This article was downloaded by: [Marcos Mafra]

On: 12 June 2013, At: 06:04

Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK

Environmental Technology
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/tent20

Adsorption of glycerol, monoglycerides and

diglycerides present in biodiesel produced from
soybean oil
a b a
Érika de Castro Vasques , Célia R. Granhen Tavares , Carlos Itsuo Yamamoto , Marcos
a a
Rogério Mafra & Luciana Igarashi-Mafra
a
Chemical Engineering Department , Federal University of Paraná , Curitiba , Brazil
b
Chemical Engineering Department , University of Maringá , Paraná , Brazil
Published online: 03 Jun 2013.

To cite this article: Érika de Castro Vasques , Célia R. Granhen Tavares , Carlos Itsuo Yamamoto , Marcos Rogério Mafra &
Luciana Igarashi-Mafra (2013): Adsorption of glycerol, monoglycerides and diglycerides present in biodiesel produced from
soybean oil, Environmental Technology, DOI:10.1080/21622515.2013.770558

To link to this article: http://dx.doi.org/10.1080/21622515.2013.770558

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions

This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to
anyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representation that the contents
will be complete or accurate or up to date. The accuracy of any instructions, formulae, and drug doses should
be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims,
proceedings, demand, or costs or damages whatsoever or howsoever caused arising directly or indirectly in
connection with or arising out of the use of this material.
 Environmental Technology, 2013
http://dx.doi.org/10.1080/21622515.2013.770558

Adsorption of glycerol, monoglycerides and diglycerides present in biodiesel produced from

soybean oil
Érika de Castro Vasquesa , Célia R. Granhen Tavaresb , Carlos Itsuo Yamamotoa , Marcos Rogério Mafraa
and Luciana Igarashi-Mafraa∗
a Chemical Engineering Department, Federal University of Paraná, Curitiba, Brazil; b Chemical Engineering Department,
University of Maringá, Paraná, Brazil.
(Received 6 August 2012; final version received 21 January 2013 )

The most common methods currently used for the removal of waste glycerol, monoglycerides and diglycerides remaining
after phase separation during biodiesel production involve wet processes. These procedures are not environmentally viable
because they require large volumes of water and thus generate significant quantities of effluent. In this study, adsorption was
employed to replace this purification step. Some commercial activated carbons were tested along with adsorbents chemically
modified with HNO3 . A kinetics study was conducted at 30◦ C and adsorption isotherms were obtained at 20◦ C, 30◦ C and
40◦ C. The results indicated that the adsorption of glycerol increased with the use of chemically-modified activated carbon,
Downloaded by [Marcos Mafra] at 06:04 12 June 2013

showing that pH has a strong influence on glycerol adsorption. The pseudo-first-order kinetic model provided the best fit with
the experimental data for the monoglycerides while the pseudo-second-order model showed a better fit for the glycerol and
diglycerides. The Freundlich model had the best fit with experimental data on the adsorption equilibrium for all temperatures.
The thermodynamic study indicated that the adsorption process is endothermic and thus adsorption is favoured by increasing
the temperature. The adsorption process using chemically-modified activated carbon was therefore very effective for the
removal of waste glycerol resulting from biodiesel production, which is of considerable significance given the legal limits
imposed.
Keywords: adsorption; kinetics; isothermal; thermodynamics; biodiesel

1. Introduction a large volume of water generating considerably quantities

Interest in alternative energy is on the increase because
most of the energy consumed worldwide comes from non-
renewable resources such as oil, coal and natural gas.[1]
Biodiesel is a biofuel produced from renewable biolog-
ical sources such as vegetable oils and fats.[1,2] It is
predominantly produced by the transesterification reac-
tion, where an alcohol (usually methanol or ethanol) reacts
with triglycerides in the presence of a catalyst, forming a
methyl ester (or ethyl), glycerol (co-product), monoglyc-
erides, diglycerides (intermediate reaction products) and
non-transesterified triglycerides.[3]
Glycerides present in residues can cause severe
burning.[4] Moreover, depending on the temperature
and storage time, such residues tend to precipitate in
biodiesel.[5] Another contaminant that serves as a parame-
ter for the quality of biodiesel is glycerol. This component
attracts compounds such as water, increasing corrosion of
the engine and reducing its useful life. The process most
frequently used to remove the contaminants present in
biodiesel requires the use of water. However, this step uses
of effluent. The reduction of these effluents in the biodiesel
industry is important for environmental and economic
reasons.[6]
Liu et al. [7] studied the efficiency of the adsorption
process in the removal of traces of glycerol present in
biodiesel wastewater using different adsorbents, and acti-
vated carbon reportedly had the highest adsorption capacity.
The glycerol adsorption capacity was increased by thermal
treatment in an N2 atmosphere and decreased by chemi-
cal treatment through HNO3 oxidation. This result shows
that thermal treatment was more efficient for a hydrophilic
glycerol solution. In this study we evaluated the efficiency
of the adsorption process for the removal of a hydrophobic
solution of glycerol from biodiesel.
The adsorption process has emerged as a promising
technique for the removal of contaminants from biodiesel
because of its low cost, simplicity of design and opera-
tion and the selectivity of some adsorbents. The aim of
this study was to evaluate the adsorption of some contami-
nants present in biodiesel onto commercial-grade granular

∗ Corresponding author. Email: luciana.igarashi@ufpr.br

© 2013 Taylor & Francis

 2 É. de Castro Vasques et al.
activated carbon in equilibrium batch experiments. The data
obtained regarding the adsorption equilibrium could be used
to improve large-scale separation processes.
2. Materials and methods
2.1. Biodiesel production
The materials used for the biodiesel production were soy-
bean oil (in natura), 99.8% methanol (Vetec , Rio de
Janeiro, Brazil), potassium hydroxide (Vetec , Rio de
Janeiro, Brazil), 37% hydrochloric acid fuming (Merck ,
Darmstadt, Germany), sodium chloride (Biotec , Pinhais,
Brazil), sodium sulphate (Biotec , Pinhais, Brazil) and dis-
tilled water. The methodology for the preparation of the
biodiesel was adapted from the steps proposed by Geris
et al. [8] A sample of the crude biodiesel was purified with
hot water to produce biodiesel according to European Norm
EN 14105.[9] The crude biodiesel and B100 were mixed in
order to obtain different percentages of glycerol.
Downloaded by [Marcos Mafra] at 06:04 12 June 2013
2.2. Analysis of glycerol and glycerides
The glycerol, monoglyceride and diglyceride concentra-
tions in the biodiesel solution were determined according to
the appropriate European norm [9] in a gas chromatograph
(Shimadzu GC-2010, Tokyo, Japan). The equipment has
a capillary column stationary phase with 95% dimethyl-
(5% diphenyl)-polysiloxane, length 25 m, inner diameter
0.32 mm and film thickness 0.10 μm. The samples were
injected into the gas chromatograph in duplicate.
2.3. Adsorbent materials and chemical modification
The adsorbent materials examined included two commer-
cial activated carbons (labelled A and B) and HNO3 was
used as an oxidizing agent for their chemical modification.
In this modification 5 g of activated carbon was placed in
a beaker with 50 ml of 65% nitric acid. The mixture was
heated to 60◦ C on a hot plate and held at this temperature
with constant stirring for 1 h. It was then washed with dis-
tilled water five times to remove any residual acid and the
solid was dried at 110◦ C for approximately 1 h.[10] The pH
values of aqueous solutions of the adsorbents were deter-
mined according to standard ASTM D3838 [11] before and
after chemical modification.
2.4. Characterization of the adsorbent
The characteristics of the activated carbons were investi-
gated morphologically by scanning electron microscopy
(SEM) (Jeol JSM-636, Tokyo, Japan) and chemically by
energy dispersive spectroscopy (EDS) (Noran System Six,
Thermo Scientific, East Grinstead, UK). Also, X-ray photo-
electric spectroscopy (XPS) was performed (ESCA-3000,
Thermo VG Scientific, East Grinstead, UK) under ultra-
vacuum (4.0 × 10−9 mbar). The spectra of the Fourier trans-
form infrared spectroscopy (FTIR) samples were recorded
between 4000 and 400 cm−1 using an Excalibur Bio Rad
spectrophotometer (Philadelphia, USA). All of these analy-
ses were performed on two samples, i.e. untreated activated
carbon (AC) and after modification with HNO3 (ACM).
2.5. Adsorption experiments
The adsorption experiments were carried out in a
temperature-controlled orbital shaker (Tecnal TE-421,
Piracicaba, Brazil) at a constant speed of 150 r/min. For all
tests 20 mL of biodiesel solution was placed into a 125 mL
conical flask with 0.300 g of adsorbent material. Adsorp-
tion of glycerol was determined at 30◦ C for 24 h in order to
choose the most efficient adsorbent. The adsorption kinetics
was then investigated at 30◦ C after different time inter-
vals for up to 48 h using only crude biodiesel. Adsorption
isotherms were obtained at 20◦ C, 30◦ C and 40◦ C for glyc-
erol and 20◦ C and 40◦ C for monoglycerides. Solutions with
concentrations ranging from 0.4% (wt) to 0.02% (wt) of
glycerol were prepared by diluting the crude biodiesel with
purified biodiesel (B100). It was not possible to evaluate
the thermodynamics of the diglycerides due to technical
limitations related to their quantification.
The adsorption capacity of the adsorbents was calcu-
lated, based on the changes in the concentrations of glycerol,
monoglycerides and diglycerides in the solution, according
to Equation (1), where q is the adsorption capacity (g g−1 ),
V is the volume of the solution, m is the amount of adsor-
bent and C0 and Ce are the glycerol, monoglyceride and
diglyceride concentrations in the biodiesel solution before
adsorption and at the adsorption equilibrium, respectively.
V · (C0 − Ce )
q= (1)
m
The experimental kinetics data were modelled using
pseudo-first-order and pseudo-second-order adsorption
kinetics. The pseudo-first-order model is described by
Equation (2) where qe is the amount of component adsorbed
at equilibrium (g g−1 ), qt is the amount of component
adsorbed at time t (g g−1 ) and k1 is the rate constant of
pseudo-first-order adsorption (h−1 ).
k1
log(qe − qt ) = log(qe ) − t (2)
2.303
The pseudo-second-order model is presented by Equation
(3) where k2 is the pseudo-second-order rate constant
(g g−1 h−1 ). Values for the rate constants (k1 and k2 ),
equilibrium adsorption capacity (qe ) and the determination
coefficient (R2 ) were calculated for the adsorption of the
glycerol, monoglycerides and diglycerides present in the
crude biodiesel.
t 1 1
= + t (3)
qt k2 · q e
2 qe
Equation (4) represents the Elovich model where α is the
initial adsorption rate and β is the desorption constant. This
model is used to describe the kinetics of the chemisorption
 Environmental Technology 3

of gases on solids. It has also been used to describe the
adsorption of pollutants in aqueous media.[12]
1 1
qt = ln(αβ) + ln(t) (4)
β β
The equilibrium time determined in the kinetics study
was used to obtain the adsorption isotherms. The glycerol
was evaluated at 20◦ C, 30◦ C and 40◦ C and monoglyc-
erides at 20◦ C and 40◦ C, under agitation at 150 r/min. The
adsorption isotherms were obtained applying the Langmuir,
Freundlich and Redlich-Peterson models. Equations (5)
and (6) presents the Langmuir and Freundlich isotherms
respectively, where KL (L g−1 ) is the Langmuir adsorption
constant, KF (L g−1 ) is the Freundlich adsorption constant,
qmax (g g−1 ) is the maximum adsorption capacity, qe (g g−1 )
is the adsorption capacity at the equilibrium concentration
Ce (g L−1 ) and n is a measure of the deviation from linearity.
qmax KL Ce
qe = (5)
Downloaded by [Marcos Mafra] at 06:04 12 June 2013
of qe can be extrapolated to zero, obtaining the Ka value.
For each temperature of the adsorption process there is one
adsorption equilibrium constant and hence one G 0 value.
Equation (10) gives the Gibbs energy of adsorption where
R is the universal ideal gas constant and T (K) is the process
temperature.
Gads = −RT ln(Ka ) (10)
The values for H 0 and S 0 are obtained through the angu-
lar coefficient between ln(Ka ) versus (1/T) obtained from
the Van’t Hoff equation (Equation (11)).
S 0 H 0
ln(Ka ) = − (11)
R RT
Thus, adsorption is a spontaneous process and thermo-
dynamically favourable when the free energy of adsorp-
tion (Gads ) is negative. When an adsorption process is
endothermic the adsorption capacity of a given compo-
nent increases with increasing temperature. Thus, a positive

1 + K L Ce value for H 0 indicates an endothermic adsorption process

qe = KF Ce1/n (6) and a positive value for S 0 suggests an affinity between
the adsorbent and the adsorbate.[14]
The essential characteristics of a Langmuir isotherm can
also be expressed in terms of a dimensionless constant-
3. Results and discussion
separation factor (RL ) given by Equation (7) where C0 is the
initial concentration of components in the biodiesel (g L−1 ). 3.1. Composition of biodiesel produced
The RL value indicates the shapes of the isotherm and hence Table 1 presents the characteristics of the crude biodiesel.
the type of adsorption: irreversible (RL = 0), favourable The purified biodiesel presents a clear yellow appearance
(0 < RL < 1), linear (RL = 1) or unfavourable (RL > 1). while the crude biodiesel has a cloudy aspect indicating the
presence of these contaminants.
1
RL = (7) The concentration of glycerol in the crude biodiesel did
1 + K L C0 not meet the parameters specified by legislation. However,
Equation (8) presents the Redlich-Peterson isotherm, where the values for the monoglycerides and diglycerides were
A, B and g are the constants of the model. These three param- within the specified limits. Although only the amount of
eters of this isotherm incorporate features of the Langmuir glycerol was above the legally established limit, studies
and Freundlich isotherms. The constant g can be used to on the kinetics and thermodynamics of monoglycerides
characterize the isotherm: if g = 1 the Langmuir isotherm and diglycerides were also carried out to evaluate their
is predominant and if g = 0 the Freundlich isotherm is behaviour in the adsorption process, since this data was
predominant. not available in the literature.
ACe
qe = g (8)
1 + BCe 3.2. Adsorption of glycerol by different adsorbents
Table 2 shows the pH values obtained experimentally for
2.6. Thermodynamics of adsorption the adsorbents tested in the adsorption of glycerol present
In the study on the adsorption process, factors such as the
Gibbs energy and entropy must be evaluated to determine Table 1. Characteristics of crude biodiesel.
if the process occurs spontaneously.[13] The thermody-
namic equilibrium constant of adsorption (Ka ) is used in Experimental
the calculations, where qe (g g−1 ) is the equilibrium concen- value in crude Value established by
Characteristic biodiesel law (UE-EN 14214)
tration in the solid phase and Ce (g L−1 ) is the equilibrium
concentration in the solution, as shown in Equation (9). Glycerol 0.4354% Maximum 0.02%
qe Monoglycerides 0.3860% Maximum 0.8%
Ka = (9) Diglycerides 0.0320% Maximum 0.2%
Ce Triglycerides Not detected Maximum 0.2%
pH 2.26
This equilibrium constant is obtained from a plot of (qe /Ce )
Density (g mL−1 ) 0.855 ± 0.0009 0.850–0.900
or ln (qe /Ce ) versus qe . From the linear equation the value
 4 É. de Castro Vasques et al.

Table 2. pH of the adsorbents materials tested in the Table 3. Surface area and pore volume of acti-
adsorption process of glycerol. vated carbon from untreated organic material
(AC) and after modification with HNO3 (ACM).
Adsorbent pH
Surface area Pore volume
Activated carbon A (coconut material) 9.36 Adsorbent (m2 g−1 ) (cm3 g−1 )
Activated carbon B (organic material) 7.98
Activated carbon A modified with HNO3 3.35 AC 945.93 0.504
Activated carbon B modified with HNO3 2.26 ACM 848.50 0.432

100
pH=2,3
pH=3,4

80

60
% adsorbed

pH=7,9

40
Downloaded by [Marcos Mafra] at 06:04 12 June 2013

pH=9,4

20

0
A B A with HNO3 B with HNO3

Adsorbent
Figure 1. Adsorption of glycerol by different adsorbents at 50◦ C
for 24 h.

in crude biodiesel. It can be observed that the untreated

activated carbons (AC) have an alkaline pH. After treatment
with HNO3 the surface chemistry of the AC is modified
and the pH becomes acidic. This chemical modification is
a factor that significantly influences the adsorption process,
as shown in Figure 1.
The results showed that the untreated AC had a low
adsorption capacity. However, the glycerol adsorption
capacity of both ACs when modified with HNO3 increased
Figure 2. Scanning electron micrographs of activated carbon
considerably, that is, from 29% to 83% and 53% to 86% for
before adsorption, magnification ×3000. (a) Untreated activated
carbons A and B, respectively. This increase in the adsorp- carbon (AC); (b) activated carbon modified with HNO3 (ACM).
tion capacity is a strong indication that this process is highly
dependent on the pH of the adsorbent used.
The adsorbent selected for the adsorption experiments On these micrographs it is clear that the disintegration of
was the activated carbon sample B modified with HNO3 , larger particles into smaller particles occurred. According
which was labelled ACM. to a study by Abdel-Nasser and El-Hendawy [10] the ero-
sion of the particles caused by oxidation with HNO3 could
increase the size of macropores due to the distortion and/or
3.3. Characterization of the adsorbent removal of adjacent micropores. This may contribute to the
The surface area and pore volume of the selected adsor- increase in the adsorption capacity of the modified activated
bent are presented in Table 3. The analysis shows that both carbon.
the surface area and volume of the pores reduced with the Table 4 presents the results obtained from the chemical
chemical treatment. analysis by EDS for AC and ACM. There was a significant
The SEM images showed the morphology of the increase in the oxygen present in ACM compared to AC,
untreated AC (Figure 2(a)) and after modification with demonstrating the presence of a high density of acidic oxy-
HNO3 (ACM) (Figure 2(b)). gen groups on the surface of the adsorbent material oxidized
 Environmental Technology
Table 4. Elemental analysis of AC and ACM by energy disper-
sive spectroscopy (EDS).
Elemental analysis (% atomic)
Adsorbent O P K Ca
AC 25.71 12.26 62.03
ACM 91.38 1.63 6.99
Downloaded by [Marcos Mafra] at 06:04 12 June 2013
Figure 3. X-ray photoelectric spectroscopy (XPS) spectrum of
activated carbon before adsorption. (a) Untreated activated carbon
(AC); (b) activated carbon modified with HNO3 (ACM).
with HNO3 . It was not possible to detect nitrogen atoms by
EDS since their levels were below the detection limit of the
device.
On comparing the XPS spectrum for AC (Figure 3(a))
and ACM (Figure 3(b)) a considerable increase in the inten-
sity of oxygen can be noted, corroborating the EDS results.
Furthermore, it is possible to observe the incorporation of
nitrogen, represented by the peak present at a binding energy
of 400 eV.
The FTIR spectra for the AC and ACM samples can
be seen in Figure 4. It presents the presence of functional
groups in a molecule according to their wavenumber. Peaks
can be observed in the region 2340 to 2360 cm−1 and at
670 cm−1 , which correspond to the carbon dioxide present
during analysis. In addition, there is the presence of organic
material in the samples, represented by the region between
1600 and 2530 cm−1 . The presence of hydroxyl groups seen
5
Figure 4. FTIR spectrum of untreated activated carbon (AC) and
activated carbon modified with HNO3 (ACM).
100
80
60
% adsorbed
40
20
0
0 10 20 30 40 50
Time (h)
Figure 5. Time of equilibrium adsorption in concentrations
of crude biodiesel at 30◦ C and 150 r/min. (•) Glycerol; ()
monoglycerides; () diglycerides.
at 3445 and 3450 cm−1 is due to the adsorption of water
molecules. The peak at 1380 cm−1 appears only in the spec-
trum after the chemical treatment, suggesting the presence
of anions derived from nitric acid.[15]
3.4. Adsorption kinetics
The adsorption of glycerol, monoglycerides and diglyc-
erides by the ACM sample analysed at 30◦ C is shown in
Figure 5. The equilibrium for glycerides is reached quickly
in the initial stage, with around 50% being adsorbed within
1 h. Equilibrium was achieved in around 30 h for glyc-
erol, 10 h for monoglycerides and 15 h for diglycerides.
Therefore, the equilibrium time selected for the remaining
experiments was 48 h in order to ensure thermodynamic
equilibrium.
Figures 6, 7 and 8 show the experimental data
on the adsorption kinetics for glycerol, monoglycerides
 6 É. de Castro Vasques et al.

(a) 0.35 (a) 0.25

0.30
0.20
0.25

qt (gmonoglycerides/gACM )
qt (gglycerol/gACM )

0.20 0.15

0.15
0.10

0.10

0.05
0.05

0.00 0.00
0 10 20 30 40 50 0 10 20 30 40 50
Time (h) Time (h)
(b)
(b) 0.35 0,35

0.30
0.30
Downloaded by [Marcos Mafra] at 06:04 12 June 2013

qt (gmonoglycerides/gACM )
0.25
0.25
qt (gglycerol/gACM )

0.20
0.20

0.15
0.15

0.10
0.10

0.05
0.05

0.00
0.00 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 40
ln (t)
In (t)
Figure 7. Modelling of the experimental data for adsorption
Figure 6. Modelling of the experimental data for adsorp- kinetics of monoglycerides using ACM. (a) Pseudo-first-order
tion kinetics of glycerol using ACM. (a) Pseudo-first-order and pseudo-second-order models, (•) experimental data, (---)
and pseudo-second-order models, (•) experimental data, (---) pseudo-first order model and (—) pseudo-second order model;
pseudo-first order model and (—) pseudo-second order model; (b) (•) experimental data and (—) Elovich model.
(b) (•) experimental data and (—) Elovich model.

fit, mainly with the adsorption kinetics data for glycerol

and diglycerides, respectively. These were fitted with
pseudo-first-order (Equation (2)), pseudo-second-order
(Equation (3)) and the Elovich (Equation (4)) models.
The data in Table 5 give the parameters of the pseudo-
first-order, pseudo-second-order and Elovich models as
well as data on the amount of solute adsorbed at equilib-
rium calculated by the models (qcalc ) and experimentally
(qexp ). Table 6 presents the kinetic model parameters of the
intra-particle diffusion model for glycerol, monoglycerides
and diglycerides at 30◦ C.
The theoretical values obtained applying the pseudo-
first-order and pseudo-second-order models (qe(calc) )
showed good agreement with the experimental data
(qe(exp) ). However, the pseudo-first-order model provided a
better determination coefficient for monoglycerides (R2 =
0.92) while the pseudo-second-order model showed a better
fit for glycerol (R2 = 0.97) and diglycerides (R2 = 0.95).
Despite being used to determine the kinetics of chemisorp-
tion of gases in a solid, the Elovich model had a good
(R2 = 0.99).
3.5. Adsorption isotherm
Figures 9(a)–(c) show the adsorption isotherms for glycerol
at 20◦ C, 30◦ C and 40◦ C, respectively, obtained applying
the Langmuir, Freundlich and Redlich-Peterson models.
These adsorption isotherms indicate unfavourable adsorp-
tion, since the amount adsorbed is very low. The param-
eters of the Langmuir, Freundlich and Redlich-Peterson
isotherms were obtained by nonlinear regression using
Statistic 7.0 software (StatSoft , Tulsa, USA), as shown
in Table 6.
On observing the value of qmax it can be noted that
the higher the temperature the larger the amount of glyc-
erol adsorbed. This could be attributed to the fact that
the viscosity of the system decreases with increasing
temperature, which favours the adsorption process. The
Freundlich and Redlich-Peterson isotherms provided the
 Environmental Technology 7

(a) 0.25 best fit to the experimental data for glycerol for all tem-
peratures (R2 > 0.95). Moreover, the parameter n being
lower than 1 indicates unfavourable adsorption and the
0.20
parameter g indicates the predominance of the Freundlich
isotherm.
qt (gdiglycerides/gACM )

0.15 The adsorption isotherms obtained for the monoglyc-

erides obtained at 20◦ C and 40◦ C, applying the Langmuir,
0.10
Freundlich and Redlich-Peterson models, are shown in
Figure 10(a) and (b), respectively.
As observed for glycerol, the profiles of the isotherms
0.05
for monoglycerides also indicate unfavourable adsorption.
The temperature appeared to have no influence on the ACM
0.00 adsorption capacity for glycerol within the range evaluated.
0 10 20 30 40 50
The parameters for the Langmuir, Freundlich and Redlich-
Time (h) Peterson isotherms are reported in Table 6.
(b) 0.35
The Freundlich and Redlich-Peterson isotherms pro-
vided the best fit with experimental equilibrium data for
0.30
monoglycerides (R2 > 0.96). The n parameter for the Fre-
undlich isotherm was lower than 1, indicating unfavourable
Downloaded by [Marcos Mafra] at 06:04 12 June 2013

0.25
qt (gdiglycerides/gACM )

conditions for adsorption, in agreement with the isotherm

0.20 profile obtained. The g value for the Redlich-Peterson
isotherm tended toward zero, indicating the predominance
0.15
of the Freundlich isotherm in this model.
0.10
In this study, it was not possible to detect the adsorption
of diglycerides because the concentration of this component
0.05 was below the detection limits of the equipment used.

0.00
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0

ln (t) 3.6. Thermodynamic adsorption parameters

Figure 8. Modelling of the experimental data of adsorp- The thermodynamic parameters for the adsorption of glyc-
tion kinetics of diglycerides using ACM. (a) Pseudo-first-order erol and monoglycerides using ACM are given in Table 7.
and pseudo-second-order models, () experimental data, (---) The effect of increased adsorption capacity with increas-
pseudo-first-order model and (___ ) pseudo-second-order model; ing temperature was more pronounced at higher concen-
(b) () experimental data and (___ ) Elovich model.
trations of glycerol. This behaviour is observed due to

Table 5. Kinetic model parameters of pseudo-first-order, pseudo-second-order and Elovich models for glycerol, monoglycerides and
diglycerides at 30◦ C.

Pseudo-first-order Pseudo-second-order Elovich model

Conditions at 30◦ C qe(exp) k1 qe(calc) R2 k2 qe(calc) R2 α β R2

Glycerol (C0 = 0.4354%) 0.294 0.63 0.281 0.89 3.58 0.294 0.97 4.56 28.7 0.99
Monoglycerides (C0 = 0.3869%) 0.179 0.40 0.184 0.92 2.62 0.202 0.86 3.86 254 0.91
Diglycerides (C0 = 0.0320%) 0.218 3.09 0.215 0.71 75.3 0.217 0.95 4.39 303 0.97

Table 6. Langmuir, Freundlich and Redlich-Peterson constants for glycerol (20◦ C, 30◦ C and 40◦ C) and monoglycerides (20◦ C and
40◦ C) using ACM.

Langmuir Freundlich Redlich-Peterson

Conditions qmax KL RL R2 KF n R2 A B g R2

Glycerol (20◦ C) 503 0.004 0.99 0.94 8900 0.93 0.95 0.05 −0.9 0.0001 0.95
Glycerol (30◦ C) 4170 0.005 0.99 0.75 8092 0.19 0.99 0.47 −1.1 0.0971 0.99
Glycerol (40◦ C) 6480 0.003 0.99 0.91 7019 0.39 0.96 0.65 −1.7 0.8022 0.97
Monoglycerides (20◦ C) 139 0.0006 0.99 0.69 4.10 0.27 0.96 0.04 −10 0.001 0.96
Monoglycerides (40◦ C) 243 0.0004 0.99 0.71 7.71 0.26 0.98 0.03 −14 0.004 0.99
 8 É. de Castro Vasques et al.

(a) 0.40 (a) 0,06

0.35
0,05
0.30

qe (gmonoglycerides/gACM )
0,04
qe (gglycerol/gACM )

0.25

0.20 0,03

0.15
0,02
0.10

0.05 0,01

0.00
0.000 0.005 0.010 0.015 0.020 0.025 0,00
0,14 0,16 0,18 0,20 0,22 0,24 0,26
Ce (% wt)
Ce (% wt)
(b) 1.0 (b) 0,06

0.8 0,05

qe (gmonoglycerides/gACM )
qe (gglycerol/gACM )

0,04
Downloaded by [Marcos Mafra] at 06:04 12 June 2013

0.6

0,03
0.4

0,02

0.2
0,01

0.0
0.000 0.005 0.010 0.015 0.020 0.025 0,00
0,14 0,16 0,18 0,20 0,22 0,24 0,26 0,28
Ce (% wt)
(c) 1.0
Ce (% wt)

Figure 10. Adsorption isotherm of monoglycerides. (a) 20◦ C;

0.8 (b) 40◦ C. () Experimental data, (___ ) Langmuir, (---) Freundlich
and (-·-·-) Redlich-Peterson models.
qe (gglycerol/gACM )

0.6
the unfavourable adsorption conditions associated with the
isotherm. The value of H 0 was positive, indicating that
0.4 the adsorption process is endothermic. The positive value
of S 0 indicates the affinity of the ACM for glycerol. Also,
0.2 in this case, negative values of G 0 at temperatures of 20◦ C
and 40◦ C indicate that the process is spontaneous and ther-
0.0 modynamically favourable. The variation in the Gibbs free
0.000 0.008 0.016 0.024 0.032
energy for both temperatures evaluated was negative and
Ce (% wt)
did not change with increasing temperature, indicating that
Figure 9. Adsorption isotherm of glycerol. (a) 20◦ C; (b) 30◦ C; the temperature did not affect the adsorption of monoglyc-
(c) 40◦ C. (•) Experimental data, (___ ) Langmuir, (---) Freundlich erides onto ACM. Thus, it was not possible to determine
and (-·-·-) Redlich-Peterson models. the H and S 0 values for this component.

Table 7. Thermodynamic parameters adsorption of glycerol and monoglycerides using ACM.

Temperature (K)
Thermodynamic
parameter 293.15 303.15 313.15

Glycerol Ka 5.61 11.35 17.13

G 0 (kJ mol−1 ) −4203.27 −6122.12 −7396.01
H 0 (kJ mol−1 ) 42.70
S 0 (J mol−1 K−1 ) 160.34
Monoglycerides Ka 0.0374 0.0395
G 0 (kJ mol−1 ) −8.0107 −8.4133
 Environmental Technology
4. Conclusions
(1) Experimental data indicated that the adsorption
process is strongly dependent on the pH of the
adsorbent in aqueous solution since the adsorp-
tion capacity of glycerol greatly increased with
the use of activated carbon chemically modified
with HNO3 even through the surface area and pore
volume of the adsorbent decreased.
(2) The application of the adsorption process for the
removal of glycerol from crude biodiesel using
ACM was very effective. After 30 h the amount of
glycerol remaining in the biodiesel was approxi-
mately 0.02% (wt), which meets the specifications
required by legislation. The adsorption of mono-
glycerides was also effective but it was not pos-
sible to evaluate the adsorption of diglycerides
because the concentrations of this compound after
the adsorption process were below the detection
Downloaded by [Marcos Mafra] at 06:04 12 June 2013
limit.
(3) Experimental data on the adsorption equilibrium
for glycerol and monoglycerides indicated an
unfavourable isotherm. The Freundlich model pro-
vided the best fit with the experimental data since
it is suitable for the working concentration ranges.
(4) The thermodynamic study of glycerol indicated that
the process is favoured by increasing temperature
and is endothermic. However, the equilibrium data
for monoglycerides indicated that the process was
not affected by temperature.
(5) In conclusion, the use of the adsorption process
in the purification of biodiesel was found to be a
promising operation, reaching glycerol concentra-
tions below the limit stipulated by legislation.
References
[1] Yusuf NNAN, Kamarudin SK, Yaakub Z. Overview on
the current trend in biodiesel production. Energy Convers
Manag. 2011;52:2741–2751.
9
[2] Ma F, Hanna MA. Biodiesel production: A review. Bioresour
Technol. 1999;70:1–15.
[3] Lin L, Cunshan Z, Vittayapadung S, Xiangqian S, Ming-
dong D. Opportunities and challenges for biodiesel fuel.
Appl Energy. 2011;88:1020–1031.
[4] Özgül-Yücel S, Türkay S. Purification of FAME by Rice
hull ash adsorption. American Oil Chemists’ Society.
2003;80:373–376.
[5] Yori JC, D’ippolito SA, Benitez VM, Pieck CL,
Vera, CR. Deglycerolization of biodiesel streams by
adsorption over silica beds. Energy Fuels. 2007;21: 347–
353.
[6] Atadashi IM, Aroua MK, Abdul Aziz A. Biodiesel sepa-
ration and purification: A review. Renew Energ. 2011;36:
437–443.
[7] Liu S, Musuku SR, Adhikari S, Fernando S. Adsorption
of glycerol from biodiesel washwaters. Environ Technol.
2009;30:505–510.
[8] Geris R, Santos, NAC, Amaral BA, Maia IS, Castro VD,
Carvalho JRM. Biodiesel from soybeans – The transesteri-
fication reaction for practical classes of organic chemistry.
New Chem. 2007;30:1369–1373.
[9] EN 14105:2003. Fats and oils derived from fatty acid methyl
esters (FAME). Determination of free and total glycerol and
mono-, di and triglycerides.
[10] El-Hendawy ANA. Influence of HNO3 oxidation on the
structure and adsorptive properties of corncob-based acti-
vated carbon. Carbon. 2003;41:713–722.
[11] ASTM D 3838. Standard test method for pH of activated
carbon.
[12] Ranjana D, Talat M, Hasana SH. Biosorption of arsenic from
aqueous solution using agricultural residue ‘rice polish’. J
Hazard Mater. 2009;166:1050–1059.
[13] Ho YS, Mackay G, Yeung LL. Selective adsorbents
from ordered mesoporous silica. Langmuir. 2003;19:3019–
3024.
[14] Rajoriya RK, Prasad B, Mishra IM, Wasewar KL. Adsorp-
tion of benzaldehyde on granular activated carbon: Kinetics,
equilibrium and thermodynamic. Chem Biochem Eng Q.
2007;21:219–226.
[15] Terzyk AP. The influence of activated carbon surface chemi-
cal composition on the adsorption of acetaminophen (parac-
etamol) in vitro, Part II. TG, FTIR, and XPS analysis of
carbons and the temperature dependence of adsorption kinet-
ics at the neutral pH. Colloids Surf A Physicochem Eng Asp.
2001;177:23–45.