Accepted Manuscript

Fixed-bed column and batch reactors performance in removal of diazinon

pesticide from aqueous solutions by using walnut shell-modified activated
carbon

Mohammad Bayat, Abolghasem Alighardashi, Azam Sadeghasadi

PII: S2352-1864(18)30138-X
DOI: https://doi.org/10.1016/j.eti.2018.08.008
Reference: ETI 267

To appear in: Environmental Technology & Innovation

Received date : 23 March 2018

Revised date : 22 August 2018
Accepted date : 31 August 2018

Please cite this article as:, Fixed-bed column and batch reactors performance in removal of
diazinon pesticide from aqueous solutions by using walnut shell-modified activated carbon.
Environmental Technology & Innovation (2018), https://doi.org/10.1016/j.eti.2018.08.008

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form.
Please note that during the production process errors may be discovered which could affect the
content, and all legal disclaimers that apply to the journal pertain.
Fixed-bed column and batch reactors performance in removal of diazinon pesticide from
aqueous solutions by using walnut shell-modified activated carbon
Mohammad Bayat a, Abolghasem Alighardashi a, *, Azam Sadeghasadi b
a
Department of Water and Wastewater Engineering, Faculty of Civil, Water and Environmental
Engineering, Shahid Beheshti University, Tehran, Iran
b
Department of Environmental Engineering, Faculty of Marine Science and Technology,
Islamic Azad University, Tehran, Iran
Corresponding author: Abolghasem Alighardashi
E-mail address: a_ghardashi@sbu.ac.ir
Corresponding author. Tel.:+982173932467
Abstract
This paper evaluates adsorption of diazinon pesticide from aqueous solution onto walnut shell-
modified activated carbon using fixed-bed column and batch system. Contact time, initial
concentration, and pH were considered as variable parameters to obtain maximum adsorption
capacity using batch technique. The pseudo-first-order, the pseudo-second-order, and Webber
and Morris models were chosen to analyze adsorption kinetic data. The results indicated that
adsorption process followed closely the pseudo-second-order kinetic model. The equilibrium
adsorption data fitted the Freundlich isotherm well, with Kf = 53.82, n= 1.98, and R2= 0.9966.
For continuous fixed-bed column studies, the effect of inlet concentration (15-40 mg.lit-1), flow
rate (9.5-16.5 lit.h-1), and bed height (10-30 cm) was investigated. Results showed that increase
in inlet concentration and also bed height lead to increase of bed capacity. In addition, changes in
flow rate from 9.5 to 14.5 lit.h-1 caused a decrease in bed capacity. However, a minor increase in
bed capacity was observed via an increase in flow rate from 14.5 to 16.5 lit.h-1. The highest bed
capacities were 34.98 and 34.31 mg.g-1 for the same initial concentration of 40 mg.lit-1, flow
rates of 12 and 9.5 lit.h-1, and bed heights of 30 and 20 cm, respectively. Dynamic behavior of
column was investigated by Bohart-Adams and Yoon-Nelson models. According to obtained
results, Yoon-Nelson model can describe well the adsorption process. The findings of the current
study show that the walnut shell-modified activated carbon can be successfully applied to
remove diazinon pesticide from aqueous solution.

Highlights
 Increase in surface area of commercial activated carbon with a simple modification.
 Significant removal of diazinon from aqueous solution in the pH ranges of 1.5-10.
 The pseudo-second-order model is the best to analyze adsorption kinetic.
 Flow rate, initial concentration, and bed height were the experimental conditions.
 The highest bed capacity was 34.98 mg.g-1 with flow rate of 12 lit.h-1.

Keywords: Diazinon, Walnut shell, Adsorption, Fixed-bed column, Isotherms

1 Introduction

Pesticides are known as hazardous pollutants in surface water and groundwater and their
excessive utilization is a major threat to human health (Mostafalou and Abdollahi, 2013).
Effluent from agricultural drainage may be polluted by pesticides, which are released into rivers,
lakes, and groundwater. These effluents also lead to contamination of drinking water (Peters et
al., 2013; Taha et al., 2014). Diazinon is an organophosphate insecticide with a relatively high
solubility in water. Moreover, it is moderately mobile, resistant in soil, and non-polar. Mentioned
characteristics show that diazinon is a high-risk pollutant for surface water and groundwater. In
this regard, diazinon and its derivatives have frequently been detected in water bodies throughout
the world (Aggarwal et al., 2013; Shemer and Linden, 2006).
To tackle this problem, despite of various methods, adsorption on activated carbon has been
proven one of the most practical and the best available procedure for removal of pesticides
(Bhatnagar et al., 2013; Hokkanen et al., 2016; Ngo et al., 2015). Much scientific research has
focused on removal of pesticides using activated carbon derived from agricultural by-products
and renewable sources including coconut frond (Njoku et al., 2014), cannabis sativa (Vukčević et
al., 2015), pumpkin seed hull (Njoku et al., 2013), and langsat empty fruit bunch (Njoku et al.,
2015).
According to FAO statistics (2017) walnut is one of the main agricultural products in Iran with
an annual production of 405,281 tons, which is equivalent to 11% of the worldwide production.
Attractive features of the walnut shell are low cost, considerable available amount, good
mechanical strength, and renewable precursor that make it a beneficial precursor for activated
carbon production (Dahri et al., 2014).
In recent years, few scientific papers have been reported on the removal of diazinon pesticide
from aqueous solution using different methods. Moussavi et al. (2014) investigated the effect of
UVC (ultraviolet C), UVC/ZnO (zinc oxide), and vacuum UV (VUV) processes for oxidation of
diazinon pesticide in water. They reported that the highest diazinon degradation is related to
UVC/ZnO process (93.3%) at pH=5 and reaction time of 30 min. Batch adsorption of diazinon
on bentonite was studied by Ouznadji et al. (2016). The results showed that adsorption capacity
is 5.56 mg.g-1 at pH=5.2, room temperature, and a contact time of 60 min. Moussavi et al. (2013)
studied the removal of diazinon pesticide from water in a batch mode process using adsorption
onto NH4Cl-induced activated carbon. The results indicated that maximum adsorption capacity is
97.5% of 20 mg.lit-1 diazinon with a contact time of 30 min at neutral pH.
Considering frequent detection of diazinon pesticide in freshwater ecosystems and also lack of
enough studies about the removal of diazinon pesticide from aqueous solution, the present study
was conducted to evaluate the removal efficiency of diazinon from water using walnut shell-
modified activated carbon in batch and fixed-bed system for the first time. The purpose of this
study is to optimize diazinon adsorption using experimental parameters including the initial
concentration of diazinon, initial solution pH, contact time, flow rate, and bed height. In addition,
adsorption kinetic and dynamic models and isotherm were studied to evaluate the adsorption
mechanisms.
2 Material and Method

2.1 Chemicals
Analytical grade diazinon (Molecular formula = C12H21N2O3PS, Molecular weight = 304.35
g.mol-1, Solubility in water = 60 mg.lit-1 at 25 , CAS number= 333-41-5, pKa=2.6) was
procured from Sigma Co. as an adsorbate. Deionized water was used for the preparation of all
solutions and other agents (chlorobenzene and acetone) were obtained from Merck Co.
(Germany). The diazinon chemical structure is presented in Fig. 2 (a).
2.2 Adsorbent
Commercial activated carbon (AC) derived from walnut shell was purchased from Part Chemical
Co., Tūyserkān, Iran. For modification, AC was thoroughly blended with 0.1 M H3PO4 for 30
min and the phosphoric acid/carbon ratio was 1:1(w/w). In order to attain modified activated
carbon (MAC), the impregnated sample with phosphoric acid was washed with hot deionized
water consecutively until the pH of the solution reached 6-7 and then overspread outdoor for 4
days.
2.3 Characterization of adsorbent
Barrett-Joyner-Halenda method (BJH) (Barrett et al., 1951) and micropore analysis method (MP
method) (Mikhail et al., 1968) were used to measure pores size in the range of 1-70 nm and
micropore size distribution using Bel Japan Inc., Belsorp-mini-II instrument, respectively. The
functional groups on MAC and AC surface were detected by Fourier transform infrared (FTIR)
spectroscopy (Bruker Tensor 27). The surface morphology of MAC and AC was observed using
a scanning electron microscope (Tescan 8745 R).
2.4 Batch adsorption tests
The batch adsorption experiments were performed in a series of 250 mL Erlenmeyer flasks as a
batch reactor. Certain amounts of MAC (0.02 g) were added to 0.2 L diazinon solutions with
varying concentration of 1-40 mg.lit-1 and then each flask was placed inside a water bath shaker
at 30 and 120 rpm for 2 hours to reach an equilibrium state. To attain optimum pH, the pH
ranges of 1.5-10 of diazinon solution with initial concentration 20 mg.lit-1 was adjusted by drop-
wise adding of 0.1 M HCl or 0.1 M NaOH solution. After determination of optimum pH,
adsorption isotherm was determined with initial concentrations of 1-40 mg.lit-1.
The diazinon uptake at equilibrium, qe (mg.g-1), was calculated by Eq. (1):

= (1)

The quantity adsorbed at time t, qt (mg.g-1), was calculated by Eq. (2):

= (2)

where C0, Ce, and Ct are the liquid-phase concentrations of diazinon at initial, equilibrium, and
time t, respectively. V (lit) is the volume of the solution and M (g) is the weight of the adsorbent.
2.5 Experimental set-up
The pilot shown in Fig. 1 is totally made of Plexiglass with a column bed height of 60 cm and an
internal diameter of 4 cm. Three valves are installed through the sidewall of the column to
sample at various heights of activated carbon (10, 20, and 30 cm). Two flanges are attached
bottom and top of column and inlet is installed to the bottom flange of the column. A safety
valve is connected to dosing pump by a tee to control pressure pump (5 bar). The column was
packed with an activated carbon of 2-5 mm particle size between two supporting layers of sand
(Fig. 1). Different initial concentrations of diazinon solution (15, 30, and 40 mg.lit-1) were passed
in the up-flow mode through the column at the known flow rate of 9.5, 12, 14.5, and 16.5 lit.h-1.
Before the operation, the bed was rinsed with deionized water to purge ash.

Fig. 1. Photograph of pilot: (1) flange, (2) inlet tube, (3) sampling port, (4) sand layer, (5) fixed-bed
column, (6) table, (7) inlet water, (8) water tee, (9) dosing pump, (10) safety valve, (11) reservoir.
2.6 Analytical method
The liquid-liquid micro-extraction method was used to determine diazinon concentration based
on the study of Brijani et al. (2006). The effluents samples (5.00 mL) were collected using 10
mL screw cap glasses test tube with a conic bottom. To extract diazinon pesticide from water, 20
µL chlorobenzene (extraction solvent) and 1 mL acetone (disperse solvent) were mixed and the
obtained liquid was fast injected into 5 mL synthetic solution. A cloudy solution was
immediately formed and then the sample was placed in centrifuging for 2 min at 5000 rpm. The
visible droplet of chlorobenzene (13 µL) was precipitated in the bottom of conic test pipe.
Finally, 1 µL from the deposited phase was injected into GC-MS (Agilent GC 6890N) for
determination of diazinon.
2.7 Fixed-bed column data analysis
In this column, the initial diazinon concentration, flow rate, and bed height were as the
experimental conditions, which their influence on diazinon adsorption was evaluated using
breakthrough (BT) curve. BT curve is obtained from Ct (mg.lit-1) or Ct/C0 versus Vt (lit) or t (h)
curves (Auta and Hameed, 2014); where Ct, C0, Vt, and t are the effluent concentrations at any
time t, influent concentration, treated effluent volume, and the service time, respectively. The
effluent concentration usually starts to increase fast at Ct/Co = 0.05, which expressed as
breakthrough point and also exhaustion point is defined as Ct/Co = 0.95. The defined ratio at the
breakthrough point is arbitrarily and in this study, it was taken as Ct/C0=0.10. The total capacity
of the column (qtotal, mg), defined as the total adsorbed quantity of diazinon in the column, can be
calculated using integrating from Cad (Cad =C0-Ct) for a given inlet concentration, flow rate, and
fixed height (Tan et al., 2008) by Eq. (3):

= (3)

where Q (lit.h-1) and ttotal (h) are the volumetric flow rate and the total flow time, respectively.
Cad is the difference the inlet pollutant concentration and outlet pollutant concentration of
equilibrium uptake and qbed (mg.g-1) is the total adsorbed quantity of diazinon in the column per
mass of activated carbon in the bed (Tan et al., 2008) using Eq. (4):
= (4)

3 Results

3.1 Characterization of adsorbent

As shown in Fig. 2b, modification process led to an increase in uniformity of MAC area
distribution, decrease from 3.23 to 2.08 nm in average pore diameter, and also reduction in total
pore volume from 0.86 to 0.74 cm3.g-1. Accordingly, modification process increased surface area
from 1076 to 1434 m2.g-1. As can be seen in Fig. 2c, area distribution is unstable between 0 and 1
nm and also rpeak occurs at 0.3 nm.
The Fig. 3a shows three peaks at 1194, 1571, and 3433 . The peak at 1194
corresponds to C-O stretch, which demonstrates the existence of alcohols, carboxylic acids,
esters, and ether functional groups. With respect to modification process using H3 PO4, peak at
1194 can be attributed to the stretching of hydrogen-bonded P=OOH or the P-O-C linkage
(Xu et al., 2014). The band appeared at 1571 is assigned to C=C vibration in aromatic
rings while the peak around 3433 is assigned to O-H stretching vibration of hydroxyl groups
(Njoku et al., 2015; Xu et al., 2014). As seen, modification process using phosphoric acid leads
to the elimination of several functional groups. Two peaks observed at 572 and 873
can be ascribed to the O-H out of plane bending vibration and the C-H out of plane bending
vibration in benzene derivative (Yang and Qiu, 2010). The band at 2918 indicates the
symmetric and asymmetric C-H stretching vibration of aliphatic acids (Hashemian et al., 2014).
Fig. 2. (a) Chemical structure of diazinon, (b) Barrett-Joyner-Halenda (BJH) plot of MAC and AC and
(C) MP plot of MAC.

The surface morphologies of MAC and AC were investigated by scanning electron microscopy.
It can be seen from Fig. 3b that the surface of AC is rough and also white droplets could be
obviously seen on MAC surface from Fig. 3c that can likely be related to adding phosphoric acid
during the modification process.

Fig. 3. (a) FTIR spectra of MAC and AC, (b) SEM image of AC and (c) SEM image of MAC.

3.2 Batch adsorption studies

3.2.1 Effect of solution pH on diazinon adsorption
Influence of initial pH on adsorption process was evaluated at a fixed dose of 0.04 g.lit-1 with 20
mg.lit-1 diazinon concentration to obtain optimal pH. The diagram of pH vs. qe was quite steady
so pH changes have no impact on diazinon adsorption process. The adsorbed amount in the pH
ranges of 1.5-10 was fixed and around 98 mg.g-1. Two important points can be mentioned in this
regard. First, MAC adsorbent is able to remove diazinon pesticide from water with the highest
efficiency. Second, MAC is completely affordable as it can be used for real contaminated water
to diazinon pesticide without any changes in real water pH. According to the dissociation
constant of diazinon (2.6), there is no valid reasoning for being a steady trend in pH ranges of
1.5-10.
3.2.2 Effect of initial concentration and contact time
The effect of contact time and initial diazinon concentration is shown in Fig 4. An increase in the
quantity of diazinon adsorption with the rise of initial concentration and also extremely steep
slope of all curves at the initial stage are easily observable. The amount adsorbed at equilibrium
(qe) increased from 4.951 to 151.96 mg.g-1 with the rise in initial concentration from 1 to 40
mg.lit-1. This increase can be attributed to an increase in concentration gradient at a higher initial
concentration that creates driving force during the adsorption process. In addition, greater
concentration gradient caused a faster transport due to an increased mass transfer coefficient.
High adsorption rates in the first 10 to 20 min can be due to numerous vacant surface sites to
adsorb diazinon molecules. However, after this because of repulsive forces between diazinon
molecules on the walnut shell-modified activated carbon and those in solution, removal rate
progressively decreases with time until equilibrium was finally reached (Fu et al., 2015; Salameh
et al., 2015).
Fig. 4. Plot of pseudo-second-order for the adsorption capacity of diazinon on MAC (30 ; agitator; 120
rpm, contact time; 120 min).

3.2.3 Adsorption kinetic modeling

Three models, namely, the Lagergren pseudo-first order (PFO), the Ho pseudo-second-order
(PSO), and Weber and Morris were chosen to study the mechanism of the adsorption of diazinon
onto MAC. The nonlinear forms of the pseudo-first (Lagergren, 1898) and pseudo-second-order
(Ho and McKay, 1999) models are expressed by Eq. (5) and Eq. (6).
= (1‐ ) (5)

= (6)

where t (min) is the contact time and (1.min-1) and (g.mg-1 min-1) are the PFO and the PSO
rate constants, respectively.
In this study, Webber and Morris intra-particle diffusion model (1963) was employed to analyze
diffusion process between adsorbent particle and solution as follows:

= +C (7)

where (mg.g-1 ) is the intra-particle diffusion constant and C is an intercept that

describes the boundary layer effects.
All parameters of aforementioned models are summarized in Table 1. The best model was
verified using coefficient of determination ( average relative error (ARE). The
comparison between coefficients of determination and average relative errors in three models
shows that the PSO kinetics was the most appropriate model for adsorption of diazinon onto
MAC. Reduction of the PSO rate constant value with increasing diazinon concentration indicated
that adsorbed diazinon was held by chemical bonds and the chemical process could play a rate-
limiting role during adsorption (Nguyen et al., 2015).
According to , C, and ARE values achieved from plots of qt versus t0.5, the experimental data
do not fit linearly to intra-particle diffusion model. The first sharp section presents the adsorption
of diazinon by boundary layer diffusion and the second section indicates that the adsorption rate
decreased gradually and intra-particle diffusion acts as a dominant mechanism. Therefore, intra-
particle diffusion is not the only rate control mechanism and rather other mechanisms also are
involved in diazinon adsorption process (Danmaliki and Saleh, 2017; Wang et al., 2013).
Fig. 5. Intra-particle diffusion model for diazinon adsorption onto MAC.

Table 1
Parameters of kinetic models at different initial concentration.
parameters C0 (mg.lit-1)

1 5 10 20 40
(mg.g-1) 4.95 24.01 46.76 88.4 151.97

Pseudo-first-order
(mg.g-1) 0.35 5.21 20.37 43.47 80.08

(1.min-1) 0.0479 0.0738 0.0691 0.0577 0.061

0.59 0.85 0.94 0.95 0.97

ARE 95.63 84.12 65.83 64.28 53.99

Pseudo-second-order
(mg.g-1) 4.95 24.213 47.62 90.09 156.25

(g.mg-1 min-1) 0.615 0.0516 0.0114 0.0049 0.0022

0.9999 0.9999 0.9995 0.9995 0.9991

ARE 6.20 5.82 9.14 7.08 6.75

 parameters C0 (mg.lit-1)

1 5 10 20 40
Intra-particle diffusion
C(mg.g-1) 3.02 11.702 19.517 31.283 54.012
(mg.g-1 ) 0.362 1.499 3.26 6.65 11.41

0.36 0.53 0.6 0.69 0.68

ARE 15.66 16.86 21.31 18.81 20.09

3.2.4 Adsorption isotherms

In this study, Langmuir and Freundlich isotherm models were chosen to describe diazinon
adsorption process onto MAC in the equilibrium. The nonlinear form of the Langmuir and
Freundlich isotherms and parameters of both models including Kf and n as Freundlich constants
and qm and KL as the adsorption capacity and Langmuir constant respectively are presented in
Table 2. Also, the nonlinear plot of the models is illustrated in Fig 6. The regression coefficient
(R2) and the root mean square error (RMSE) (Njoku et al., 2014) were used to identify the best
fitting model between Langmuir (1918) and Freundlich (1906) isotherms. Obtained data
represent that Freundlich isotherm is a better fit to experimental data. Fitting of Freundlich
isotherm data with experimental data suggests that diazinon adsorption follows a multi-layer
adsorption onto the heterogeneous surface of MAC. The amount 1/n <1 shows that diazinon
adsorption process is a physisorption so adsorption sites are decreased and become occupied by
adsorbate molecules during adsorption process (Ouali et al., 2015).

Fig. 6. Non-linear plot of Freundlich and Langmuir isotherm for diazinon adsorption onto MAC at 30 .
 Table 2
Freundlich and Langmuir isotherm parameters for diazinon adsorption onto MAC at 30

Freundlich model: Langmuir model: =

Kf n R2 RMSE: qmax Ka R2 RMSE:

(mg.g-1 (L.mg-1)1/n) (mg.g-1) (lit.mg-1)

53.82 1.98 0.996 9.01 169.49 0.77 0.9724 11.86

3.3 Fixed bed adsorption studies

3.3.1 Effect of inlet diazinon concentration
The effect of inlet diazinon concentration on BT curves at different initial concentrations of 15,
30, and 40 mg.lit-1 with the same bed height of 20 cm and solution flow rate of 12 lit.h-1 is shown
in Fig. 7. Moreover, bed capacities, breakthrough times, and exhaustion times are listed in Table
3. The increase in inlet diazinon concentration caused adsorption sites became more quickly
saturated and subsequently led to a reduction in breakthrough and exhaustion time. Therefore,
inlet diazinon concentration affected directly termination of BT curves (Fig. 7). As can be seen,
bed capacity was changed from 15.75 to 28.35 mg.g-1 when influent diazinon concentration
increased from 15 to 40 mg.lit-1. Such a change can be explained by the fact that the greater
concentration gradient caused the transport within the pores occurs quickly due to increase in
diffusion coefficient, which plays driving force role during the adsorption process (Nazari et al.,
2016). The highest adsorption capacity of 28.35 mg.g-1 was found for the greatest influent
diazinon concentration.

Fig. 7. Breakthrough curves for diazinon adsorption on MAC at different initial concentrations of
diazinon (flow rate, 12 lit.h-1; bed height, 20 cm).
3.3.2 Effect of the flow rate
Effect of flow rates of 9.5, 12, 14.5, and 16.5 lit.h-1 on the diazinon adsorption using walnut
shell-modified activated carbon was investigated while initial diazinon concentration and bed
height were kept constant at 40 mg.lit-1 and at 20 cm, respectively. Fig. 8 shows that changes in
flow rate from 9.5 to 16.5 made BT curves steeper and led to a decrease in column saturation
time or exhaustion time. The decrease in bed capacity at high flow rate can be attributed to
increasing solution velocity, which caused a reduction in residence time of solution in the
column and allowed diazinon molecules leave the bed before adhering well to the adsorbent
(Gong et al., 2015). However, as shown in Table 3, with increased flow rate from 14 to 16.5 lit.h-
1
, bed capacity decreased. This observation can be explained by the increase in flow rate that
caused gradient concentration growth, which limits the influence of velocity solution on bed
capacity.

Fig. 8. Breakthrough curves for diazinon adsorption on MAC at different flow rates (diazinon initial
concentration, 40 mg.lit-1; bed height, 20 cm).

3.3.3 Effect of bed height

Fig. 8 shows BT curves for bed heights of 10, 20, and 30 cm while the initial concentration of
diazinon and initial flow rate were held constant at 40 mg.lit-1 and at 12 lit.h-1, respectively. With
the same flow rate and inlet concentration, rise in bed height led to increase of breakthrough
times. This observation can be ascribed to the increase in height of mass transfer zone that could
adsorb extra escaped diazinon molecules form underlay. Thus, a decrease occurred in the
termination of BT curves. As expected, there is a direct relationship between the mass of
adsorbent and bed capacity, as an increase in mass led to the expansion of specific surface area,
which provided more available adsorption sites. Comparison of exhaustion times at different bed
heights showed that exhaustion time strongly decreased at 10 cm bed height, due to the axial
dispersion phenomena (Auta and Hameed, 2014).
 Fig. 9. Breakthrough curves for diazinon adsorption on MAC at different bed heights (diazinon initial
concentration, 40 mg.lit-1; flow rate, 12 lit.h-1).

Table 3
Fixed-bed adsorption data for diazinon adsorption by MAC.

Initial diazinon Flow rate bed height Breakthrough Exhaustion Bed capacity
concentration time time
(lit.h-1) (cm) (mg.g-1)
-1
(mg.lit ) (min) ( min)
15 12 20 440 1120 15.75
30 12 20 340 870 24.5
40 12 20 290 740 28.35
40 9.5 20 455 1100 34.31
40 14.5 20 195 560 24.86
40 16.5 20 210 470 26.29
40 12 10 75 290 18.2
40 12 30 600 1270 34.98

3.3.4 Modeling of fixed bed adsorption

Bohart-Adams (1920) is a widely used model to evaluate the initial part of BT curve. This model
assumes that the rate of adsorption is commensurate with a concentration of adsorbate molecules
and the remaining capacity of activated carbon. Bohart-Adams equation has been incorrectly
 referred in several articles to Thomas equation and therefore this misattribution led to wrong
analysis about the dynamic behavior of adsorption columns (Chu, 2010; Tan and Hameed, 2017).
Bohart-Adams equation can be expressed as Eq. (8) and (9), where Eq. (9) is valid for C/C0 << 1.

Ln ( = - (8)

Ln ( = - (9)

where kBA (L.mg-1 min-1) is Bohart-Adams rate constant, N0 (mg.lit-1) is the saturation
concentration, Z (cm) is the bed depth, and U (cm.min-1) is linear velocity.
The parameters of Adams-Bohart model are listed in Table 4 in terms of R2 and ARE. As can be
seen, the value of KBA decreased with increasing the initial concentration but increased with the
increase of both flow rate and bed height. As a result, the external mass transfer is dominant in
the initial part of the adsorption process. Also, a lower bed height, lower initial concentration,
and higher flow rate cause adsorption of more diazinon in the column (Gong et al., 2015; Nazari
et al., 2016).
Table 4
Bohart-Adams model parameters at different conditions for diazinon adsorption

C(mg.lit-1) Q (lit.h-1) h(cm) Bohart-Adams R2 ARE (%)

KBA N0
15 12 20 0.000433 11031.13 0.83 2.4
30 12 20 0.000313 17278.46 0.68 5.15
40 12 20 0.0002 21291.2 0.70 0.45
40 9.5 20 0.000202 21718.81 0.77 4.67
40 14.5 20 0.000323 18211.78 0.79 2.58
40 16 20 0.000283 18957.77 0.85 0.05
40 12 10 0.00045 14919.11 0.85 1.69
40 12 30 0.00014 23318.48 0.83 2.6

Yoon-Nelson model (1984) is a simple mathematical model to investigate the 50% breakthrough
time. This model is based on the proportionality between “rate of decrease in the probability of
adsorption for each adsorbate molecule” and “the probability of adsorbate adsorption and the
probability of adsorbate breakthrough on the adsorbent” (Nazari et al., 2016). Linearized form of
the model is defined as Eq. (10):
 Ln ( = – τ) (10)

Where KYN (min-1) is the Yoon-Nelson rate constant and τ (min) is the time required for 50%
adsorbate breakthrough.
As presented in Table 5, the value of KYN increased with increasing the flow rate and influent
concentration but decreased with increasing bed height.

Bohart-Adams (Eq. 8) and Yoon-Nelson model illustrate linear plots of Ln ( against t and

Ln ( versus t, respectively. Both models have the equivalent R2 and ARE because the left
sides of both equations are reverse of each other with similar parameters. Therefore, comparison
of Eq. (8) and Eq. (10) as two different models does not make any sense. According to R2 and
ARE values, Yoon-Nelson model better describes the continuous removal of diazinon pesticide
onto MAC so the validity of Yoon-Nelson model is higher than that of Adams-Bohart model for
this study.

Table 5
Yoon-Nelson model parameters at different conditions for diazinon adsorption

Q (lit.h-1) C(mg.lit-1) H (cm) Yoon-Nelson R2 ARE (%)

KYN τ
15 12 20 0.009 729.28 0.94 0.33
30 12 20 0.0121 594.34 0.80 0.46
40 12 20 0.0113 510.04 0.82 0.55
40 9.5 20 0.0102 760.64 0.86 0.60
40 14.5 20 0.017 363.86 0.88 0.46
40 16 20 0.017 322.32 0.91 0.16
40 12 10 0.0288 163.28 0.96 0.33
40 12 30 0.0071 907.46 0.91 0.23

Conclusion
This study showed that chemical modification of commercial activated carbon using H3PO4 was
successful as a modification process to increase surface area. Modified activated carbon was able
to remove diazinon in pH ranges of 1.5-10 with the equal efficiency and thus it can be used as an
affordable and promising adsorbent. Effect of initial concentration of diazinon and contact time
was investigated and found that bed capacity increases as inlet concentration increases.
Considerable removal of diazinon and also noncompliance between batch technique and
industrial scale have made researchers carry out continuous fixed-bed studies via experimental
set-up with a column height of 60 cm and an internal diameter of 4 cm. Inlet concentration, flow
rate, and bed height were chosen as experimental conditions to investigate the trend of change in
breakthrough time, exhaustion time, and bed capacity. The highest bed capacity was 34.98 mg.g-
1
for breakthrough time and exhaustion time of 600 and 1270 min, respectively. The ARE values
less than 1% suggest that Yoon-Nelson model described the experimental data of breakthrough
curves for a successful design better than Bohart-Adams model. Furthermore, MAC or set-up of
this study can be used to investigate the removal of organophosphate pesticides from aqueous
solutions.

Acknowledgment
The authors acknowledge Mr. Amireza Rashidi and Mr. Farhad Khalili for their unwavering
support.
‘This research did not receive any specific grant from funding agencies in the public,
commercial, or not-for-profit sectors’
‘Declarations of interest: none’
2017. http://www.fao.org/faostat/en/#compare. 
Aggarwal,  V.,  Deng,  X.,  Tuli,  A.,  Goh,  K.S.,  2013.  Diazinon—Chemistry  and  Environmental  Fate:  A 
California Perspective, Reviews of Environmental Contamination and Toxicology Volume 223. Springer, 
pp. 107‐140. 
Auta,  M.,  Hameed,  B.,  2014.  Chitosan–clay  composite  as  highly  effective  and  low‐cost  adsorbent  for 
batch and fixed‐bed adsorption of methylene blue. Chemical Engineering Journal 237, 352‐361. 
Barrett, E.P., Joyner, L.G., Halenda, P.P., 1951. The determination of pore volume and area distributions 
in  porous  substances.  I.  Computations  from  nitrogen  isotherms.  Journal  of  the  American  Chemical 
society 73(1), 373‐380. 
Berijani,  S.,  Assadi,  Y.,  Anbia,  M.,  Hosseini,  M.‐R.M.,  Aghaee,  E.,  2006.  Dispersive  liquid–liquid 
microextraction  combined  with  gas  chromatography‐flame  photometric  detection:  Very  simple,  rapid 
and  sensitive  method  for  the  determination  of  organophosphorus  pesticides  in  water.  Journal  of 
Chromatography A 1123(1), 1‐9. 
Bhatnagar, A., Hogland, W., Marques, M., Sillanpää, M., 2013. An overview of the modification methods 
of activated carbon for its water treatment applications. Chemical Engineering Journal 219, 499‐511. 
Bohart, G., Adams, E., 1920. Some aspects of the behavior of charcoal with respect to chlorine. Journal 
of the Franklin Institute 189(5), 669. 
Chu,  K.H.,  2010.  Fixed  bed  sorption:  setting  the  record  straight  on  the  Bohart–Adams  and  Thomas 
models. Journal of hazardous materials 177(1), 1006‐1012. 
Dahri, M.K., Kooh, M.R.R., Lim, L.B., 2014. Water remediation using low cost adsorbent walnut shell for 
removal of malachite green: equilibrium, kinetics, thermodynamic and regeneration studies. Journal of 
Environmental Chemical Engineering 2(3), 1434‐1444. 
Danmaliki,  G.I.,  Saleh,  T.A.,  2017.  Effects  of  bimetallic  Ce/Fe  nanoparticles  on  the  desulfurization  of 
thiophenes using activated carbon. Chemical Engineering Journal 307, 914‐927. 
Freundlich,  H.,  1906.  Uber  die  adsorption  in  losungen  [Adsorption  in  solution]”  Zeitschrift  für 
Physikalische Chemie, 57. 
Fu, J., Chen, Z., Wang, M., Liu, S., Zhang, J., Zhang, J., Han, R., Xu, Q., 2015. Adsorption of methylene blue 
by  a  high‐efficiency  adsorbent  (polydopamine  microspheres):  kinetics,  isotherm,  thermodynamics  and 
mechanism analysis. Chemical Engineering Journal 259, 53‐61. 
Gong, J.‐L., Zhang, Y.‐L., Jiang, Y., Zeng, G.‐M., Cui, Z.‐H., Liu, K., Deng, C.‐H., Niu, Q.‐Y., Deng, J.‐H., Huan, 
S.‐Y., 2015. Continuous adsorption of Pb (II) and methylene blue by engineered graphite oxide coated 
sand in fixed‐bed column. Applied Surface Science 330, 148‐157. 
Hashemian,  S.,  Salari,  K.,  Yazdi,  Z.A.,  2014.  Preparation  of  activated  carbon  from  agricultural  wastes 
(almond shell and orange peel) for adsorption of 2‐pic from aqueous solution. Journal of Industrial and 
Engineering Chemistry 20(4), 1892‐1900. 
Ho,  Y.‐S.,  McKay,  G.,  1999.  Pseudo‐second  order  model  for  sorption  processes.  Process  biochemistry 
34(5), 451‐465. 
Hokkanen, S., Bhatnagar, A., Sillanpää, M., 2016. A review on modification methods to cellulose‐based 
adsorbents to improve adsorption capacity. Water research 91, 156‐173. 
Lagergren,  S.,  1898.  Zur  theorie  der  sogenannten  adsorption  geloster  stoffe.  Kungliga  svenska 
vetenskapsakademiens. Handlingar 24, 1‐39. 
Langmuir, I., 1918. The adsorption of gases on plane surfaces of glass, mica and platinum. Journal of the 
American Chemical society 40(9), 1361‐1403. 
Mikhail,  R.S.,  Brunauer,  S.,  Bodor,  E.,  1968.  Investigations  of  a  complete  pore  structure  analysis:  I. 
Analysis of micropores. Journal of Colloid and Interface Science 26(1), 45‐53. 
Mostafalou, S., Abdollahi, M., 2013. Pesticides and human chronic diseases: evidences, mechanisms, and 
perspectives. Toxicology and applied pharmacology 268(2), 157‐177. 
Moussavi, G., Hossaini, H., Jafari, S.J., Farokhi, M., 2014. Comparing the efficacy of UVC, UVC/ZnO and 
VUV  processes  for  oxidation  of  organophosphate  pesticides  in  water.  Journal  of  Photochemistry  and 
Photobiology A: Chemistry 290, 86‐93. 
Moussavi,  G.,  Hosseini,  H.,  Alahabadi,  A.,  2013.  The  investigation  of  diazinon  pesticide  removal  from 
contaminated water by adsorption onto NH4 Cl‐induced activated carbon. Chemical Engineering Journal 
214, 172‐179. 
Nazari, G., Abolghasemi, H., Esmaieli, M., Pouya, E.S., 2016. Aqueous phase adsorption of cephalexin by 
walnut shell‐based activated carbon: A fixed‐bed column study. Applied Surface Science 375, 144‐153. 
Ngo, H.H., Guo, W., Zhang, J., Liang, S., Ton‐That,  C.,  Zhang, X., 2015. Typical low cost biosorbents for 
adsorptive removal of specific organic pollutants from water. Bioresource technology 182, 353‐363. 
Nguyen,  T.C.,  Loganathan,  P.,  Nguyen,  T.V.,  Vigneswaran,  S.,  Kandasamy,  J.,  Naidu,  R.,  2015. 
Simultaneous  adsorption  of  Cd,  Cr,  Cu,  Pb,  and  Zn  by  an  iron‐coated  Australian  zeolite  in  batch  and 
fixed‐bed column studies. Chemical Engineering Journal 270, 393‐404. 
Njoku,  V.,  Foo,  K.,  Hameed,  B.,  2013.  Microwave‐assisted  preparation  of  pumpkin  seed  hull  activated 
carbon  and  its  application  for  the  adsorptive  removal  of  2,  4‐dichlorophenoxyacetic  acid.  Chemical 
Engineering Journal 215, 383‐388. 
Njoku,  V.,  Islam,  M.A.,  Asif,  M.,  Hameed,  B.,  2014.  Preparation  of  mesoporous  activated  carbon  from 
coconut  frond for the  adsorption of carbofuran  insecticide. Journal  of  Analytical  and  Applied Pyrolysis 
110, 172‐180. 
Njoku,  V.,  Islam,  M.A.,  Asif,  M.,  Hameed,  B.,  2015.  Adsorption  of  2,  4‐dichlorophenoxyacetic  acid  by 
mesoporous activated carbon prepared from H 3 PO 4‐activated langsat empty fruit bunch. Journal of 
environmental management 154, 138‐144. 
Ouali,  A.,  Belaroui,  L.S.,  Bengueddach,  A.,  Galindo,  A.L.,  Peña,  A.,  2015.  Fe  2  O  3–palygorskite 
nanoparticles, efficient adsorbates for pesticide removal. Applied Clay Science 115, 67‐75. 
Ouznadji,  Z.B.,  Sahmoune,  M.N.,  Mezenner,  N.Y.,  2016.  Adsorptive  removal  of  diazinon:  kinetic  and 
equilibrium study. Desalination and Water Treatment 57(4), 1880‐1889. 
Peters, K., Bundschuh, M., Schäfer, R., 2013. Review on the effects of toxicants on freshwater ecosystem 
functions. Environmental Pollution 180, 324‐329. 
Salameh,  Y.,  Albadarin,  A.B.,  Allen,  S.,  Walker,  G.,  Ahmad,  M.,  2015.  Arsenic  (III,  V)  adsorption  onto 
charred dolomite: charring optimization and batch studies. Chemical Engineering Journal 259, 663‐671. 
Shemer, H., Linden, K.G., 2006. Degradation and by‐product formation of diazinon in water during UV 
and UV/H 2 O 2 treatment. Journal of hazardous materials 136(3), 553‐559. 
Taha,  S.M.,  Amer,  M.E.,  Elmarsafy,  A.E.,  Elkady,  M.Y.,  2014.  Adsorption  of  15  different  pesticides  on 
untreated  and  phosphoric  acid  treated  biochar  and  charcoal  from  water.  Journal  of  Environmental 
Chemical Engineering 2(4), 2013‐2025. 
Tan, I., Ahmad, A., Hameed, B., 2008. Adsorption of basic dye using activated carbon prepared from oil 
palm shell: batch and fixed bed studies. Desalination 225(1), 13‐28. 
Tan, K., Hameed, B., 2017. Insight into the adsorption kinetics models for the removal of contaminants 
from aqueous solutions. Journal of the Taiwan Institute of Chemical Engineers 74, 25‐48. 
Vukčević,  M.M.,  Kalijadis,  A.M.,  Vasiljević,  T.M.,  Babić,  B.M.,  Laušević,  Z.V.,  Laušević,  M.D.,  2015. 
Production of activated carbon derived from waste hemp (Cannabis sativa) fibers and its performance in 
pesticide adsorption. Microporous and Mesoporous Materials 214, 156‐165. 
Wang,  H.,  Yuan,  X.,  Wu,  Y.,  Huang,  H.,  Zeng,  G.,  Liu,  Y.,  Wang,  X.,  Lin,  N.,  Qi,  Y.,  2013.  Adsorption 
characteristics  and  behaviors  of  graphene  oxide  for  Zn  (II)  removal  from  aqueous  solution.  Applied 
Surface Science 279, 432‐440. 
Weber, W.J., Morris, J.C., 1963. Kinetics of adsorption on carbon from solution. Journal of the Sanitary 
Engineering Division 89(2), 31‐60. 
Xu,  J.,  Chen,  L.,  Qu,  H.,  Jiao,  Y.,  Xie,  J.,  Xing,  G.,  2014.  Preparation  and  characterization  of  activated 
carbon from reedy grass leaves by chemical activation with H 3 PO 4. Applied Surface Science 320, 674‐
680. 
Yang,  J.,  Qiu,  K.,  2010.  Preparation  of  activated  carbons  from  walnut  shells  via  vacuum  chemical 
activation and their application for methylene blue removal. Chemical Engineering Journal 165(1), 209‐
217. 
Yoon,  Y.H.,  NELSON,  J.H.,  1984.  Application  of  gas  adsorption  kinetics  I.  A  theoretical  model  for 
respirator cartridge service life. The American Industrial Hygiene Association Journal 45(8), 509‐516.