyst surface than the saturated fats and thus show a lowering in the conversion.

[Link]. Miscibility of Water with Oil

Miscibility of water with oil is a major issue in this reaction. Eklesberg et al [22]
reported that the heterogeneous reaction between water and oil at the interface is of
minor importance in the hydrolysis as compared with the homogenous reaction
between oil and the water dissolved in the oil phase. Since water has limited
solubility in oil, the reaction proceeds very slowly at the beginning, and a significant
induction period is usually observed (Fig. 3.2). Miscibility of water with oil can be
enhanced by adding suitable phase transfer agents or solvents. The following
methods have been explored:
1) Addition of oleic acid
2) Addition of reaction product containing mono- and diglycerides along with
fatty acids
3) Use of solvent
4) Use of surfactants
Table 3.7 shows the influence of the addition of oleic acid in different amounts to
improve the miscibility of oil and water and thereby the conversion. Addition of 1 – 2
wt% of oleic acid is sufficient to enhance the catalytic activity significantly. Addition
of oleic acid significantly reduced the induction period. But, after 10% oleic acid
addition there is not much increase in FA yield (Fig. 3.6). Higher efficiency and FA
yield observed with non-edible oils compared to edible oils (Table 3.6) can therefore
be attributed to the presence of free fatty acids in those oils. The fatty acids improve
the miscibility of the oil phase with water. It is likely that the hydrolysis reactions
occur in the oil phase. The low concentration of water in the oil will lead to slow
reaction rates in the early stages of the reaction. An induction period (of about 4 h) is
observed (Fig. 3.2).

Ph.D. Thesis, University of Pune Page 95

 Chapter 3 : Hydrolysis of Vegetable Oils and Animal Fat

Table 3.7. Influence of the addition of oleic acid on the hydrolysis of soybean oil
Oleic No catalyst Fe-Zn DMC catalyst (5 wt% of oil)
a
acid Product distribution, wt% Acid valuea Product distribution,a wt% Acid valuea
(%) FA MG DG TG (mgKOH/g) FA MG DG TG (mgKOH/g)
0 1.2 0.0 0.8 98.0 2.4 2.6 0.0 2.5 94.9 5.2
1 6.5 0.3 6.4 86.8 12.8 50.7 8.0 28.6 12.7 100.2
2 10.3 0.9 13.9 74.9 20.3 50.3 8.9 28.5 12.3 99.4
5 17.4 2.0 22.6 58.0 34.3 54.6 9.0 26.7 9.8 107.8
10 28.1 5.5 30.8 35.6 55.4 54.6 8.8 27.7 8.9 107.6
15 28.0 9.6 32.7 29.7 55.0 53.3 10.0 29.4 7.3 104.9
20 32.6 8.1 35.3 24.0 64.0 52.4 9.4 31.4 6.8 103.0
a
After subtracting the corresponding value due to added oleic acid.
Reaction conditions: (Soybean oil + oleic acid ) = 10 g, (oil+OA) : water molar ratio = 1 : 20, temperature = 463 K

35

FA yield-OA added (Wt.%)

100
30

80 1% OA 25
5% OA
TG (wt%)

10% OA 20
60 0% OA

15
40
10

20
5

0 0
0 1 2 3 4 0 5 10 15 20

Reaction time(h) OA% Added

Fig. 3.6. Effect of oleic acid addition on induction time and FFA yield

Ph.D. Thesis, University of Pune Page 96

 Chapter 3 : Hydrolysis of Vegetable Oils and Animal Fat

Fig. 3.7 and Table 3.8 shows the effect of addition of 20 wt% of a partially
hydrolyzed product (composition: FA - 35.3%, MG - 6.9%, DG - 21.2% and TG -
36.6%) to a fresh batch of reactants mixture. Reaction time was reduced
considerably. The normal induction period of about 3 - 4 h (Fig. 3.2) was eliminated
when 20 wt% of hydrolyzed product was added to the reaction mixture (Fig. 3.7).
While the FA yield is about 2.6% at the end of 4 h in the absence of the added
products, it increases to 60.9% when 20% of the hydrolyzed product was added.
Water is a hydrophilic solvent. Oil and fat (TG) are aprotic (hydrophobic) and so
immiscible with water. Mono and diglycerides are less hydrophobic than the starting
triglycerides and are more miscible with water. As a result, the mono- and
diglycerides function as emulsifiers and enhance mixing of the triglyceride layer with
water and, thereby, the rate of the hydrolysis reaction.

Table 3.8. Hydrolysis of soybean oil over Fe-Zn DMC catalyst: influence of the
addition of a part of hydrolysis product (20 wt%) to the reaction mixture

Reaction Product distribution (wt%)

time (h)
Without adding reaction After addition of 20 wt% hydrolyzed
mixture (oil: 10 g of soybean product (feed: oil: 8 g of soybean oil +
oil) 2 g of hydrolyzed product)a

FA MG DG TG FA MG DG TG

1 0.3 0.00 0.2 99.5 10.2 1.0 7.8 80.9

2 0.8 0.1 0.3 98.8 24.1 5.1 12.7 58.1

3 1.1 0.1 0.9 97.9 45.4 6.0 37.6 10.9

4 2.6 0.0 2.5 94.9 60.9 13.1 19.3 6.7

Reaction conditions: oil : water molar ratio = 1 : 30, catalyst = 5 wt% of oil,
temperature = 463 K, wt% composition of hydrolyzed product added = FA(35.3) +
MG (6.9) + DG (21.2) + TG (36.6). aProduct composition includes the acid present
in the added hydrolyzed product

Ph.D. Thesis, University of Pune Page 97

 Chapter 3 : Hydrolysis of Vegetable Oils and Animal Fat

60

50

40
FA (wt%)
30

20

10

0
0 1 2 3 4
Reaction time (h)
Fig. 3.7. Hydrolysis of soybean oil with (●) and without (■) the addition of 20 wt%
hydrolysis product
Similar effects were also found when suitable solvents, phase transfer agents and
surfactants were added (Table 3.9). Among the solvents investigated, DMF was the
best. TG conversion was lower with acetone (7.4%) than without any solvent
(12.6%). A similar observation was found also with Triton X 100 (8.8%). DMF was
miscible with water in all proportions and up to 25% with soybean oil at 298 K.
Acetone and THF are 100% miscible with both water and oil. When a known
quantity of solvent was added to the reactants mixture (soybean oil: water molar ratio
= 1: 20) at 298 K, a turbid layer formed at the interface of the reactants. The amount
of such turbid layer is more in the case of DMF than with THF. When DMF was
added the miscibility of these phases increased to 10 wt%, otherwise vegetable oil
and water are immiscible. No such turbidity was detected with acetone solvent. The
boiling point of different solvents decreases in the following order: DMF (426 K) >
THF (339 K) > acetone (329 K). At our reaction conditions, in contract to acetone,
DMF is mostly in the liquid phase in contact with the reactants enhancing their
miscibility. This enhancement in reactants miscibility, formation of interface and
high boiling point are the causes for the higher conversions observed in DMF than in
THF and acetone solvents. Use of solvents, products containing partially hydrolyzed
glycerides or fatty acids increase the miscibility of oil and water (form an active
interface region) and greatly accelerate the reaction so that high conversion is
reached in shorter time periods. Recycle of a portion of the products as a miscibility
enhancer is more practical and beneficial.

Ph.D. Thesis, University of Pune Page 98

 Chapter 3 : Hydrolysis of Vegetable Oils and Animal Fat

Table 3.9. Influence of addition of solvent or surfactanta

Solvent TG TONb Products selectivity (wt%) Acid

or conv. value
surfactant (wt%) FA MG DG TG

No solvent 12.6 3 5.2 0.2 7.1 87.4 10.4

DMF 40.7 11 13.1 29.9 16.3 80.5 40.7
THF 32.7 9 16.5 3.4 12.8 67.3 32.6
Acetone 7.4 2 3.6 0.1 3.7 92.6 7.2
TPAB 43.8 12 21.0 3.2 19.5 56.2 41.6
Triton X 100 8.8 2 3.6 1.5 3.8 91.2 7.1
a
Reaction conditions: Soybean oil = 5 g, oil : water molar ratio = 1 : 20, Fe-Zn
DMC = 5 wt% of oil, solvent or surfactant = 5 wt% of oil, temperature = 463 K,
reaction time = 5 h, reaction mode = batch process in a teflon-lined stainless steel
reactor (100 mL).bTurnover number (TON) = moles of TG converted per mole of
acid sites (NH3-TPD) on the catalyst.

Table 3.10. Hydrolysis of soybean oil over Fe-Zn DMC : catalyst recyclability studya

Number of TG TONb Products selectivity (wt%) Acid value

recycles FA MG DG TG of the
conv.
product
(wt%)
(mgKOH/g)

0th (Fresh) 93.9 25 71.8 4.5 17.7 6.1 141.7

1st 96.5 26 70.8 10.2 15.5 3.5 139.9

2nd 97.1 26 72.8 9.8 14.5 2.9 143.8

3rd 96.2 26 71.2 10.0 15.0 3.8 140.6

4th 97.1 26 72.9 10.1 14.1 2.9 143.9

5th 97.9 26 72.8 10.9 14.2 2.1 143.8

a
Reaction conditions: Soybean oil = 10 g, oil : water molar ratio = 1 : 20, catalyst =
5 wt% of oil, temperature = 463 K, reaction time = 8 h, bTurnover number (TON) =
moles of TG converted per mole of acid sites (NH3-TPD) on the catalyst.

Ph.D. Thesis, University of Pune Page 99

 Chapter 3 : Hydrolysis of Vegetable Oils and Animal Fat

[Link] Recyclability of Fe-Zn DMC Catalyst

Recyclability is perhaps the most important parameter for any heterogeneous
catalyst. Fe-Zn DMC catalyst was reused in five recycle experiments without any
pretreatment (Table 3.10). No major loss in activity and selectivity for fatty acids
were observed. This shows that DMC is a recyclable catalyst and does not get
deactivated due to water (reactant in hydrolysis) and glycerol (a product of
hydrolysis).

3.3.3. Structure–Activity Correlation

Adsorption and activation of the triglyceride molecule on the catalyst surface are
the key steps in the hydrolysis of fats and oils over solid surfaces. Therefore, the
nature of the catalyst surface plays crucial role in the hydrolysis of vegetable oils. In
the case of hydrophilic catalysts, the surface is covered mostly with the water
molecules and accessibility of the active sites for oil molecules is hindered. Hence,
catalysts even with stronger acid sites will not be able to show higher activity. On the
contrary, oil molecules will have greater access to acidic active sites on hydrophobic
catalyst surfaces and therefore higher conversions are expected on those catalyst
surfaces. The adsorption of this non-polar molecule will be facilitated on
hydrophobic surfaces. To compare the relative surface hydrophobicity of our
catalysts, a known amount of the powdered catalyst (DMC, Hβ, HY, SAPO-11,
sulfated zirconia or MoOx/Al2O3) was suspended in equal volumes of water-soybean
oil mixture. Water and soybean oil are immiscible, with water forming the bottom
and oil the top layers (Fig. 3.8). Interestingly, the DMC catalyst remained suspended
in the oil layer, Hβ settles fast at the interface of the layers and the rest of the
catalysts went into the water layer. The reason for this could be that the surface of
DMC is hydrophobic and hence, it prefers to be in the hydrophobic oil layer. Hβ is
less hydrophobic so settles at the interface and rest of the catalysts including sulfated
zirconia are relatively hydrophilic and go into the water layer. Because of their
hydrophilic nature their surface is covered with water molecules during the catalytic
reaction thereby limiting access to the surface for oil molecules. On the contrary, on
hydrophobic DMC surface, the hydrophobic oil/fat molecules are preferentially
adsorbed and activated. The higher TG conversions and FA yield observed over Fe-
Zn DMC are therefore due to its hydrophobic surface structure in addition to its
higher acid site density.

Ph.D. Thesis, University of Pune Page 100

 Chapter 3 : Hydrolysis of Vegetable Oils and Animal Fat

Fig. 3.8. Preferential dispersion of DMC in non-polar, soybean oil layer, Hβ near
the interface and HY, SAPO-11, sulfated zirconia (SZ) and MoOx/Al2O3 in polar
water layer

3.4. Conclusions
Hydrolysis of vegetable oils and fats has been investigated over different solid
catalysts including acidic, macroporous polymeric resin (AmberlystTM70), zeolites
(SAPO-11, HY and Hβ), supported oxide (MoOx/Al2O3), sulfated zirconia, DMC and
basic ZnO. Fe-Zn DMC exhibited the highest conversion and selectivity. Complete
conversion of triglycerides and fatty acid yields greater than 73% were obtained at
moderate temperature and pressure. The highly efficient activity of DMC has been
attributed to its hydrophobic surface structure enabling it to adsorb and activate the
triglyceride molecules selectively. The catalyst is reusable. Induction periods in
hydrolysis reactions can be considerably curtailed / eliminated by including some
amounts of the products (containing some fatty acids and mono- / diglycerides) in the
initial reaction mixture. This is the first report on the application of DMC catalyst to
hydrolysis of triglycerides.

3.5. References

1. M. Di Serio, R. Tesser, L. Pengnice, E. Santacesaria, Energy Fuels 22

(2008) 207.
2. M. DiSerio, M. Cozzolino, M. Giordano, R. Tesser, P. Patrono, E.
Santacesaria, Ind. Eng. Chem. Res. 46 (2007) 6379.

Ph.D. Thesis, University of Pune Page 101

 Chapter 3 : Hydrolysis of Vegetable Oils and Animal Fat

3. E. Lotero, Y. Liu, D.E. Lopez, K. Suwannakarn, D.A. Bruce, J.G. Goodwin,

Jr., Ind. Eng. Chem. Res. 44 (2005) 5353.
4. U. Schuchardt, R. Sercheli, R.M. Vargas, J. Braz. Chem. Soc. 9 (1998) 199.
5. D. Kusdiana, S. Saka, Appl. Biochem. Biotech. 113-116 (2004) 781.
6. An Ullmann’s Encyclopedia, Industrial Organic Chemicals: Starting
Materials and Intermediates, Wiley-VCH, Weinheim, Germany, Vol. 4 pp.
2481–2532 (1999).
7. T.A. Patil, D.N. Butala, T.S. Raghunathan, H.S. Shankar, Ind. Eng. Chem.
Res. 27 (1988) 727.
8. H.L. Barnebey, JAOCS (1948) 95.
9. R.L. Holliday, J.W. King, G.R. List, Ind. Eng. Chem. Res. 36 (1997) 932.
10. L. Hartman, JAOCS (1953) 349.
11. U.S. patent No. 6,646,146 B1 (2003) assigned to Haltermann Ascot GmbH,
Hamburg, GE.
12. US patent No. 5,383,678 (1995) assigned to Church & Dwight Co., Inc.,
Princeton, N.J.
13. US patent No. 6,933,139 B2 (2005) assigned to T + T Oleochemie GmbH,
Alzeunau, DE.
14. W.M. Linfield, D.J. O’Brien, S. Serota, R.A. Barauskas, JAOCS 61 (1984)
1067.
15. U.S. patent No. 6,258,575 B1 (2001) assigned to Kao Corp., Tokyo, JP.
16. K. Ngaosuwan, E. Lotero, K. Suwannakaru, J.G. Goodwin Jr., P.
Praserthdam, Ind. Eng. Chem. Res. 48 (2009) 4757.
17. E. Minami, [Link], Fuel 85 (2006) 2479.
18. P.S. Sreeprasanth, R. Srivastava, D. Srinivas and P. Ratnasamy, Appl. Catal.
A: Gen. 314 (2006) 148.
19. “Lange’s Handbook of Chemistry,” Ed. By J.A. Dean, McGraw-Hill Inc.,
New York, 15th ed. pp. 6.05-6.50 (1999)..
20. P. Khuwijitjaru, T. Fujii, S. Adachi, Y. Kimura, R. Matsuno, Chem. Eng. J.
99 (2004) 1.
21. A.E. Bailey, Industrial Oil and Fat products, Interscience publishers, New
York (1945).
22. O.J. Ackelsberg, J. Am. Oil chem. Soc. 35 (1958) 635.

Ph.D. Thesis, University of Pune Page 102

 Chapter 3 : Hydrolysis of Vegetable Oils and Animal Fat

23. E. Minami, S. Saka, Fuel 85 (2006) 2479.

24. R.J. Madon, M. Boudart, Ind. Eng. Chem. Fundam. 21 (1982) 438.
25. R.M. Koros, E.J. Nowak, Chem. Eng. Sci. 22 (1967) 470.

Ph.D. Thesis, University of Pune Page 103