Waste Management 26 (2006) 487–494

www.elsevier.com/locate/wasman

Production of biodiesel from waste frying oils

Pedro Felizardo a, M. Joana Neiva Correia a,*, Idalina Raposo b, João F. Mendes c,
Rui Berkemeier d, João Moura Bordado a
a
Department of Chemical Engineering, Instituto Superior Técnico, Av. Rovisco Pais, 1049-001 Lisboa, Portugal
b
Escola Secundária Alfredo dos Reis Silveira, Torre da Marinha, Av. 25 de Abril, 2840 Seixal, Portugal
c
SPACE – Eco-Combustı́veis. Rua 8 de Dezembro No. 494, 4760-016 Vila Nova de Famalicão, Portugal
d
Quercus – Associação Nacional Conservação da Natureza, Apartado 4333, 1503-003 Lisboa, Portugal

Accepted 9 February 2005

Available online 17 June 2005

Abstract

Waste frying oils transesterification was studied with the purpose of achieving the best conditions for biodiesel production.
Transesterification reactions were carried out for 1 h using waste frying oils (WFOs), methanol, and sodium hydroxide as catalyst.
In order to determine the best conditions for biodiesel production, a series of experiments were carried out, using methanol/WFO
molar ratios between 3.6 and 5.4 and catalyst/WFO weight ratios between 0.2% and 1.0%. For oils with an acid value of 0.42 mg
KOH/g, results show that a methanol/WFO ratio of 4.8 and a catalyst/WFO ratio of 0.6% gives the highest yield of methyl esters.
Furthermore, an increase in the amount of methanol or catalyst quantity seems to simplify the separation/purification of the methyl
esters phase, as showed by a viscosity reduction and an increasing purity to values higher than 98% for methyl esters phase.
Ó 2005 Elsevier Ltd. All rights reserved.

1. Introduction
The increasing production of waste frying oils
(WFOs) from household and industrial sources is a
growing problem in Portugal and all around the world.
This residue is regularly poured down the drain, result-
ing in problems for wastewater treatment plants and en-
ergy loss, or is integrated into the food chain through
animal feeding, thus becoming a potential cause of hu-
man health problems (Costa Neto et al., 2000). There
are several end-uses for this waste, such as the produc-
tion of soaps or of energy by anaerobic digestion, ther-
mal cracking (Zaher, 2003), and more recently the
production of biodiesel, a fuel that can be used as a min-
eral diesel substitute for engines. The core process of the
biodiesel production consists in the transesterification of
the raw materials, vegetable oils and fats, using a short
*
Corresponding author. Tel.: +351 21 8417344; fax: +351 21
8499242.
E-mail address: qjnc@mail.ist.utl.pt (M.J. Neiva Correia).
chain alcohol. Vegetable oils and fats are mainly consti-
tuted by triglycerides with minor amounts of mono and
diglycerides (Fig. 1).
The transesterification reaction (Reaction 1) takes
place between a lipid and an alcohol to produce an ester
and a by-product, glycerol, also known as glycerine.
This reaction occurs stepwise, with mono and diglyce-
rides as intermediate products (Ullmanns, 1992). The
fuel thus obtained performs in a similar way to fossil
fuel, with the advantage of reducing greenhouse emis-
sions because it is a renewable resource (Mittelbach
and Tritthart, 1988; Leung, 2001; Al-Widyan et al.,
2002; Dorado et al., 2003; Tashtoush et al., 2003; Ulu-
soy et al., 2004).
Meher et al. (2004) published an excellent review of
biodiesel production processes by transesterification.
The use of several low molecular weight alcohols and
homogeneous acid and basic catalysts for transesterifi-
cation has been studied in recent years with success
(Peterson et al., 1983; Schudardt et al., 1998; Khan,
2002; Knothe et al., 1997; Nye et al., 1983). Basic

0956-053X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.wasman.2005.02.025
488 P. Felizardo et al. / Waste Management 26 (2006) 487–494
H2C O COR1 H2C O COR1 H2C O COR1
HC OH HC O COR2 HC O COR2
H2C OH H2C OH H2C O COR3
Monoglyceride Diglyceride Triglyceride
Fig. 1. Structures of mono-, di- and triglycerides, where R1, R2 and R3
represent the fatty acids chains.
H 2C O R1 R1 COOR H2 C
HC O R2
catalyst
R2 COOR + HC
+ 3 ROH
H 2C O R3 R3 COOR H2 C
Reaction 1. Transesterification reaction.
homogeneous catalysts have surpassed the acid ones,
because the reaction is faster (only 30 min compared
to 1–8 h for the acid catalysis) and lead to the same
yield of transesterification (above 90% for both) (Schu-
dardt et al., 1998). However, for oils with a high con-
centration of free fatty acids (acid value higher than 2
mg KOH/g), a pre-treatment step is recommended.
This step consists in an acid catalyzed reaction with
an alcohol in order to transform the free fatty acids
into their corresponding esters. Good results concern-
ing the reduction of the acid value were achieved even
with high initial percentages of free fatty acids (Cana-
kci and Gerpen, 2001, 2003; Pereda Marı́n et al.,
2003). After this pre-treatment, glycerides in the oil/es-
ter phase are transesterified with basic catalysis. So-
dium and potassium hydroxides and sodium and
potassium methylates have been preferred as catalysts
due to high yields in the transesterification reaction.
The methylate salts are more reactive than the corre-
sponding hydroxides, but are expensive, cause forma-
tion of various by-products and require high quality
oils and water-free methanol (Schudardt et al., 1998;
Meher et al., 2004). On the other hand, sodium and
potassium hydroxides react with the alcohol to give
water. This reaction is undesirable, because water can
react with triglycerides, fatty acids, or esters in hydro-
lysis reactions (Reaction 2) and subsequent saponifica-
tion (Reaction 3) (Schudardt et al., 1998). However, in
spite of their lower reactivity, sodium and potassium
hydroxides can also lead to high reaction yields just
by increasing the catalyst quantities.
The production of biodiesel has been actively investi-
gated in the past 20 years. Several studies concerning the
production of methyl esters from ‘‘fresh’’ vegetable oils
and fats (Freedman et al., 1984; Peterson and Scarrah,
1984; Knothe et al., 1997; Conneman and Fischer,
RCOOR1 + H2O
RCOOH + R1OH
Reaction 2. Hydrolysis reaction.
H2C O R1 R1 COONa H2C OH
HC O R2 + NaOH R2 COONa + HC OH
H2C O R3 R3 COONa H2C OH
Reaction 3. Saponification reaction.
1998; Schudardt et al., 1998; Zhang et al., 2003a), as well
as low-cost raw materials such as used frying oils, soap
OH stocks and animal fats (Nye et al., 1983; Mittelbach
et al., 1992; Haas et al., 2000; Canakci and Gerpen,
OH
2001; Al-Widyan and Al-Shyoukh, 2002; González
OH
Gómez, 2002; Tomasevic and Siler-Marinkovic, 2003;
Tashtoush et al., 2004), have been conducted. Enzy-
matic catalysts like lipases are also able to catalyze the
transesterification of triglycerides. According to the lit-
erature, this reaction overcomes some of the drawbacks
of chemical transesterification, namely the difficulty of
recovering glycerol, as well as the incomplete conversion
of the free fatty acids contained in WFO into alkyl es-
ters. However, the production costs of the catalyst are
significantly greater than those of an alkaline one (Me-
her et al., 2004).
The recent concerns with sustainability, environment,
and raw material costs have made the use of waste frying
oils attractive to the industry, especially if stimulated
with economic benefits associated with the use of blends
of biodiesel and fossil diesel (Zhang et al., 2003b). Veg-
etable oils used in industrial or household frying under-
go degradation by hydrolytic and oxidative reactions,
both processes being responsible for changes in the
chemical and physical properties, as compared to the
fresh oil (Costa Neto et al., 2000). During frying, the vis-
cosity and acidity of the oil increases, and the oil be-
comes darker and develops an unpleasant odour.
This work describes the optimum conditions for bio-
diesel production using waste frying oils as raw material
in the transesterification process.
2. Materials and methods
WFO were collected in school cafeterias and local
restaurants, and from local domestic consumers. The
numerous fractions were blended in order to obtain
two different types of waste oils. Methanol chromato-
graphic grade (99.5%), sodium hydroxide commercial
grade, and dry magnesium sulphate (>99%) were sup-
plied by José M. Vaz Pereira (Lisbon, Portugal). The
reference standards for gas chromatography (GC) were
obtained from Sigma–Aldrich (Lisbon, Portugal).
2.1. Pre-treatment of the frying oil
Homogenous samples of the two WFO that were pre-
pared were dried over magnesium sulphate and subse-
quently filtered under vacuum to remove any
 P. Felizardo et al. / Waste Management 26 (2006) 487–494
Table 1
Fatty acid composition (mass%) of WFO
Fatty acid Oil sample 1 Oil sample 2
C16:0 8.65 8.29
C18:0 4.70 4.02
C18:1 30.60 27.25
C18:2 55.10 57.62
C18:3 0.30 2.28
C20:0 0.65 0.55
suspended matter and magnesium sulphate crystals. The
treated oil samples (here referred to as oil samples 1 and
2) presented, respectively, a water content of 0.07% and
0.13% by weight, an acid value of 0.42 and 2.07 mg of
KOH/g of oil, and an iodine value of 116.9 and 130.5
(determined by Hannus method – Portuguese Standard
NP 941, 1985). The fatty acid composition of both sam-
ples was obtained by GC (Table 1).
2.2. Transesterification procedure
A sample of 200 g of WFO (equivalent to 228.6 mmol
considering a molar mass of 874.8 g/mol) was trans-
ferred to a two-neck Woulf flask equipped with a ther-
mometer and a reflux condenser. A magnetic bar was
used for stirring. The oil was heated until the desired
temperature was reached (65 °C). At this point, a mix-
ture of methanol and sodium hydroxide was added to
the oil and the transesterification reaction began. The
reactor was kept at around 65 °C for 1 h (Freedman
et al., 1984; Meher et al., 2004). The study was carried
out using a molar ratio of methanol/WFO which varied
from 3.6 to 5.4 and a catalyst quantity equivalent to 0.2–
1.0 mass% of oil.
2.3. Purification of methyl esters phase
At the end of the reaction period (1 h), the glycerol
rich-phase was separated from the methyl ester layer
in a decantation funnel. The latter phase was washed
with water, with a 0.5% HCl solution and again with
water to provide a purified biodiesel. The washed methyl
esters were then dried over magnesium sulphate and fil-
tered under vacuum.
2.4. Analyses
The purified methyl esters were evaluated in a series
of tests to determine if the biodiesel produced met the
European Standard EN 14214 (2003). Measurements
of the density at 15 °C by the hydrometer method and
of the kinematic viscosity at 40 °C were carried out
according to ISO 3675 (1998) and ISO 3104 (1994),
respectively. The iodine value was calculated from the
methyl esters composition (EN 14214, 2003), while the
acid value was obtained by titration (Portuguese Stan-
489
dard NP 903, 1998) and the water content by the
Karl–Fischer method (ISO 12937, 2000). Sodium was
determined by flame photometry (EN 241, 2000). The
concentration of methanol was evaluated by heating at
75 °C a known mass of methyl esters until constant
weight was achieved. The flash point was determined
by the Pensky-Martens method (ISO 2719, 2002). The
methyl ester composition was obtained by gas chroma-
tography using a Varian Star 3400CX Gas Chromato-
graph, with a DB-WAX 52 column from Supelco. The
temperature program was as follows: 140 °C for 5
min, heating until 240 °C at a rate of 4 °C/min and hold-
ing at 240 °C for 10 min. The injector was set up for 250
°C, as well as the FID detector. Helium was used as car-
rier gas, at a pressure of 10 bar. Methyl heptadecanoate
was used as an internal standard (Portuguese Standard
NP EN ISO 5508, 1997). Finally, the sulphated ashes
were quantified according to ISO 3987 (1994).
3. Results and discussion
3.1. Effect of WFO properties on the transesterification
reaction
Freedman et al. (1984) have studied the influence of
the free fatty acids (FFA) and water content in the pro-
duction process of biodiesel. They have found that the
oil acid value should be less than 1 mg KOH/g and that
all raw materials should be anhydrous (water content
<0.3%). If these requirements are not met it is still pos-
sible to produce biodiesel, but the yield of the reaction is
reduced due to the deactivation of the catalyst and the
formation of soaps. Although high acid values may be
corrected by the addition of sodium hydroxide, this pro-
cedure results in the saponification of the free fatty
acids, thus affecting the yield in methyl esters, as well
as the glycerol separation (Freedman et al., 1984).
To determine the influence of the oil acid value on the
production of biodiesel, six experiments were conducted
using WFOs with acid values of 0.42 and 2.07. The
experimental conditions are presented in Table 2, to-
gether with the calculated mass of methyl esters (MEs)
produced per 100 g of WFO (mME/100 g WFO) and
the purity of methyl esters phases (% ME).
In the first two experiments, although the yield in
methyl esters is similar for both oil samples, the purity
of the samples is quite different. In all experiments, the
biodiesel produced from the WFO with the lower acid
value had a purity higher than the one obtained from
the more acidic frying oil, by as much as 14%. It was
also observed that those ME layers could be more easily
separated from the glycerol-rich phases than the ones
produced using the more acidic WFO. For the latter, a
pre-treatment step would be advisable in order to obtain
higher purities.
490
Table 2
Influence of acid value in methyl esters phase yield and purity
Experiment Acid value (mg KOH/g)
1 0.42
2 2.07
3 0.42
4 2.07
5 0.42
6 2.07
3.2. Effect of methanol/waste frying oil molar ratio
The methanol/WFO ratio is one of the most impor-
tant variables in methyl esters production. Nye et al.
(1983) achieved a fuel containing 74% of methyl esters
by reacting WFO and methanol (molar ratio of 1:3.6)
at 50 °C, for 24 h in the presence of 0.4% potassium
hydroxide. Freedman et al. (1984) obtained a maximum
conversion (98–99% of methyl esters) by transesterifying
refined oil with methanol, for 1 h at 60 °C, using a molar
ratio methanol/oil of 6:1 with 1% potassium hydroxide
or 0.5% sodium methoxide. De Filippis et al. (1995)
and, more recently, Tomasevic and Siler-Marinkovic
(2003) have also achieved high purity fuels using the
same molar ratio of 6:1 for methanol/frying oils. As rec-
ognized by Tomasevic and Siler-Marinkovic (2003), it is
difficult to find sufficient information about the methan-
olysis reaction of used frying oils in the literature. To
our knowledge, no information is available on the opti-
mal value of the methanol/WFO molar ratio to be used
for various NaOH/WFO mass ratios and vice versa
measured in terms of the yield and purity of the methyl
ester phase.
Fig. 2 shows that for the oil sample with an acid value
of 0.42 mg KOH/g, at constant mass ratio of catalyst/
WFO, the yield of the reactions (expressed as the mass
of methyl esters produced per 100 g of WFO after 1 h
100
95
m M E/100 g of WFO
90
85
80
75
70
65
60
3 3.6 4.2 4.8
Methanol/WFO (mol/mol)
Fig. 2. Influence of the methanol/WFO molar ratio on the yield of
biodiesel production. Experimental conditions: acid value of the oil:
0.42; reaction time: 1 h; catalyst/WFO mass ratio: r, 0.2%; }, 0.4%;
m, 0.6%; j, 0.8%; n, 1.0%.
P. Felizardo et al. / Waste Management 26 (2006) 487–494
mNaOH/mWFO (%) MeOH/WFO (mol/mol) mME/100 g WFO Purity (% ME)
1.0 3.6 80.4 95.3
1.0 3.6 80.9 80.6
0.2 5.4 92.7 83.9
0.2 5.4 86.8 70.8
1.0 5.4 87.0 99.3
1.0 5.4 80.2 84.9
of reaction) increases directly with the excess of metha-
nol. It was also observed that the increase of the meth-
anol excess facilitates the separation of the glycerol
from the MEs phase.
From Fig. 2, it may be concluded that the best cata-
lyst/WFO ratio is 0.6% because it allows the maximum
yield of methyl esters to be achieved. Furthermore, the
use of methanol/WFO ratios higher than 4.2 is not
advantageous in terms of the methyl esters yield. How-
ever, as presented in Section 3.5, using a higher value
may be of importance since this ratio influences the ex-
tent of the reaction and, consequently, the biodiesel pur-
ity and its apparent viscosity.
3.3. Effect of sodium hydroxide/frying oil mass ratio
The amount of catalyst used in the process is another
variable to take into account, not only because it deter-
mines the reaction rate, but also because it can cause
hydrolysis and saponification. Both reactions interfere
with the separation of the glycerol rich phase and with
the methyl esters purification. Freedman et al. (1984)
recommended the use of 1% NaOH or 0.5% NaOCH3
as the optimal concentrations of catalyst. In recent work
(Tomasevic and Siler-Marinkovic, 2003), concentrations
of 0.5%, 1.0%, and 1.5% of KOH were studied and the
optimum value found was 1% of potassium hydroxide.
Since no further data on the optimal sodium hydroxide
concentrations are available in the literature, the results
obtained for 0.2%, 0.4%, 0.6%, 0.8%, and 1.0% NaOH/
WFO mass ratios are presented below.
As shown in Fig. 3, the reaction yields after 1 h of
reaction are maximized when the ratio of catalyst/
WFO ranges from 0.4% to 0.8%. The influence of the
catalyst/WFO ratio seems to be less important when
the methanol/frying oil ratio is higher than 4.8. There-
fore, it is possible to reduce the consumption of sodium
hydroxide without affecting the yield of methyl esters.
3.4. Determination of the optimal conditions for biodiesel
5.4 6 production
As shown in Figs. 2 and 3, for a WFO with an acidity
of 0.42 mg KOH/g WFO and for a reaction time of 1 h,
the optimal conditions to produce biodiesel are a meth-
anol/WFO molar ratio higher than 4.2 and, depending
 P. Felizardo et al. / Waste Management 26 (2006) 487–494
95
90
m M E/ 100 g of WFO
85
80
75
70
65
60
55
50
0.0% 0.2% 0.4% 0.6% 0.8% 1.0%
NaOH/WFO (m/m)
Fig. 3. Influence of the catalyst/WFO mass ratio on the yield of
biodiesel production. Experimental conditions: acid value of the oil:
0.42; reaction time: 1 h; methanol/WFO molar ratio: r, 3.6; }, 4.2; m,
4.8; n, 5.4.
on the alcohol excess, a NaOH/WFO mass ratio
between 0.4% and 0.8%. However, a molar ratio of 4.2
may cause difficulties because the reaction yield is too
sensitive to catalyst concentration changes (Fig. 3).
Therefore, it is advisable to use a methanol/WFO ratio
of 4.8 or higher. On an industrial scale, this is not a
problem since methanol can be recovered from the glyc-
erol and methyl esters phase. However, increasing the
consumption of methanol will inevitably increase its
recovery costs. Concerning the catalyst mass ratio,
Fig. 2 clearly shows that 0.6% is the optimal value be-
cause it gives the higher yields. It is important to empha-
size that these optimal values would correspond to
methanol and catalyst consumptions lower than the
ones cited in the literature (Freedman et al., 1984; Tom-
asevic and Siler-Marinkovic, 2003).
3.5. Characterization of biodiesel
To determine if the biodiesel produced under the dif-
ferent experimental conditions met the specifications of
the standard EN 14214 (2003), several samples of bio-
diesel were submitted to a series of tests.
3.5.1. Density at 15 °C
The standard for biodiesel states that the fuel should
have a density between 0.860 and 0.900 g/cm3. This
property is important mainly in airless combustion sys-
tems because it influences the efficiency of atomization
of the fuel (Ryan et al., 1984). The results obtained
showed that for all the conditions studied, the biodiesel
produced in this study had a density in the range of
0.864–0.900 g/cm3.
3.5.2. Kinematic viscosity at 40° C
Even more than density, this is an important property
regarding fuel atomization, as well as fuel distribution.
For biodiesel to be used in diesel engines, the kinematic
viscosity must be between 3.5 and 5.0 mm2/s. The kine-
491
6.5
Vi scosity (mm2/s)
6
5.5
5
4.5
4
3.5
0.0% 0.2% 0.4% 0.6% 0.8% 1.0%
NaOH/WFO (m/m)
Fig. 4. Influence of the biodiesel production conditions on the
apparent fuel viscosity. Experimental conditions: acid value of the
oil: 0.42; reaction time: 1 h; methanol/WFO molar ratio: r, 3.6; }, 4.2;
m, 4.8; n, 5.4. The broken line (5 mm2/s) indicates the maximum value
for viscosity (EN 14214, 2003).
matic viscosity of the biodiesel samples produced in this
work ranged from 4.1 to 6.3 mm2/s. It is worth noting
that the extent of the reaction, as well as the experimen-
tal conditions used in biodiesel production, greatly influ-
ence the fuel properties and especially its apparent
viscosity (the same is to say, the viscosity of the methyl
esters phase after washing). Indeed, this quantity de-
creases when the methanol/WFO or the NaOH/WFO
ratios are increased, as shown in Fig. 4.
Different viscosities are due to the incomplete reaction
of the frying oils or to the inefficient biodiesel purifica-
tion, leaving conjugated or free glycerin in the methyl
esters phase. In fact, the presence of glycerides changes
the apparent viscosity of the methyl esters, revealing
the extent of the transesterification reaction and the
methyl esters phase purity (De Filippis et al., 1995).
As mentioned before, waste frying oils differ from
‘‘fresh’’ oils mainly due to oxidation and polymerization
of triglycerides. Since a heat-up stage is carried out in the
pre-treatment step, further polymerization can occur,
raising the viscosity of the raw material even further.
To prevent polymerization and the consequent viscosity
increase, minor quantities of radical quenching agents or
antioxidants can be used. Addition of several inhibiting
polymerization reagents to the frying oil is currently
under research in order to determine their influence.
3.5.3. Iodine value
The iodine value is an important measure that allows
determination of the unsaturation degree of the fuel.
This property greatly influences fuel oxidation and the
type of aging products and deposits formed in diesel en-
gines injectors. According to EN 14214 (2003), methyl
esters used as diesel fuel must have an iodine value be-
low 120 g I2 per 100 g of sample.
Analyses of several samples of the produced biodiesel
revealed iodine values between 123 and 130. These high
values were expected since most of the frying oils used in
492 P. Felizardo et al. / Waste Management 26 (2006) 487–494
Portugal are derived from sunflower seeds and are char-
acterized by iodine values between 120 and 140. Fur-
thermore, the iodine value rises during the later stages
of oil degradation and especially if the oil is exposed
to temperatures higher than 180 °C, due to elimination
reactions of the oxidized triglyceride molecules. Supple
et al. (2002) reported that during frying the formation
of C18:2 and C18:3 were enhanced.
Methods to lower the iodine value of biodiesel by
using reducing agents before transesterification are cur-
rently under research at Instituto Superior Técnico.
Some preliminary results are encouraging as they show
the feasibility of producing biodiesel with an iodine va-
lue lower than 120, even when the raw material is de-
rived from sunflower oils.
3.5.4. Acid value
The acid value measures the content of free fatty
acids in biodiesel. Considering that the presence of free
fatty acids influences fuel aging, the European Standard
specifies a maximum value of 0.5 mg of KOH/g of sam-
ple. The maximum acid value found for the biodiesel
produced in this work was 0.47, although for about
80% of the samples analyzed, this parameter ranged
0.15–0.25, which is well below the required value.
3.5.5. Water content
Fuel contaminated with water can cause engine cor-
rosion or react with the glycerides to produce soaps
and glycerine. Therefore, EN 14214 (2003) imposes a
maximum content of 0.05% of water in fuels. In this
study, the produced biodiesels were routinely dried over
anhydrous magnesium sulphate. This technique,
although used in laboratory practice, is not very efficient
and the water content in several samples of the biodiesel
was consequently high, ranging from 0.08% to 0.21%.
On an industrial scale, dewatering is usually carried
out by distillation under vacuum (0.05 bar) at tempera-
tures of 30–40 °C. Accordingly, two samples were also
distilled, in a bench-scale batch column, under similar
operating conditions. The water content in the biodiesel
samples was thus reduced from 0.12% and 0.21% to
0.02% and 0.04%, respectively.
3.5.6. Sodium content
Because the presence of high amounts of sodium in
biodiesel induces metal corrosion as well as saponifica-
tion of the methyl esters phase, EN 14214 (2003) speci-
fies a maximum value of 5 mg/kg. It is well known that
the content of sodium in biodiesel is mainly determined
by the efficiency of the washing step. The sodium con-
tent of several biodiesel samples produced in this work
was in the range of 2.3–5.3 mg/kg of fuel and only two
productions were out of specification. These results indi-
cate that the washing procedure used was efficient in so-
dium removal.
3.5.7. Purity of the methyl esters phase
According to EN 14214 (2003), the minimum
acceptable purity for biodiesel is 96.5% in methyl
esters. To evaluate the conformity of the biodiesel pro-
duced in this study to the European norm, all samples
were analyzed by GC. As shown in Fig. 5, the influence
of the NaOH/WFO ratio on the biodiesel purity is sig-
nificant, especially up to 0.6%. The effect of the molar
ratio of methanol/WFO is also clearly illustrated in the
same figure.
It is therefore possible to conclude that the biodiesel
purity increases when the alcohol/WFO and catalyst/
WFO ratios are augmented up to 4.8% and 0.6%, respec-
tively. The increase of the purity is the net result of an
efficient removal of glycerol and sufficient reaction ex-
tent. It is worth noting that the acceptable EN 14214 pur-
ity could be achieved with consumptions of methanol
and catalyst lower than the ones cited in the literature
(Freedman et al., 1984; Tomasevic and Siler-Marinkovic,
2003).
3.5.8. Methanol content
Methanol is another possible contaminant of the
methyl esters phase responsible for metal corrosion, par-
ticularly of aluminium, and for the decreasing of the fuel
flash point. EN 14214 (2003) specifies a maximum con-
tent of 0.2% of methanol in biodiesel. The alcohol is eas-
ily removed from the esters phase during the washing
step, and in the laboratory biodiesel the methanol con-
tent varied from 0.02% to 0.11%. Even for the less pure
samples (<80% ME), the methanol content was within
the specifications.
3.5.9. Flash point
EN 14214 (2003) specifies that biodiesel must have a
flash point higher than 120 °C. In order to determine
how the methanol content influences the fuel flash point,
this property was determined for some of the samples.
100
95
90
Purity (%)
85
80
75
70
0.0% 0.2% 0.4% 0.6% 0.8% 1.0%
NaOH/Frying Oil Ratio
Fig. 5. Influence of the biodiesel production conditions in the methyl
esters purity. Experimental conditions: acid value of the oil: 0.42;
reaction time: 1 h; methanol/WFO molar ratio: r, 3.6; }, 4.2; m, 4.8;
n, 5.4; the broken line (96.5%) indicates the minimum methyl esters
purity (EN 14214, 2003).
 P. Felizardo et al. / Waste Management 26 (2006) 487–494
The results show that the samples of biodiesel contain-
ing less than 0.1% of methanol had a flash point from
160 to 170 °C. However, for samples with a 0.5% meth-
anol content, the measured flash points (98–110 °C)
were below the minimum acceptable value. It is, there-
fore, worth noting that an increase of only 0.5% in the
methanol content leads to a 50 °C decrease of the biodie-
sel flash point.
3.5.10. Sulphated ash
EN 14214 (2003) establishes a maximum value of
0.02% for sulphated ash. This property is important
since high values are generally associated with cocking
of injectors and fuel filters clogging. The values obtained
for the several productions were between 0.002% and
0.006%, meeting the fuel specifications.
3.6. Correlation between viscosity of biodiesel and purity
of methyl esters phase
The content of methyl esters produced in the transe-
sterification reaction is a variable of the utmost impor-
tance in the industrial production of biodiesel. De
Filippis et al. (1995) have established a fast analytical
method to determine the methyl esters phase purity, just
by making a single viscosity measurement at 40 °C.
According to these authors, the mass fraction of methyl
esters in the biodiesel phase can be correlated with vis-
cosity by
w ¼ a ln m þ b; ð1Þ
where w represents the mass fraction of methyl esters, m
(mm2/s) is the measured viscosity and a and b are con-
stants, which depend on the type of oil and temperature.
The method proposed by De Filippis et al. (1995) was
used in the present work and Fig. 6 presents the correla-
tion established between the methyl ester layer purity
and its measured viscosity. The points presented in
Fig. 6 correspond to the results obtained in several bio-
diesel production experiments using the WFO with acid
values of 0.42 and 2.07.
100
95
Puri ty (% ME)
90
85
80
75 y = -45.055x + 162.85
70 R 2 = 0.9268
65
60
1.3 1.5 1.7 1.9
Ln ( )
Fig. 6. Plot of methyl esters purity vs. ln m. Experimental data, r;
linear regression, ——; slope, 45.055; interception, 162.85;
R2 = 0.9268.
493
Fig. 6 shows that, although the correlation coefficient
of the simple linear equation fitted to the experimental
data is lower than the one obtained by De Filippis et al.
(R2 = 0.998), it still allows a meaningful estimation of
the degree of transesterification. Taking into account
that the used frying oils had different properties, the pro-
posed correlation can be considered of significant value.
4. Conclusions
The production of biodiesel from used frying oils is
feasible by basic catalyzed transesterification and the
biodiesel produced has the quality required to be a diesel
substitute. The results show that after 1 h of reaction, a
methanol/waste frying oil ratio of 4.8 and a catalyst/
waste frying oil weight ratio of 0.6% give the highest
yield of methyl esters and allow an efficient separation/
purification of the methyl esters phase, as shown by
the viscosity reduction and increased purity of this layer.
Further studies will be developed to produce biodiesel
on a pilot scale to meet the specifications of EN 14214.
These studies are to be carried out in a full-automated
machine and concern the optimization of the production
of biodiesel on an industrial scale.
Acknowledgement
Thanks are due to Professor Maria de Fátima Farelo
(Instituto Superior Técnico, Lisboa) for her help in
revising the manuscript.
References
Al-Widyan, M., Al-Shyoukh, A., 2002. Experimental evaluation of the
transesterification of waste palm oil into biodiesel. Bioresour.
Technol. 85, 253–256.
Al-Widyan, M., Tashtoush, G., Abu-Qudais, M., 2002. Utilization of
ethyl ester of waste vegetable oils as fuel in diesel engines. Fuel
Process. Technol. 76, 91–103.
Canakci, M., Van Gerpen, J., 2001. Biodiesel production from oils and
fats with high free fatty acids. Trans. ASAE 44, 1429–1436.
Canakci, M., Van Gerpen, J., 2003. A pilot plant to produce biodiesel
from high free fatty acid feedstocks. Trans. ASAE 46, 945–954.
Connemann, J., Fischer, J., 1998. Biodiesel in Europe 1998 – Biodiesel
processing technologies. In: International Liquid Biofuels Con-
gress, Curitiba, Brasil.
Costa Neto, P.R., Rossi, L., Zagonel, G., Ramos, L., 2000. Produção
de Biocombustı́vel Alternativo ao Óleo Diesel Através da Transe-
sterificação de Óleo de Soja Usado em Frituras. Quim. Nova. 23,
531–537.
De Filippis, P., Giavarini, C., Scarsella, M., Sorrentino, M., 1995.
Transesterification process for vegetable oils: a simple control
method of methyl ester content. J. Am. Oil Chem. Soc. 72 (11),
1399–1404.
Dorado, M.P., Ballesteros, E., Arnal, J.M., Gómez, J., López, F.J.,
2003. Exhaust emissions from a diesel engine fuelled with transe-
sterified waste olive oil. Fuel 82, 1311–1315.
494 P. Felizardo et al. / Waste Management 26 (2006) 487–494
European Standard EN 241, 2000. Liquid petroleum products –
Determination of sodium content – Atomic absorption spectro-
metric method. CEN – European Committee for Standardization,
Brussels, Belgium.
European Standard EN 14214, 2003. Automotive fuels – Fatty acid
methyl esters (FAME) for diesel engine – Requirements and test
methods. CEN – European Committee for Standardization,
Brussels, Belgium.
Freedman, B., Pryde, E.H., Mounts, T.L., 1984. Variables affecting the
yields of fatty esters from transesterified vegetable oils. J. Am. Oil
Chem. Soc. 61 (10), 1638–1643.
González Gómez, M.E., Howard-Hildige, R., Leahy, J.J., Rice, B.,
2002. Winterization of waste cooking oil methyl ester to improve
cold temperature fuel properties. Fuel 81, 33–39.
Haas, M.J., Bloomer, S., Scott, K., 2000. Simple, high-efficiency
shynthesis of fatty acid methyl esters from soapstock. J. Am. Oil
Chem. Soc. 77 (4), 373–379.
International Standard ISO 12937, 2000. Petroleum products –
Determination of water – Coulometric Karl Fischer titration
method, International Organization for Standarization, Geneva,
Switzerland.
International Standard ISO 2719, 2002. Determination of flash point –
Pensky-Martens closed cup method. International Organization for
Standarization, Geneva, Switzerland.
International Standard ISO 3104, 1994. Petroleum Products – Trans-
parent and opaque liquids – Determination of kinematic viscosity
and calculation of dynamic viscosity. International Organization
for Standarization, Geneva, Switzerland.
International Standard ISO 3675, 1998. Crude petroleum and liquid
petroleum products – Laboratory determination of density –
Hydrometer method. International Organization for Standariza-
tion, Geneva, Switzerland.
International Standard ISO 3987, 1994. Petroleum products – Lubri-
cating oils and additives – Determination of sulfated ash. Interna-
tional Organization for Standarization, Geneva, Switzerland.
Khan, A.K., 2002. Research into Biodiesel Kinetics & Catalyst
Development. Brisbane, Australia.
Knothe, G., Dunn, R.O., Bagby, M.O., 1997. Biodiesel: The use of
vegetable oils and their derivatives as alternative diesel fuels. In:
ACS Symposium Sereis 666, pp. 172–208.
Leung, D., 2001. Development of a clean biodiesel fuel in Hong Kong
using recycled oil. Water Air Soil Poll. 130, 277–282.
Meher, L.C., Sagar, D.V., Naik, S.N., 2004. Technical aspects of
biodiesel production by transesterification – A review. Renew.
Sustain. Energy Rev. XX, 1–21.
Mittelbach, M., Tritthart, P., 1988. Diesel fuel derived from vegetable
oils, III. Emission tests using methyl esters of used frying oil. J.
Am. Oil Chem. Soc. 65 (7), 1185–1187.
Mittelbach, M., Pokits, B., Silberholz, A., 1992. Production and fuel
properties of fatty acid methyl esters from used frying oil. In:
Liquid Fuels Renewable Resource, Proceedings of the Alternative
Energy Conference, pp. 74–78.
Nye, M.J., Williamson, T.W., Deshpande, J.H., Schrader, W.H.,
Snively, W.H., Yurkewich, T.P., French, C.L., 1983. Conversion of
used frying oil to diesel fuel by transesterification: preliminary tests.
J. Am. Oil Chem. Soc. 60 (8), 1598–1601.
Pereda Marı́n, J., Barriga Mateos, F., Álvarez Mateos, P., 2003.
Aprovechamiento de las oleinas residuales procedentes del pro-
cesso de refinado de los aceites vegetales comestibles, para la
fabricación de biodiesel. Grasas y Aceites 54, 130–137.
Peterson, G.R., Scarrah, W.P., 1984. Rapeseed oil transesterification by
heterogeneous catalysis. J. Am. Oil Chem. Soc. 61 (8), 1593–1597.
Portuguese Standard NP 903, 1998. Gorduras e óleos comestı́veis –
Determinação do ı́ndice de acidez e da acidez. Método titrimétrico
(Fats and edible oils – Acidity and acid number determination by
titrimetric procedure). Instituto Português da Qualidade, Caparica,
Portugal.
Portuguese Standard NP 941, 1985. Gorduras e óleos comestı́veis –
Determinação do ı́ndice de iodo pela técnica de Hanus. Processo
corrente (Fats and edible oils – Iodine number determination by
Hannus method. Current procedure). Instituto Português da
Qualidade, Caparica, Portugal.
Portuguese Standard NP EN ISO 5508, 1997. Animal and vegetable
fats and oils – Analysis by gas chromatography of methyl esters of
fatty acids. Instituto Português da Qualidade, Caparica, Portugal.
Ryan III, T.W., Dodge, L.G., Callahan, T.J., 1984. The effects of
vegetable oil properties on injection and combustion in two
different diesel engines. J. Am. Oil Chem. Soc. 61, 1610–1619.
Schuchardt, U., Sercheli, R., Vargas, R., 1998. Transesterification of
vegetable oils: a review. J. Braz. Chem. Soc. 9 (1), 199–210.
Supple, B., Howard-Hildige, R., Gonzalez-Gomez, E., Leahy, J.J.,
2002. The effect of steam treating waste cooking oil on the yield of
methyl ester. J. Am. Oil Chem. Soc. 79, 174–178.
Tashtoush, G., Al-Widyan, M., Al-Shyoukh, A., 2003. Combustion
performance and emissions of ethyl ester of a waste vegetable oil in
a water-cooled furnace. Appl. Therm. Eng. 23, 285–293.
Tashtoush, G., Al-Widyan, M., Al-Jarrah, M., 2004. Experimental
study on evaluation and optimization of convertion of waste
animal fat into biodiesel. Energy Conv. Manage. 45, 2697–2711.
Tomasevic, A.V., Siler-Marinkovic, S.S., 2003. Methanolysis of used
frying oil. Fuel Process. Technol. 81, 1–6.
Ullmanns Encyclopedia of Industrial Chemistry, 1992Fats and Fatty
Oils, vol. A10. VCH, New York, pp. 173–244.
Ulusoy, Y., Tekin, Y., Cetinkaya, M., Karaosmanoglu, F., 2004. The
engine tests of biodiesel from used frying oil. Energy Sources 26,
927–932.
Zaher, F., 2003. Utilization of used frying oil as diesel engine fuel.
Energy Sources 25, 819–826.
Zhang, Y., Dubé, M.A., McLean, D.D., Kates, M., 2003a. Biodiesel
production from waste cooking oil: 1. Process design and techno-
logical assessment. Bioresource Technol. 89, 1–6.
Zhang, Y., Dubé, M.A., McLean, D.D., Kates, M., 2003b. Biodiesel
production from waste cooking oil: 2. economic assessment and
sensitivity analysis. Bioresource Technol. 90, 229–240.