Journal of the Taiwan Institute of Chemical Engineers 61 (2016) 166–173

Contents lists available at ScienceDirect

Journal of the Taiwan Institute of Chemical Engineers

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

Chestnut shells to mitigate pesticide contamination

M. Cobas, J. Meijide, M.A. Sanromán, M. Pazos∗
Department of Chemical Engineering, University of Vigo, Campus As Lagoas, Marcosende, 36310 Vigo, Spain

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

Article history: In the last years, the presence of pesticides in aquatic environments has increased mainly as a result of the
Received 10 June 2015 intensification of agricultural activities. Among the different techniques that can be used to remove these
Revised 28 October 2015
contaminants, the adsorption process can disclose as an appropriate treatment. In this work, chestnut shells
Accepted 13 November 2015
were evaluated as a low-cost biosorbent for the pesticides pirimicarb, imidacloprid, acetamiprid and thi-
Available online 28 December 2015
amethoxam, widely used in agricultural activities. Furthermore, economic acid pretreatments were evaluated
Keywords: in order to increase the removal efficiency. The highest removal was obtained when chestnut shells were pre-
Sorption treated with citric acid, increasing a 15% of adsorption capacity in comparison to raw chestnut shells. Scan-
Pesticides ning electron microscopy images, energy disperse spectroscopy analysis, Fourier transform infrared spec-
Chestnut troscopy and Boehm method were performed in order to characterize the biosorbent. The new biosorbent
Acid pretreatments was analysed in deep by kinetic and isotherm studies. The values of the correlation coefficients and stan-
Low-cost biosorbent
dard error of estimate indicated that the Freundlich isotherm and pseudo-second-order model fitted well to
Removal
the experimental data. In addition, the biosorbent showed a good behaviour working in continuous mode.
The reported results open a new alternative for the restoration of polluted aquatic environments through the
application of this new low-cost biosorbent developed.
© 2015 Taiwan Institute of Chemical Engineers. Published by Elsevier B.V. All rights reserved.

1. Introduction numerous advantages, such as eco-friendly, renewable, cheaper, and

easily available, as compared to commercial adsorbents [8]. Lignocel-
Pesticides are considered as one of the persistent organic pollu- lulosic materials have appropriate properties to act as adsorbent, due
tants, and their production and application in agriculture, forestry, to their porosity and the presence of carboxylic and phenolic groups
and domestic activities have progressively increased worldwide dur- in their structure [9]. These materials have a rather polar surface and
ing recent decades [1]. These compounds reach the water resources will prefer polar and ionic adsorbates. Furthermore, simple chemical
by leachate runoff when they are used in agricultural activities. treatments can improve their affinity to compounds with different
Adsorption process is one of the most attractive techniques for re- structure [4,10–12].
moving contaminants in aquatic environments. This treatment has Among the different industrial wastes that can produce valuable
advantages over the other physical–chemical methods because of adsorbent, in this work chestnut shells were selected. This biomate-
simple design; however, its use has been found to be limited because rial is a waste of different food industries and nowadays it has nei-
it can involve high investment related to the adsorbent material cost. ther added value nor reuses. This waste was extensively studied to
Many years ago, activated carbon was used as the adsorbent par ex- adsorb inorganic pollutants such as heavy metals, but there is not so
cellence because it can remove a variety of pollutants [2], but its pro- much bibliography about its capacity of organic pollutants adsorption
duction process is expensive. Therefore, the scientific community is [13]. For this reason, this study evaluates the pesticides adsorption
focusing its attention in the search of new low-cost adsorbent mate- capacity of chestnut shells. Several pesticides (pirimicarb, imidaclo-
rials [3,4]. In recent times, materials easily available such as wastes prid, acetamiprid and thiamethoxam), commonly used in agroindus-
from forestry, agricultural or industrial activities are being studied as trial activities, were selected to evaluate the adsorption capacity of
low-cost adsorbents. Among them, agricultural lignocellulosic mate- chestnut shells. Initially, the efficiency of the waste was evaluated as
rials, commonly used for preparing activated carbon [2,5] appear as raw material and after chemical treatments. After that, physical and
promising biosorbents [6,7]. These lignocellulosic materials possess chemical analyses were carried out and kinetics and isotherms stud-
ies were performed in order to characterize the biosorbent. Finally,
∗
column assays were carried out to validate the efficacy of the new
Corresponding author. Tel.: +34 986812304; fax: +34 986812380.
biosorbent for environmental applications.
E-mail address: [email protected] (M. Pazos).

http://dx.doi.org/10.1016/j.jtice.2015.11.026
1876-1070/© 2015 Taiwan Institute of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
 M. Cobas et al. / Journal of the Taiwan Institute of Chemical Engineers 61 (2016) 166–173 167

2. Material and methods the time to evaluate the kinetic profile and concentration of equilib-
rium. For the isotherms studies only a final sample was taken when
2.1. Pesticides equilibrium was reached.
The samples were centrifuged at 10,0 0 0 rpm for 5 min. The super-
Different pesticides were tested to evaluate the adsorption capac- natant was separated to analyse pH and pollutant removal. The assays
ity of the chestnut shells. Pirimicarb, imidacloprid, acetamiprid and were performed twice, and the experimental error was calculated as
thiamethoxam were assayed as model pesticides. All of them were the standard deviation, which was below 3% in all cases.
provided by Sigma-Aldrich as Pestanal analytical standard.
2.6. Column adsorption system
2.2. Biosorbent
A glass column with dimensions of 16 cm height and 2 cm of
Chestnut shells were collected from Cuevas & Cía, S.A., a food in- internal diameter was employed. The fixed-bed column was filled
dustry placed in Ourense, Spain. In this company, the chestnuts are with 7.2 g of biosorbent. As shown in Fig. 1, different polyamide
used to elaborate Marrón Glacé and the shells were removed with sponge discs (2 cm of diameter and 0.5 cm of thickness) were placed
steam. The chestnut shells were collected and dried in oven at 60 ºC along the column to prevent the biosorbent caking. A continuous flow
for 48 h. After that, they were crushed in a grinder at a particle size (0.62 mL min−1 ) was provided thought a peristaltic pump (Master
lower than 0.5 mm. Flex L/S Model 77202-60). The polluted solution was a mixture of the
four pesticides (5 mg L−1 each one), and it was kept in an amber bot-
2.3. Acid pretreatments tle to prevent light breakdown. The assays were performed at room
temperature. Samples were periodically taken from the outflow, and
Three different acid pretreatments were performed to modify the they were processed as described previously.
surface of chestnut shells. The concentrations of acid solutions were Pesticide distribution along the column was obtained using
based on the literature thus; 1.1 M was used for citric acid and oxalic methanol extraction. For this purpose, 1 g of biosorbent was mixed
acid pretreatments [15] and 0.1 M was used for nitric acid pretreat- with 50 mL of methanol in 250 mL Erlenmeyer flasks during
ment [10]. For each pretreatment, dried chestnut shells were mixed 48 h/25 ºC at 120 rpm (Thermo Scientific MaxQ800) and Teflon screw
with acid solution in a ratio of 1:20 (g:mL) and the mixture was main- caps were used to avoid evaporation. After that, samples were taken
tained at 25 ºC for 3 h under shaking (120 rpm) in a shaker (Thermo and processed as it was described previously.
Scientific MaxQ800). After that, the liquid was decanted and chestnut
shells were dried in an oven at 60 ºC during 24 h and subsequently the
temperature of the oven was increased up 120 ºC and maintained at 2.7. Pesticides analysis
this temperature for 90 min. The obtained biosorbents were repeat-
edly washed with NaHCO3 and with distillate water to remove the Pesticides concentration in aqueous samples was determined by
acid excess and finally they were dried in oven at 60 ºC for 24 h [15]. HPLC (Agilent 1100) equipped with an XDB-C8 reverse-phase column
(150 × 4.6 mm2 i.d., 5 μm). Prior to injection, the samples were fil-
2.4. Biosorbent characterization tered through a 0.45 μm Teflon filter. The injection volume was set at
5 μL, and a gradient of eluent (acetonitrile/water) was pumped at a
Scanning Electron Microscopy and Energy Dispersive Spectrome- rate of 1 mL min−1 for 16 min. Detection was performed with a diode
try (SEM/EDS) were performed on a JEOL JSM-6700F SEM equipped array detector from 200 to 600 nm, and the column temperature was
with an EDS Oxford Inca Energy 300 SEM using an accelerating volt- maintained at 20 ºC.
age of 20 kV. The biosorbent surface was observed and the elemental
analysis was determined. This analysis can be used to obtain informa- 3. Results and discussion
tion about the physical characteristics and the elemental composition
of the biosorbent. 3.1. Acid pretreatments
Fourier transform infrared spectroscopy (FTIR) and Boehm
method are different techniques to obtain information about the In this study, several acid pretreatments [10,12] using citric acid,
functional groups. FTIR analyses were performed in a JASCO FT/IR- oxalic acid, and nitric acid were tested to modify the surface of the
4100 spectrometer to evaluate the changes in the functional groups biosorbent for increasing its sorption capacity. It is expected that the
on the biosorbent surface. Solid samples were dried at 60 ºC in an surface charge of the chemically-treated chestnut shells, particularly
oven overnight, crushed and analysed between 400 and 4000 cm−1 . cellulose fraction, will be transform to negative charge.
On the other hand, the Boehm method [16] was used to measure The adsorption capacity of the raw chestnut shells for the removal
the amount of surface functional groups. 1 g of biosorbent was added of imidacloprid was evaluated and compared with the obtained when
to series of 250 mL Erlenmeyer flasks containing 50 mL of 0.1 N: chestnut shells were pretreated with the different acids. It was found
sodium bicarbonate, sodium carbonate, sodium hydroxide and hy- that the raw material reached a pesticide removal nearby 70%; how-
drochloric acid. The mixtures were shaken 120 rpm (Thermo Scien- ever, the pretreatment with citric and oxalic acids increased their ad-
tific MaxQ800) for 24 h at 25 ºC, after which the resulting suspensions sorption capacity at values around 85%. This can be explained by the
were filtered and 10 mL of aliquot were titrated by 0.1 N hydrochloric fact that these organic acids have carboxylic groups in their structure
acid or 0.1 N sodium hydroxide in order to estimate the residual base and they can be linked to the surface of the chestnut shells increasing
or acid. the number of carboxylic groups [17]. These groups are well-known
as the main groups in the adsorption processes in lignocellulosic ma-
2.5. Flask adsorption studies terials to obtain activated carbon [18] or their direct use [19,20].
On the other hand, the pretreatment realized with nitric acid
The adsorption of the pesticides was studied individually and as reached the lowest removal of pesticide, around 50%. It was expected
mixture. Adsorption assays were carried out in 250 mL Erlenmeyer that the chemically treated biosorbent showed an enhancement in
flasks by mixing 1 g of biosorbent with 50 mL of the pollutant solu- the pollutant removal due to the protonation of the biomass surface
tion. Flasks were shaken in a shaker (Thermo Scientific MaxQ800) at by the acid attack [21]. However, the removal reached was lower than
120 rpm and 25 ºC. During kinetics assays, samples were taken along obtained by the raw material. This can be explained by the fact that
168 M. Cobas et al. / Journal of the Taiwan Institute of Chemical Engineers 61 (2016) 166–173

Fig. 1. Schematic diagram of experimental set-up of column adsorption system: (1) pesticides solution inlet, (2) peristaltic pump, (3) glass beads, (4) polyamide sponge,
(5) biosorbent CA-CS and (6) treated solution outlet.

this acid is a well-known strong oxidant and its chemical attack re- the carboxyl groups with the pretreatment explains the increase of
duced the hemicellulose content of the biosorbent [22] thus the con- the adsorption capacity of the developed biosorbent.
centration of active sites was also reduced. To study in deep of the functional groups associated to the ad-
Based on the previous results, both organic acids can be consid- sorption process, the basic and acidic functional groups on the biosor-
ered appropriate by their use in the pretreatment of the chestnut bent surface were determined by Boehm method and the results are
shells. However, from an economic point of view, citric acid is cheaper shown in Table S1. As can be seen, the citric acid pretreatment in-
than oxalic acid. Therefore citric acid was selected as an appropri- creased notably the number of carboxylic groups on chestnut shells
ate pretreatment for chestnut shells in order to obtain an improved surface, which is according with the FTIR results, while the basic-
biosorbent. ity groups decreased. In addition, the treatment also increased other
oxygenated functional groups as lactones and phenolics. According to
Farahani et al. [24], this fact suggests that the majority of functional
groups on the biosorbent surface are acidic.
3.2. Chestnut shells characterization
SEM images of CA-CS were performed in order to observe the sur-
face morphology of the biosorbent. From these images (Fig. 3), it is
The chemical composition of used chestnut shells, according to
observed that the biosorbent surface has an irregular pattern within
Vázquez et al. [14], was 29.15% acid insoluble lignin, 2.83% acid sol-
the CA-CS particles, meaning a heterogeneous surface. The presence
uble lignin, 19.2% cellulose and 33.82% total sugars. The high per-
of macroporous structures and channels favour a high surface area
centages of cellulose and lignin predict high adsorption ability by the
that could lead to high pollutant uptake. The elemental analysis per-
presence of carbonyl and carboxyl functional groups.
formed with the EDS coupled to SEM showed that the CA-CS was ap-
Different analyses were carried out to determine the modification
proximately composed of 60% carbon and 40% oxygen (Fig. 3).
of functional groups on the chestnut shells surface after the citric acid
pretreatment. The FTIR spectra of raw chestnut shells (named raw-
CS) and the biosorbent pretreated with citric acid (named CA-CS) 3.3. Adsorption kinetic studies
are presented in Fig. 2. In this figure the characteristic bands of the
lignocellulosic materials can be observed. As it is shown, CA-CS in- The adsorption kinetics of obtained biosorbent CA-CS were car-
creased, regarding with the raw-CS, the adsorption band correspond- ried out, with each pesticide individually and in a mixture (Fig. 4), for
ing to the stretch vibration of the carboxyl group around 1700 cm−1 initial pesticide concentration of 10 mg L−1 each one.
(A in Fig. 2) which is in agreement with Sajab et al. [23]. They re- It can be seen for all pesticides, that the equilibrium was reached
ported a similar behaviour in the carboxylic band when kenaf core after 190 min and the maximum uptakes were achieved by pirim-
fibres pretreated with citric acid were characterized. This increase of icarb followed by acetamiprid and imidacloprid. The lowest uptake
 M. Cobas et al. / Journal of the Taiwan Institute of Chemical Engineers 61 (2016) 166–173 169

110

105

% Transmitance
C
100
B
A

95

Raw-CS
90
CA-CS
CA-CS after adsorption

85
4000 3000 2000 1000
-1
Wavenumber (cm )

Fig. 2. FTIR spectrum of raw-CS and CA-CS before and after adsorption of a mixture of the four studied pesticides (A – carboxyl groups, B – aromatic rings and C – amines band).

Fig. 3. SEM images and EDS analysis.

around 0.28 mg g−1 , both individually and in mixture, was obtained equation [25]:
for thiamethoxam. To examine the mechanism of sorption process,
several kinetic models were used. The kinetic studies of the adsorp- 
tion of pesticides were carried out by using the pseudo-first-order (qe exp − qe cal )2
model (Eq. (1)) and pseudo-second-order model (Eq. (2)). SEE = (3)
n−2
qt = qe (1 − exp(−K1 t ) ) (1)
where n is the number of measurements, qe exp is the experimental qe
and qe cal is the calculated qe by the different models.
qt = qe − qe /(1 + qe K2 t ) (2) The parameters for the different models (Table 1), for the pes-
ticides individually and in a mixture, were determined from the
where qt (mg g−1 ) and qe (mg g−1 ) are the amount of adsorbed pes- corresponding plots by the use of Sigma Plot software. The pseudo-
ticide at equilibrium and at time t (min), respectively and K1 and K2 second-order model was the best model to explain the experimen-
are the equilibrium rate constants of pseudo-first-order (min−1 ) and tal data. This provides satisfactory prediction adsorption kinetics as
pseudo-second-order (g (mg min)−1 ), respectively [3]. well as the lowest values of SEE. Accordingly, it can be established
For each model, the standard error of estimate (SEE) was de- that the main dominant adsorption process is the chemisorption.
termined because it brings a measure of fit between theoretical This fact is in accordance with the previous results obtained from
and experimental data. SEE was evaluated through the following the CA-CS characterization, specifically the FTIR analysis and the
170 M. Cobas et al. / Journal of the Taiwan Institute of Chemical Engineers 61 (2016) 166–173

Table 1
Kinetic parameters for the different kinetic models for the studied pesticides individually and in a mixture.

Model Parameters Pesticides individually Pesticides mixture

Thiamethoxam Acetamiprid Imidacloprid Pirimicarb Thiamethoxam Acetamiprid Imidacloprid Pirimicarb

Pseudo-first-order qe (mg g−1 ) 0.2667 0.3114 0.3346 0.3919 0.2595 0.4031 0.3563 0.3947
K1 (min−1 ) 0.076 0.099 0.1149 1.2826 0.1046 0.1065 0.1329 0.5448
R2 0.9877 0.9785 0.9764 0.9916 0.9648 0.9584 0.9656 0.9272
SEE (mg g−1 ) 0.2127 0.2418 0.2580 0.3636 0.2533 0.3914 0.3308 0.2842
Pseudo-second-order qe (mg g−1 ) 0.2777 0.3227 0.3462 0.4196 0.2728 0.4242 0.3715 0.4068
K2 (g (mg min)−1 ) 0.5253 0.6414 0.6997 5.1568 0.5785 0.3752 0.5728 2.4293
R2 0.9944 0.9946 0.9976 0.9996 0.9808 0.9807 0.9788 0.9735
SEE (mg g−1 ) 0.0075 0.0081 0.0063 0.0030 0.0137 0.0212 0.0191 0.0188

confirmed the presence of the organic pollutants on the CA-CS sur-

face.

3.4. Intra-particle diffusion

The intra-particle diffusion model (Eq. (4)) was also used to anal-
yse the adsorption kinetic data which permits the determination of
the rate-limiting step in an adsorption process [27].

qt = Kid t 1/2 + I (4)

where Kid mg(g min1/2 )−1 is the intra-particle diffusion rate constant,
and value of I indicates the thickness of the boundary layer, i.e., the
larger intercept, the greater the boundary layer effect.
If the straight line passes through the origin, then intra-particle
diffusion is the rate-controlling step. Otherwise, other mechanisms
are also involved. An evaluation of intra-particle model of the stud-
ied pesticides (Fig. 5) confirmed that the plot was multilinear, which
indicates that two or more process were involved in the adsorption
of pesticides [2,26]. This implies that the external surface adsorption,
1st step, is relatively very fast and the step of intra-particle diffusion,
2nd step, is rapidly attained and equilibrium adsorption, 3rd step,
starts after 120 min for all pesticides, except pirimicarb, for which
the two first steps are faster and reach the equilibrium after 20 min.

3.5. Adsorption isotherm studies

In this study, the adsorption isotherms were performed using a

mixture of pesticides changing their concentrations in a range of 10–
30 mg L−1 . Higher concentrations of pesticides were not tested be-
cause there are not expected in the environment [28]. The equilib-
Fig. 4. Profiles of pesticide uptakes onto CA-CS to pesticides individually (a) and in a rium data of CA-CS in the adsorption of pirimicarb was not reached
mixture (b) (solid lines are guides to the eye).
at the studied concentrations; therefore the following studies were
only performed with the pesticides thiamethoxam, acetamiprid and
imidacloprid. Different isotherms, Langmuir, Freundlich and Temkin
were used to evaluate the data (Table 2).
Boehm method. Thus, the higher adsorption capacity can be directly Based on the coefficient correlations as well as the SEE values, it
related with carboxyl groups, so the adsorption process is due to can be stated that Freundlich isotherm was the best model of the
binding of the sorbate in these functional groups. This behaviour is studied models (Table 2). The Freundlich isotherm model assumes
also in accordance with the reported results for pesticide adsorption that the adsorption occurs on heterogeneous surfaces or surfaces
by different agricultural wastes [26]. Furthermore, the kinetic con- supporting sites with varying affinities [29]. SEM images demon-
stants, K2 , obtained when the pesticides are in a mixture were lower strated that the biosorbent had heterogeneous surface, so there was a
than when they are individually; therefore, it can be established correlation between the physical characteristics, provided by the SEM
that there is a competition in the adsorption of the different studied images, and the theoretical characteristics, obtained by the isotherms
pesticides [26]. analysis. The isotherm constants are KF (mg g−1 )(L g−1 )1/ n and n, this
To evaluate the surface modification of the CA-CS after the ad- second one characterizes the affinity between the sorbent and the so-
sorption process, FTIR analyses were conducted. After adsorption, the lute, its value should be less than 10 and more than 1 for favourable
band corresponding with aromatic rings, around 30 0 0 cm−1 (B in adsorption conditions. In this study, all n values are higher than 1
Fig. 2), as well as, the amines band, around 2400 cm−1 (C in Fig. 2) which is in agreement with the reported by ElGouzi et al. [30], in their
were increased. As it was reported by Wang et al. [15], these facts study of pesticide adsorption onto different soils.
 M. Cobas et al. / Journal of the Taiwan Institute of Chemical Engineers 61 (2016) 166–173 171

Fig. 5. Intra-particle diffusion plots for adsorption of acetamiprid, imidacloprid, thiamethoxam and pirimicarb on CA-CS. Solids line represent the intra-particle model data.

Table 2
Parameters for the different adsorption isotherm models.

Isotherm Equation Parameters Thiamethoxam Acetamiprid Imidacloprid

Freundlich qe = KF Ce1/n -1
KF (mg g )( L mg ) -1 1/ n
0.0903 0.2012 0.2478
n 1.0661 1.2303 1.1121
R2 0.9761 0.9813 0.9860
SEE (mg g-1 ) 0.0491 0.0495 0.0544
Langmuir qe = qmax 1+bLbCeCe qmax (mg g-1 ) 14.310 4.6984 8.5070
L
bL ( L mg-1 ) 0.0058 0.0378 0.0284
R2 0.9747 0.9727 0.9857
SEE (mg g-1 ) 0.0500 0.0599 0.0551
Temkin qe = RT
bT
ln(KT Ce ) KT (L mg-1 ) 0.5197 0.8769 1.0338
bT (kJ mol-1 ) 5.2137 4.5847 3.7877
R2 0.9395 0.9352 0.9677
SEE (mg g-1 ) 0.0779 0.0923 0.0823

Ce (mg L−1 ) is the liquid phase equilibrium concentration of the pesticides.

1.0

0.8

0.6
C/C0

0.4

0.2 Thiamethoxam
Acetamiprid
Imidacloprid

0.0
0 10 20 30 40 50

t (h)

Fig. 6. Breakthrough curves for adsorption of thiamethoxam, imidacloprid and acetamiprid onto CA-CS fixed-bed column. (Symbols represent experimental data and solid lines
are Thomas model.)
172 M. Cobas et al. / Journal of the Taiwan Institute of Chemical Engineers 61 (2016) 166–173

Fig. 7. Pesticides distribution in the CA-CS after column adsorption assay.

Table 3 biosorbent of the fixed-bed column was divided in three different

Breakthrough parameters for the Thomas model.
sections (lower, middle and upper), as it can see in Fig. 1.
Parameters Thiamethoxam Acetamiprid Imidacloprid Fig. 7 shows the percentages of the concentration of the pesti-
−1 −3 −4
cides in each section of the column after the continuous assay. As it is
KTh (mL (mg min) ) 1.2 × 10 3.8 × 10 4.8 × 10−4
qe (mg g−1 ) 0.2923 0.5517 0.8077
depicted, thiamethoxam, imidacloprid and acetamiprid were almost
R2 0.9869 0.9708 0.9947 homogeneously distributed along the column. However, the 86% of
SEE (mg g−1 ) 0.2073 0.1821 0.1223 the pirimicarb was adsorbed in the lower section, and only the 0.08%
was found in the upper section. This fact was previously explained
because pirimicarb did not reach the saturation in the assay; thereby
more pirimicarb can be adsorbed in the middle and upper section. In
3.6. Column adsorption system
the case of the other tested pesticides, the CA-CS bed was saturated,
and the pesticides were homogenously distributed along the column.
To evaluate the behaviour of the developed biosorbent in a contin-
uous adsorption process various adsorption assays were performed
using a fixed-bed column showed in Fig. 1. For these experiments, 4. Conclusions
the breakthrough curves were determined (Fig. 6). Thiamethoxam,
acetamiprid and imidacloprid reached the saturation after 38–45 h. In this work, chestnut shells were evaluated as a low-cost ad-
On the other hand, pirimicarb did not reach the saturation in 50 h, sorbent for mitigating pesticide pollution. It was found that this in-
so data are not represented in Fig. 6. This behaviour is in accordance dustrial waste has appropriate adsorptive properties for the studied
with the previous adsorption assays where no equilibrium concen- agricultural pesticides. Furthermore, the citric acid pretreatment in-
tration was reached for the adsorption of pirimicarb onto CA-CS. The creased the removal efficiency of this waste by the presence of more
breakthrough curves of acetamiprid, imidacloprid and thiamethoxam carboxylic groups in their structure. The increase of the carboxylic
were evaluated by using Thomas model. As previous studies, the cor- groups after the pretreatment has been demonstrated by the char-
relation coefficients as well as SEE were determined. Thomas model acterization of the biosorbent, both FTIR analysis and the Boehm
is the most widely used in processes of adsorption in fixed-bed col- method. The adsorption process of the studied pesticides using this
umn; it can be explained through the followed equation: new biosorbent is well-represented by the pseudo-second-order ki-
netic model. Freundlich isotherm describes the equilibrium of the ad-
C 1
=  KT h   (5) sorption of thiamethoxam, acetamiprid and imidacloprid onto de CA-
C0 1 + exp Q qe m − C0Ve f CS. The assumption of this isotherm with regard to the adsorption on
heterogeneous surface was confirmed by the SEM images. In addition,
where C0 is the influent concentration and C is the effluent concen- column adsorption system assays demonstrated the appropriate per-
tration at time (mg L−1 ), KTh (mL(min mg)−1 ) is the velocity constant formance of the new biosorbent for working on continuous mode.
of Thomas, qe (mg g−1 ) is the maximum uptake, Q is volumetric feed Therefore, the present study opens a new way for the use of this de-
rate (mL min−1 ), m is biosorbent amount (g) and Vef is volume of ef- veloped biosorbent for restoration of polluted aquatic environment.
fluent (mL). The obtained constants are shown in Table 3. The model
was applied to predict the dynamic behaviour of the adsorption pro-
cess over the range 0.1 < C/C0 < 0.9 corresponding to breakthrough Acknowledgements
and saturation time, respectively. It is showed an appropriate fit be-
tween the experimental and predicted data reaching correlation co- This research has been financially supported by the Xunta de Gali-
efficients higher than 0.98 for the studied pesticides. cia and ERDF Funds (GRC 2013/003). The authors are grateful to the
Previous studies showed that the methanol reaches values of 100% University of Vigo for financial support of the researcher Marta Cobas
of pesticides extraction from CA-CS. Thus, methanol extraction of the under a predoctoral grant and Spanish Ministry of Economy and Com-
CA-CS was performed to evaluate the distribution of the pollutants petitiveness for the financial support of Marta Pazos under the Ramón
along the column after the adsorption process. For this purpose, the y Cajal program.
 M. Cobas et al. / Journal of the Taiwan Institute of Chemical Engineers 61 (2016) 166–173 173

Supplementary materials [14] Vázquez G, Fontenla E, Santos J, Freire MS, González-Álvarez J, Antorrena G. An-
tioxidant activity and phenolic content of chestnut (Castanea sativa) shell and
eucalyptus (Eucalyptus globulus) bark extracts. Ind Crops Prod 2008;28:279–85.
Supplementary material associated with this article can be found, [15] Wang H, Yuan X, Zeng G, Leng L, Peng X, Liao K, et al. Removal of malachite green
in the online version, at doi:10.1016/j.jtice.2015.11.026. dye from wastewater by different organic acid-modified natural adsorbent: kinet-
ics, equilibriums, mechanisms, practical application, and disposal of dye-loaded
References adsorbent. Environ Sci Pollut Res 2014;21:11552–64.
[16] Boehm HP. Chemical identification of surface groups. Adv Catal 1966;16:179–274.
[17] Zhu B, Fan T, Zhang D. Adsorption of copper ions from aqueous solution by citric
[1] Buchanan I, Liang HC, Razaviarani V. Pesticides and herbicides. Water Environ Res
acid modified soybean straw. J Hazard Mater 20 08;153:30 0–8.
2012;84:1533–67.
[18] Chowdhury ZZ, Hamid SBA, Das R, Hasan MR, Zain SM, Khalid K, et al. Preparation
[2] Bouhamed F, Elouear Z, Bouzid J. Adsorptive removal of copper(II) from aqueous
of carbonaceous adsorbents from lignocellulosic biomass and their use in removal
solutions on activated carbon prepared from Tunisian date stones: equilibrium,
of contaminants from aqueous solution. BioResources 2013;8:6523–55.
kinetics and thermodynamics. J Taiwan Inst Chem Eng 2012;43:741–9.
[19] Zuorro A, Lavecchia R, Medici F, Piga L. Spent tea leaves as a potential low-
[3] Cobas M, Sanromán MA, Pazos M. Box-Behnken methodology for Cr (VI) and
cost adsorbent for the removal of azo dyes from wastewater. Chem Eng Trans
leather dyes removal by an eco-friendly biosorbent: F. vesiculosus. Bioresour
2013;32:19–24.
Technol 2014;160:166–74.
[20] Marshall WE, Wartelle LH, Boler DE, Toles CA. Metal ion adsorption by soy-
[4] Kiruba UP, Kumar PS, Prabhakaran C, Aditya V. Characteristics of thermodynamic,
bean hulls modified with citric acid: a comparative study. Environ Technol
isotherm, kinetic, mechanism and design equations for the analysis of adsorption
20 0 0;21:601–7.
in Cd(II) ions-surface modified Eucalyptus seeds system. J Taiwan Inst Chem Eng
[21] Janoš P, Coskun S, Pilarová V, Rejnek J. Removal of basic (Methylene Blue) and
2014;45(6):2957–68.
acid (Egacid Orange) dyes from waters by sorption on chemically treated wood
[5] Temdrara L, Khelifi A, Addoun A, Spahis N. Study of the adsorption properties of
shavings. Bioresour Technol 20 09;10 0:1450–3.
lignocellulosic material activated chemically by gas adsorption and immersion
[22] Tutt M, Kikas T, Olt J. Influence of different pretreatment methods on bioethanol
calorimetry. Desalination 2008;223:274–82.
production from wheat straw. Agron Res 2012;10:269–76.
[6] Fenoll J, Garrido I, Hellín P, Flores P, Vela N, Navarro S. Use of different organic
[23] Sajab MS, Chia CH, Zakaria S, Jani SM, Ayob MK, Chee KL, et al. Citric acid modified
wastes in reducing the potential leaching of propanil, isoxaben, cadusafos and
kenaf core fibres for removal of methylene blue from aqueous solution. Bioresour
pencycuron through the soil. J Environ Sci Health Part B Pestic Food Contam Agric
Technol 2011;102:7237–43.
Wastes 2014;49:601–8.
[24] Farahani M, Abdullah SRS, Hosseini S, Shojaeipour S, Kashisaz M. Adsorption-
[7] Fenoll J, Ruiz E, Flores P, Vela N, Hellín P, Navarro S. Use of farming and agro-
based cationic dyes using the carbon active sugarcane bagasse. Proc Environ Sci
industrial wastes as versatile barriers in reducing pesticide leaching through soil
2011;10:203–8.
columns. J Hazard Mater 2011;187:206–12.
[25] Elkhatib EA, Bennett OL, Wright RJ. Kinetics of arsenite sorption in soils. Soil Sci
[8] Han R, Zhang L, Song C, Zhang M, Zhu H, Zhang L. Characterization of modified
Soc Am J 1984;48:758–62.
wheat straw, kinetic and equilibrium study about copper ion and methylene blue
[26] Rojas R, Vanderlinden E, Morillo J, Usero J, El Bakouri H. Characterization of sorp-
adsorption in batch mode. Carbohydr Polym 2010;79:1140–9.
tion processes for the development of low-cost pesticide decontamination tech-
[9] Gupta VK, Carrott PJM, Ribeiro Carrott MML, Suhas. Low-cost adsorbents: grow-
niques. Sci Total Environ 2014;488–489:124–35.
ing approach to wastewater treatmenta review. Crit Rev Environ Sci Technol
[27] McKay G, Poots VJP. Kinetics and diffusion processes in colour removal from ef-
2009;39:783–842.
fluent using wood as an adsorbent. J Chem Technol Biotechnol 1980;30:279–92.
[10] Zuhra Memon G, Bhanger MI, Akhtar M. The removal efficiency of chestnut
[28] Postigo C, Barceló D. Synthetic organic compounds and their transformation
shells for selected pesticides from aqueous solutions. J Colloid Interface Sci
products in groundwater: occurrence, fate and mitigation. Sci Total Environ
2007;315:33–40.
2015;503-504:32–47.
[11] Sidiras D, Batzias F, Ranjan R, Tsapatsis M. Simulation and optimization of batch
[29] Cobas M, Ferreira L, Sanromán MA, Pazos M. Assessment of sepiolite as a low-cost
autohydrolysis of wheat straw to monosaccharides and oligosaccharides. Biore-
adsorbent for phenanthrene and pyrene removal: kinetic and equilibrium studies.
sour Technol 2011;102:10486–92.
Ecol Eng 2014;70:287–94.
[12] Sajab MS, Chia CH, Zakaria S, Khiew PS. Cationic and anionic modifications of oil
[30] ElGouzi S, Mingorance MD, Draoui K, Chtoun EH, Peña A. Assessment of pheny-
palm empty fruit bunch fibers for the removal of dyes from aqueous solutions.
lurea herbicides sorption on various Mediterranean soils affected by irrigation
Bioresour Technol 2013;128:571–7.
with wastewater. Chemosphere 2012;89:334–9.
[13] Kyzas GZ, Kostoglou M. Green adsorbents for wastewaters: a critical review. Mater
2014;7:333–64.