Journal of Food Engineering 185 (2016) 9e16

Contents lists available at ScienceDirect

Journal of Food Engineering

journal homepage: www.elsevier.com/locate/jfoodeng

Kinetics and thermodynamics adsorption of carotenoids and

chlorophylls in rice bran oil bleaching
R.S. Pohndorf a, b, T.R.S. Cadaval Jr. a, L.A.A. Pinto a, *
a
School of Chemistry and Food, Unit Operations Laboratory, Federal University of Rio Grande, 96203-900 Rio Grande, RS, Brazil
b
Department of Agroindustrial Science and Technology, Federal University of Pelotas, 96010-900 Pelotas, RS, Brazil

a r t i c l e i n f o a b s t r a c t

Article history: The adsorption characteristics of carotenoids and chlorophylls in the rice bran oil bleaching at different
Received 25 October 2015 temperatures (100e120
C) and activated earth concentrations (0.5e2.5% w/w) were investigated. The
Received in revised form kinetic study showed that, after 20 min the adsorption capacity tends to stay constant. Pseudo-second
20 February 2016
order model was more appropriate to describe the adsorption kinetics for both pigments, especially at
Accepted 30 March 2016
Available online 1 April 2016
higher temperatures. The use of 1% (w/w) of activated earth at 120
C led to a high decrease in chlo-
rophyll content, while the decrease in carotenoids content was less pronounced. Freundlich model was
suitable to represent the equilibrium experimental data for the pigments. The activation energy values
Keywords:
Rice bran oil
showed that the chlorophyll molecules required more energy to be removed from the oil. The rice bran
Bleaching oil bleaching was considered an endothermic, favorable and spontaneous process, and the isosteric heat
Adsorption of adsorption indicated that the activated surface of the earth was heterogeneous.
Isosteric heat © 2016 Elsevier Ltd. All rights reserved.

1. Introduction 0.14% (w/w) (Lerma-García et al., 2009). According to Wilson et al.

World production of rice (Oryza sativa L.) has increased in recent
years, mainly, due to the expansion of planted area and increased
productivity. The south of Brazil is the largest producer of rice
outside Asia and has an average productivity about 20% above the
worldwide. This is mainly due to climatic factors, availability of
lowlands and water. However, with the fluctuation of commodity
prices, a way to add value is the utilization of byproducts of culture,
such as rice bran (FAO, 2015). This byproduct corresponds from 8 to
11% (w/w) of the grain, and it has interesting characteristics for the
food oil industry (Gopinger et al., 2015; Pourali et al., 2009). The rice
bran oil contains proteins (12e13%, w/w), carbohydrates (48e60%,
w/w) and lipids (18e25%, w/w) (Silva et al., 2006). Although being
obtained from a by-product of rice processing, the rice bran oil has a
high content of bioactive phytochemicals and monounsaturated
fatty acids, which confers to the oil high oxidative stability. Rice
bran oil contains too ɣ-oryzanol, which is a natural antioxidant
composed mainly of esters of phytosterols with trans-ferulic acid.
The ɣ-oryzanol content found in the oil ranges from 0.9 to 2.9% (w/
w), and the tocopherols (vitamin E) content ranges from 0.10 to
* Corresponding author.
E-mail addresses: ricardoscherer.eng@gmail.com (R.S. Pohndorf), titoeq@gmail.
com (T.R.S. Cadaval), dqmpinto@furg.br (L.A.A. Pinto).
(2000), the rice bran oil may assist in reducing LDL cholesterol due
to its fatty acid profile and high quantity of unsaponifiable
compounds.
The vegetable oil refinement, including rice bran oil, can be done
by chemical or physical methods. The steps of chemical refinement
typically include degumming, neutralization, bleaching, dewaxing
and deodorizing. The degumming is typically carried out using
phosphoric acid or hot water to remove phospholipids and muci-
laginous gums (Kreps et al., 2014). In the neutralization, removal of
free fatty acids is realized through the addition of an alkali solution
of sodium hydroxide (Marrakchi et al., 2015). Bleaching is an
adsorption operation that aims at removing pigments, oxidation
products and traces of metals using adsorbing substances (Ribeiro
et al., 2001). Some vegetable oils have a high content waxes
which are separated from the oil in the winterization step (dew-
axing) through the oil cooling and crystallization of waxes or via
solvents (Baümler et al., 2007). Finally, the deodorization removes
volatile compounds, such as, aldehydes, ketones and alcohols by
vacuum distillation (Silva et al., 2014). The physical refinement has
been reported in rice bran oil to reduce losses ɣ-oryzanol during the
chemical refinement, especially in the neutralization step (Pestana-
Bauer et al., 2012).
Bleaching is an important step because it removes the excess of
pigments, metals and oxidation products. The main pigments found

http://dx.doi.org/10.1016/j.jfoodeng.2016.03.028
0260-8774/© 2016 Elsevier Ltd. All rights reserved.
10 R.S. Pohndorf et al. / Journal of Food Engineering 185 (2016) 9e16

in vegetable oils are the carotenoids and chlorophylls. Carotenoids carotenoids content was obtained by absorbance at 446 nm and
are red pigment, precursors of vitamin A and have antioxidant expressed in mg kg1 (Eq. (1)), according to MPOB (2005).
activity. They can be divided into carotenes and xanthophylls. The
carotenes most known are a-carotene, b-carotene, ɣ-carotene and 383A446
C¼ (1)
lycopene. The b-carotene is normally the most abundant carotenoid Lc
found in vegetable oils (Rodriguez-Amaya, 1996). The carotenoids
where C is the carotenoids content (mg kg1), A is the absorbance at
removal is important only to improve the visual appearance to the
446 nm, L is the length of cuvette (cm), c is the oil concentration in
consumer, since its permanence assists in oxidative stability of the
hexane (g 100 mL1) and 383 is the extinction coefficient for
oil and provide health benefits. However, more important than the
carotenoids.
carotenoids is the removal of chlorophyll, which by the heat action
The chlorophyll content (predominantly pheophytin-a) was
decomposes into pheophytins, which give the oil an opaque and
determined by absorbance at wavelengths of 630, 670 and 710 nm
dark color (Sabah, 2007).
(Sabah, 2007), and calculated by Eq. (2).
Several studies have investigated the bleaching step of oils from
different sources, such as, hazelnut oil (Bayrak, 2003), soybean oil ½A670  ðA360 þ A710 Þ=2V
(Liu et al., 2008), rapeseed oil (Su et al., 2013), palm oil (Silva et al., Cl ¼ (2)
0; 0964ML
2013), sardine oil (García-Moreno et al., 2013) and cottonseed oil
(Caglayan et al., 2005). However, few studies have reported the being, Cl the chlorophyll content (mg kg1), A630, A670 and A710 the
kinetic behavior and thermodynamic aspects associated with the absorbances (nm), V the volume of hexane (mL), L the length of
compounds adsorption of oils. cuvette (cm) and M the mass of oil (g).
The activated earths are widely used in bleaching of vegetable
oil. They are mainly composed of clay minerals with surface acti-
2.4. Adsorption assays
vated by acid treatment. The use of bentonite, attapulgite and
sepiolite has been reported in the literature for the adsorption of
Samples (40 g) of neutralized rice bran oil were heated under
pigments in oils (Liu et al., 2008; Worasith et al., 2011). After the
constant stirring (40 rpm) using a magnetic stirrer with heating.
bleaching step, the extraction of adsorbent (spent bleaching earth)
The heating rate was 7
C min1, and the oil was kept at absolute
and the use of oil in the production of renewable diesel as also have
pressure of 70 mmHg. After reaching the desired temperature, the
been proposed in literature (Kuuluvainen et al., 2015).
adsorbent was added to the oil. The kinetic curves were obtained
The adsorbent-adsorbate interactions is a fundamental property
during 60 min using 1% (w/w) of activated earth at 120 ± 1
C. For
in adsorption, and can be elucidated by equilibrium isotherms. The
the equilibrium curves were used different concentrations of
thermodynamic parameters are used in identify the nature of the
adsorbent (from 0.5 to 2.5% w/w) at different temperatures (from
process, and through of kinetic curves is possible to determine the
100 to 120
C). The contact time between adsorbent and the oil was
processing time (Bayrak, 2003).
of 120 min (Bayrak, 2003). The adsorbent was immediately sepa-
In this context, the aim was to study the kinetics, equilibrium
rated from the oil by centrifugation (3500  g for 3 min). The pa-
and thermodynamics of the bleaching step of rice bran oil. The
rameters used in this study were based on preliminary tests and
influence of temperature on adsorption kinetics of carotenoids and
literature (Ribeiro et al., 2001; Silva et al., 2013). The experiments
chlorophylls was evaluated by the Arrhenius relationship, and ki-
were performed in triplicate. The samples were analyzed for ca-
netic models of pseudo-first order, pseudo-second order and Elo-
rotenoids and chlorophylls contents.
vich were fitted to experimental data. The equilibrium isotherms
The adsorption capacities at any time (qt) and in equilibrium (qe)
were obtained using different adsorbents concentrations, and the
were determined by Eqs. (3) and (4):
changes in thermodynamic parameters (Gibbs free energy,
enthalpy, entropy and isosteric heat) were estimated. Mo ðCi  Ct Þ
qt ¼ (3)
Ma
2. Material and methods
Mo ðCi  Ce Þ
2.1. Materials qe ¼ (4)
Ma
The neutralized rice bran oil was obtained from a local industry
being, Mo the oil quantity (kg), Ma the amount of adsorbent (kg), Ci
and stored at 20
C to avoid oxidation. Activated earth (Tonsil
the initial adsorbate concentration (mg kg1), Ct and Ce are the
Supreme 110 FF) was used as commercial adsorbent in bleaching
adsorbate concentrations (mg kg1) at time t and at equilibrium,
experiments.
respectively. Usually the concentration is expressed in mg L1, but
in this study the concentration of the solution is equivalent to the
2.2. Adsorbent characterization
pigments content in the oil, which in literature is expressed in basis
weight.
Activated earth was characterized by scanning electron micro-
scopy (SEM) (JEOL JSM-6060 model, Japan) and energy dispersive
spectroscopy (EDS) (JEOL JSM-5800, Japan). The surface area of the 2.5. Kinetic and equilibrium models
adsorbent was determined by a volumetric adsorption analyzer
(Quantachrome Instruments, New 2 Win, USA) using the method of The adsorption kinetics of the pigment was obtained by fit of
Brunauer, Emmett and Teller (BET). pseudo first order (Equation (5)), pseudo-second order (Equation
(6)) and Elovich (Equation (7)) models. These models are based on
2.3. Carotenoids and chlorophylls analyses l the adsorption capacity rather than the solution concentration
(Silva et al., 2013).
The determination of the carotenoids content was carried out by
qt ¼ q1 ð1  expð  k1 tÞÞ (5)
spectrophotometry (Quimis, Q108 model, Brazil). Rice bran oil
samples were diluted in hexane (10% w/v) and filtered. The
 R.S. Pohndorf et al. / Journal of Food Engineering 185 (2016) 9e16 11

and adapted for pigments adsorption in oils (Bayrak, 2003).

t
qt ¼
 (6)
1=k2 q22 þ ðt=q2 Þ vlnðCe Þ DHa
¼ (13)
vT RT 2
1 By integrating Eq. (13) and assuming that the isosteric heat is
qt ¼ þ lnð1 þ abtÞ (7)
a temperature independent, can be obtained the isosteric heat to the
pigment concentrations at equilibrium (Ce) by Eq. (14), being c the
being, k1 and k2 the adsorption rate constants for pseudo-first order
integration constant.
(min1) and pseudo-second order (kg mg1 min1) models,
respectively, q1 and q2 the theoretical values for maximum DHa 1
adsorption capacities (mg kg1) at time t, a the parameter of the ln Ce ¼ þc (14)
R T
Elovich model associated with the initial velocity (mg kg1 min1)
and b the desorption constant (mg kg1).
The effect of temperature on the adsorption rate constant was
by fit to the Arrhenius relationship (Eq. (8)) (Dogan and Alkan, 2.7. Statistical analysis
2003).
Kinetic and equilibrium models were fitted to experimental data
k ¼ ka expðEa =RTÞ (8) through nonlinear regression using Quasi-Newton method. The fit
quality was obtained through determination coefficient (R2),
where, k is the adsorption rate constant associated with the process adjusted determination coefficient (R2adj) and average relative error
quickness (min1), ka is the Arrhenius constant (s1), Ea is the (ARE), as showed in Eqs. (15)e(17), respectively (El-Khaiary and
activation energy (kJ mol1), R is the universal gas constant Malash, 2011). The parameters were calculated by the Statistic 7.0
(8.314  103 kJ mol1 K1) and T the absolute temperature (K). software (StatSoft, USA) and the graphics were built by SigmaPlot
The equilibrium isotherms were adjusted by Langmuir and 10.0 software (Systat, USA).
Freundlich models. The Langmuir isotherm (Eq. (9)) assumes that
0P  2 P  2 1
the adsorption occurs in monolayer, where the surface of the i i
q
B n i;exp  q i;exp  n qi;exp  q i;model C
adsorbent is homogeneous and the active sites have the same af- R2 ¼ @ A (15)
finity and energy (Langmuir, 1918). The Freundlich isotherm (Eq. Pi  2
n q i;exp  q i;exp
(10)) assumes that the surface of the adsorbent is heterogeneous
and, that the stronger binding sites are occupied first (Freundlich,
  n  1 
1906). R2adj ¼ 1  1  R2 (16)
np
qm kL Ce
qe ¼ (9)  
1 þ ðkL Ce Þ
n  qi;model  qi;exp 
100 X  
ARE ¼ (17)
1=nF n i¼1  qi;exp 

qe ¼ kF Ce (10)

being, qm is the maximum adsorption capacity (mg kg1adsorbent), kL where qi;model is the each value of q predicted by the adjusted
is the Langmuir constant (kg adsorbent mg1), kF the Freundlich model, qi;exp is the each value of q experimentally measured, qi;exp
constant (mg kg1adsorbent) (mg kg1oil)1/nF and 1/nF the hetero- the average of q experimentally measured, n is the number of
geneity factor. experimental points and p is the parameters number of the model.

2.6. Thermodynamic evaluation 3. Results and discussion

The pigments adsorption from rice bran oil were evaluated by 3.1. Adsorbent characteristics
Gibbs free energy (DG0, kJ mol1), enthalpy (DH0, kJ mol1) and
entropy (DS0, kJ mol1 K1) changes. The thermodynamic param- The Fig. 1 shows the SEM image and the EDS spectrum of the
eter DG0 was estimated using the parameters obtained in the best activated earth used in this study for the rice bran oil bleaching. In
fit of the isotherm according to Eq. (11), and DH0 and DS0 were Fig. 1(a), it can be observed that the activated earth particles shows
determined by the van't Hoff plot, adjusting data to Eq. (12) and different shapes, irregular surface and a diameter predominantly
obtaining a slope DH0/RT and interception DS0/R (Milonji

c, 2007). lower than 50 mm. The elemental composition shown in the EDS
spectrum (Fig. 1(b)) indicates the percentage mass of silicon (51%),
DG0 ¼ RT lnðke Þ (11) aluminum (13%), oxygen (22%), iron (5%) and magnesium (2%),
besides other elements in lower concentration. The high content of
DG0 DH 0 DS0 silicates in the adsorbent material tends to improve bleaching ca-
¼  (12) pacity (Díaz and Santos, 2001).
RT RT R
The surface area to activated earth was 240.1 ± 5.9 m2 g1,
where, ke is the equilibrium constant obtained from the isotherms determined by BET. Didi et al. (2009) characterized five commercial
that showed better fit (dimensionless), T is the absolute tempera- adsorbents (Tonsil Optimum, Tonsil EX096, Maghnia, Turkey and
ture (K) and R is the universal gas constant Indonesia) used in bleaching of vegetable oil, and the authors found
(8.314  103 kJ mol1 K1). that the surface area ranged from 165 to 340 m2 g1. The results
The heat in the adsorption, at a constant surface area, or isosteric indicated that the lowest bentonite content was found in Turkish
heat (DHa, mol1 kJ) is an important thermodynamic parameter and Indonesian adsorbents. The Indonesian material showed high
used to elucidate the nature of the pigments adsorption, and can be surface area (340 m2 g1bentonite), but its low surface acidity (no acid
obtained by Eq. (13), derived from the Clausius-Clapeyron equation treatment) resulted in lower bleaching capacity. The specific
12 R.S. Pohndorf et al. / Journal of Food Engineering 185 (2016) 9e16

Fig. 2. Influence of temperature on kinetic curves for the carotenoids adsorption from
rice bran oil onto earth activated.

77 and 87% of the maximum adsorption capacity were achieved for

the temperatures 100, 110 and 120
C, respectively. After 10 min
bleaching, the adsorption rate of carotenoids decreases consider-
ably, and stabilized after 45 min. The increase in the bleaching
temperature from 100 to 120
C led to an increase of 20% in ca-
rotenoids adsorption capacity at 60 min. Although high tempera-
tures increase the adsorption rate, can also cause degradation of
Fig. 1. Characteristics of activated earth: (a) SEM image and (b) EDS spectrum. minor compounds, hydrolysis of fatty acids and oxidation in the oil.
In this sense, the maximum temperature bleaching depends on the
characteristics of oil. Ma and Lin (2004) suggest that at lower
surface area in both Tonsil (Optimum and EX096) were higher than temperatures than 100
C, the adsorption is controlled by forces
200 m2 g1, suggesting a pronounced acid activation. The authors van der Waals (physisorption), and that the increase of temperature
concluded that between the commercial adsorbents tested, the favors the adsorption of b-carotene in both outer layers and in the
Tonsil had the highest bleaching capacity and that adsorbents pores of the adsorbent.
based on acid activated bentonite are recommended for bleaching The kinetic curves of chlorophyll adsorption are shown in Fig. 3.
oil. As occurred in the carotenoids adsorption, a high initial adsorption
The surface area depends mainly on the intensity of the acid rate was found and progressively reduced until the end of the
treatment and is proportional to bleaching capacity. Acid activation process. The chlorophyll adsorption capacity was higher than the of
promotes replacement of cations by protons. The metal ions (Feþ2, carotenoids adsorption in 60 min. With the raising the temperature
Mgþ2, Alþ3) are replaced by protons which constitute active sites of from 100 to 120
C the maximum adsorption capacity of chloro-
interaction between the adsorbent particle and the adsorbed phyll was increased by about 35%. According to Tong et al. (2008),
molecules (Christidis et al., 1997). The bleaching capacity is not only
related to the surface area and the intensity of the acid treatment,
several other physical and chemical factors may interfere, such as,
relative polarity, surface active sites, porosity, particle size, pH value
in water, and moisture content (Didi et al., 2009).
The oil industry widely uses the activated earth as adsorbent
because of its low cost. Among the adsorbents used in the
bleaching, bentonite (calcium-montmorillonite) is the most popu-
lar, but the use of other minerals, such as, sepiolite, attapulgite and
kaolinite has been reported (Worasith et al., 2011; Sabah, 2007; Liu
et al., 2008). However, the activated carbon is important by to
present high surface area and porosity. The use of blends of
different adsorbents can to improve the characteristics of the
bleached oil. Van Hoed et al. (2010) reported the use of a mixture of
activated earth/activated carbon (90/10), for rice bran oil bleaching.

3.2. Kinetic studies

Fig. 2 shows the kinetic curves of adsorption carotenoids for rice

bran oil at different temperatures (100e120
C), using 1% (w/w) of
adsorbent in relation to the oil mass. The carotenoids adsorption Fig. 3. Influence of temperature on kinetic curves for the chlorophyll adsorption from
occurred rapidly within the first 10 min of the bleaching, where 69, rice bran oil onto earth activated.
 R.S. Pohndorf et al. / Journal of Food Engineering 185 (2016) 9e16 13

chlorophylls are adsorbed and the equilibrium is reached more

rapidly on the surface of activated earth, indicating increased
chemical interaction between adsorbent and adsorbate.
The models of pseudo-first order, pseudo-second order and
Elovich were fitted to the experimental data, and the results are
presented in Table 1. Based on the highest values of determination
coefficient (R2) and adjusted determination coefficient (R2adj), and
lowest values of average relative error (ARE), it can be affirmed that
the most suitable model to describe the adsorption kinetics of
pigments was the pseudo-second order model, especially for
higher temperatures. The Elovich model also showed a good fit for
the adsorption of carotenoids. This model has been applied in
heterogeneous systems, so, its use in oils can be interesting.
The adsorption rate constants (k1 and k2) are related to the
speed of adsorption. In this regard, the carotenoids were adsorbed
more rapidly than chlorophylls, but the values were similar. The
increase in the oil temperature from 100
C to 120
C doubled the
adsorption rate constant for both pigments. Similar results were
Fig. 4. Arrhenius plot for the pigments adsorption from rice bran oil onto earth
found by Silva et al. (2013), where both models (pseudo-first order
activated.
and pseudo-second order) were suitable to describe the kinetic
behavior of palm oil bleaching. Tong et al. (2008) reported that the
model of pseudo-second order was appropriated to describe the peroxidation, hydrolysis and structural changes.
adsorption of carotenoids and chlorophylls by activated bentonite. The rice bran oil presents different characteristics in relation to
Other models are used in the literature for adsorption of pigments other vegetable oils. Although showing a lower carotenoids con-
in oils (Brimberg, 1982; Monte et al., 2015). tent, when compared to more traditional vegetables oils (palm oil,
The temperature influence on the rice bran oil bleaching was soybean or canola), rice bran oil contains natural antioxidants,
evaluated by the Arrhenius relationship (Fig. 4). The activation which ensures a good thermal stability (Van Hoed et al., 2006).
energy of carotenoids adsorption was of 36.25 ± 3.68 kJ mol1 and Thus, the continuous optimization of the process leads to the
for chlorophyll adsorption was of 42.21 ± 2.83 kJ mol1, indicating reducing costs and maintenance of oil quality. Shahidi (2005) re-
that the chlorophyll molecules require more energy to be removed ported that the optimum temperature range for the bleaching of
from the oil. Mbah et al. (2005) studied the adsorption of shea traditional vegetable oil is between 85 and 110
C, but some ex-
butter pigments (Butyrospernum parkii), and they found values for ceptions such as palm oil require temperatures up to 120
C and
the activation energy from 47 to 73 kJ mol1, concluding that the higher bleaching time. The conditions of time, temperature and
process is predominantly chemical. amount of adsorbent depend on the oil type and adsorbent char-
Pestana-Bauer et al. (2012) reported that the temperature of rice acteristics. Thus, the desired oil quality and energy consumption
bran oil bleaching industrially processed is 110
C. The increase of and economic evaluations must be considered when choosing the
the bleaching temperature for 120
C favors the removal of pig- parameters used in the bleaching.
ments, permitting the use of a lower amount of adsorbent or
reducing the bleaching time, however, the ejectors costs to keeping
3.3. Equilibrium studies
vacuum conditions will be increased (Landucci et al., 2013). The
vacuum system is important because the presence of air at high
The equilibrium isotherms were obtained at different temper-
temperatures favors the oil adverse reactions, such as,
atures (100e120
C) and different concentrations of adsorbent in

Table 1
Kinetic parameters for pigments adsorption from rice bran oil onto earth activated.

Adsorbate Carotenoids Chlorophyll

Temperature (
C) 100 110 120 100 110 120

Pseudo-first order model

q1 (mg kg1) 644 ± 6 709 ± 9 803 ± 19 809 ± 13 907 ± 3 1073 ± 22
k1 (min1) 0.16 ± 0.01 0.22 ± 0.01 0.29 ± 0.01 0.10 ± 0.01 0.14 ± 0.01 0.20 ± 0.01
R2 0.968 0.972 0.995 0.995 0.983 0.997
R2adj. 0.957 0.963 0.993 0.993 0.977 0.996
ARE (%) 7.3 6.3 2.52 3.03 4.69 1.9

Pseudo-second order model

q2 (mg kg1) 714 ± 10 770 ± 12 848 ± 21 919 ± 15 1020 ± 11 1160 ± 18
k2  103 (kg mg1 min1) 0.33 ± 0.02 0.74 ± 0.02 0.69 ± 0.03 0.12 ± 0.02 0.19 ± 0.01 0.28 ± 0.02
R2 0.992 0.992 0.999 0.979 0.997 0.997
R2adj. 0.989 0.989 0.999 0.972 0.996 0.996
ARE (%) 3.59 3.31 0.49 6.02 1.86 2.13

Elovich model
a  103 (mg kg1 min1) 8.46 ± 0.11 10.22 ± 0.56 13.20 ± 0.88 4.78 ± 0.40 5.39 ± 0.20 6.39 ± 0.45
b (kg mg1) 707 ± 55 3952 ± 102 87,254 ± 3256 215 ± 13 607 ± 15 3902 ± 158
R2 0.998 0.998 0.996 0.987 0.995 0.985
R2adj. 0.997 0.997 0.995 0.983 0.993 0.980
ARE (%) 1.35 1.65 2.46 5.86 2.71 4.70
14 R.S. Pohndorf et al. / Journal of Food Engineering 185 (2016) 9e16
relation to the mass of rice bran oil (0.5e2.5%, w/w). The adsorption
equilibrium curves for carotenoids and chlorophylls are shown in
Figs. 5 and 6, respectively.
In Fig. 5 can be seen that the carotenoids adsorption capacity
increased with the rising temperature. This can be explained by the
activation of more adsorption sites on earth activated with
increasing temperature. Another relevant factor is that the increase
in temperature leads to a reduction in viscosity, which increases the
fluidity of the oil. The effect of temperature on the adsorption was
less pronounced than the adsorbent concentration. The same
behavior was found by Boki et al. (1992) in the vegetable oils
bleaching. The initial carotenoids content in rice bran oil was
13.88 ± 0.99 mg kg1 and initial chlorophyll content was
14.10 ± 0.70 mg kg1. The high concentrations of the two pigments
in rice bran oil causes a competition between them in the
adsorption process (multi-component) (Ribeiro et al., 2001; Sabah
et al., 2007).
Chlorophylls showed more affinity with the earth activated than
that carotenoids, as indicated by adsorption capacity values (Fig. 6).
Unlike carotenoids, chlorophylls reached the near-zero values in
highest concentrations of the adsorbent. At 120
C, the use of 1%
(w/w) of activated earth caused a high decrease in chlorophyll
content, while the decrease in carotenoids content was less pro-
nounced. Being the main type of carotenoid found in vegetable oil,
b-carotene is a tetraterpene (C40H56) non-polar and highly soluble
in oil, especially at high temperatures, which makes more difficult
its adsorption when compared to the chlorophylls. Heat and
exposure time are factors which can promote the pheopigments
formation from chlorophylls. Pheophorbide and pheophytin are
Mg2þ free chlorophylls. In pheophytin a/b molecules, the Mg2þ are
replaced by 2Hþ, forming divalent cations protonated which facil-
itate electrostatic interaction with the active sites of bleaching
earth (Sabah, 2007; Ma €ki-Arvela et al., 2014). This elucidates the
fact that the chlorophylls are absorbed more easily than the ca-
rotenoids in the rice bran oil bleaching. Similar behavior was
observed by Ribeiro et al. (2001), which studied the pigments
adsorption and oxidation products in the olive oil.
The equilibrium experimental curves were fitted to the Lang-
muir and Freundlich models, and the results are shown in Table 2.
Both, Langmuir and Freundlich models were efficient to describe
the carotenoids adsorption at different temperatures. However, for
the chlorophyll adsorption, the Freundlich model showed best fit to
the experimental data than the Langmuir model, evidenced by high
Fig. 5. Equilibrium curves for the carotenoids adsorption from rice bran oil onto earth
activated.
Fig. 6. Equilibrium curves for the chlorophylls adsorption from rice bran oil onto earth
activated.
values of determination coefficient (R2) and adjusted determination
coefficient (R2adj), and the smallest values of average relative error
(ARE). These results are in agreement with the literature (Bayrak,
2003; Liu et al., 2008), since the Langmuir model assumes that
the surface area of the adsorbent is homogeneous while the
Freundlich model is based on the energy dependence of sites
adsorption. The heterogeneity factor of Freundlich model (1/n)
showed values from 1 to 10 indicating high affinity of the adsorbate
with the adsorbent, which confirm that the adsorption process is
favorable. The highest kF values showed that the adsorption is more
effective at 120
C.
Temperature and amount of adsorbent in the rice bran oil
bleaching are basic parameters in choosing the optimal operating
conditions by the oil refineries. With respect to energy consump-
tion, the bleaching has the lowest thermal requirements among
refinement steps, especially compared to deodorization, which
requires large quantities of steam and cold water. In contrast, the
adsorbents costs are usually high, which can lead to very expensive
bleaching step. Thus, the use of higher temperatures in industrial
bleaching (>110
C), it is important to perform economic and
technical studies. In this way, important tools can be used to opti-
mize the refining process, such as computer simulation. Addition-
ally, advanced thermodynamic models have been proposed in
order to reproduce the vapor/liquid equilibrium of oil-residual
solvent system, since the residual hexane content in the oil can
decrease the efficiency of refinement steps, in terms for the con-
sumption of steam ejector system (Landucci et al., 2013).
Regarding the oil features, one of the problems faced by oil re-
fineries refers to the quality of the rice bran before extraction. The
rice bran storage conditions in the rice processing industries, as
time and temperature may accelerate the oxidation process and oil
acidity by enzymatic action (Gopinger et al., 2015). During the oil
extraction and refinement, the use of higher temperatures favors
the formation of primary oxidation products (peroxides) and its
conversion into secondary oxidation products (aldehydes, ketones
and alcohols) (Silva et al., 2014). The amount of oxidation products
in the neutralized oil may be changed. However, the amount of
pigments and fatty acids profile should not undergo abrupt changes
unless oil source is changed (Monte et al., 2015). Neutralization is
an important step in rice bran oil refinement, being responsible for
the removal of free fatty acids, but the soap formed carries a large
amount of g-oryzanol. Similarly, the carotenoids which remain in
the oil after bleaching improve the oxidative stability, although
 R.S. Pohndorf et al. / Journal of Food Engineering 185 (2016) 9e16 15

Table 2
Isotherms parameters for pigments adsorption from rice bran onto earth activated.

Adsorbate Carotenoids Chlorophyll

Temperature (ºC) 100 110 120 100 110 120

Langmuir model
qm (mg kg1ads) 1183 ± 62 1164 ± 70 1245 ± 72 1273 ± 52 1503 ± 80 1838 ± 95
KL (kgoil mg1) 0.18 ± 0.01 0.25 ± 0.01 0.33 ± 0.02 3.03 ± 0.04 2.21 ± 0.09 1.73 ± 0.13
R2 0.993 0.990 0.999 0.870 0.822 0.786
R2adj. 0.991 0.987 0.999 0.827 0.763 0.715
ARE (%) 1.59 1.73 0.76 13.62 21.45 28.67

Freundlich model
kF (mg kg1ads) (mg kg1)1/n 258 ± 8 319 ± 6 394 ± 11 771 ± 11 915 ± 12 1064 ± 20
1/nF 2.09 ± 0.07 2.34 ± 0.09 2.49 ± 0.05 3.12 ± 0.02 3.02 ± 0.01 2.85 ± 0.03
R2 0.981 0.998 0.982 0.974 0.981 0.965
R2adj. 0.975 0.997 0.976 0.965 0.974 0.953
ARE (%) 2.61 3.18 3.14 5.62 6.22 9.82

most of these carotenoids and other phytochemicals are lost in 4. Conclusion

deodorization, by the action of high temperatures. After bleaching,
the most undesirable compounds have been removed, leaving only In this work the kinetics, equilibrium and thermodynamics of
minor and volatile compounds to make appreciable oil the bleaching step of rice bran oil were studied. The adsorbent
consumption. particles of activated earth showed different shapes, irregular sur-
faces, diameters lower than 50 mm and surface area of 240 m2 g1.
The kinetic curves indicated that after 20 min of bleaching, the
adsorption rates decreases considerably leading to equilibrium. The
3.4. Thermodynamic results pseudo-second order model was appropriated to represent the
bleaching kinetics for carotenoids and chlorophylls. There was an
Through thermodynamic parameters, the pigments adsorption increase in the adsorption capacity of pigments at 120
C, and the
from rice bran oil was evaluated by the equilibrium constant (ke), chlorophyll content was drastically reduced using 1% (w/w)
Gibbs free energy changes (DG0), enthalpy (DH0) and entropy (DS0)
changes. The results are shown in Table 3.
As can be seen in Table 3, the equilibrium constant (ke) increased
with the rising temperature, indicating that the pigments adsorp-
tion from rice bran oil was favored at 120
C. Negative values of DG0
showed that the process was spontaneous, and based on the pos-
itive values of DH0 it is possible to infer that the process is endo-
thermic. Based on the enthalpy values, the chemisorption controls
the bleaching step of rice bran oil. According to Bayrak (2003), DH0
values above 20 kJ mol1 indicate that the process is predominantly
governed by chemical interactions. The positive values of DS0
indicate that a randomness increase at the solid-liquid interface
during the adsorption.
The isosteric heat of adsorption (DHa), at different concentra-
tions of adsorbents, where in each concentration indicates a same
surface area is shown in Fig. 7. It can be seen that the lowest DHa
values were found in lowest adsorbent concentrations, for both
pigments. This is due to the loading of the adsorbent surface, where
the increased concentration of the adsorbent leads to an area in-
crease, and pigments are preferentially adsorbed at sites with
higher adsorption energy. When the concentration of the adsor-
bent is decreased, the pigments are adsorbed by the weaker energy Fig. 7. Isosteric heat for the pigments adsorption from rice bran oil onto earth
sites, showing lowest values of DHa (Silva et al., 2013). activated.

Table 3
Thermodynamic parameters for the pigments adsorption from rice bran oil onto earth activated.

Temperature (ºC) ke (dimensionless)* DG0 (kJ mol1)* DH0 (kJ mol1)* DS0 (kJ mol1 K1)*
Carotenoids
100 251 ± 7 17.13 ± 0.21 25.42 ± 4.44 0.11 ± 0.01
110 309 ± 7 18.25 ± 0.09
120 381 ± 17 19.41 ± 0.19
Chlorophyll
100 772 ± 12 20.62 ± 0.07 19.38 ± 2.56 0.11 ± 0.01
110 914 ± 15 21.71 ± 0.11
120 1061 ± 22 22.76 ± 0.15
* mean value ± standard deviation (n¼3).
16 R.S. Pohndorf et al. / Journal of Food Engineering 185 (2016) 9e16
adsorbent. Freundlich model was the most suitable to describe
equilibrium curves of bleaching step for both pigments. The rise in
temperature from 100
C to 120
C (in 60 min) increased the
maximum adsorption capacities around 20% and 35%, respectively,
for carotenoids and chlorophyll. The thermodynamic parameters
showed that the process was endothermic, favorable and sponta-
neous. The activation energy values for carotenoids and chlorophyll
adsorption were 36 kJ mol1 and 42 kJ mol1, respectively. The
lowest values of isosteric heat were found in the lowest adsorbent
concentrations for the both pigments.
Acknowledgements
The authors would like to thank CAPES/Brazil (Coordination for
the Improvement of Higher Education Personnel) and CNPq/Brazil
(National Council for Scientific and Technological Development) for
the financial support. CEMESUL/FURG (Electron Microscopy Center
of South/Rio Grande/RS/Brazil) for the scanning electron
microscopy.
References
Bayrak, Y., 2003. Adsorption isotherms in bleaching hazelnut oil. J. Am. Oil Chem.
Soc. 80 (11), 1143e1146. http://dx.doi.org/10.1007/s11746-003-0833-7.
Baümler, E.R., Crapiste, G.H., Carelli, A.A., 2007. Sunflower-oil wax reduction by seed
solvent washing. J. Am. Oil Chem. Soc. 84, 603e608. http://dx.doi.org/10.1007/
s11746-007-1074-0.
Boki, K., Kubo, M., Kawasaki, N., Mori, H., 1992. Adsorption isotherms of pigments
from alkali-refined vegetable oils with clay minerals. J. Am. Oil Chem. Soc. 69
(4), 372e378. http://dx.doi.org/10.1007/BF02636071.
Brimberg, U.I., 1982. Kinetics of bleaching of vegetable oils. J. Am. Oil Chem. Soc. 59
(2), 74e78. http://dx.doi.org/10.1007/BF02678716.
Caglayan, M.O., Kafa, S., Yigit, N., 2005. Al-pillared clay for cottonseed oil bleaching:
an optimization study. J. Am. Oil Chem. Soc. 82, 599e602. http://dx.doi.org/
10.1007/s11746-005-1115-0.
Christidis, G.E., Scott, P.W., Dunham, A.C., 1997. Acid activation and bleaching ca-
pacity of bentonites from the islands of Milos and Chios, Agean, Greece. Appl.
Clay Sci. 12 (4), 329e347. http://dx.doi.org/10.1016/S0169-1317(97)00017-3.
Díaz, F.R.V., Santos, P.S., 2001. Studies on the acid activation of brazilian smectitic
clays. Quim. Nova 24 (3), 345e353. http://dx.doi.org/10.1590/S0100-
40422001000300011.
Didi, M.A., Makhoukhi, B., Azzouz, A., Villemin, D., 2009. Colza oil bleaching through
optimized acid activation of bentonite. A comparative study. Appl. Clay Sci. 42
(3e4), 336e344. http://dx.doi.org/10.1016/j.clay.2008.03.014.
Dogan, M., Alkan, M., 2003. Adsorption kinetics of methyl violet onto perlite.
Chemosphere 50 (4), 517e528. http://dx.doi.org/10.1016/S0045-6535(02)
00629-X.
El-Khaiary, M.I., Malash, G.F., 2011. Common data analysis errors in batch adsorp-
tion studies. Hydrometallurgy 105 (3e4), 314e320. http://dx.doi.org/10.1016/j.
hydromet.2010.11.005.
FAO, 2015. Food and Agriculture Organization of the United Nations. Food Outlook
e Biannual Report on Global Food Markets. Italy, Rome. http://www.fao.org/3/
a-i5003e.pdf.
Freundlich, H.M.F., 1906. Over the adsorption in solution. J. Phys. Chem. 57 (A),
385e470, 1906.
García-Moreno, P.J., Guadix, A., Go
mez-Robledo, L., Melgosa, M., Guadix, E.M., 2013.
Optimization of bleaching conditions for sardine oil. J. Food Eng. 116 (2),
606e612. http://dx.doi.org/10.1016/j.jfoodeng.2012.12.040.
Gopinger, E., Ziegler, V., Catalan, A.A.S., Krabbe, E.L., Elias, M.C., Xavier, E.G., 2015.
Whole rice bran stabilization using a short chain organic acid mixture. J. Stored
Prod. Res. 61, 108e113. http://dx.doi.org/10.1016/j.jspr.2015.01.003.
Kreps, F., Vrbikova  2014. Influence of industrial physical refining on

, L., Schmidt, S.,
tocopherol, chlorophyll and beta-carotene content in sunflower and rapeseed
oil. Eur. J. Lipid Sci. Technol. 116, 1572e1582. http://dx.doi.org/10.1002/
ejlt.201300460.
Kuuluvainen, V., Ma €ki-Arvela, P., Er€ €
anen, K., Holappa, A., Hemming, J., Osterholm, H.,
Toukoniitty, B., 2015. Extraction of spent bleaching earth in the production of
renewable diesel. Chem. Eng. Technol. 38, 769e776. http://dx.doi.org/10.1002/
ceat.201500052.
Landucci, G., Pannocchia, G., Pelagagge, L., Nicolella, C., 2013. Analysis and simu-
lation of an industrial vegetable oil refining process. J. Food Eng. 116 (4),
840e851. http://dx.doi.org/10.1016/j.jfoodeng.2013.01.034.
Langmuir, I., 1918. The adsorption of gases on plane surface of glass, mica and
platinum. J. Am. Chem. Soc. 40 (9), 1361e1403. http://pubs.acs.org/doi/pdf/10.
1021/ja02242a004.
Lerma-García, M.J., Herrero-Martínez, J.M., Simo
-Alfonso, E.F., Mendonça, C.R.B.,
Ramis-Ramos, G., 2009. Composition, industrial processing and applications of
rice bran g-oryzanol. Food Chem. 115 (2), 389e404. http://dx.doi.org/10.1016/j.
foodchem.2009.01.063.
Liu, Y., Huang, J., Wang, X., 2008. Adsorption isotherms for bleaching soybean oil
with activated attapulgite. J. Am. Oil Chem. Soc. 85 (10), 979e984. http://dx.doi.
org/10.1007/s11746-008-1278-y.
Ma, M.H., Lin, C.I., 2004. Adsorption kinetics of b-carotene from soy oil using re-
generated clay. Sep. Purif. Technol. 39 (3), 201e209. http://dx.doi.org/10.1016/j.
seppur.2003.12.007.
€ki-Arvela, P., Hachemi, I., Murzin, D., Yu, 2014. Extraction of carotenoids from
Ma
algae. J. Chem. Technol. Biotechnol. 89, 1607e1626. http://dx.doi.org/10.1002/
jctb.4461.
Marrakchi, F., Kriaa, K., Hadrich, B., Kechaou, N., 2015. Experimental investigation of
processing parameters and effects of degumming, neutralization and bleaching
on lampante virgin olive oil's quality. Food Bioprod. Process. 94, 124e135.
http://dx.doi.org/10.1016/j.fbp.2015.02.002.
Mbah, J.B.B., Kamga, R., Nguetnkam, J.P., Fanni, J., 2005. Adsorption of pigments and
free fatty acids from shea butter on activated Cameroonian clays. J. Lipid Sci.
Technol. 107 (6), 387e394. http://dx.doi.org/10.1002/ejlt.200501149.
Milonji
c, S.K., 2007. A consideration of the correct calculation of thermodynamic
parameters of adsorption. J. Serbian Chem. Soc. 72 (12), 1363e1367. http://dx.
doi.org/10.2298/JSC0712363M.
Monte, M.L., Monte, M.L., Pohndorf, R.S., Crexi, V.T., Pinto, L.A.A., 2015. Bleaching
with blends of bleaching earth and activated can reduces color and oxidation
products of carp oil. European Journal of Lipid Science Technology. v 117 (6),
829e836. http://dx.doi.org/10.1002/ejlt.201400223.
MPOB, 2005. MPOB Test Methods. Determination of Carotene Content. Malaysian
Palm Oil Board, pp. 194e197. Method p 2.6.
Pestana-Bauer, V.R., Zambiazi, R.C., Mendonça, C.R.B., Beneito-Cambra, M., Ramis-
Ramos, G., 2012. ɣ-oryzanol and tocopherol contents in residues of rice bran oil
refining. Food Chem. 134, 1479e1483. http://dx.doi.org/10.1016/j.foodchem.
2012.03.059.
Pourali, O., Asghari, F.S., Yoshida, H., 2009. Simultaneous rice bran oil stabilization
and extraction using sub-critical water medium. J. Food Eng. 95 (3), 510e516.
http://dx.doi.org/10.1016/j.jfoodeng.2009.06.014.
Ribeiro, M.H.L., Lourenço, P.A.S., Monteiro, J.P., Ferreira-Dias, S., 2001. Kinetics of
selective adsorption of impurities from a crude vegetable oil in hexane to
activated earths and carbons. Eur. Food Res. Technol. 213, 132e138. http://dx.
doi.org/10.1007/s002170100347.
Rodriguez-Amaya, D.B., 1996. Assessment of the provitamin A contents of foods e
Brazilian experience. J. Food Compos. Anal. 9 (3), 196e230. http://dx.doi.org/10.
1006/jfca.1996.0028.
Sabah, E., 2007. Decolorization of vegetable oils: chlorophyll-a adsorption by acid-
activated sepiolite. J. Colloid Interface Sci. 310 (1), 1e7. http://dx.doi.org/10.
1016/j.jcis.2007.01.044.
Sabah, E., Cinar, M., Çelik, M.S., 2007. Decolorization of vegetable oils: adsorption
mechanism of b-carotene on acid-activated sepiolite. Food Chem. 100 (4),
1661e1668. http://dx.doi.org/10.1016/j.foodchem.2005.12.052.
Shahidi, F., 2005. Edible Oil and Fat Products: Edible Oils In: Baileys Industrial Oil
and Fat Products 6th Ed., sixth ed., vol. 2. Wiley & Sons, New YorkSilva, M.A.
Silva, M.A., Sanches, C., Amante, E.R., 2006. Prevention of hydrolytic rancidity in rice
bran. J. Food Eng. 75 (4), 487e491. http://dx.doi.org/10.1016/j.jfoodeng.2005.
03.066.
Silva, S.M., Sampaio, K.A., Ceriani, R., Verhe
, R., Stevens, C., De Greyt, W.,
Meirelles, A.J.A., 2013. Adsorption of carotenes and phosphorus from palm oil
onto acid activated bleaching earth: equilibrium, kinetics and thermodynamics.
J. Food Eng. 118 (4), 341e349. http://dx.doi.org/10.1016/j.jfoodeng.2013.04.026.
Silva, S.M., Sampaio, K.A., Ceriani, R., Verhe
, R., Stevens, C., De Greyt, W.,
Meirelles, A.J.A., 2014. Effect of type of bleaching earth on the final color of
refined palm oil. LWT e Food Sci. Technol. 59 (2), 1258e1264. http://dx.doi.org/
10.1016/j.lwt.2014.05.028.
Su, D., Xiao, T., Gu, D., Cao, Y., Jin, Y., Zhang, W., Wu, T., 2013. Ultrasonic bleaching of
rapeseed oil: effects of bleaching conditions and underlying mechanisms.
J. Food Eng. 117 (1), 8e13. http://dx.doi.org/10.1016/j.jfoodeng.2013.01.039.
Tong, J., Wu, Z., Sun, X., Xu, X., Li, C., 2008. Adsorption kinetics of b-carotene and
chlorophyll onto acid-activated. Chin. J. Chem. Eng. 16 (2), 270e276. http://dx.
doi.org/10.1016/S1004-9541(08)60074-9.
Van Hoed, V., Depaemelaere, G., Vila Ayala, J., Santiwattana, P., Verhe
, R., De
Greyt, W., 2006. Influence of chemical refining on the major and minor com-
ponents of rice bran oil. J. Am. Oil Chem. Soc. 83, 315e321. http://dx.doi.org/
10.1007/s11746-006-1206-y.
Van Hoed, V., Vila Ayala, J., Czarnowska, M., De Greyt, W., Verhe
, R., 2010. Opti-
mization of physical refining to produce rice bran oil with light color and high
oryzanol content. J. Am. Oil Chem. Soc. 87, 1227e1234. http://dx.doi.org/
10.1007/s11746-010-1606-x.
Wilson, T.A., Ausman, L.M., Lawton, C.W., Hegsted, D.M., Nicolosi, R.J., 2000.
Comparative cholesterol lowering properties of vegetable oils: beyond fatty
acids. J. Am. Coll. Nutr. 19 (5), 601e607. http://dx.doi.org/10.1080/07315724.
2000.10718957.
Worasith, N., Goodman, B.A., Jeyashoke, N., Thiravetyan, P., 2011. Decolorization of
rice bran oil using modified kaolin. J. Am. Oil Chem. Soc. 88 (12), 2005e2014.
http://dx.doi.org/10.1007/s11746-011-1872-2.