Journal of Water Process Engineering 3 (2014) 18–25
Contents lists available at ScienceDirect
Journal of Water Process Engineering
journal homepage: www.elsevier.com/locate/jwpe
Petrochemical wastewater treatment by electro-Fenton process using
aluminum and iron electrodes: Statistical comparison
Reza Davarnejad a,∗ , Mohsen Mohammadi a , Ahmad Fauzi Ismail b
a
Department of Chemical Engineering, Faculty of Engineering, Arak University, Arak 38156-8-8349, Iran
Department of Gas Engineering, Faculty of Chemical and Natural Resources Engineering, Universiti Teknologi Malaysia, 81310 UTM Skudai, Johor,
Malaysia
b
a r t i c l e
i n f o
Article history:
Received 21 May 2014
Received in revised form 1 August 2014
Accepted 7 August 2014
Keywords:
COD
Electro-Fenton
Optimization
Petrochemical wastewater
a b s t r a c t
Petrochemical manufacturing wastewaters often contain a high concentration of biodegradable compounds that possess either toxicity or activity inhibition to the biological unit. In this paper, a comparison
between aluminum and iron plate electrodes on COD and color removal from Petrochemical wastewaters
by electro-Fenton process was studied. The experiments were conducted to evaluate the effects of reaction time, current density, pH, H2 O2 /Fe2+ molar ratio, and H2 O2 of petrochemical wastewater (PW) (ml/l)
on the performance of the process. Response surface methodology (RSM) was employed to assess individual and interactive effects of the five main independent parameters on the COD and color removal. The
results show that COD and color removal efficiencies of iron electrode (67.3% and 71.58%, respectively)
were more than those of aluminum electrode (53.94% and 67.35%, respectively).
© 2014 Elsevier Ltd. All rights reserved.
1. Introduction
most popular EAOP is the electro Fenton (E-Fenton) process [11]
which can proceed by the following chain reactions [11–13]:
The production stages of a petroleum industry, such as extraction and refining, are potentially responsible for generating large
volumes of effluent to be discarded in the environment [1–3]. The
waste generated in oil refineries contains many different chemical compositions, depending on the complexity of the refinery, the
existing processes and the type of oil used [1,4].
The physical–chemical and bioremediation methods utilized for
the degradation of these compounds have shown various operational problems, such as: partial degradation of the effluent, toxic
intermediates production, energy consumption and secondary
phases generation that impose extra cost in the process [5–7].
The traditional Fenton process, one of the advanced oxidation
processes (AOPs), is widely used as a suitable treatment method
for highly concentrated wastewaters due to its effectiveness in
producing hydroxyl radicals [8,9]. Applicability of traditional Fenton process is limited by its acidic pH requirements, the formation
of iron sludge and high cost of hydrogen peroxide [8,10]. Electrochemical advanced oxidation processes (EAOPs) based on Fenton’s
reaction chemistry are eco-friendly methods that have recently
received much attention for wastewaters remediation [10]. The
H2 O2 + Fe2+ → Fe3+ + OH• + OH−
∗ Corresponding author. Tel.: +98 9188621773; fax: +98 86 34173450.
E-mail addresses: R-Davarnejad@araku.ac.ir, reza davarnejad@yahoo.com.ph,
redavarnejad@yahoo.com (R. Davarnejad).
http://dx.doi.org/10.1016/j.jwpe.2014.08.002
2214-7144/© 2014 Elsevier Ltd. All rights reserved.
(1)
Hydroxyl radicals are also generated at the surface of a highoxygen overvoltage anode from water oxidation:
H2 O → H+ + OH• + e−
(2)
Also the produced ferric ion from Eq. (1) can be reduced to
ferrous ion by electrochemical regeneration of Fe2+ ions on the
cathode surface:
Fe3+ + e− → Fe2+
(3)
Since iron and aluminum electrodes have not been compared
in detail for the treatment of petrochemical wastewaters, it is
the purpose of this study is to compare the treatment of petrochemical wastewaters by electro-Fenton using aluminum and iron
electrodes. The response surface methodology (RSM) is an excellent
tool for optimization and statistical analysis [14]. It allows considerable reduction of experiments number and a rapid interpretation
[11,15]. Furthermore, it is possible to study a large number of factors and to detect the possible interactions between them [15,16].
The RSM is a useful statistical method for the optimization of
chemical reactions and/or industrial processes and widely used for
experimental design [17]. In this paper, Optimizations of E-Fenton
was carried out by the RSM which was used to develop a mathematical technique to describe the effects of main independent variables
19
R. Davarnejad et al. / Journal of Water Process Engineering 3 (2014) 18–25
Table 1
Independent variables and their levels obtained from the statistical software.
Symbol
Factor
X1
X2
X3
Reaction time
Current density
pH
H2 O2
PW
H2 O2
Coded levels of variables
−1
X4
X5
Fe2+
10
25
2
0
+1
50
52.5
3.5
90
80
5
0.3
1.22
2.14
0.5
2.75
5
as shown in Table 1. COD and color removal efficiencies (Y1 and Y2,
respectively) were considered as the dependent factors (response).
The response was expressed as removal (%) which could be calculated by using the following equation.
Removal (%) =
Ci − C0
× 100
Ci
(4)
where Ci and C0 are initial and final COD or color concentrations.
Table 2 shows the matrix design obtained with the Design Expert
software for both experimental systems (iron and the aluminum
electrodes). Accordingly, 47 experiments were conducted with 32
factorial points, 10 axial points and 1 central point.
(such as reaction time, current density, pH, H2 O2 /Fe2+ molar ratio
and H2 O2 of PW (ml/l)), to maximize COD and color removal.
3. Results and discussion
2. Experimental
3.1. Regression models and statistical testing
2.1. Materials and methods
In this paper, correlations between the responses and the independent variables were obtained by the following second-order
model with a least-squares method [18]:
2.1.1. Wastewater sampling and characterization
The study was conducted into an industrial wastewater
obtained from Shazand Petrochemical Company (Arak, Iran). Sample is taken from the equalization basin (EQU). EQU is a place where
materials are separated based on the density in API separator. 40 l
of sample from the EQU is taken and saved in a plastic container.
It immediately transported to Arak University Chemical Engineering Research Laboratory and stored in a refrigerator at 4 ◦ C before
further analysis. The applied petrochemical wastewater had COD
1400–1700 mg/l, color 100 color unit, BOD/COD 0.4–0.6 and pH
6–6.7.
2.1.2. Electro-Fenton experiments
The experiments were conducted at room temperature
(25 ± 2 ◦ C) and atmospheric pressure in an open cylindrical glass
cell with 400 ml capacity. pH of sample was adjusted with H2 SO4 or
NaOH and measured by pH meter (METTLER-TOLEDO 320). Before
measurements, the pH meter was calibrated with the standard
buffers at room temperature.
In each run, 250 ml of wastewater was placed in an electrolytic
cell and desired amounts of iron (Fe2+ ) and hydrogen peroxide
(H2 O2 ) were added before the electrical current was turned on.
Then, electrodes were placed in the reactor and solutions were
mixed at 350 rpm. The current density (CD) was adjusted by a digital DC power supply (fabricated by Kala Gostaran-e-Farda supplier,
30 V and 3 A) operated at galvanostatic mode and the experiment
was started. Both electrodes (anode and cathode) were in square
shape and made from plates with dimensions of 2 cm × 0.5 cm. The
effective electrode area was 1 cm2 and the spacing between electrodes was 3 cm.
At the end of the run, the samples were allowed to stand for
30 min (for solids sedimentation) and the supernatant was then
taken for wastewater quality measurements. The electrodes were
washed thoroughly with water to remove any solid residues on the
surfaces. Color and COD were respectively measured at 475 nm and
605 nm wavelength using a UV–Vis spectrophotometer (HACH, US).
2.1.3. Experimental design
In this study, the optimization of experimental conditions for
petrochemical wastewater mineralization and decolorization by
electro-Fenton process was conducted using the central composite design (CCD) technique under RSM. The software Design Expert
8.0.7.1 Trial was used for the experimental design, data analysis,
quadratic model extraction, and graph plotting.
The independent variables of reaction time (X1), current density (X2), pH (X3), H2 O2 ml/l of PW (X4) and H2 O2 /Fe2+ molar
ratio (X5). They were coded with low and high levels in the CCD
Y = ˇ0 +
k
ˇj xj +
j=1
ˇij xi xj +
i<j
k
ˇjj xj2 + ∈
(5)
j=1
where Y is the response, ˇ0 is a constant coefficient, ˇj , ˇij and ˇjj
are the coefficients for the linear, quadratic and interaction effects,
respectively. xi and xj are the coded levels for the independent variables. k is the number of independent variables and ε is the random
error.
Reduced models for describing the COD and color removal using
aluminum (Eqs. (6) and (7)) and iron (Eqs. (8) and (9)) electrodes
after excluding the insignificant coefficients can be presented:
EF-Al process:
COD removal (%) = 46.62 + 6.61A + 4.27B − 2.78C + 3.98D
+ 3.01E − 1.14AC + 1.48AE + 1.18BC
+ 1.11BD + 2.32BE + 1.32DE − 2.78A2
− 5.65B2 − 3.64C 2 − 4.17D2 − 4.33E 2
(6)
Color removal (%) = 51.56 + 7.08A + 4.93B − 3.23C + 4.06D
+ 3.73E − 1.35AC + 1.06AD + 2.14AE
+ 1.22BC + 0.74BD + 2.66BE − 0.69CD
+ 1.68DE − 7.06A2 − 8.46B2 − 6.11C 2
(7)
EF-Fe process:
COD removal (%) = 63.28 + 7.1A + 4.35B − 3.5C + 3.56D
+ 2.34E + 0.63AB + 1.32AD + 2.65AE
+ 0.88BE − 0.62CD − 0.79CE + 0.82DE
− 4.56A2 − 8.72B2 − 6.81C 2 − 7.19D2
− 7.47E 2
(8)
Color removal (%) = 64.51 + 8.29A + 4.32B − 5.13C + 3.74D
+ 5.39E + 1.99AD + 2.49AE − 0.07BC
+ 1.78BD + 0.61BE − 1.1CE + 0.69DE
− 4.39A2 − 6.59B2 − 7.12C 2 − 7.2D2
− 6.27E 2
(9)
20
R. Davarnejad et al. / Journal of Water Process Engineering 3 (2014) 18–25
Table 2
Experimental matrix design for EF-Al and EF-Fe.
Run
Block
Type
Reaction time
Current density
pH
H2 O2 /PW
H2 O2 /Fe2+
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
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Block 1
Center
Center
Center
Center
Center
Factorial
Factorial
Factorial
Factorial
Factorial
Factorial
Factorial
Factorial
Factorial
Factorial
Factorial
Factorial
Factorial
Factorial
Factorial
Factorial
Factorial
Factorial
Factorial
Factorial
Factorial
Factorial
Factorial
Factorial
Factorial
Factorial
Factorial
Factorial
Factorial
Factorial
Factorial
Factorial
Axial
Axial
Axial
Axial
Axial
Axial
Axial
Axial
Axial
Axial
50
50
50
50
50
10
90
10
90
10
90
10
90
10
90
10
90
10
90
10
90
10
90
10
90
10
90
10
90
10
90
10
90
10
90
10
90
10
90
50
50
50
50
50
50
50
50
52.5
52.5
52.5
52.5
52.5
25
25
80
80
25
25
80
80
25
25
80
80
25
25
80
80
25
25
80
80
25
25
80
80
25
25
80
80
25
25
80
80
52.5
52.5
25
80
52.5
52.5
52.5
52.5
52.5
52.5
3.5
3.5
3.5
3.5
3.5
2
2
2
2
5
5
5
5
2
2
2
2
5
5
5
5
2
2
2
2
5
5
5
5
2
2
2
2
5
5
5
5
3.5
3.5
3.5
3.5
2
5
3.5
3.5
3.5
3.5
1.22
1.22
1.22
1.22
1.22
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.3
2.14
2.14
2.14
2.14
2.14
2.14
2.14
2.14
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.3
2.14
2.14
2.14
2.14
2.14
2.14
2.14
2.14
1.22
1.22
1.22
1.22
1.22
1.22
0.3
2.14
1.22
1.22
2.75
2.75
2.75
2.75
2.75
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
2.75
2.75
2.75
2.75
2.75
2.75
2.75
2.75
0.5
5
The experimental and predicted data (obtained from Eqs.
(6)–(9)) for COD and color removals using the iron and aluminum
electrodes are presented in Table 3. As shown in this table, the initial
removal in the iron electrode was more than that of the aluminum
electrode under the same operating conditions.
The model adequacy includes some tests for the regression
model consideration. For significance on model, some coefficients
and tests for lack of fit can be obtained [19,20]. For this purpose,
analysis of variance (ANOVA) was performed and the results were
shown in Table 4. The fit of model was checked by the determination of coefficient (R2 ). A model can properly predict the response
when R2 is close to 1 [18,20]. As shown in Table 4, high correlation
coefficients R2 > 0.92, R2 adjusted > 0.88 and R2 predicted > 0.80 for
all responses in EF-Al and EF-Fe imply that the regression model
fits to the experimental value and it can provide an excellent explanation of the relationships between the independent variable and
the response. The F-statistics values were high and its values for
EF-Al and EF-Fe were 42–55 for COD and 24–21 for color removals,
respectively. The large F-values indicated that most of the variation
in the response could be explained by the regression model [16,20].
The Prob > F values for the EF-Al and EF-Fe from the ANOVA was less
than 0.05 showed that the model was considered to be statistically
significant [20]. In this study, values of Prob > F were <0.0001 for
both of the models. The coefficient of variance (CV) as the ratio of
the standard error of estimate to the mean value of the observed
response (as a percentage) was a measure of reproducibility of the
model [19,20]. CV was considered to be reproducible when it is not
greater than 10%. As shown in Table 4, values of CV for EF-Al and EFFe were 9.84–10.07% for COD and 12.94–14.67% for color removals.
The adequate precision (AP) measures the signal to noise ratio and
AP was compared for the range of the predicted values at the design
points to the average prediction error [20]. The AP values greater
than 4 indicate an adequate signal for all responses of EF-Al and
EF-Fe. Therefore, the quadratic model could be used to navigate
the design space.
3.2. Three-dimensional plots
3D plots for EF-Al and EF-Fe in Figs. 1 and 2 were shown to
visualize the effects of experimental factors on removal efficiencies
of color and COD responses. Some interactions among variables
were significant. So the curvature of three-dimensional surfaces
was obvious as shown in Figs. 1 and 2. The optimum conditions
21
R. Davarnejad et al. / Journal of Water Process Engineering 3 (2014) 18–25
Table 3
Observed (actual) and predicted data for EF-Al and EF-Fe.
Run
EF-Al (aluminum electrodes)
COD removal (%)
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
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
EF-Fe (iron electrodes)
Color removal (%)
COD removal (%)
Color removal (%)
Act
Pre
Act
Pre
Act
Pre
Act
Pre
47.18
47.32
46.98
47.35
46.58
21.32
28.74
18.32
19.47
10.21
18.74
15.32
31.16
17.35
35.51
25.39
38.93
15.75
22.53
19.78
29.98
13.35
31.97
19.28
45.27
9.53
18.74
18.31
33.5
20.57
38.02
32.75
53.84
18.73
29.93
35.63
45.17
35.83
51.25
33.98
47.36
46.16
39.21
38.09
46.21
38.65
45.33
46.61
46.61
46.61
46.61
46.61
17.21
29.76
16.53
29.07
11.57
19.56
15.6
23.59
20.31
32.86
24.07
36.62
14.66
22.66
23.14
31.14
13
31.45
21.59
40.03
7.36
21.25
20.66
34.56
21.37
39.82
34.41
52.85
15.73
26.63
33.48
47.38
37.22
50.44
36.69
45.22
45.57
40.19
38.46
46.42
39.27
45.28
48.47
48.58
47.89
48.53
47.65
25.63
30.69
21.36
23.73
12.43
18.39
21.73
28.16
19.71
38.71
26.95
42.93
16.25
25.31
21.75
35.73
16.95
29.34
23.08
52.17
11.56
24.13
21.75
43.04
23.78
50.6
40.21
63.17
21.26
33.73
39.71
48.52
38.82
55.32
39.69
51.65
50.13
45.91
56.26
50.08
56.6
49.73
51.56
51.56
51.56
51.56
51.56
20.83
31.28
21.43
31.88
16.0
21.06
21.49
26.56
23.36
38.05
26.93
41.62
15.77
25.08
24.24
33.54
15.32
34.32
26.56
45.58
10.49
24.12
26.62
40.25
24.57
47.83
38.77
62.17
16.98
34.85
36.08
53.95
37.42
51.58
38.17
48.02
48.67
42.22
47.5
55.62
47.83
55.28
63.15
63.85
63.98
64.05
63.48
21.03
23.03
33.73
23.71
12.59
22.93
18.25
32.21
19.75
37.51
28.93
43.98
17.83
27.53
22.31
33.95
15.58
35.71
21.95
48.31
10.28
23.87
20.32
35.72
23.53
41.98
37.58
56.71
21.83
32.71
18.25
49.31
50.23
66.67
48.98
59.6
59.23
53.16
52.38
59.25
52.37
58.35
63.27
63.27
63.27
63.27
63.27
19.56
24.56
25.23
32.75
15.38
20.38
21.06
28.57
23.63
33.91
29.30
42.10
16.99
27.27
32.67
35.46
17.12
23.74
26.32
44.44
9.78
25.39
18.97
37.10
24.47
45.37
33.66
57.07
14.66
35.56
23.85
47.27
51.61
65.82
50.21
58.9
59.95
52.96
52.52
59.64
53.46
58.19
64.53
64.73
64.89
65.16
64.18
25.46
39.53
38.21
18.25
10.25
21.93
16.71
26.71
11.85
38.56
27.83
53.72
17.53
26.71
26.58
41.78
20.75
46.71
31.75
57.21
18.21
31.93
22.71
42.58
28.93
53.21
51.39
66.95
27.03
41.07
19.63
53.21
51.63
68.39
52.23
63.39
61.59
52.98
52.18
62.23
52.32
63.95
64.51
64.51
64.51
64.51
64.51
22.01
29.64
27.43
35.06
15.53
23.16
17.82
25.45
20.58
36.17
33.11
48.69
14.1
29.68
23.49
39.08
27.42
45.01
35.29
52.88
16.25
34.11
21.26
38.85
28.74
54.29
43.72
69.27
17.85
43.39
29.69
55.24
51.83
68.41
53.60
62.24
62.52
52.27
53.58
61.05
52.85
63.64
Table 4
Quadratic model ANOVA results for EF-Al and EF-Fe.
Variable
Standard deviation
Mean
R2
R2 adjusted
R2 predicted
Coefficient of variance (CV)
Press
Adequate precision
F-value
Prob > F
EF-Al (aluminum electrodes)
EF-Fe (iron electrodes)
COD removal (%)
Color removal (%)
COD removal (%)
Color removal (%)
3.20
31.73
0.9578
0.9353
0.8925
10.07
780.37
23.678
42.55
<0.0001
4.65
35.91
0.9279
0.8895
0.8354
12.94
1479.04
18.450
24.15
<0.0001
3.75
38.14
0.97
0.9524
0.9172
9.84
1127.24
24.123
55.13
<0.0001
6.12
41.69
0.9262
0.883
0.7967
14.67
2990.42
14.573
21.42
<0.0001
22
R. Davarnejad et al. / Journal of Water Process Engineering 3 (2014) 18–25
Fig. 1. Effects of the variables on COD and color removal in EF-Al: (a) reaction time and current density, (b) reaction time and pH, (c) reaction time and H2 O2 /PW (ml/l), (d)
reaction time and H2 O2 /Fe2+ , (e) current density and pH, (f) current density and H2 O2 /PW (ml/l), (g) current density and H2 O2 /Fe2+ , (h) pH and H2 O2 /PW (ml/l), (i) pH and
H2 O2 /Fe2+ , (j) H2 O2 /PW (ml/l) and H2 O2 /Fe2+ for COD removal, (k) reaction time and current density, (l) reaction time and pH, (m) reaction time and H2 O2 /PW (ml/l), (n)
reaction time and H2 O2 /Fe2+ , (o) current density and pH, (p) current density and H2 O2 /PW (ml/l), (q) current density and H2 O2 /Fe2+ , (r) pH and H2 O2 /PW (ml/l), (s) pH and
H2 O2 /Fe2+ , (t) H2 O2 /PW (ml/l) and H2 O2 /Fe2+ for color removal.
for maximum values of the responses are attributed to all variables
(reaction time, current density, pH, H2 O2 /PW (ml/l) and H2 O2 /Fe2+
molar ratio).
3.3. Optimization and validation
Numerical optimization was used to determine the optimum process parameters for maximum waste mineralization and
decolorization. Based on response surface and desirability functions, the optimum conditions for COD and color removals were
obtained. In this case, all variables were targeted to be in range. Furthermore, COD and color removals were goaled to be maximized. In
order to confirm the accuracy of the predicted models and the reliability of the optimum conditions, an additional experiment was
carried out at optimum conditions. Table 5 compared the experimental data under the optimum conditions with the predicted
23
R. Davarnejad et al. / Journal of Water Process Engineering 3 (2014) 18–25
Fig. 2. Effects of the variables on COD and color removal in EF-Fe: (a) reaction time and current density, (b) reaction time and pH, (c) reaction time and H2 O2 /PW (ml/l), (d)
reaction time and H2 O2 /Fe2+ , (e) current density and pH, (f) current density and H2 O2 /PW (ml/l), (g) current density and H2 O2 /Fe2+ , (h) pH and H2 O2 /PW (ml/l), (i) pH and
H2 O2 /Fe2+ , (j) H2 O2 /PW (ml/l) and H2 O2 /Fe2+ for COD removal, (k) reaction time and current density, (l) reaction time and pH, (m) reaction time and H2 O2 /PW (ml/l), (n)
reaction time and H2 O2 /Fe2+ , (o) current density and pH, (p) current density and H2 O2 /PW (ml/l), (q) current density and H2 O2 /Fe2+ , (r) pH and H2 O2 /PW (ml/l), (s) pH and
H2 O2 /Fe2+ , (t) H2 O2 /PW (ml/l) and H2 O2 /Fe2+ for color removal.
Table 5
Optimum conditions found by design expert and experimental verification for EF-Al and EF-Fe.
Electrodes
Reaction time
Current density
pH
H2 O2 /PW
H2 O2 /Fe2+
Aluminum
Iron
78.97
73.19
68.65
59.7
3.06
2.67
2.14
1.23
4.99
3.65
COD removal (%)
Color removal (%)
Act
Pre
Act
Pre
51.23
66.85
53.94
67.3
66.71
69.89
67.35
71.58
24
R. Davarnejad et al. / Journal of Water Process Engineering 3 (2014) 18–25
ones (calculated from Eqs. (6)–(9)). The low error in the experimental and predicted data indicated a good agreement between
the experimental data and predicted ones.
The optimized data for the maximum removals of COD and color
were around 53.94% and 67.35% at reaction time of 78.97 min, current density of 68.65 mA/m2 , pH of 3.06, H2 O2 /PW of 2.14 ml/l and
H2 O2 /Fe2+ molar ratio of 4.99 for EF-Al. They were 67.3% and 71.58%
at reaction time of 73.19 min, current density of 59.7 mA/m2 , pH of
2.67, H2 O2 /PW of 1.23 ml/l and H2 O2 /Fe2+ molar ratio of 3.65 for
EF-Fe.
Figs. 1 and 2 show the effects of various variables on COD and
color removal: (a) reaction time and current density, (b) reaction
time and pH, (c) reaction time and H2 O2 /PW (ml/l), (d) reaction
time and H2 O2 /Fe2+ , (e) current density and pH, (f) current density and H2 O2 /PW (ml/l), (g) current density and H2 O2 /Fe2+ , (h) pH
and H2 O2 /PW (ml/l), (i) pH and H2 O2 /Fe2+ , (j) H2 O2 /PW (ml/l) and
H2 O2 /Fe2+ for COD removal, (k) reaction time and current density,
(l) reaction time and pH, (m) reaction time and H2 O2 /PW (ml/l), (n)
reaction time and H2 O2 /Fe2+ , (o) current density and pH, (p) current
density and H2 O2 /PW (ml/l), (q) current density and H2 O2 /Fe2+ , (r)
pH and H2 O2 /PW (ml/l), (s) pH and H2 O2 /Fe2+ , (t) H2 O2 /PW (ml/l)
and H2 O2 /Fe2+ for color removal in EF-Al and EF-Fe, respectively.
The current density plays an important role in the electroFenton process because it controls the formation of the hydroxyl
radical and the concentration of ferrous ions in the solution [8,11].
It means that an increase in the current density will improve the
removal efficiency up to the optimum value. As shown in Fig. 1(a)
and (k), when the current density rises from 36 to 58 mA/m2 , COD
and color removals respectively increase from 44.33% to 55.36%
and from 42.63% to 53.49% (at reaction time of 90 min). As shown
in Fig. 2(a) and (k), the COD and color removal respectively increase
from 57.21% to 67.93% and from 61.39% to 69.71%, when the current density rises from 36 to 58 mA/m2 in EF-Fe. This is due to
more production of OH• on the surface of anode and more electroregeneration of ferrous ions from ferric ions at higher current
density. These increase the efficiency of Fenton chain reactions.
The continuous conversion of Fe3+ into Fe2+ is a great advantage
compared with the chemical Fenton systems [8,11]. As illustrated
in Table 5, the optimum current density in the EF-Al and EF-Fe are
68.65 and 59.7, respectively. These results are in good agreement
with the literature [11,13,20].
Furthermore, the value of pH solution is an important factor for
the E-Fenton process. Most of the studies report that the optimal
value of pH is between 2 and 5 [8–11]. The pH enhancement (for
pH > 5) decreases the E-Fenton efficiency, rapidly. This is due to
the fact that H2 O2 is unstable in the basic solution. H2 O2 rapidly
decomposes to oxygen and water (Eq. (10)). When pH decreases
(for pH < 2), H2 O2 changes to H3 O2 + (H2 O2 cannot be changed to
OH• by Fe2+ ) [8,9].
H2 O2 + 2H+ + 2e− → 2H2 O
(10)
As shown in Fig. 1(b) and (l), when the pH rises from 2.6 to
4.4, COD and color removals respectively decreases from 52.33%
to 40.36% and from 53.63% to 42.49% (at reaction time of 90 min)
in EF-Al. As shown in Fig. 2(b) and (l), the COD and color removal
respectively decreases from 64.21% to 52.93% and from 69.39% to
54.71%, when the pH rises from 2.6 to 4.4 in EF-Fe. These outputs
show that pH also has a significant effect on the COD and color
removal in EF-Al and EF-Fe. As illustrated in Table 5, the optimum
pH for COD and color removals in EF-Al and EF-Fe are around 3.06
and 2.67, respectively. These results are in good agreement with
the other studies on the oxidation of organic compounds [8,11,21].
In order to maximize the removals, the H2 O2 /Fe2+ molar ratio
and H2 O2 /PW (ml/l) should be optimized, as well. Fig. 1(c) and (d)
shows that an increase in H2 O2 /Fe2+ molar ratio (H2 O2 /Fe2+ > 4.1)
and H2 O2 /PW (H2 O2 /PW > 1.73 ml/l) decreases the COD removal at
reaction time of 90 min. Fig. 1(m) and (n) shows that an increase
in H2 O2 /Fe2+ molar ratio and H2O2/PW (ml/l) improves the color
removal. According to Fig. 2(c), (d), (m) and (n), an increase in
H2 O2 /Fe2+ molar ratio (2.8 and 4.1 for COD and color removals,
respectively) and H2 O2 /PW (1.53 ml/l and 1.73 ml/l for COD and
color removals, respectively) decreases the removal efficiency. As
illustrated in Table 5, the optimum H2 O2 /PW (ml/l) and H2 O2 /Fe2+
molar ratio are 2.14 and 4.99, respectively in the EF-Al. These data
are 1.23 and 3.65 in the EF-Fe. These results are in good agreement
with the literature [11,14]. This may be due to that Fenton’s reaction
mechanisms would change and some side reactions would occur.
It seems that excessive hydrogen peroxide has a scavenging effect
on hydroxyl radicals (Eq. (11)) [10,11]:
•
•
H2 O2 + OH → OOH + H2 O
(11)
This reaction leads to the production of hydroperoxyl radical, a
species with much weaker oxidizing power compared to hydroxyl
radical [22].
According to Table 5, the optimum removal data in EF-Fe (67.3%
and 71.58% for COD and color removals, respectively) were more
than those of the EF-Al (53.94% and 67.35%for COD and color
removals, respectively). According to the literature, El-Naas et al.
found 53% COD removal by adsorption process on Date-Pit activated carbon [23], Saien and Nejati investigated 61% COD removal
by a photocatalytic process [24], and El-Naas et al. discovered 57%
COD removal by electrocoagulation process [25] for the petroleum
wastewaters. Furthermore, Hami et al. obtained 72% COD removal
by dissolved air flotation process [26], Ting et al. found 50% COD
removal by Fenton process [27], and Basheer Hasan et al. investigated 98.1% COD removal by Fenton-like process [28] for the
petroleum wastewaters. Therefore, it seems that EF-Fe is a fast and
acceptable pre-treatment process for the petroleum wastewaters.
A regeneration iron ion on the cathode surface (Eq. (3)) is one of
the most important reactions in the electro-Fenton process. When
the iron electrodes were used, iron ions are also produced (with
corrosion at the anode surface (Eq. (12))):
Fe → Fe2+ + 2e−
(12)
Iron ions amount can be calculated from Faraday’s law [29]:
m=
MItEC
nf
(13)
where m is the mass of substance produced at the electrode. M
is the molar mass of substance. I is the total electric charge that
passes through the solution. tEC and n are reaction time and valence
number of the substance (as an ion in solution), respectively. f is the
Faraday’s constant (=96,485 c/mole).
It seems that the regeneration and production of iron ions (Fe2+ )
increases when iron electrodes are used. So, the requested iron
salt (as a source of Fe2+ ) decreases and hydrogen peroxide is proportionality consumed. Furthermore, current density and reaction
time are more than those of the iron electrodes to cover iron ions
leakage effect in the EF-Al although these conditions cannot provide
Fe2+ ions as are produced in the EF-Fe. On the other hand, reaction
time and current density increment (in the aluminum electrodes)
cause Eq. (14) on the anode surface. These produced metallic ions
and excess hydrogen peroxide (resulted from the Fe2+ ions leakage in the solution) intensify the color and decrease color removed
[30–33]. Therefore, COD and color removal will be decreased.
Al → Al3+ + 3e−
(14)
4. Conclusions
In this study, a comparison between aluminum and iron electrodes on COD and color removal from petroleum industry by
R. Davarnejad et al. / Journal of Water Process Engineering 3 (2014) 18–25
electro-Fenton process was investigated. The effect of various
parameters such as reaction time (10–90 min), current density
(25–80 mA/m2 ), pH (2–5). The propos quadratic model was statistically fitted with the experimental data (with R2 > 0.92 for all
responses) in EF-Al and EF-Fe. According to the statistical and
experimental analysis, current density had a very significant effect
on the removals (in especial in the EF-Fe case). Its reason was due
to the material electrode (which increased iron ions production in
the solution). The optimized data for the maximum removals of
COD and color were around 53.94% and 67.35% at reaction time of
78.97 min, current density of 68.65 mA/m2 , pH of 3.06, H2 O2 /PW
of 2.14 ml/l and H2 O2 /Fe2+ molar ratio of 4.99 for EF-Al. They were
67.3% and 71.58% at reaction time of 73.19 min, current density of
59.7 mA/m2 , pH of 2.67, H2 O2 /PW of 1.23 ml/l and H2 O2 /Fe2+ molar
ratio of 3.65 for EF-Fe.
[14]
[15]
[16]
[17]
[18]
Acknowledgments
[19]
The authors are very thankful to Arak University (Research ViceChancellery) for their kind help and support. This research was
financially supported by Arak University (Grant no. 90/11326).
[20]
[21]
References
[1] J.V.F.L. Cavalcanti, C.A.M. Abreu, M.N. Carvalho, M.A.M. Sobrinho, M. Benachour,
O.S. Baraúna, Removal of effluent from petrochemical wastewater by adsorption using organoclay, in: V. Patel (Ed.), Petrochemicals, Intech, Rijeka, 2012,
pp. 277–294.
[2] Y. Yavuz, A.S. Koparal, Ü. Öğütveren, Treatment of petroleum refinery wastewater by electrochemical methods, Desalination 258 (2010) 201–205.
[3] J.H.B. Rocha, M.M.S. Gomes, N.S. Fernandes, D.R. da Silva, C.A. Martínez-Huitle,
Application of electrochemical oxidation as alternative treatment of produced
water generated by Brazilian petrochemical industry, Fuel Process. Technol. 96
(2012) 80–87.
[4] A. Coelho, A.V. Castro, M. Dezotti, G.L. Sant’Anna Jr., Treatment of petroleum
refinery sourwater by advanced oxidation processes, J. Hazard. Mater. 137
(2006) 178–184.
[5] A.J.C. da Silva, E.V. dos Santos, C.C. de Oliveira Morais, C.A.M. nez-Huitle, S.S.L.
Castro, Electrochemical treatment of fresh, brine and saline produced water
generated by petrochemical industry using Ti/IrO2 –Ta2 O5 and BDD in flow
reactor, Chem. Eng. J. 233 (2013) 47–55.
[6] P. Cañizares, A. Beteta, C. Sáez, L. Rodríguez, M.A. Rodrigo, Use of electrochemical technology to increase the quality of the effluents of biooxidation processes.
A case studied, Chemosphere 72 (2011) 1080–1085.
[7] F. Fu, Q. Wang, B. Tang, Effective degradation of C.I. Acid Red 73 by advanced
Fenton process, J. Hazard. Mater. 174 (2010) 17–22.
[8] P.V. Nidheesh, R. Gandhimathi, Trends in electro-Fenton process for water and
wastewater treatment: an overview, Desalination 299 (2012) 1–15.
[9] E. Neyens, J. Baeyens, A review of classic Fenton’s peroxidation as an advanced
oxidation technique, J. Hazard. Mater. 98 (2003) 33–50.
[10] H. Lee, M. Shoda, Removal of COD and color from livestock wastewater by the
Fenton method, J. Hazard. Mater. 153 (2008) 1314–1319.
[11] S. Mohajeri, H. Abdul Aziz, M. Hasnain Isa, M. Ali Zahed, M.N. Adlan, Statistical optimization of process parameters for landfill leachate treatment using
electro-Fenton technique, J. Hazard. Mater. 176 (2009) 749–758.
[12] N.R. Mohanty, I.W. Wei, Oxidation of 2,4-dinitrotoluene using Fenton’s reagent:
reaction mechanisms and their practical applications, Hazard. Waste Hazard.
Mater. 10 (1993) 171–183.
[13] E. Guinea, C. Arias, P.L. Cabot, J.A. Garrido, R.M. Rodríguez, F. Centella, E. Brillas,
Mineralization of salicylic acid in acidic aqueous medium by electrochemical
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
25
advanced oxidation processes using platinum and boron-doped diamond as
anode and cathodically generated hydrogen peroxide, Water Res. 42 (2008)
499–511.
M. Azama, M. Bahram, S. Nouri, A. Naseri, A central composite design for the
optimization of the removal of the azo dye, methyl orange, from waste water
using the Fenton reaction, J. Serb. Chem. Soc. 77 (2012) 235–246.
S. Garcia-Seguraa, L.C. Almeidab, N. Bocchib, E. Brillas, Solar photoelectroFenton degradation of the herbicide 4-chloro-2-methylphenoxyacetic acid
optimized by response surface methodology, J. Hazard. Mater. 194 (2011)
109–118.
A.R. Khataee, M. Zarei, L. Moradkhannejhad, Application of response surface methodology for optimization of azo dye removal by oxalate catalyzed
photoelectro-Fenton process using carbon nanotube-PTFE cathode, Desalination 258 (2010) 112–119.
J. Virkutyte, E. Rokhina, V. Jegatheesan, Optimization of electro-Fenton denitrification of a model wastewater using a response surface methodology,
Bioresour. Technol. 101 (2010) 1440–1446.
A. El-Ghenymy, S. Garcia-Segura, R.M. Rodríguez, E. Brillas, M.S. El Begrani,
B.A. Abdelouahid, Optimization of the electro-Fenton and solar photoelectroFenton treatments of sulfanilic acid solutions using a pre-pilot flow
plant by response surface methodology, J. Hazard. Mater. 221–222 (2012)
288–297.
B.K. Korbahti, N. Aktas, A. Tanyolac, Optimization of electrochemical treatment
of industrial paint wastewater with response surface methodology, J. Hazard.
Mater. 148 (2007) 83–90.
N. Daneshvar, A. Oladegaragoze, N. Djafarzadeh, Decolorization of basic dye
solutions by electrocoagulation: investigation of the effect of operational
parameters, J. Hazard. Mater. B 129 (2006) 116–122.
E. Gengec, M. Kobya, E. Demirbas, A. Akyol, K. Oktor, Optimization of baker’s
yeast wastewater using response surface methodology by electrocoagulation,
Desalination 286 (2012) 200–209.
M.I. Badawy, M.E.M. Ali, Fenton’s peroxidation and coagulation processes for
the treatment of combined industrial and domestic wastewater, J. Hazard.
Mater. 136 (2006) 961–966.
M.H. El-Naas, S. Al-Zuhair, M.A. Alhaija, Reduction of COD in refinery wastewater through adsorption on date-Ppit activated carbon, J. Hazard. Mater. 173
(2009) 750–757.
J. Saien, H. Nejati, Enhanced photocatalytic degradation of pollutants in
petroleum refinery wastewater under mild conditions, J. Hazard. Mater. 148
(2007) 491–495.
M.H. El-Naas, S. Al-Zuhair, A. Al-Lobaney, Treatment of petroleum refinery
wastewater by continuous electrocoagulation, Int. J. Eng. Res. Technol. 2278
(2013) 2–10.
M.L. Hami, M.A. Al-Hashimi, M.M. Al-Doori, Effect of activated carbon on BOD
and COD removal in dissolved air flotation unit treating refinery wastewater,
Desalination 216 (2007) 116–122.
W.P. Ting, Y.H. Huang, M.C. Lu, Catalytic treatment of petrochemical wastewater by electroassisted Fenton technologies, React. Kinet. Catal. Lett. 92 (2007)
41–48.
D. Basheer Hasan, A.R. Abdul Aziz1, W.M.A. Wan Daud, Oxidative mineralisation
of petroleum refinery effluentusing Fenton-like process, Chem. Eng. Res. Des.
90 (2012) 298–307.
H. Zhang, D. Zhang, J. Zhou, Removal of COD from landfill leachate by electroFenton method, J. Hazard. Mater. 135 (2006) 106–111.
F. Ilhan, U. Kurt, O. Apaydin, M. Talha Gonullu, Treatment of leachate by electrocoagulation using aluminum and iron electrodes, J. Hazard. Mater. 154 (2008)
381–389.
M. Kobya, O. Taner Can, M. Bayramoglu, Treatment of textile wastewaters by
electrocoagulation using iron and aluminum electrodes, J. Hazard. Mater. B100
(2003) 163–178.
.T. Ü, A.S. Ün, Ü.B. Koparal, Öğütveren, Hybrid processes for the treatment
of cattle-slaughterhouse wastewater using aluminum and iron electrodes, J.
Hazard. Mater. 164 (2009) 580–586.
G. Sayiner, F. Kandemirli, A. Dimoglo, Evaluation of boron removal by electrocoagulation using iron and aluminum electrodes, Desalination 230 (2008)
205–212.