Professional Documents
Culture Documents
ScienceDirect
journal homepage: www.elsevier.com/locate/watres
Civil Engineering Department, Universiti Teknologi PETRONAS, 31750 Tronoh, Perak, Malaysia
Department of Civil Engineering, King Abdulaziz University, Jeddah, Saudi Arabia
article info
abstract
Article history:
This work investigated the removal of boron from wastewater and its recovery by elec-
design was developed using Box-Behnken Model. An initial study was performed based on
9 December 2013
four preselected variables (pH, current density, concentration and time) using synthetic
wastewater. Response surface methodology (RSM) was used to evaluate the effect of pro-
cess variables and their interaction on boron removal. The optimum conditions were obtained as pH 6.3, current density 17.4 mA/cm2, and time 89 min. At these applied optimum
Keywords:
conditions, 99.7% boron removal from an initial concentration of 10.4 mg/L was achieved.
Adsorption kinetics
The process was effectively optimized by RSM with a desirability value of 1.0. The results
Boron
showed that boron removal efficiency enhanced with increase in current density and
Electrocoagulation
treatment time. Removal efficiency also increased when pH was increased from 4 to 7 and
Hydrothermal mineralization
subsequently decreased at pH 10. Adsorption kinetics study revealed that the reaction
Produced water
followed pseudo second order kinetic model; evidenced by high correlation and goodness
of fit. Thermodynamics study showed that mechanism of boron adsorption was chemi-
Thermodynamics
sorption and the reaction was endothermic in nature. Furthermore, the adsorption process
was spontaneous as indicated by negative values of the adsorption free energy. Treatment
of real produced water using electrocoagulation resulted in 98% boron removal. The hydrothermal mineralization study showed that borate minerals (Inyoite, Takadaite and
Nifontovite) can be recovered as recyclable precipitate from electrocoagulation flocs of
produced water.
2013 Elsevier Ltd. All rights reserved.
1.
Introduction
boric acid and borate ions in aquatic environment. The functions of boron in plants include degradation of carbohydrates,
sugar translocation, and hormonal action. Boron deficiency
causes stunted growth, yield loss and even death of plant
(Yilmaz et al., 2008b). High boron concentration in irrigation
water, however, can cause severe environmental problem
because boron compounds form complexes with heavy metals
present in soil and increase the potential toxicity of these
* Corresponding author. Tel.: 966 (0)2 6402000x68239; fax: 966 (0)2 6952179.
E-mail addresses: zkhan@kau.edu.sa, himatali@gmail.com (Z. Ahmed).
0043-1354/$ e see front matter 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.watres.2013.12.024
114
w a t e r r e s e a r c h 5 1 ( 2 0 1 4 ) 1 1 3 e1 2 3
(1)
(2)
Al s /Al
Cathode:
4
7
5
Al7(OH)4
17 , Al8(OH)20 , Al13O4(OH)24 , Al13(OH)34 , etc. transform
finally into Al(OH)3(s) according to the following simplified
equation (Ghosh et al., 2008).
Al
aq 3H2 O l/AlOH3 s 3H aq
(3)
2.
Experimental method
2.1.
Characteristics of wastewater
A preliminary study was conducted with synthetic wastewater prepared with appropriate amount of boric acid (H3BO3)
and dissolved in 1 L distilled water to yield varying boron
concentrations of 10, 20 and 30 mg/L. The produced water was
collected from a local Crude Oil Terminal in Malaysia and was
characterized with atomic absorption spectrometry (AAS) and
ion chromatography (IC). A pH meter (Hach Sension 2 pH
meter) and a conductivity meter (Myron L conductivity meter)
were used to measure the pH and conductivity of the sample
respectively. The produced water characteristics are shown in
Table 1.
15
7.84
TSS
TDS
Conductivity
Turbidity
Aluminum
136
15,829
30,000 mS/cm
72 NTU
0.65
Iron
Chloride
Sodium
Calcium
Magnesium
Sulphate
Potassium
Hardness
1.66
7546
3952
357
600
168
284
957
Parameter
Concentration
Bromine
Total
phosphate
COD
BOD
Nitrite
Copper
Ammonia
nitrogen
TKN
Sulphate
Nitrate
Sulphide
Phenol
Total alkalinity
Zinc
Fluoride
31.2
12
1560
883
0.03
2.98
16.5
60.7
168
1.9
0.21
15
1546
0.04
0.61
w a t e r r e s e a r c h 5 1 ( 2 0 1 4 ) 1 1 3 e1 2 3
2.2.
Electrocoagulation setup
2.3.
Adsorption kinetics
V
W
ITM
ZF
(4)
(5)
where C0 is initial boron concentration (mg/L), Ce is equilibrium boron concentration (mg/L), V is volume of sample (L), W
is mass of adsorbent (g), I is current (A), T is time (s), M is molar
mass of electrode, Z is number of electrons involved in the
redox reaction, and F is Faradays constant (C/mol).
The suitability of both pseudo first order and pseudo second order kinetic models was further evaluated using the chisquare (c2) represented as follows (Sundaram et al., 2008):
c2
exp
qe
qcal
e
qcal
e
2
(6)
exp
2.4.
Adsorption thermodynamics
115
ln Kc
DS
R
(7)
DH
RT
(8)
2.5.
(HM)
3.
116
w a t e r r e s e a r c h 5 1 ( 2 0 1 4 ) 1 1 3 e1 2 3
1
0
1
4
7
10
6.25
12.5
18.75
10
20
30
30
60
90
k
X
j1
Xi
bj Xj
k
X
j1
bjj X2j
XX
bij Xi Xj ei
(9)
i <j
xi x0
Dx
(10)
4.
4.1.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
10
4
7
7
7
4
7
7
7
7
7
7
4
7
10
7
7
10
10
4
7
4
7
10
7
7
7
10
4
6.25
12.5
12.5
12.5
18.75
12.5
12.5
12.5
18.75
18.75
12.5
6.25
12.5
12.5
12.5
18.75
6.25
12.5
12.5
18.75
12.5
6.25
6.25
18.75
6.25
12.5
12.5
12.5
12.5
20
10
30
20
20
20
20
20
20
10
30
20
20
20
20
30
30
30
20
20
10
20
10
20
20
10
20
10
30
60
60
30
60
90
30
60
60
30
60
90
30
90
60
30
60
60
60
90
60
90
60
60
60
90
30
60
60
60
26
50
55
76
94
37.5
74
73
68
97
82.7
49.5
50
71.5
26.5
87.5
52
33.6
44
53
98
34
67
43.5
64
78
70
41
42.3
(11)
227.07
1887.52
505.70
1164.27
7292.14
120.40
1
1
1
1
1
1
227.07
1887.52
505.70
1164.27
7292.14
120.40
11.13
92.50
24.78
57.05
357.35
5.90
<0.0049
<0.0001
<0.0002
<0.0001
<0.0001
<0.0292
117
w a t e r r e s e a r c h 5 1 ( 2 0 1 4 ) 1 1 3 e1 2 3
Model
Residual
Lack of fit
Pure error
Sum of squares
DF
Mean square
F value
Prob > F
12069.28
285.69
264.49
21.20
14
14
10
4
862.09
20.41
26.45
5.30
42.25
<0.0001
4.99
0.0676
Significant
Not significant
Standard deviation 4.52, R2 0.9769, R2 adj 0.9538, Pred R2 0.8740, Adequate precision 23.088, Press 1556.58.
4.2.
118
w a t e r r e s e a r c h 5 1 ( 2 0 1 4 ) 1 1 3 e1 2 3
4.4.
Adsorption kinetics
To understand the adsorption kinetics of boron using electrocoagulation, four different adsorption models were
evaluated.
4.4.1.
(12)
where qe and qt are the amount of boron adsorbed at equilibrium (mg/g) on Al(OH)3 and at any time (t) respectively, and k1
(min1) is the calculated pseudo first order rate constant of
adsorption.
If the adsorption follows the pseudo first order kinetics, a
plot of log (qe qt) versus t should be linear. qe and k1 are
calculated from the intercept and slope of the plot of log
(qe qt) versus t respectively. As shown in Fig. 4, the data did
not completely conform to a linear plot. The points deviated
from the straight line. Though the correlation coefficient was
high, the calculated equilibrium adsorption capacity qcal
e
deviated from the experimental equilibrium adsorption caexp
pacity qe and the chi-square was high. This implies that
the adsorption did not completely follow the pseudo first
order kinetics. The kinetic constants of pseudo first order kinetics is shown in Table 6.
4.4.2.
t
1
t
qt k2 q2e qe
(13)
4.3.
Optimization
119
w a t e r r e s e a r c h 5 1 ( 2 0 1 4 ) 1 1 3 e1 2 3
exp
qe
(mg/g)
6.09
11.22
16.68
Pseudo-first order
qcal
e (mg/g)
R2
c2
k2 (g/mg min)
qcal
e (mg.g)
R2
c2
0.078
0.029
0.037
4.45
6.56
14.06
0.918
0.898
0.957
0.62
3.31
0.49
0.017
0.084
0.037
6.80
11.90
17.86
0.998
0.99
0.989
0.08
0.04
0.08
where qe and qt are the amount of boron adsorbed at equilibrium (mg/g) on Al(OH)3 and at any time (t) respectively, and k2
is the calculated pseudo second order rate constant of
adsorption.
If the adsorption kinetics follows the pseudo second order,
the plot of t/qt versus t should be linear. The qe and k2 can be
calculated from the slope and intercept of plot of t/qt versus t.
As shown in Fig. 5, the plot of t/qt versus t was linear for all
concentrations studied. The correlation coefficient was high,
the chi-square was very low and the calculated equilibrium
adsorption capacity qcal
e was in agreement with the experiexp
mental equilibrium adsorption capacity qe . This implies
that the adsorption of boron followed a second order kinetics
as shown in Table 6.
4.4.3.
Pseudo-second order
k1 (min 1)
(14)
where Kid (mg/g min0.5) is a measure of the diffusion coefficient and Ci is the intra-particle diffusion constant (mg/g).
If intra particle diffusion is the rate limiting step, a plot of
fraction of solute adsorbed against the square root of time (t0.5)
should be linear, passing through the origin. Kid and Ci are
obtained from the slope and intercept of the graph respectively. Ci is directly proportional to the boundary layer thickness. From the result (Figure not shown), the plot of qt against
(t0.5) was observed to be linear but did not pass through the
origin. This could be due to the boundary layer effect. Additionally, the values of Ci (mg/g) for all boron concentrations
4.4.4.
Elovich model
1
1
lnab lnt
b
b
(15)
4.5.
Adsorption thermodynamics
10
20
30
0.36
0.59
1.28
3.132 0.97
4.809 0.98
5.017 0.98
1.06
1.72
3.77
1.55
0.99
0.499
0.98
0.94
0.97
120
w a t e r r e s e a r c h 5 1 ( 2 0 1 4 ) 1 1 3 e1 2 3
adsorption was spontaneous for the temperature range evaluated and the degree of spontaneity of the reaction increases
with increasing temperature (Li et al., 2010). The positive value
of the standard enthalpy change (DH0 44.8 kJ/mol) implies that
the adsorption process is endothermic. Values of enthalpy
change less than 40 kJ/mol indicates that the process is
physiosorption while values above 40 kJ/mol indicate a
chemisorption process (Ma et al., 2011). Accordingly, in the
present case, the process is chemisorption. The positive value
of entropy change (DS0 125.04 J/mol) indicates increased
randomness of the solution interface during the adsorption of
boron on the electrocoagulant Al(OH)3.
4.6.
DG
(kJ/mol)
3.60
6.16
8.75
12.86
3.18
4.66
5.73
6.96
DH
(kJ/mol)
DS (J/mol.K)
44.8
125.04
(16)
where
Cenergy and Celectrode are amount of consumed electricity
(kWh/m3) and amount of consumed electrode material (kg
electrode/m3) respectively. Csludge (kg/m3) is the amount of
sludge generated during electrocoagulation. Unit prices used
in this study (for Malaysia market) were expressed as (a) unit
prices for electrical energy 35 cents Malaysia Ringgits (MYR)/
kWh (Tenaga National Malaysia, 2013), (b) unit price for electrode material (Bayramoglu et al., 2004) ($1.8; current exchange in MYR) RM 5.8/kg (c) Cost associated with sludge
handling and disposal 50 cents MYR/kg (MIDA, 2013).
The electrical energy consumption (Cenergy) is an important
parameter which defines the energy used up for a process. The
electrical energy consumptions at a constant potential of 2 V
was 1.2 kWh/m3, 2.4 kWh/m3 and 3.6 kWh/m3 for 6.25 mA/
cm2, 12.5 mA/cm2 and 18.75 mA/cm2 respectively as calculated using the equation below. Increase in current density
increased the electrical energy consumption and also
increased the overall cost.
ECC
Kc
IUT
V
(17)
where
ECC energy consumption (kWh/m3); I current (A);
U voltage (V); T time (h); V volume (L)
The aluminum electrode consumption (AI consumption)
with a unit of (g Al/g of Boron) removed was investigated using
the Faraday law as expressed below. The electrode consumption increased from 0.4 g Al/g at 6.25 mA/cm2 to 0.81 g Al/
g at 12.5 mA/cm2. At 18.75 mA/cm2, there was further increase
in aluminum electrode consumption to 1.22 g Al/g making
current density and the corresponding voltages important
controls of operational cost.
121
w a t e r r e s e a r c h 5 1 ( 2 0 1 4 ) 1 1 3 e1 2 3
Al consumption
ItM
ZFW
(18)
where
I current in amperes (A); t time (s); M molecular
weight of the electrode material (Aluminum 26.98);
Z number of electrons involved in the redox reaction
(zAl 3); F Faraday constant (96,500 C/mol); W weight of
treated wastewater (g).
The amount of sludge generated at 6.25 mA/cm2, 12.5 mA/
cm2 and 18.75 mA/cm2 was 0.037 kg, 0.029 kg and 0.021 kg
respectively. Sludge generation is influenced by electric
charge. The unit cost of NaOH (50%) and H2SO4 (98%) used in
this study was about 0.018 cents MYR/kg. 0.3e0.5 ml of 1 M
NaOH and H2SO4 was used to control the pH of each experiment. The operational cost for the removal of boron in this
study at optimum conditions is RM 0.88/m3 as depicted in
Table 9. Electrical energy consumption was the largest
contributor to the total operational cost for boron removal.
The unit cost for the electrodialytic treatment of boron
containing wastewater was estimated at $1.27/m3 for a twostep process removing boron and salinity (Turek et al., 2007)
and MYR 6.33/m3 for adsorption-flocculation (Chong et al.,
2009). With an operational cost of MYR 0.88/m3 at optimum
condition, electrocoagulation has comparative advantage in
terms of cost over some boron removal processes.
In large installations, the operation is feasible using an
electrocoagulation reactor equipped with a secondary sedimentation tank. Installation of parallel perforated rhombus
shaped electrode materials into the reactor will enhance
movement of metal ions, improve coagulation, reduce the
number of electrodes, improve hydrogen bubble generation
and increase floatation of formed flocs. However, residual
aluminum concentration should be controlled by operating at
low electric potential and current. This could increase treatment time but will not affect operational cost because the
increased time is compensated by the low electrical energy
consumption (kWh/m3) and electrode consumption.
4.7.
Boron recovery
Energy consumption
Electrode plate consumption
Chemicals
Sludge disposal
Total
Cost (MYR/m3)
6.25 mA/cm
12.5 mA/cm
18.75 mA/cm
6.25 mA/cm
12.5 mA/cm2
18.75 mA/cm2
1.2 kWh/m3
0.4 g Al/g
10 ml
0.037 kg
2.4 kWh/m3
0.81 g Al/g
10 ml
0.029 kg
3.6 kWh/m3
1.22 g Al/g
10 ml
0.021 kg
0.42
0.0024
0.018
0.006
0.45
0.84
0.0049
0.018
0.014
0.88
1.26
0.0073
0.018
0.019
1.3
122
w a t e r r e s e a r c h 5 1 ( 2 0 1 4 ) 1 1 3 e1 2 3
references
5.
Conclusions
Acknowledgement
This work was funded by Deanship of Scientific Research
(DSR), King Abdulaziz University (KAU), Jeddah, under
research grant (no. 135-006-D1434) and Universiti Teknologi
w a t e r r e s e a r c h 5 1 ( 2 0 1 4 ) 1 1 3 e1 2 3
Li, Q., Yue, Q.-Y., Su, Y., Gao, B.-Y., Sun, H.-J., 2010. Equilibrium,
thermodynamics and process design to minimize adsorbent
amount for the adsorption of acid dyes onto cationic polymerloaded bentonite. Chem. Eng. J. 158 (3), 489e497.
Linder, R.E., Strader, L.F., Rehnberg, G.L., 1990. Effect of acute
exposure to boric acid on the male reproductive system of the
rat. J. Toxicol. Environ. Health 31 (2), 133e146.
Ma, J., Jia, Y., Jing, Y., Yao, Y., Sun, J., 2011. Kinetics and
thermodynamics of methylene blue adsorption by cobalthectorite composite. Dyes Pigments 93 (1e3), 1441e1446.
Malakootian, M., Yousefi, N., 2009. The efficiency of
electrocoagulation process using aluminum electrodes in
removal of hardness from water. Iran. J. Environ. Health Sci.
Eng. 6 (2), 131e136.
Masion, A., Thomas, F., Tchoubar, D., Bottero, J.Y., Tekely, P.,
1994. Chemistry and structure of Al(OH)/Organic precipitates.
A small-Angle x-ray scattering study. 3. Depolymerization of
the Al13 polycation by organic ligands. Langmuir 10 (11),
4353e4356.
Malaysian Investment Development Authority (MIDA), 2013.
Scheduled Waste Treatment Rates. Available: http://www.
mida.gov.my/env3/index.php?pagescheduled-wastetreatment-rates.
Mohajeri, L., Aziz, H.A., Isa, M.H., Zahed, M.A., 2010. A statistical
experiment design approach for optimizing biodegradation of
weathered crude oil in coastal sediments. Biores. Technol. 101
(3), 893e900.
Morris, W., Weber, W., 1963. Kinetics of adsorption on carbon
from solution. J. Sanit. Eng. Div. 89 (2), 31e60.
Olmez-Hanci, T., Kartal, Z., Arslan-Alaton, I. Electrocoagulation of
commercial naphthalene sulfonates: process optimization
and assessment of implementation potential. J. Environ.
Manag. 99(0), 44e51.
Pouet, M.F., Grasmick, A., 1995. Urban wastewater treatment by
electrocoagulation and flotation. Water Sci. Technol. 31 (3e4),
275e283.
Seki, Y., Seyhan, S., Yurdakoc, M., 2006. Removal of boron from
aqueous solution by adsorption on Al2O3 based materials
using full factorial design. J. Hazard Mater 138 (1), 60e66.
Shen, X.E., Shan, X.Q., Dong, D.M., Hua, X.Y., Owens, G., 2009.
Kinetics and thermodynamics of sorption of nitroaromatic
123
All in-text references underlined in blue are linked to publications on ResearchGate, letting you access and read them immediately.